journal of applied business and economics€¦ · ©journal of applied business and economics 2012...

Journal of Applied Business and Economics

North American Business Press

Atlanta - Seattle – South Florida - Toronto

Journal of Applied Business and Economics

Editors Dr. Adam Davidson Dr. William Johnson

Editor-In-Chief

Dr. David Smith

NABP EDITORIAL ADVISORY BOARD

Dr. Andy Bertsch - MINOT STATE UNIVERSITY Dr. Jacob Bikker - UTRECHT UNIVERSITY, NETHERLANDS Dr. Bill Bommer - CALIFORNINA STATE UNIVERSITY, FRESNO Dr. Michael Bond - UNIVERSITY OF ARIZONA Dr. Charles Butler - COLORADO STATE UNIVERSITY Dr. Jon Carrick - STETSON UNIVERSITY Dr. Mondher Cherif - REIMS, FRANCE Dr. Daniel Condon - DOMINICAN UNIVERSITY, CHICAGO Dr. Bahram Dadgostar - LAKEHEAD UNIVERSITY, CANADA Dr. Deborah Erdos-Knapp - KENT STATE UNIVERSITY Dr. Bruce Forster - UNIVERSITY OF NEBRASKA, KEARNEY Dr. Nancy Furlow - MARYMOUNT UNIVERSITY Dr. Mark Gershon - TEMPLE UNIVERSITY Dr. Philippe Gregoire - UNIVERSITY OF LAVAL, CANADA Dr. Donald Grunewald - IONA COLLEGE Dr. Samanthala Hettihewa - UNIVERSITY OF BALLARAT, AUSTRALIA Dr. Russell Kashian - UNIVERSITY OF WISCONSIN, WHITEWATER Dr. Jeffrey Kennedy - PALM BEACH ATLANTIC UNIVERSITY Dr. Jerry Knutson - AG EDWARDS Dr. Dean Koutramanis - UNIVERSITY OF TAMPA Dr. Malek Lashgari - UNIVERSITY OF HARTFORD Dr. Priscilla Liang - CALIFORNIA STATE UNIVERSITY, CHANNEL ISLANDS Dr. Tony Matias - MATIAS AND ASSOCIATES Dr. Patti Meglich - UNIVERSITY OF NEBRASKA, OMAHA Dr. Robert Metts - UNIVERSITY OF NEVADA, RENO Dr. Adil Mouhammed - UNIVERSITY OF ILLINOIS, SPRINGFIELD Dr. Roy Pearson - COLLEGE OF WILLIAM AND MARY Dr. Sergiy Rakhmayil - RYERSON UNIVERSITY, CANADA Dr. Robert Scherer - CLEVELAND STATE UNIVERSITY Dr. Ira Sohn - MONTCLAIR STATE UNIVERSITY Dr. Reginal Sheppard - UNIVERSITY OF NEW BRUNSWICK, CANADA Dr. Carlos Spaht - LOUISIANNA STATE UNIVERSITY, SHREVEPORT Dr. Walter Amedzro ST-Hilaire - HEC, MONTREAL, CANADA Dr. Ken Thorpe - EMORY UNIVERSITY Dr. Robert Tian - MEDIALLE COLLEGE Dr. Calin Valsan - BISHOP'S UNIVERSITY, CANADA Dr. Anne Walsh - LA SALLE UNIVERSITY Dr. Thomas Verney - SHIPPENSBURG STATE UNIVERSITY Dr. Christopher Wright - UNIVERSITY OF ADELAIDE, AUSTRALIA

Volume 13(2) ISSN 1499-691X Authors have granted copyright consent to allow that copies of their article may be made for personal or internal use. This does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Any consent for republication, other than noted, must be granted through the publisher:

North American Business Press, Inc. Atlanta - Seattle – South Florida - Toronto ©Journal of Applied Business and Economics 2012 For submission, subscription or copyright information, contact the editor at: [email protected] Subscription Price: US$ 360/yr Our journals are indexed by UMI-Proquest-ABI Inform, EBSCOHost, GoogleScholar, and listed with Cabell's Directory of Periodicals, Ulrich's Listing of Periodicals, Bowkers Publishing Resources, the Library of Congress, the National Library of Canada. Our journals have been used to support the Academically Qualified (AQ) faculty classification by all recognized business school accrediting bodies.

This Issue

Does Crowd Out Hamper Government Stimulus Programs In Recessions? ...................................... 11 John J. Heim In well controlled statistical tests, crowd out was found related to government deficits financed by borrowing. Roughly equal effects were found for both recession and non-recession periods. Tax cut deficits were found were found more detrimental than spending deficits Private borrowing systematically declined with the growth of government deficits, and explained most variation in consumer and investment spending. Crowd out may be avoided by foreign borrowing or if M2 money increases prior to the deficit. These findings offer a plausible explanation for the failure of recent U.S. government stimulus programs to offset the 2008 recession. Testing the Utility of Licensing: Evidence from a Field Experiment on Occupational Regulation ................................................................................................ 28 Dick M. Carpenter II This study examines hypothesized benefits associated with occupational licensing in one long-regulated industry in Louisiana—floristry—in order to determine to what extent licensing results in theorized benefits that might justify the costs associated with licensure systems. Results indicate the regulation appears not to result in a statistically significant difference in quality of product. Moreover, florist-judges, whether licensed or not, appeared consistent in how they rated 50 floral arrangements from randomly chosen floral retailers. Human and Economic Development in China and India: A Comparative MDG Assessment ........................................................................................................... 42 Bruce A. Forster This paper assesses economic and human development in China and India using the United Nations set of Millennium Development Goals (MDGs) as the framework for analyzing the relative performance of the two countries. The MDG framework was used by Rausch & Kostyshak (2009) to assess the development of Arab countries in the Middle East and Africa, and by Forster (2010) to examine development in Sub-Saharan Africa. In summary, the paper shows that China has been more successful in improving its positions relative to the set of MDGs than has India. China also leads several comparator groups while India lags them.

Leadership Styles and Organizational Citizenship Behavior: The Mediating Effect of Subordinates’ Competence and Downward Influence Tactics .............................................. 59 Lee Kim Lian, Low Guan Tui The objective of this study is to test a theory-based model predicting the relationships between leadership styles, subordinates’ competence, downward influence tactics and outcome of organizational citizenship behavior in Malaysian-based organizations. Data was collected from 347 respondents that represent major industries like services, manufacturing, mining and construction companies. Path analysis technique was used to test the model developed. The results show that the transformational leadership style has significant positive relationship with subordinates’ organizational citizenship behavior, whereas the transactional leader style is negatively related to organizational citizenship behavior. This result illustrates the direct effects of leadership styles on the subordinates’ outcome. In addition, inspirational appeals and consultation tactics, as downward influence tactics, were found to mediate the relationship between transformational leadership and organizational citizenship behavior. Likewise, subordinates’ competence mediates the relationship between transformational leadership and consultation tactics. These results only partially support the efficacy of the influence theory, and therefore lend support to contingency theories of leadership. Implications for research and direction for future research are also discussed. A New Normal for Portfolios: Construction, Assessment and Evaluation .......................................... 97 Jeffry Haber The financial crisis found many investors questioning what went wrong and how they might structure their portfolio going forward to avoid similar catastrophe. Such re-examination has led to new buzz-words and what seems like new paradigms. An example is the new normal. Other frequently seen headlines deals with whether the endowment model is broken. This paper looks at the investing issues in portfolio construction, assessment and evaluation in terms of the “new normal” and the “broken endowment model” and finds that the new normal is no different than the old normal and the endowment model is actually the blind following and replication of a few industry leaders. Despite the ample rhetoric, nothing much has changed or will change. The Impact of the Sarbanes-Oxley Act (SOX) on the Cost of Equity Capital of S&P Firms ............................................................................................................................. 102 Sheryl-Ann K. Stephen, Pieter J. de Jong This study examines the impact of SOX on the cost of equity capital for small and large S&P firms. The provisions of SOX aim to improve internal control systems and reduce information asymmetry by improving corporate governance systems and increasing transparency. Using a fixed-effects regression model, our findings suggest that the cost of equity capital has decreased post-SOX for the overall sample of firms, but more specifically for the small firms, which are usually associated with poor internal control systems and high information asymmetry. Collectively, our results provide evidence that SOX has had a positive impact on firms.

Comparing GDP Indexed Bonds to Standard Government Bonds .................................................... 116 Lillian Kamal, Malek Lashgari This paper discusses the construction of a new financial instrument whose payoffs are linked to the overall performance of the U.S. economy. The idea behind this financial instrument, known as a GDP bond, has been explored by financial economists for several years. Since there is no historical data for GDP bonds in the U.S., we construct a simulation process through which the payoff function for a GDP bond could be compared with other high quality bonds during the 1947-2010 period. It is observed that the GDP bond has a superior performance to other high quality bonds in a risk return framework. The Influence of Winning on Mid-Major College Football Attendance ............................................ 129 Tim Padgett, C. Shane Hunt A strong positive correlation exists between winning percentage and attendance within a given season. Understanding this correlation is important to sports marketers and sports business managers. Considering that marketers have no impact on win-loss record, this impact can be extracted before evaluating an organization’s marketing performance. Sports business managers have to consider the uncertainty of a team’s performance during an upcoming season when they attempt to accurately project revenues. Risk Behavior and Performance in Chinese Private Firms: From Regulatory Impact to Owner-Managers .................................................................................... 133 Chuanyin Xie Given the background that private firms in China were risk taking in once an unfavorable regulatory environment, this study examines their risk behavior and performance implications in a new era when the “invisible hand” begins to exert its influence. In the new era, private firms have choices as to whether or not to take risk, so risk behavior needs to be examined beyond the regulatory impact. I shift the research attention to owner-managers who make risk decisions. I argue that owner-managers’ characteristics, including who they are, why they run their own business, and to what degree they control their business, have direct implications for firm risk behavior. These characteristics can also affect the relationship between risk behavior and performance. An Analysis of the Behavior of Alternative Employment Indicators: Was the Great Recession Different? ...................................................................................................... 146 Bill Seyfried From 2007-2009, the United States experienced its worst economic downturn since the Great Depression. Though economic growth experienced a significant decline, the labor market was particularly hard hit. Some economists expressed surprise as to the severity of job losses. Unlike traditional models, this study develops a model that estimates the responsiveness of the labor market to economic growth as well as a risk premium to account for the impact of the availability of credit on the labor market. Alternative employment indicators as well as economic growth are considered. Empirical findings indicate that the behavior of employment during the Great Recession fit the historical experience based on the model developed.

Important Theories of Unemployment and Public Policies ................................................................ 156 Adil H. Mouhammed This paper intends to analyze the most important theories of unemployment. These theories are scientifically developed and confirmed by economists representing various schools of economic thought such as the Keynesian and the classical schools of political economy. These theories are used to develop some essential public policies that can be employed to reduce the unemployment rate. Estimating the Current Value of Time-Varying Beta .................................................................................. 167 Joseph Cheng, Elia Kacapyr This paper proposes a special type of discounted least squares technique and applies it to the Capital Asset Pricing Model. There is evidence that the value of beta, the measure of risk in the model, varies over time. The technique, entropic least squares, detects differences in the past and present standard error of the model. The rate of change in this standard error is referred to as the entropy rate. Unlike discounted least squares where the discount rate must be assumed in an ad hoc manner, entropic least squares estimates the entropy rate simultaneously with the parameters of the model. Specialty Funds vs. General Mutual Funds and Socially Responsible Investment (SRI) Funds: An Intriguing Risk/Return Paradigm ........................................................ 175 Brian D. Fitzpatrick, Joshua Church, Christopher H. Hasse This paper examines the benefits and detriments of specialty funds (sector funds) during both bull and bear markets. We created a contrasting analysis between general global mutual funds, specialty funds and Socially Responsible Investment (SRI) funds, incorporating 1 year, 3 year, 5 year and 10 years annualized returns for fifteen popular funds (period ending 5/7/2010), and found that specialty funds exhibited superior risk/return tradeoffs. PFSA and BaFin - Comparison of Institutional Framework in Dynamic Financial Markets ................................................................................................................................... 188 Agata Kocia, Grazyna Szymanska At present, theoretical debate in academic communities is ongoing concerning the structures and aims of the centralised supervisory offices. There are some arguments in favour of mergers of individual sector supervisors to create mega institutions and there are also some against it. The article presented below adds to this debate by presenting a more empirical approach aimed at comparing two super-supervisors the Polish PFSA and the German BaFin. The goal of this work is to compare and contrast both institutions and note any differences between, seemingly similar supervisory structures. The article begins with a theoretical explanation of most important terms and concepts related to the subject of financial supervision. Next, we describe the present day structure, aims and economic environment of the Polish PFSA and German BaFin. The following section is dedicated to comparison and contrast of peripheral supervisors’ activities. The final section is dedicated to drawing final conclusions.

GUIDELINES FOR SUBMISSION

Journal of Applied Business and Economics (JABE)

Domain Statement

The Journal of Applied Business and Economics is dedicated to the advancement and dissemination of business and economic knowledge by publishing, through a blind, refereed process, ongoing results of research in accordance with international scientific or scholarly standards. Articles are written by business leaders, policy analysts and active researchers for an audience of specialists, practitioners and students. Articles of regional interest are welcome, especially those dealing with lessons that may be applied in other regions around the world. This would include, but not limited to areas of marketing, management, finance, accounting, management information systems, human resource management, organizational theory and behavior, operations management, economics and econometrics, or any of these disciplines in an international context.

Focus of the articles should be on applications and implications of business, management and economics. Theoretical articles are welcome as long as their focus is in keeping with JABE’s applied nature. Objectives Generate an exchange of ideas between scholars, practitioners and industry specialists Enhance the development of the Business and Economic disciplines Acknowledge and disseminate achievement in regional business and economic development thinking Provide an additional outlet for scholars and experts to contribute their ongoing work in the

area of applied cross-functional business and economic topics. Submission Format

Articles should be submitted following the American Psychological Association format. Articles should not be more than 30 double-spaced, typed pages in length including all figures, graphs, references, and appendices. Submit two hard copies of manuscript along with a disk typed in MS-Word.

Make main sections and subsections easily identifiable by inserting appropriate headings and sub-headings. Type all first-level headings flush with the left margin, bold and capitalized. Second-level headings are also typed flush with the left margin but should only be bold. Third-level headings, if any, should also be flush with the left margin and italicized.

Include a title page with manuscript which includes the full names, affiliations, address, phone, fax, and e-mail addresses of all authors and identifies one person as the Primary Contact. Put the submission date on the bottom of the title page. On a separate sheet, include the title and an abstract of 200 words or less. Do not include authors’ names on this sheet. A final page, “About the authors,” should include a brief biographical sketch of 100 words or less on each author. Include current place of employment and degrees held.

References must be written in APA style. It is the responsibility of the author(s) to ensure that the paper is thoroughly and accurately reviewed for spelling, grammar and referencing. Review Procedure

Authors will receive an acknowledgement by e-mail including a reference number shortly after receipt of the manuscript. All manuscripts within the general domain of the journal will be sent for at least two reviews, using a double blind format, from members of our Editorial Board or their designated reviewers. In the majority of cases, authors will be notified within 60 days of the result of the review. If reviewers recommend changes, authors will receive a copy of the reviews and a timetable for submitting revisions. Papers and disks will not be returned to authors. Accepted Manuscripts

When a manuscript is accepted for publication, author(s) must provide format-ready copy of the manuscripts including all graphs, charts, and tables. Specific formatting instructions will be provided to accepted authors along with copyright information. Each author will receive two copies of the issue in which his or her article is published without charge. All articles printed by JABE are copyrighted by the Journal. Permission requests for reprints should be addressed to the Editor. Questions and submissions should be addressed to:

North American Business Press 301 Clematis Street, #3000

West Palm Beach, FL USA 33401 [email protected]

866-624-2458

Does Crowd Out Hamper Government Stimulus Programs In Recessions?

John J. Heim Rensselaer Polytechnic Institute

In well controlled statistical tests, crowd out was found related to government deficits financed by borrowing. Roughly equal effects were found for both recession and non-recession periods. Tax cut deficits were found were found more detrimental than spending deficits Private borrowing systematically declined with the growth of government deficits, and explained most variation in consumer and investment spending. Crowd out may be avoided by foreign borrowing or if M2 money increases prior to the deficit. These findings offer a plausible explanation for the failure of recent U.S. government stimulus programs to offset the 2008 recession. INTRODUCTION

When government finances deficits by borrowing from the savings pool, the reduction in loanable funds available to consumers and businesses is “crowd out”. Borrowing by both consumers and businesses is extensively used even in recessions to supplement the purchasing power of income, for example, when buying a house or new machinery. Government borrowing to finance deficits may crowd out private borrowing, and hence, spending, offsetting some or all of the deficit’s stimulus effects. Whether it actually does or not is an empirical issue, examined in this paper.

The paper tests whether consumer or business borrowing is negatively impacted by deficits, in either recessions or non-recession periods. It also tests if any declines in spending are systematically related to declines in borrowing, as theorized in the crowd out model. U.S. data for the period 1960-2000 are tested. Findings are tested for robustness by dropping the first or last ten year period from the sample, and then retesting. THE NO - CROWD OUT MODEL

Standard simple models of the economy do not allow for borrowing - related crowd out. In such models the impact of taxes and government spending are derived using the GDP identity:

GDP = Y = C + I + G + (X-M) (1)

A simple consumption function might be given as a linear function of disposable income (Y-T)

C = β(Y-T) (2) substituting C into (1) gives

Journal of Applied Business and Economics vol. 13(2) 2012 11

_ _ Y = | 1 | * [ - βT + I +G + X-M) ] (3)

|_ (1-β) _|

The clear expectation of standard model demand theory is that tax changes in are expected to be negatively related to the GDP, with a multiplier effect -β /(1-β). Changes in government spending and net exports are related to GDP in the positive direction, with a multiplier effect 1/(1-β).

THE CROWD OUT MODEL

However, to test the hypothesis that savings otherwise used to finance consumer credit is diverted to

finance government deficits, the consumption function must be modified to add the crowd out - causing factor, the deficit (T-G), where (T-G) = taxes minus government spending:

C = β (Y-T) + λ(T-G) (4)

where lambda (λ) represents the marginal effect of deficit spending on consumer demand. With this function, the model becomes

GDP = Y = β (Y-T) + λ(T-G) + G + I + (X-M) = [1/(1- β)] [ (-β+ λ) T + (1- λ) G + I + (X-M) ] (5)

The impact of a change in T or G on the GDP depends on λ as well as β. The tax multiplier, is now (-β+ λ)/ (1- β). The spending multiplier, is now (1-λ)/(1-β). Both T and G marginal effects on the GDP will be smaller (in absolute terms) than they would have been without crowd out effects.

If crowd out has different effects in recession (Rec) and non-recession periods (NonRec), the formulation becomes

GDP = Y= β (Y-T) + λRec(T-G) + λNonRec(T-G) + G + I + (X-M) = [1/(1- β)] [ (-β+ λRec)T + (-β+ λNonRec)T + (1- λRec) G + (1- λNonRec) G + I + (X-M)] (6)

We can see the impact of a change in T or G on the GDP depends on λRec or λNonRec as well as β. The

tax multiplier, is now (-β+ λRec)/ (1- β) in recessions or (-β+ λNonRec)/ (1- β) in non-recessions. If crowd out is less in recessions, the tax cut multiplier effects will be larger than in non-recessions. The spending multiplier, is now (1-λRec)/(1-β) or (1-λNonRec)/(1-β) and if crowd out is less in recessions, the spending multiplier will be larger in recessions than in non-recessions.

We can expand this model to include effects of crowd out on investment spending. Assume a simple investment model in which investment is determined by real interest rates (r) and access to credit, which varies with the government deficit (T-G).

I = γ(T-G) - θ r (7) where gamma (γ) indicates the marginal effect of crowd out (the government deficit) on investment spending, and (θ) represents the marginal effect of real interest rates (r).

If we replace investment in the GDP identity with its hypothesized determinants, we obtain a typical “IS” equation:

GDP = Y = = [1/1- β] [ (-β+ λ+ γ) T + (1- λ-γ) G - θ r + (X-M) ] (8)

In this IS equation, the normal stimulating impact of tax cuts on the GDP (-β) is offset by the effects

of deficit – induced changes in credit available to consumers and investors (λ+γ). Tax stimulus effects may switch from negative to positive if the crowd out effects (λ+ γ) are larger than the disposable income

12 Journal of Applied Business and Economics vol. 13(2) 2012

effect (-β). The normal stimulating effect of government spending is reduced from (1) to (1- λ-γ), and stimulus effects are either reduced or become negative. The net exports multiplier effect stays the same, now becoming an even stronger stimulus relative to government spending or tax cuts. Results are shown in Table 1. Crowd out reduces the stimulus of spending deficits less than tax cut deficits, because the spending stimulus effect (1) to be offset by crowd out is larger than the tax stimulus effect (-β). Results are shown in Table 1.

Crowd out may or may not be a problem in recessions. It can be argued consumers and businesses borrow less in recessions, leaving savings available to finance new government deficits without causing crowd out. However, national savings may also drop in recessions due to falling incomes. If savings decline as much or more that private borrowing demand, new deficits will still cause new crowd out. Hence, arguments for and against crowd out in recessions can be made theoretically.

If crowd out effects are different in recessions than in non-recessions, the investment and IS functions change as follows:

I = - θ r + γRec (T-G) + γNonRec (T-G) (9)

Y = [1/1- β] [ (-β+ λRec+ γRec) T or (-β+ λNonRec+ γNonRec) T + (1- λRec-γRec) G

or (1- λNonRec-γNonRec) G - θ r + (X-M) ] (10)

Hence, the stimulus effect of tax cuts (-β) is offset by either the recession effect of crowd out (λRec+ γRec) or the non-recession effect (λNonRec+ γNonRec). The stimulus effect of government spending (+1) is offset by either the recession effect of crowd out (-λRec- γRec) or the non-recession effect (-λNonRec- γNonRec). These results are also summarized in Table 1.

TABLE 1 EFFECTS OF CROWD OUT ON TAXES AND GOVERNMENT SPENDING STIMULUS

Without With Without With

Crowd Out Crowd Out Crowd Out Crowd Out Tax coefficient (-β) -β+ (λ+ γ) Spending Coefficient. 1 1-(λ+γ) (-β) -β+ (λ+ γ)Rec 1 1-(λ+γ) Rec (-β) -β+ (λ+ γ)NonRec 1 1-(λ+γ)NonRec Tax Multiplier (-β) -β+ (λ+ γ ) Spending Multiplier 1 1-(λ+γ) (Average-All Per.) (1-β) (1- β) (1-β) (1-β) Tax Multiplier (-β) -β+ (λ+ γ)Rec Spending Multiplier 1 1-(λ+γ)Rec (Recession Period) (1-β) (1- β) (1-β) (1-β) Tax Multiplier (-β) -β+ (λ+ γ )NonRec Spending Multiplier 1 1-(λ+γ)NonRec (Non-Recession) (1-β) (1- β) (1-β) (1-β)

Several conclusions follow from Table1:

a) If the signs of the crowd out variable coefficients (λ, γ) are positive, the stimulus effect of tax cuts on the GDP will be smaller than a standard model would predict. Reducing taxes has a net stimulus effect only if (β) is larger than the appropriate variant of (λ+γ). If (λ+γ) is equal to or greater than (β), there is complete, or more than complete, crowd out.

b) The government spending multiplier of (1/1- β) in the “no - crowd out” model, also declines in

the presence of crowd out. It is now (1-λ)/(1- β), (1-λRec)/(1- β), or (1-λNonRec)/(1- β). Stimulus due


to increased government spending is now partially or wholly offset by reductions in consumer spending caused by crowd out

c) The multiplier effect of net export spending stays the same. Relatively speaking, this means that if crowd out exists, a dollar increase in net exports should have a larger multiplier effect than a dollar of government spending, a testable hypothesis.

The model we shall test later in this paper is a slightly different form of the model shown above. The

model above was based on the GDP identity

Y = C + I + G + (X-M) (11)

But we can just as accurately write

Y = CD + ID + GD + X (where subscript d denotes domestically produced goods ) (12)

This is an important distinction in calculating multipliers because only spending on domestically produced consumer or investment goods generates the multiplier effect on the GDP. Hence, the last formulation of the GDP identity may be the better form to use when calculating IS curve parameter estimates, since multiplier effects inherent in the coefficients are more correctly estimated. (We abstract from effects on exports of growth in import demand).

Because the data available does not allow separation of government purchases into domestic produced goods and imports, the form of the model we test is:

Y = CD + ID + G + X) (13) OTHER STUDIES

The ability of deficits to stimulate the economy hangs heavily on whether or not the government borrowing to finance deficits crowds out private borrowing, and therefore private spending. No macroeconomic models were found which directly tested borrowing for this relationship. No studies were found that specifically test for differences in effects in recession and non-recession periods. Recession periods are when deficits are most commonly adopted tools for stimulating the economy.

Some work, discussed below, has been done to test crowd out indirectly, by testing the average relationship between private spending and deficits over the course of the business cycle. If a negative relationship was found, it was assumed to be due to crowd out.

The popular press is full of discussion of crowd out effects that are based on the assumption that crowd out does or does not work. For example:

1. Chan, S. (NY Times, 2/7/10, p.A16): reported the I.M.F. warned on Jan. 26 that rising sovereign debt "could crowd out private sector credit growth, gradually raising interest rates for private borrowers and putting a drag on the economic recovery."

2. Barley, R. (Wall Street Journal, 2/24/10 p.C14): “any government-bond buying by banks is another form of crowding out, potentially reducing supply of consumer and corporate lending”

3. Krugman (New York Times, 9/28/09) notes that in recessions, the accelerator effect is likely to dominate any crowd out effect, leaving a net stimulus effect of government spending increases or tax cuts.

In the professional literature, studies examining crowd out have been entirely, or principally literature

reviews. For example, Spencer and Yohe, (1970), in reviewing the literature, found that the dominant view the past two hundred years from all types of studies has been that government deficits cause crowding out. Friedman’s work (1978) is largely theoretical, though it contains some references to his and


others’ empirical work. He shows portfolio theory suggests the LM curve may shift in response to an IS shift due to a fiscal stimulus like a government deficit, and that elasticity of substitution between bonds and stocks when interest rates rise (due to deficit borrowing) is key: elasticities less than one lead to crowd out; greater than one: crowd in. Therefore crowd out effects are indeterminate theoretically. Friedman’s own empirical results, based on money demand models, were ambiguous.

Gale and Orszag’s work (2004) does include some empirical testing indicating crowd out matters. Consumer demand was hypothesized to be a function of current and one period lagged Net National Product (NNP), government purchases, taxes, transfer payments, interest payments and the size of the government debt. A negative relationship between taxes and GDP were taken as a sign that crowd out, if it existed, was not complete. That said, their tested hypothesis did not include the government deficit as an explanatory variable. This can result in stimulus effects of tax cuts being overstated (Heim 2010). Other tests also indicated a positive relationship between interest rates and deficits. This was taken as an indicator of crowd out. However, some studies show the interest rates most systematically related to the GDP are not supply and demand driven rates, but exogenously determined: the federal funds and prime interest rates, (Heim, 2007). These would not systematically move upward as deficits increased, and in fact might move in the opposite direction, if the controllers of these rates followed some sort of Taylor Rule, since deficits seem most likely to grow in recessions.

Using a VAR methodology, Montford and Uhlig (2008) found investment falls in response to both spending and tax increases, a finding inconsistent with both standard stimulus theory and crowd out theory. The VAR specified consumption or investment as being a function of six lagged values of each of ten variables: GDP, C, G, Taxes, real wages, private non-residential I, adjusted reserves, the PPI index and the GDP deflator. Since the tested hypotheses in VAR models are somewhat atheoretical, findings can be difficult to interpret. Blanchard and Perrotti (2002), also using a VAR model, obtained the same result for investment when testing taxes and government spending, but more Keynesian results for total output, and non-Keynesian results for consumption.

Furceri and Sousa (2009) examine 145 countries using a VAR methodology to determine if government spending as a % of GDP was related to consumption and investment spending as a % of GDP, They conclude government spending is adversely related. Fundamentally the model tests consumption and investment spending against right - side variables fixed effects variables for the individual countries and the current and four lagged values of the government spending/ GDP variable. While many of the government spending variables had statistically significant adverse effects, the lack of controls for other structural variables makes it difficult to be sure the finding truly represent the government spending effect, and not perhaps be proxying for non-included variables.

METHODOLOGY

1960 - 2000 data on the determinants of consumption and investment spending were taken from the Economic Report of the President 2002 and 2010. Flow of Funds Accounts of the Federal Reserve were used to obtain data on consumer and business debt, changes in which were taken as measures of net consumer or business borrowing. The specific variables are identified in later sections.

Two -stage least squares regression was used since both consumption and investment are driven in part by income related variables (disposable income or the accelerator), and therefore 2SLS was needed to avoid issues of simultaneity. The remaining variables in the consumption and investment equations are lagged, or considered exogenous (e.g. the prime interest rate is rigidly set to reflect the federal funds rate, which is exogenously determined by the Fed; depreciation allowances are determined by prior year investment levels, etc.). They were used as 1st Stage regressors. Variables were tested in first differences instead of levels to address nonstationarity, serial correlation and multicollinearity issues commonly found in time series data. Newey West corrections to standard errors were used to avoid heteroskedasticity problems. Durbin - Watson measures of serial correlation are shown with each regression. Hausman endogeneity tests failed to show significant endogeneity between the government spending and tax variables, and either consumption or investment. The 2SLS form of instrumental


variables and the D.W. test for serial correlation are considered the most appropriate for use with time series data (Griffiths, Hill, Lim, 2011).

It is difficult to separate consumer imports out of total imports in the Economic Report of the President. The Bureau of Economic Analysis (BEA) has confirmed it does not categorize import and export data into same “C” and “I” and “G” categories used elsewhere in the national GDP accounts. Absent official determinations by BEA, economists must make their own evaluations of how to divide the data. For example, it is not clear from Table 104 in the Economic Report of the President how much of the value of motor vehicle imports or petroleum imports should be treated as inventory investment vs. sales to consumers. Data on imported services (Table B-106) does not distinguish between imports of services by businesses and consumers, though one might suspect the former dominate. Nor do the services data extend back beyond 1974. Hence, no deduction from total imports for business services imports could be made in calculating consumer imports.

Following Heim (2010), we take as our definition of consumer imports all imports except imports of capital goods and industrial supplies and materials. The theory behind this choice was that the best definition of “consumer” imports was the one whose variation was best explained (highest R2) by the variables theoretically thought to drive demand for consumer imports. Other definitions of consumer imports, did not explain consumer behavior as well.

To obtain separate deficit, tax or government spending variables for recession periods, they are multiplied by a dummy variable taking the value (1) when there is a depression at some time during the data year, and (0) in non-recession years. For non-recession years, the dummy variable is reversed. National Bureau of Economic Research estimates (NBER 2009) were used to define recession years.

VARIABLES INCLUDED IN CONSUMPTION AND INVESTMENT MODELS

Theory suggests a wide range of variables are determinants of consumer demand. Individual studies have provided some empirical support for many variables, though not always controlling adequately for the rest. The consumption functions tested in this paper control for an extensive list of possible factors that might influence consumption, including crowd out, thereby helping ensure the correctness of the crowd out results. Lagged or average values are used with some of these variables, reflecting the findings of earlier studies. The variables used include:

CT = real consumer goods and services (Y-T) = real disposable income (T-G) = real government deficit, defined as one variable PR = real prime interest rate DJ-2 = a measure of wealth (Dow Jones Composite Index), lagged two years XRAV = exchange rate average for current and past three years POP16 = ratio of young (16 -24 year olds) to old (over 65) in population POP = population size ICC-1 = Index of Consumer Confidence (Conference Board), lagged one year M2AV = real M2 money supply, average of past three years (CBOR) = Net annual consumer borrowing = change in consumer debt (∆CDEBT) that period.

Theory also suggests many variables are determinants of investment demand. Here again, individual

studies have provided some empirical support for many different variables, though results are not always obtained controlling adequately for other possible factors. The variables used here include:

IT = total spending on investment goods, both domestically produced and imported IM = Spending on imported investment goods ID = Spending on domestically produced investment goods ACC = a Samuelson accelerator variable measuring the economy’s growth rate (∆ Real GDP)


DEP = real business depreciation allowances CAP-1 = industrial capacity utilization, lagged one period r-2 = real Prime interest rate, lagged two periods DJ-2 = Dow Jones Composite Average, a proxy for Tobin’s q PROF-2 = real corporate profits, lagged two periods BBOR = Net annual business borrowing = change in business debt (∆BDEBT) that period.

Other variables in the investment model are as defined in the consumption function.

TEST RESULTS, FINDINGS Overview

The below equations develop “no crowd out” and “average crowd out” models and their test results for the 40 year period 1960-2000. Average crowd out models have coefficients on the deficit variable that represent the average effect on consumption or investment over the period tested. No separate recession/non-recession coefficients are estimated.

In addition, the same models allowing for different crowd out effects in recession and non-recession (R/NR) period effects are compared to no crowd out models. Such models will include crowd out variables: ∆(T-G)Rec , and ∆(T-G)NoRec .

Lastly, we compare average crowd out results to (R/NR) models results to see which of these two models better predict actual IS curve coefficients. Baseline Comparisons: “Average Crowd Out” Versus ”No Crowd Out” Models

The “no crowd out” model and the average crowd out model ∆(T -G) are tested below. Results are shown for domestically produced consumer and investment goods models, because they are used to predict IS curve coefficients. Results are presented for total consumption and investment, which includes imports, and compared with total consumption and business borrowing. Separate results for consumer and investment imports can be obtained by subtracting domestic production coefficients from total model coefficients.

The regression results for consumer and investment spending are:

Consumption Functions - No Crowd Out) ΔCD =.43Δ(Y-TG) – .80ΔPR+.46 ΔDJ-2 + .63 ΔXRAV -414.54ΔPOP16+.006ΔPOP+.37ΔICC-1+32.45ΔM2AV R2=81.3% (14) (t =) (7.1) (-0.3) (2.3) (0.4) (-1.5) (1.7) (1.1) (4.2) D.W.= 1.8 where CD represents spending on domestically produced consumer goods. Investment Functions - No Crowd Out ΔID =.36ΔACC + .83ΔDEP + 2.21ΔCAP-1 - 11.07Δr-2 +.07 ΔDJ-2 +.51 ΔPROF-2 + 4.55 ΔXRAV0123 - .00ΔPOP R2=.75 (15) (t =) (8.7) (1.5) (1.2) (-3.9) (0.3) (2.9) (4.8) (-0.2) DW =2.5 where ID represents spending on domestically produced investment goods. Consumption Functions With “Average” Crowd Out ΔCD =.36Δ(Y-TG) +.27Δ(T -G) -4.49ΔPR+.24 ΔDJ-2+1.30 ΔXRAV -375.09ΔPOP16+.01ΔPOP+.23ΔICC-1+37.75ΔM2AV (t =) (6.4) (3.0) (-1.6) (1.3) (1.3) (-1.9) (2.8) (0.9) (4.9) (16) R2=86.0% D.W.= 1.9 Investment Functions With “Average” Crowd Out ΔID =+.51 Δ(T-G) + .23ΔACC + .16ΔDEP - .37ΔCAP-1 - 8.22Δr-2 -.28 ΔDJ-2 +.44 ΔPROF-2 + 5.59ΔXRAV + .008ΔPOP R2=.90 (17) (t =) (7.6) (9.0) (0.5) (-0.3) (-6.6) (-1.3) (4.1) (5.6) (3.6) DW =2.3


Predicted IS Curve (No Crowd Out) ∆Y= -.75∆T +1.75∆G+1.75∆X -1.40PR + .93∆DJ-2 + 9.07XRAV0123 -725.45 ∆POP16 + .01∆POP + .65∆ICC+56.79∆M2 (18)

+.63∆ACC+1.45∆DEP +3.87∆CAP-1 -19.37r-2+ .89∆PROF-2 Predicted IS Curve (With Average Crowd Out) ∆Y=+.65∆T + .34∆G+1.56∆X -7.00PR - .06∆DJ-2 +10.75XRAV0123 -585.14 ∆POP16 + .03∆POP +.36∆ICC-1+58.89∆M2

+.36∆ACC+ .25∆DEP - .58∆CAP-1 -12.82r-2 + .69∆PROF-2 (19) Actual test Results (The Same Hypothesis Tests No Crowd Out and Average Crowd Out) ∆Y= +.78∆T - .20∆G + .61∆X -6.69∆PR +.30∆DJ-2 + 4.38XRAV +505.70∆POP16 +.05∆POP +1.42∆ICC-1+ 45.43∆M2 (t=) (6.0) (-0.6) (-2.1) (2.4) (0.8) (2.4) (1.4) (6.7) (2.8) (3.0) +.58∆ACC+ .16∆DEP +7.97∆CAP-1 + .04r-2 +.21∆PROF-2 R2=97.6% (20) (10.0) (0.3) (2.2) (0.0) (0.8) DW=2.3

The predicted IS curve coefficients for these two models, as well as the actual regression estimates obtained testing the IS curve are summarized in Table 2 below:

Findings

• Adding crowd out to the consumption model increases explained variance from 81.3% to 86.0%. For investment, explained variance increases from 75% to 90. Crowd out variables are significant at the 1% level.

• IS curve predictions from the no - crowd out model show positive net economic effects for both tax cut and spending deficits (as expected)

• IS curve predictions from the crowd out model indicate more than complete crowd out for tax cut and partial crowd out for spending deficits. Actual regression results match predictions, showing more than complete crowd out for tax cut deficits, and complete crowd out for spending deficits.

• The crowd out model predicted actual IS curve test results far better than the no crowd out model, as shown in Table 2.

TABLE 2 IS CURVE COEFFICIENTS ON TAX AND GOVERNMENT SPENDING VARIABLES

Tax Spending No Crowd Out model Prediction - 0.75 + 1.75 Av. Crowd Out Model Prediction +0.65 + 0.34 Actual Regression Result +0.78 - 0.20 (6.0) (-0.6)

• The government spending variable in the IS curve test has a smaller coefficient than the exports

coefficient, as predicted by crowd out theory in earlier. In no crowd out models, these predicted coefficients are the same and equal to the multiplier.

• The average crowd out model predicts 10 of 15 actual IS coefficients more than by the no crowd out model, including the crowd out variable coefficients.

• The predicted IS curve results indicate a $0.65 drop in GDP for every dollar of deficit-financed tax cut financed by borrowing. The predicted IS curve results also show a net stimulus effect of $+0.34 per dollar of government spending deficit incurred if financed by borrowing. The actual IS curve regression results show nearly the same effect for tax cuts ($0.78), but a result for net spending deficits of ($-0.20), a result insignificantly different from zero.

Conclusion

Average crowd out models explain substantial variance in consumption, investment and GDP that no-crowd out models leave unexplained. Average crowd out models are much better at predicting IS curve coefficients, the key determinants of GDP effects. Findings indicate that the crowd out effects of tax cut


deficits more than fully offset stimulus effects, resulting in net negative economic effects. Regression results for government spending deficits indicate full crowd out, with no net economic effect of the deficit either way, though predicted results were for a small positive net effect, about one third the size of the stimulus. Baseline Comparisons: “Recession/No Recession” Versus “No Crowd Out” Models

Below, consumption and investment models are tested for separate recession and non - recession crowd out effects. The domestic consumption and investment models tested are the same as previously used. Regression results for the crowd out model are: Domestically Produced Consumer Goods ΔCD =.36Δ(Y-TG) +.20Δ(T -G)Rec + .35Δ(T -G)NonRec - 4.07ΔPR+.22 ΔDJ-2+1.24 ΔXRAV - 410.32ΔPOP16 +.01ΔPOP (21) (t =) (6.8) (2.2) (2.2) (-1.5) (1.1) (1.5) (-2.4) (2.3)

+.13ΔICC-1+40.02ΔM2AV R2=86.4% (0.4) (4.2) D.W.= 2.1

Domestically Produced Investment Goods: ΔID =+.50 Δ(T-G)Rec =+.52 Δ(T-G)NonRec + .23ΔACC + .13ΔDEP - .35ΔCAP-1 - 8.15Δr-2 -.27 ΔDJ-2 +.44 ΔPROF-2 + 5.58ΔXRAV (22) (t =) (8.3) (3.5) (9.0) (0.5) (-0.3) (-7.0) (-1.4) (4.0) (5.7)

+ .008ΔPOP R2=89.8% (3.7) DW =2.3

Using the parameter estimates from these domestic consumption (CD ) and investment (ID ) equations,

the IS curve parameters are predicted to be: Predicted IS Curve (Separate Crowd Out Variable (T, G) For Recessions/Non-recessions) ∆Y= +.53∆Trec + .80∆TNonRec + .47∆GRec + .20∆GNonRec +1.56∆X -6.35PR - .08∆DJ-2 +10.64XRAV0123 -640.10 ∆POP16 +.03∆POP+.20∆ICC1+62.43∆M2 +.36∆ACC+ .20∆DEP -.55∆CAP-1 -12.71r-2 + .69∆PROF-2 (23) Actual IS Curve Test Results (With Separate (T) And (G) Variables For Recession/Non-Recessions) ∆Y= +.87∆TRec +.60∆TNonRec - .65∆GRec - .23∆GNonRec + .63∆X -8.00∆PR +.24∆DJ-2 + 4.97XRAV +445.43∆POP16 (24) (t=) (5.3) (2.3) (-1.1) (-0.6) (2.0) (2.2) (0.6) (2.6) (1.3) +.05∆POP +1.59∆ICC-1+ 44.51∆M2 +.59∆ACC+ .68∆DEP +8.36∆CAP-1 - .40r-2 +.22∆PROF-2 R2=97.8% (5.6) (3.5) (2.3) (10.4) (0.7) (2.3) (-0.1) (0.8) DW=2.5

The results indicate that • Adding crowd out to the consumption model increases explained variance from 81.3% to 86.4%,

slightly more than the average effect model (86.0%). For investment, adding crowd out increased explained variance about the same as the average crowd out model: from 75% to 89.8%, slightly less than the average model (90.0%). Though there are differences in specific recession/non-recession coefficients for taxes and spending, they are each within the confidence intervals of the other, suggesting they may be the same. This indicates dividing crowd out into recession and non-recession variables adds little information not already available in the average crowd out model. This actual IS curve results again suggest more than complete crowd out for tax cut deficits, and complete crowd out for spending deficits. The predicted IS curve results again indicate the same for tax cuts, but only partial crowd out for spending deficits. The lack of a significant difference in crowd out effects in the two periods is further explored below.

• All four crowd out variables in the consumption and investment models are statistically significant at the 1% or 3% level. Statistical significance in both recession and non-recession periods strongly supports the hypothesis crowd out has negative effects in both periods.

• Deficits appears to have about the same marginal effect in recession and non-recession periods on investment (.50 vs.52). For consumption, point estimates of effects are larger for non-recession periods, but well within the other estimates confidence intervals, suggesting the differences may not be significant.


• The recession/non-recession crowd out model, predicted 9 of 17 IS curve coefficients more accurately than the no crowd out model. The no crowd out model predicted 7 better. The standard for judging results was an IS model tested with separate (T) and (G) variables for recession and non-recessions.

• When the standard for judging results was an IS model tested with only one set of (T) and (G) variables (average crowd out), the recession/non-recession model predicted 10 of 17 coefficients better than the no crowd out model.

• The regression coefficient estimates of IS curve crowd out effects, indicate more than complete crowd out of government spending deficit stimulus, but the coefficient is insignificantly different from zero, suggesting only full crowd out. Results do indicate tax cuts more than fully crowd out stimulus effects. Results are shown in Table 3 below. R/NR model predictions of actual IS coefficients were also found to be far more accurate than no crowd out model predictions.

TABLE 3

IS CURVE COEFFICIENTS FOR TAX AND GOVERNMENT SPENDING VARIABLES Tax Spending Rec. No Rec. Rec. No Rec. No Crowd Out model Prediction - 0.75 - 0.75 + 1.75 + 1.75 R/NR Crowd Out Model Prediction + 0.53 + 0.80 + 0.47 + 0.20 Actual Regression Result + 0.87 + 0.60 - 0.65 - 0.23 (t=) (5.3) (2.3) (-1.1) (-0.6)

• Actual IS curve regression coefficients suggest substantially different recession and non-recession

period effects (+.87∆T Rec +.60∆TNonRec - .65∆GRec - .23∆GNonRec), with deficits causing a substantially worse crowd out problem in recession than non-recessions. This may occur because savings fall faster than borrowing in recessions, necessitating a much larger cutback in credit based spending by both consumers and businesses than just that caused by the crowd out effects. This decline would be coincidental with, but not part of, the crowd out effect. Our models do not provide an easy way of disentangling the two effects. That said, these differences also cannot be considered different with any statistical certainty; 5% confidence intervals around each estimate contain the other as a possibility.

Conclusions

The findings indicate crowd out has a statistically significant negative effect on domestic consumer and investment spending in both recession and non-recession periods.

Though the IS curve coefficients indicate both tax cuts and government spending had larger crowd out effects in recessions than non-recessions, the confidence intervals around these estimates were large enough to indicate there may not be a real difference. The likely reason we find crowd out in recessions is because loanable funds availability (savings) drops as much or more than borrowing demand, a topic more fully examined later. This is a key finding, for it is sometimes argued that crowd out is not a problem in recessions, when deficit stimulus is needed most, because private demand for loanable funds declines.

Actual regression results for the IS curve in both recession and non-recession periods indicate crowd out more than fully offset all tax stimulus, but only fully offset government spending stimulus in both periods. Predictions were the same for tax cuts, but only suggested partial crowd out for spending deficits.

IS curve coefficients are predicted more accurately from crowd out models than from no crowd out models.


Crowd Out Comparisons: “Average” Versus “Recession/No Recession” Models In preceding sections, both average and recession/non-recession (R/NR) crowd out models show

deficits to be negatively related to private spending, better at predicting IS curve parameters, and better at explaining variance in consumer and investment spending than no - crowd out models. Is either the average or R/NR model a better crowd out model than the other? Does one explain economic activity significantly better than the other?

To evaluate this, two sets of consumption and investment regressions, and there IS curve predictions, are compared to actual IS curve coefficients. One set contain the average crowd out, taken from previous equations and the other has recessions and non-recession variables, also seen earlier.

Which actual IS curve coefficients to use as the standard for judging the success of predictions is a problem: we could use either the average or R/NR regression coefficients. Will we bias results if we use one instead of the other? To deal with this, we shall evaluate each set of predictions separately against each set of actual regression results, to see if there are significant differences in the results. Predicted and actual regression results, repeated from earlier sections, are given below:

IS Curve Predictions (Repeated From Previous Section): Predicted IS Curve (With Average Crowd Out) ∆Y=+.65∆T + .34∆G+1.56∆X -7.00PR - .06∆DJ-2 +10.75XRAV0123 +(NA)∆HSE -585.14 ∆POP16 + .03∆POP +.36∆ICC-1 (25)

+58.89∆M2 +.36∆ACC+ .25∆DEP - .58∆CAP-1 -12.82r-2 + .69∆PROF-2 Predicted IS Curve (Separate Crowd Out Variable (T-G) Included For Recessions/Non-recessions) ∆Y= +.53∆Trec + .80∆TNonRec + .47∆GRec + .20∆GNonRec +1.56∆X -6.35PR - .08∆DJ-2 +10.64XRAV0123 -640.10 ∆POP16 (26) + .03∆POP+.20∆ICC1+62.43∆M2 +.36∆ACC+ .20∆DEP -.55∆CAP-1 -12.71r-2 + .69∆PROF-2 Actual IS Curve Test Results (From Previous Section): Actual Test Results (The Same Hypothesis Tests No Crowd Out and Average Crowd Out Models) ∆Y= +.78∆T - .20∆G + .61∆X -6.69∆PR +.30∆DJ-2 + 4.38XRAV +505.70∆POP16 +.05∆POP +1.42∆ICC-1+ 45.43∆M2 (27) (t=) (6.0) (-0.6) (-2.1) (2.4) (0.8) (2.4) (1.4) (6.7) (2.8) (3.0) +.58∆ACC+ .16∆DEP +7.97∆CAP-1 + .04r-2 +.21∆PROF-2 R2=97.6% (10.0) (0.3) (2.2) (0.0) (0.8) DW=2.3 Actual Test Results (With Separate T And G Variables For Recession/Non-Recessions) ∆Y= +.87∆TRec +.60∆TNonRec - .65∆GRec - .23∆GNonRec + .63∆X -8.00∆PR +.24∆DJ-2 + 4.97XRAV + 445.43∆POP16 (28) (t=) (5.3) (2.3) (-1.1) (-0.6) (2.0) (2.2) (0.6) (2.6) (1.3) +.05∆POP +1.59∆ICC-1+ 44.51∆M2 +.59∆ACC+ .68∆DEP +8.36∆CAP-1 - .40∆r-2 +.22∆PROF-2 R2=97.8% (5.6) (3.5) (2.3) (10.4) (0.7) (2.3) (-0.1) (0.8) DW=2.5

The average crowd out model predicted 8 of 17 coefficients more accurately than the R/NR model, even using the R/NR regression coefficients as the measure of correctness. The R/NR better predicted 5, using the R/NR coefficients as the standard of correctness. Differences in predicted values generally were small and tended to be closer to each other than to actual regression coefficients, suggesting differences in predictive power were minor, at best.

The R/NR model better predicted 7 of 17, the average model 6, using the average coefficients as the standard of correctness. Again, predictions from both models were closer to each other than to the actual results, indicating they are more alike than different in predicting IS curve coefficients.

These results indicate that separating the single average estimate of the deficit’s effect on private spending into recession and non-recession parts adds little to our total information, implying crowd out effects are the same in both periods. (R2 only increased from 97.6% to 97.8% using the R/NR IS model)


Conclusions Because recession and non - recession effect estimates were reasonably close, this model does not

predict actual IS curve coefficients significantly better than average crowd out models, which assume crowd out effects are the same in both periods.

There may be differences in crowd out effects in recession and non-recession periods, but the effects are more alike than different. In both periods, government tax cut deficits are associated with declines in the economy, and spending deficits result in no net effect on the economy, positive or negative.

• The findings are robust to the time period selected for testing. Dropping either the first or last ten years from the sample and re-estimating total consumption and investment, yields statistically significant crowd out effects for both recession and non-recession periods. Estimates of marginal effects are similar in the different samples.

• Average crowd out coefficients for consumption, are generally more significant (1% level) than the statistical significance of the R/NR coefficients (3%).

• Average and R/NR crowd out coefficients for investment were both significant at the 1% level. • Virtually no additional variance is explained in the IS curve using the recession/non-recession

model (97.8%) of crowd out compared to average crowd out model (97.6%). • Adding crowd out in its average form to a no crowd out model increases explained variance in

consumption from 81.3% to 86.0%. When R/NR is added, it grows slightly more to 86.4%. Average crowd out and R/NR add about the same amount to investment’s explained variance: raising it from 75% to 90%.

• These results suggest using the R/NR form adds little additional information to our knowledge of crowd out behavior, compared to average model results. This suggests crowd out significantly negatively affects the economy in recession and non-recession periods, and to about the same extent. Therefore, the average crowd out formulation probably provides as accurate an estimate of crowd out impact as the formulation with separate R/NR variables.

• The results do indicate crowd out is a significant factor offsetting stimulus in both recession and non-recession phases of the business cycle, as shown by t-statistics on crowd out variables in the domestically produced consumption and investment goods spending equations above, reproduced in Table 4 below:

TABLE 4 t-STATISTICS FOR CROWD OUT VARIABLES*,**

(T-G)Rec (T-G)NonRec (T-G)Average .

Consumption “t” (2.2) (2.2) (3.0) Investment “t” (8.3) (3.5) (7.6) . *1.5=15% level; t = 1.7 =10% level; 1.8=8% level; t=2.0 = 5% level; t= 2.7 = 1% level.

**t-statistics taken from equations 16, 17, 21, and 22 above

In addition, the average crowd out model better predicted actual IS curve coefficients (from IS curve tested with average crowd out) than the R/NR model, and even did nearly as well when the standard used was the R/NR model, as shown in Table 5


TABLE 5 AVERAGE CROWD OUT VS. R/NR PREDICTION RECORD

Average Crowd Out (C/O) Predicts Better Than R/NR When Tested Against Actual IS, Using Average Crowd Out Standard (8,5,4)* Actual IS, Using R/NR Crowd Out Standard (6,7,4)* * Variables better predicted given in the following order: (Av., R/NR, Tie). Results from Equations 25 - 28.

DIRECT EVIDENCE OF CROWD OUT: THE IMPACT OF DEFICITS ON PRIVATE BORROWING

Developed models clearly show a negative relationship between deficits and spending, ceteris paribus. It has been assumed that the mechanism causing the negative relationship between deficits and private spending is government borrowing to finance deficits, reducing the amount of national savings left for consumers and businesses to borrow. Hence, crowd out theory hangs on whether the data on consumer and business borrowing show the same decline we see in spending in the presence of deficits. To test the relationship of deficits to borrowing, Federal Reserve Flow of Funds data on business and consumer debt is used. Changes in debt measure net borrowing. Using borrowing, not spending, as the dependent variable, we retest the same consumption and investment models previously tested. Theory suggests some or all of the same factors that drive consumer spending drive consumer borrowing, since borrowing is but one way of manifesting desired spending. If the decline in spending is caused by the decline in borrowing, we should find spending and borrowing have about the same relationship to deficits, and in the same period.

Relationship of Borrowing to Average Crowd Out

Below are regression results for models of consumer and business spending on domestically produced and imported goods. They include estimates of crowd out’s average effect over the course of the business cycle. Consumers and businesses borrow money because they intend to spend it, but not all variables that influence spending necessarily influence borrowing. However, we expect to find some similarities

Total Investment Spending Function With “Average” Crowd Out ΔIT =+.60 Δ(T-G) + .27ΔACC + .29ΔDEP + .72ΔCAP-1 - 6.79Δr-2 +.08 ΔDJ-2 +.32 ΔPROF-2 + 5.16ΔXRAV0123 + .011ΔPOP (28) (t =) (6.4) (8.2) (0.9) (0.6) (-4.0) (0.3) (1.9) (6.2) (4.0) R2=91%; DW =2.5 Total Consumption Spending Function With “Average” Crowd Out ΔCT =.50Δ(Y-TG) +.54Δ(T -G) -10.28ΔPR+.59 ΔDJ-2+4.32 ΔXRAV -360.95ΔPOP16+.01ΔPOP+.55ΔICC-1+30.34ΔM2AV (29) (t =) (14.7) (11.3) (6.1) (3.6) (5.1) (-1.9) (4.7) (2.2) (4.7) R2=96.2% D.W.= 2.1

The estimated total negative effect on private consumption and investment spending per dollar of deficit incurred is $1.14, ceteris paribus, i.e., holding stimulus effects on income constant. Unlike IS curves estimates, which combine stimulus and crowd out effects, and therefore show smaller net effects of changes in deficits, the consumption and investment function estimates measure crowd out effects alone.

The same functions, with borrowing, not spending, as the dependent variable, yielded the following results:


Total Investment Borrowing Function (ΔIB) With “Average” Crowd Out ΔIB =+.48 Δ(T-G) + .09ΔACC +1.43ΔDEP - .59ΔCAP-1 - 13.64Δr-2 -1.10 ΔDJ-2 +.56 ΔPROF-2 +12.39ΔXRAV0123 +.006ΔPOP (30) B) (2.6) (0.7) (1.1) (-0.2) (-2.7) (-1.9) (1.4) (4.0) (-0.5) R2=.91; DW =1.9 Total Consumption Borrowing Function (ΔCB) With “Average” Crowd Out ΔCB =.39Δ(Y-TG) +.42Δ(T -G) -9.28ΔPR -.91 ΔDJ-2+7.89 ΔXRAV +223.05ΔPOP16-.02ΔPOP+.88ΔICC-1+11.55ΔM2AV (31) (t =) (4.5) (1.8) (1.9) (-2.8) (4.3) (0.5) (-3.1) (1.1) (0.4) R2=63.2% D.W.= 1.7

The total estimated drop in consumer and business borrowing per dollar of tax cut induced deficit is $0.90, compared to the spending drop of $1.14. The confidence intervals around the borrowing estimates, are calculated from the estimated standard error of the total borrowing point estimate (.90): √(Var .48 +Var .42). = .30. The 5% level confidence intervals do not allow us to reject the hypothesis the drop in borrowing and spending are equal. This suggests crowd out of private borrowing is either the major or only monetary channel through which deficits negatively affects consumer and business spending.

Relationship of Borrowing to Crowd Out in Recession and Non-Recession Periods

Below are regression results for models of total consumer and business spending. We expect to later find some of them are also key determinants of borrowing. Total Investment Spending Function With Recession/Non-Recession Crowd Out ΔIT =+.57 Δ(T-G)Rec +.65 Δ(T-G)NonRec + .27ΔACC + .15ΔDEP + .84ΔCAP-1 - 6.46Δr-2 + .10 ΔDJ-2 +.31 ΔPROF-2 + 5.10ΔXRAV (32) (t =) (6.3) (3.5) (8.1) (0.5) (0.6) (-3.4) (0.5) (1.8) (6.2)

+ .01ΔPOP R2=.90.7 (3.9) DW =2.6

Total Consumption Spending Function With Recession/Non-Recession Crowd Out ΔCT =.50Δ(Y-TG) +.51Δ(T -G)Rec + .59Δ(T -G)NonRec - 10.06ΔPR+.58 ΔDJ-2+4.29 ΔXRAV - 379.52ΔPOP16 +.01ΔPOP (33) (t =) (15.7) (10.0) (7.0) (-5.9) (3.5) (5.5) (-2.0) (4.3)

+.50ΔICC-1+ 31.54ΔM2AV R2=96.3% (1.9) (4.4) D.W.= 2.2

and the borrowing functions for the same models are as follows: Total Investment Borrowing Function With Recession/Non-Recession Crowd Out ΔIB =+.13 Δ(T-G)Rec +.99 Δ(T-G)NonRec + .09ΔACC - .02ΔDEP + .65ΔCAP-1 - 10.08Δr-2 - .89 ΔDJ-2 +.51 ΔPROF-2 + 11.80ΔXRAV (34) (t =) (0.5) (3.3) (0.9) (-0.0) (0.2) (-2.0) (-1.8) (1.4) (4.5)

+ .00ΔPOP R2=.615 (0.0) DW =2.0

Total Consumption Borrowing Function With Recession/Non-Recession Crowd Out ΔCB=.41Δ(Y-TG) +.59Δ(T -G)Rec + .19Δ(T -G)NonRec - 10.44ΔPR-.85 ΔDJ-2+8.05 ΔXRAV + 320.12ΔPOP16 -.01ΔPOP (35) (t =) (4.4) (2.5) (0.8) (-2.0) (-2.6) (5.3) (0.6) (-2.5)

+1.16ΔICC-1+5.27.54ΔM2AV R2=65.1% (1.4) (0.2) D.W.= 1.8

In recession periods, the deficit’s effect on combined consumption and investment borrowing

compared to spending is somewhat similar ($0.72 vs. $1.08), but the spending decline is noticeably larger. The larger drop in private spending in recessions, about equal to the increase in deficit, suggests there may be factors not adequately controlled for in the model affecting spending on consumption and investment in recessions more severely than borrowing (or affects borrowing less severely). In non-recessions, the combined borrowing and spending effects are very close: ($1.18 vs. 1.24).

The relatively small negative impact deficits have on investment in recessions may be because loanable funds are freed up because consumer borrowing in recessions is reduced to a much greater degree ($0.59) than in non-recessions ($0.19).


Conclusion Crowd out affects borrowing and spending in recession and non recession periods. Deficits were

negatively related to spending and borrowing in both periods. The decline in borrowing explains virtually all the decline in spending associated with crowd out deficits in non-recessions. However, the decline in borrowing explains most of the decline in spending in recessions (2/3), but not all of it. The divergence suggests something else beside crowd out , not controlled for in the model, may be causing the additional spending decline. It is not clear what this might be, and warrants further research.

Evidence also shows from the 1981-83 recession period, that the reason crowd out also occurs in recessions, when loan demand by consumers and investors is relatively low, may be that national savings declines at least as much. Therefore, any recession deficit financed by borrowing, i.e., by drawing on remaining savings, will create a crowd out effect.

OTHER WAYS TO FINANCE DEFICITS Foreign Borrowing as a Means of Supplementing Domestic Savings

The NIPA accounts require public and private investment must always equal saving. That said, foreign saving borrowed by the U.S. government or U.S. private parties is included as part of the available savings pool. For example, during the 1981-83 recession period, domestic saving fell considerably more than investment, but was more than offset by the growth in foreign borrowing to compensate for lost domestic savings (Economic Report of the President, 2010, Table 32). Adjusting for the statistical discrepancy between savings and investment numbers, the actual figures (in billions) were

$ - 8.1 ∆(Private Investment) +16.0 ∆(Government Investment) = $ + 7.9 ∆(Investment) (36) $ + 38.4 ∆(Foreign Borrowing) - 30.5 ∆(Domestic Saving) = $ + 7.9 ∆(Savings) (37)

Hence, government deficits may not cause crowd out if the government can borrow foreign funds to finance the deficit, thereby avoiding tapping the domestic savings pool, or if private borrowers can borrow foreign funds to replace domestic savings used to pay for government deficits.

Financing Deficits by Monetary Expansion as a Means of Avoiding Crowd Out

In theory, deficits can be financed by monetary expansion as well as by borrowing. This should avoid the crowd out problem. In fact, (Heim 2010) showed that if we do not hold M2constant when examining the effect of spending deficits on consumption, we do not see an effect, i.e., there is evidence money expansion can offset crowd out. However, when tested directly, M1 did not seem to have this effect. Regression effects on variables in the consumption and IS models varied little using an (M2-M1) formulation for the money variable instead of (M2), i.e., controlling for M1 alone did not seem to make a difference. This suggests M2 affects the crowd out problem only through its non-M1 components, i.e., the parts that represent savings growth. Monetary expansion, to finance a deficit, may only offset crowd out effects if it is saved in advance of the deficit occurring and used to replace savings lost to financing the deficit. This was the formulation used in this study successfully, taken from the Heim (2010), where it is noted that without M2 controlled for, government spending deficits seem to have no effect on consumption. Only when (the savings component of) M2 is controlled for - held constant - do we see government spending deficits have a crowd out effect. SUMMARY AND CONCLUSIONS

This paper has attempted to bring the best science possible to bear on the issue of whether crowd out exists, and whether it occurs in recessions as well as non-recession periods. To do so, we have explicitly controlled extensively for variables (or reasonable proxies for them) other studies have found might affect consumption, investment and GDP. The test results indicate


• Crowd out is a statistically significant problem in both recessions and non-recession periods. • Declines in private borrowing were found to occur in the same period deficits increased.

Investment spending declined about as much as borrowing, consumption spending actually declined more. This is evidence crowd out is the underlying mechanism responsible for the negative relationship between private spending and deficits observed in this study and others.

• The IS curve regression tests showed crowd out completely offsets the stimulus effects of government spending deficits, resulting in the stimulus attempt having no net effect on the economy, though predicted IS curve results only suggested partial crowd out from government spending deficits.

• Crowd out more than completely offsets stimulus effects of tax cut deficits, resulting in the stimulus attempt having a net negative effect on the economy, consistent with predictions of crowd out theory and IS curve predictions resulting from actual consumption and investment function regression models.

• Point estimates from actual IS curve tests(eq. 28) indicate crowd out has a more negative effect in recessions than non recession periods, but the confidence intervals around the estimates are large enough that we cannot reject the hypothesis they are equal. If effects are more severe in recessions, this may be because national savings drops as much or more than private loan demand in recessions. It may leave only enough savings to cover reduced private borrowing needs. Any additional borrowing by government to finance a deficit will have crowd out effects.

• Foreign borrowing appears to be an alternative way of financing deficits which does not have domestic crowd out effects. Monetary expansion is more problematic, depending on a pre-deficit build up of savings that can be used to offset the decline in loanable funds associate d with the deficit.

REFERENCES Barley, R. (2010) “The Hidden Risks Governments Create by Delaying Cuts”. Wall Street Journal, February 24, p.C14. Blanchard, O. and Perotti, R. (2002) “An Empirical Characterization of the Dynamic Effects of Changes in Government Spending and Taxes on Output” The Quarterly Journal of Economics, 117, ( 4), pp. 1329-1368. Chan, S. (2010) “Group of 7’s Finance Ministers Stand By Stimulus Programs” NY Times, February 7, p.A16. ______. Economic Report of the President, 2002 and 2010. Washington, D.C. U.S. Government Printing Office. 2003, 2011. Furceri, D. and Sousa, R. (2009) “The Impact of Government Spending on the Private Sector: Crowding Out versus Crowding In Effects”. NIPE WP 6/2009, Nucleo De Investigaca Politicas Economicas, Universidade Do Minho, Portugal. ______. Flow of Funds Accounts of the United States, Data Series TODNS, TBSDODNS, CMDEBT, FGSDODNS, SLGSDODNS, AND DODFS. Washington, D.C. Board of Governors of the Federal Reserve System. March 11, 2011 Update. Available at www.federalreserve.gov/releases/z1/current/ Friedman, B. (1978). “Crowding Out or Crowding In? Economic Consequences of Financing Government Deficits”. Brookings Papers on Economic Activity, 1978, (3), pp. 593-641.


Gale, W.G. and Orszag, P.R. (2004). “Budget Deficits, National Saving, and Interest Rates”. Brookings Papers on Economic Activity, 2004, (2), pp. 101-187. Heim, John J. (2007) “How Much Does The Prime Interest Rate Affect U.S. Investment?” Journal of the Academy of Business and Economics, 7 (1),Oct. 2007 Heim, John J. (2010) “ Do Government Deficits Crowd Out Consumer and Investment Spending?” Journal of the Academy of Business and Economics, Vol. 10 (3). Also available by the same title as Working Paper # 1005, Department of Economics Working Paper Series, Rensselaer Polytechnic Institute, Troy, N.Y., July 2010. Hill, R.C., Griffiths, W., Lim, G. (2011) Principles of Econometrics. 4th ed. Hoboken, NJ: John Wiley & Sons. Krugman, P. (2009).“Crowding In” New York Times, September 28. Mountford, A. and Uhlig, H. (2008) “What Are The Effects Of Fiscal Policy Shocks?” NBER Working Paper # 14551, December. Spencer, R.W. and Yohe, W.P. (1970). “The Crowding Out of Private Expenditures by Fiscal Policy Actions” Federal Reserve Bank of St. Louis Review, 56(3).


Testing the Utility of Licensing: Evidence from a Field Experiment on Occupational Regulation

Dick M. Carpenter II

University of Colorado Colorado Springs

This study examines hypothesized benefits associated with occupational licensing in one long-regulated industry in Louisiana—floristry—in order to determine to what extent licensing results in theorized benefits that might justify the costs associated with licensure systems. Results indicate the regulation appears not to result in a statistically significant difference in quality of product. Moreover, florist-judges, whether licensed or not, appeared consistent in how they rated 50 floral arrangements from randomly chosen floral retailers. INTRODUCTION

This article examines the efficacy of a workplace phenomenon whose presence and importance continues to grow each year—occupational licensing (Timmermans, 2008). It is typically defined by state-granted legal privileges that enable practitioners of the regulated profession to hold competitors at bay, thwart off meddling third parties, and control who qualifies based on training, skill, and examination (Freidson, 1994). Kleiner (2000; 2006b) estimates that occupational licensing affects more workers in the U.S. than minimum wage legislation or unionization. Using a specially designed Gallup poll, Kleiner and Krueger (2010) find that 29% of the U.S. workforce must hold a license to work in their chosen occupation. This is up from 4.5% in the 1950s and about 20% in 2000 (Kleiner, 2006a).

As Potts (2009) and Mester et al. (2009) describe, occupational licensing is typically justified as benefiting the greater society as a “public good” or a “public welfare.” Legislative or state protection is given to occupations to protect the greater society against the possibility of rogue operatives, incompetents, quacks, charlatans, and others who might cause “public harm” through delivery of sub-standard or even dangerous standards of service. Licensing has been achieved principally through cooptation of government by the political activities of professional associations in individual states (Freidson, 1986; Halliday, 1987). Government officials typically accept such arguments with little question (Skarbeck, 2008), as Carpenter (2008) illustrated in a study of the evolution of regulation for interior designers, but does the justification have merit? More specifically, does licensing result in measurable differences in quality of service, product, or producer that might lead to significant benefits for society?

The answers to such questions carry important consequences. First is the relationship between occupations and the state. Freidson (2001) advocates for the legitimacy of licenses because they govern specialized knowledge that society values enough to want advanced and applied in socially useful ways. But occupational regulations also grant protected practitioners extraordinary authority over laypersons


(Abel, 1979) through influence over the social structure that defines and regulates the environment in which that work is accomplished (Peterson, 2001).

Second are the anti-competitive effects of occupational regulation in the market. As discussed below, some evidence indicates licensing results in greater costs to consumers and disproportionately fences out aspiring practitioners who are racial/ethnic minorities, women, and/or older. Moreover, although licensing is commonly associated with health or legal professions, Summers (2007) notes that occupational regulations also abide for a broad range of occupations ideally suited for workers with fewer technical skills, less education, or those just entering or re-entering the job market. Such occupations include, among others, upholsterer, manure applicator, motion picture projectionist, interior designer, casket seller, African hairbraider, and—the subject of this research—florist.

In the face of such effects and costs, evidence of benefits from licensing—particularly those associated with less technical occupations—takes on great importance. Yet, according to Kleiner (2000), such evidence is thin or non-existent in many occupations. Thus, to answer the questions posed above—specifically whether licensing results in measurable differences in quality of service, product, or producer in non-technical occupations—I implemented a randomized experiment designed to measure whether a license for florists created a difference in products and producers that might result in some benefit to consumers. LITERATURE REVIEW

Occupational licensing is the monopolization or control of opportunities for income and of status and work privileges in a market of services or labor (Brain, 1991; Johnson, 1972; Larson, 1977; Weisz, 1983). According to Abbott (1988) and others (Abel, 1979; Brain, 1991; Larson, 1977), such control often results from an assertion and/or recognition of a body of abstract or specialized knowledge held by the practitioners of a particular occupation. Commonly, such assertions by those working in an occupation are accompanied by calls for regulation on behalf of the public interest (Peterson, 2001). Sometimes licenses are described as a way to protect public welfare by ensuring those working in a regulated occupation possess a minimum facility with the abstract or specialized knowledge (Carpenter, 2008). Others justify licensing because it governs a special knowledge that society values enough to want advanced and applied in socially useful ways. According to proponents, the carriers of that special knowledge therefore deserve to be sheltered from market laws (Freidson, 2001).

In general, there are two divergent views on the effects of occupational licensing. One view, commonly attributed to Milton Friedman (1962), argues that licensing is primarily a means for professionals to keep wages high by restricting entry into the profession. Adherents to this view believe licensure reduces consumer welfare (Potts, 2009). A second view concedes that occupational licensing increases the wages of professionals, but argues that licensing serves as a means of solving an asymmetric information problem. Consumers have less information than practitioners, and licensing protects consumers from poor service (Leland, 1979; Shapiro, 1986).

Such effects have been studied in different countries across a variety of occupations, such as nurses (Elzinga, 1990), veterinarians (Hellberg, 1990), lawyers (Karpik, 1990), architects, psychologists (Svensson, 1990), and interior designers (Harrington & Treber, 2009). One prominent strand of research focuses on licensing as a barrier to entry. Studies on occupations such as cosmetologists (Adams, Jackson, & Ekelund, 2002), manicurists (Federman, Harrington, & Krynski, 2006), accountants (Carpenter & Stephenson, 2006; Jackson, 2006; Jacob & Murray, 2006), and mortgage brokers (Kleiner & Todd, 2007) conclude that licensing reduces the supply of practitioners, which proponents of regulation support, as it supposedly prevents entry by low-quality producers. A subset of this research finds that licenses often disproportionately exclude those who are less educated, racial/ethnic minorities, or older (Angrist & Guryan, 2008; Dorsey, 1983; Federman, et al., 2006; Harrington & Treber, 2009; Kleiner & Krueger, 2008).

A second line of research examines the relationship between licensure and wages. In general, results from studies of occupations such as radiologic technologists (Timmons & Thornton, 2008), school


teachers (Angrist & Guryan, 2008), clinical lab personnel (White, 1978), nurses (White, 1980), dentists (Kleiner & Kudrle, 2000) and cosmetologists (Adams, et al., 2002) indicate licensing increases the wages of practitioners. Others, however, find no evidence of such a relationship between licensing and wages (Lueck, Olsen, & Ransom, 1995).

To the extent that licenses result in greater costs to consumers as reflected in higher wages or extracted rents (Chevalier & Morton, 2008; Harrington & Krynski, 2002; Kleiner & Kudrle, 2000), a third line of research considers whether such costs are offset by greater benefits in the form of increased quality of product, service, or producer. Many studies find little evidence of the purported benefits. Such conclusions come from research on a diversity of occupations, such as school teachers (Angrist & Guryan, 2008; Buddin & Zamarro, 2008; Kleiner & Petree, 1988), interior designers (Carpenter, 2008), construction trades (Skarbek, 2008), mortgage brokers (Kleiner & Todd, 2007), dentists (Kleiner & Kudrle, 2000), physicians (Paul, 1984), and others (Carroll & Gaston, 1981).

Such findings do not mean, however, that the question of benefits from licensing is settled, particularly in non-technical occupations. Some researchers do, in fact, find a positive relationship between occupational regulation and quality of service (Johnson & Loucks, 1986; Shilling & Sirmans, 1988). Scholarly proponents make the case for benefits from regulation (Freidson, 2001), and the creation of new licenses continues a pace (Kleiner & Krueger, 2010). Finally, when advocating for new regulations or defending existing ones, industry leaders emphasize benefits to public health, safety, and welfare as a justification (Carpenter, 2008). Thus, in light of the aforementioned consequences of licensing, the question of purported benefits remains an important research endeavor. To that end, this study uses a field experiment to test the claim of public benefit in an occupation long regulated in Louisiana—florists. Study Context

The state of Louisiana has regulated the work of florists since 1936. Senate Bill 278 first regulated the industry through the creation of an oversight commission, which was given the authority to create a license, write regulations, and provide penalties for violations of the enabling act (Legislature of the State of Louisiana, 1936). Beginning in 1950, various revisions to the bill continued to increase the regulatory requirements associated with a retail florist license. By the time of the experiment reported herein, aspiring florists in Louisiana were required to pay a series of fees and pass both a written and a practical test. The written test covered basic knowledge of the industry, and the practical test required test takers to create four different types of floral arrangements within a limited period of time given a bundle of floral materials. The arrangements were judged or scored by already-licensed Louisiana florists.

As data released by the Louisiana Horticulture Commission in 2004 indicate, these tests were more than perfunctory hurdles. In 2002, only 42% of test takers passed and were awarded licenses. That number grew to 47% in 2003, but fell to 41% in 2004. Thus, the tests represented significant barriers to those who sought to work as florists in Louisiana. Moreover, data from the state also indicate licensure regulations were strictly enforced. The Horticulture Commission employed inspectors who regularly visited stores to examine, among other things, the licenses of store employees. Violations of the regulation resulted in cease and desist letters, confiscation or destruction of floral materials, and prohibitions against further work or employment.

The original motivation behind the bill’s creation and its various revisions are largely undocumented. Newspapers and legislative journals of the time included only the sparest of information. The introduction to the 1936 enabling legislation provided the clearest justification for the creation of the license: “to prevent fraudulent practices in the professional work herein defined” (Legislature of the State of Louisiana, 1936, p. 404). Apparently legislators agreed; the bill passed unanimously in both chambers (Senate of the State of Louisiana, 1936). However, contemporary proponents of the florist license have spoken often about its public benefits. Turner (2001) notes that the process of justifying a license is not finished at its creation. Rather, proponents continuously legitimate themselves to the public and other audiences. In Louisiana during the past decade, this has often taken the form of defending the license by pointing to the regulation’s benefits, particularly the maintenance of quality standards. For example, in


response to a 2004 bill to repeal the florist license (the bill was ultimately unsuccessful), the head of the state florist association argued that the licensure regime protected consumers through upholding high professional standards. The head of the state horticulture commission agreed: “If they [aspiring florists] can’t take the instruction and pass the exam, how can they do an arrangement that you and I want to buy” (Finch, 2004, p. 1).

If such justifications are correct, it should be possible to discern measurable differences between the quality of providers and producers in the Louisiana retail florist industry and those in another state as a result of the license. That is, if the licensure regulation works as intended, retail florists who work in Louisiana should produce floral arrangements of significantly greater quality, and said florists should also possess skills and knowledge at least different from or perhaps superior to practitioners in other states without a license for florists. If true, this might confirm, at least in some part, the justification of florist license proponents. The research reported herein puts such claims to the test. In an even broader context, this article also contributes to the aforementioned literature that attempts to measure the “public benefits” notion behind the idea of occupational licensing. While the study of one occupation in one U.S. state may seem limited, its importance lays in its findings vis-à-vis the collection of extant studies on licensure. Given the scope of occupational regulation, no one study can possibly prove conclusive. But as this small but growing body of literature tests the “public benefits” claims across a diverse collection of occupations, it provides researchers, policy leaders, and decision makers important evidence about the fundamental theories and claims associated with licensure and could provide guidance when policy makers consider the creation of new licensing regulations or the elimination of existing ones. However, because this body of work is small, particularly compared to the large number of licensed occupations in the U.S. (Summers, 2007), more studies like this are necessary to facilitate a fulsome understanding of the effects of licensing. METHODS

This study is guided by two primary questions: 1. Is there a significant difference in the ratings of floral arrangements between those created by

licensed florists and those that are not? 2. What is the relationship in ratings assigned to floral arrangements between licensed and

unlicensed florists? The first question tests the theory that licensing results in measurable differences in quality of goods.

If true, floral arrangements created by licensed florists should receive higher ratings than those created by unlicensed practitioners. The second question tests the theory that licensing results in measurable distinctions in skill and knowledge among occupational practitioners. Note that this question focuses on the relationship—specifically the consistency—of licensed and unlicensed florists in how they rate a bundle of goods (i.e., floral arrangements). According to the relevant sociology literature, standardization is a distinct element in licensure (Abel, 1979; Larson, 1977; Zukin, 1998). Through training requirements, testing regimes, professional standards, and other regulatory devices, producers (i.e., industry practitioners) purportedly grow more standardized within the regulated occupation and differentiated from those kept out. Applied specifically to this study, the ratings of a group of florists composed of licensed and unlicensed florists would be expected to be inconsistent, as the opinions of what constitutes quality in a product would differ for licensed practitioners adhering to a particular standard as compared to unlicensed (and unstandardized) florists. Design and Sample

To answer these questions, this research uses an experimental design in which a randomly selected sample of florists from two states (one with a license for florists and one without) rated a sample of floral arrangements created by randomly selected retail florists, half of which were in the state with the license and the other half from the state without licensure. The random sample of florist-judges in the regulated state—Louisiana—was drawn from a list of licensed florists maintained by Louisiana’s Horticulture


Commission. The sample of unlicensed florists came from Texas. These names were drawn from an online directory of retail florist establishments. The total sample size of florists was 18—eight from Texas and 10 from Louisiana. The sample of floral arrangements came from randomly drawn retail floral stores in Texas and Louisiana. The stores were taken from the same lists used to identify the florists. The sample included 50 arrangements—25 from each state. Arrangements and florist-judges did not come from the same retail floral stores. Data

The primary data used in this analysis represent the ratings assigned by the sample of florists to the floral arrangements. The florist-judges used a rating instrument developed by the researcher. The instrument was based on (a) floral design theory as captured in several prominent floral design texts frequently used in design courses (Belcher, 1993; DelPrince, 2005; Hunter, 2000; Lamancusa, 1990; Pryke, 2006), (b) floral design rating instruments used in national floral design competitions (http://www.safnow.org/images/stories/About_SAF/PFCI/SC_2009/2009safsylviacupentrypacket.pdf, http://www.newjersey.gov/agriculture/ag_ed/ffa/activity/13.003.pdf, http://www.okcareertech.org/skills usa/docs/job_readiness_contests/FLORAL%20DESIGN%20CONTEST%20GUIDELINES.pdf), and (c) the design evaluation for certified floral designers created by the American Institute of Floral Designers (n. d.). The instrument used herein prompted judges to rate the arrangements in 10 different design elements, each using a five point scale ranging from poor (1) to excellent (5). Total scores were then calculated for each arrangement by summing across the 10 elements. Thus, each judge rated each of the 50 arrangements, assigning each arrangement a score of something between 10 and 50. These scores were used in the analyses described below.

The analyses also used one control variable: the cost of the arrangements. Cost of arrangement data reflect the number of dollars paid, as indicated on each receipt. As described below, I asked for arrangements within a certain price range, but discovered upon receipt of the flowers that some stores provided arrangements outside of the range. This required that I control for the possible effects created by price disparity. Procedures

The experiment occurred over a two-day period in January 2010 in Shreveport, Louisiana and Longview, Texas. These locations were chosen because of their geographic proximity to one another across the Louisiana state line (i.e., a one hour drive). All arrangements—25 from Louisiana and 25 from Texas—were purchased from stores in or around the two cities. Likewise, all florist-judges were recruited from in or around the respective cities. The floral retailers from whom the arrangements were purchased had no idea that their designs would be judged by other florists, which means the arrangements represented typical products purchased by consumers on any given day. All retailers were given a “theme” for the arrangements—sympathy—and some general parameters in which to work: Arrangements were to be within $50 to $75 and the arrangements were to be in a basket, vase, or bowl. Otherwise they were free to be creative in designing the arrangements, which they preferred. Almost to a person, the florists asked if I wanted arrangements displayed in online catalogues through FTD, Teleflora, or other nationwide services. When I said “no,” they were pleased, since they much preferred to create their own designs using seasonal flowers or materials they had on site or could purchase easily from a local wholesaler. This is important to note, as it ameliorates concerns of lack of variation based on state. The floral judges were recruited by mail, telephone, or email, depending on the contact information provided in the respective lists. They were asked to participate as judges in a floral design experiment similar to a floral competition, but they were not told that the primary intent was to measure differences based on licensure. Judges were each paid $200 for participation.

The floral arrangements were rated in three separate sessions. The first session occurred in Longview, and the second and third occurred in Shreveport (one in the morning and one in the afternoon). All sessions were held in hotel conference rooms. The experiment looked much like a floral design competition. The 50 arrangements were randomly ordered on tables with identification numbers. Florist-


judges were given the rating sheets and asked to score all 50 arrangements based on the printed criteria, such as proportion, balance, color, form, and workmanship. Judges did not know any arrangement’s state of origin or even that arrangements came from different states. Judges could take as much or as little time as they wished to rate the arrangements, up to three hours. After the completion of the rating period, judges participated in a focus group debrief during which they were asked about their perceptions of the arrangements, opinions of quality based on state of origin, and attitudes about licensure. Analyses

For the first research question, data were analyzed in a repeated measures design using hierarchical multivariate linear modeling (HMLM; Stephen Raudenbush, Bryk, & Congdon, 2007). This type of model is typically used when time represents the repeated measure. However, according to Raudenbush, Bryk, Cheong, Congdon, and Toit (2004), this model can also be used when the same subject (a floral arrangement in this case) is exposed to different conditions (different judges herein). Thus, the model takes the following form:

FIGURE 1 HMLM EQUATION

Level-1 Model

Rating = (Judge 1)(Score 1) + (Judge 2)(Score 2) + (Judge 3)(Score 3) + (Judge 4)(Score 4) + (Judge

5)(Score 5) + (Judge 6)(Score 6) + (Judge 7)(Score 7) + (Judge 8)(Score 8) + (Judge 9)(Score 9) + (Judge 10)(Score 10) + (Judge 11)(Score 11) + (Judge 12)(Score 12) + (Judge 13)(Score 13) + (Judge 14)(Score

14) + (Judge 15)(Score 15) + (Judge 16)(Score 16) + (Judge 17)(Score 17) + (Judge 18)(Score 18) Rating = π0 + ε

Level-2 Model

π0 = β00 + β01(ArrState) + β02(Cost)

where, rating = The scores assigned to the floral arrangements Judge 1…Judge 18 = Indicator of the judge assigning a score Score 1…Score 18 = Score of the respective judge π0 = Intercept at level 1 ε = Error term at level 1 β00 = Intercept at level 2 β01 = Effect of arrangement state of origin β02 = Effect of cost ArrState = Each arrangement’s state of origin Cost = The cost of the arrangement

In this analysis, the ArrState was coded using effects coding: Louisiana = 1, Texas = -1. The cost

variable was mean centered. The intercept then takes on the value of the grand mean, and coefficients indicate deviations from the grand mean.

I also compared three different models (an unrestricted model, a homogeneous level-1 variance model, and a heterogeneous level-1 variance model) and concluded that the unrestricted model better fit the data than homogeneous (χ2 = 466.16, p < 0.000) and heterogeneous models (χ2 = 399.52, p < 0.000). For the second research question, data were analyzed using inter-rater reliability. Specifically, the consistency of judges in rating the arrangements was analyzed using intraclass correlation (ICC) with a two-way random effects model. Three sets of ICCs were run. The first examined the consistency of all judges across all arrangements, only Texas arrangements, and only Louisiana arrangements (i.e., those


created by a licensed florist). The second set considered the consistency of Louisiana (i.e., licensed) judges on all arrangements, just Louisiana arrangements, and only Texas arrangements. The third set examined the consistency of Texas judges on all arrangements, only Louisiana arrangements, and then only Texas arrangements.

Finally, focus group data were analyzed using standard qualitative coding procedures (Miles & Huberman, 1994; Straus & Corbin, 1998). Each focus group lasted between 45 and 60 minutes and was facilitated by the author. A research assistant maintained a transcript of the focus groups, and the author took detailed notes. The transcript and notes were compared to provide a reliability check. The analysis of the notes and transcript used a constant-comparative method, which produced an index of codes organized into the themes discussed below. RESULTS 1. Is there a significant difference in the ratings of floral arrangements between those created by licensed florists and those that are not?

Table 1 includes descriptive statistics. For all arrangements, the mean rating was essentially 29 points (on a 50-point scale). The average arrangement cost almost $60, which was within the price range I set for the designers. When disaggregated by state, results indicate Louisiana arrangements had slightly higher ratings—30.42 points compared to 27.56 in Texas. Louisiana arrangements also came with greater price tags—$67.11 on average compared to $52.84 in Texas. Although the Texas mean was within the specified price range, four arrangements cost less than $50. Thus, controlling for cost was potentially particularly important.

TABLE 1 DESCRIPTIVE STATISTICS FOR RATINGS AND ARRANGEMENT COSTS

N M SD All Arrangements

Rating 900 28.99 9.61 Cost 50 59.98 8.27

TX Arrangements Rating 450 27.56 9.40 Cost 25 52.84 5.11

LA Arrangements Rating 450 30.42 9.61 Cost 25 67.11 2.96

The HMLM results in Table 2 show that although cost included some variability outside the

requested range, its effect is not significant. The variable of primary interest—arrangement state of origin—was also not statistically significant. After controlling for cost, there appears to be no difference in ratings between arrangements that come from Louisiana, where the floral license has been in existence in some form since 1936, and Texas, with no license.

TABLE 2 HMLM RESULTS

β se t p Intercept 27.60 1.99 13.85 0.00 Arrangement state of origin -0.02 0.96 -0.03 0.98 Cost 0.13 0.12 1.14 0.26


2. What is the relationship in ratings assigned to floral arrangements between licensed and unlicensed florists?

The HMLM analysis examined differences in ratings based on state of origin, but the inter-rater reliability analysis provides a different perspective—the consistency of licensed versus unlicensed judges. As Table 3 illustrates, whether it was all judges or judges grouped by state, inter-rater reliability was strong or quite strong across the different arrangements, indicating strong consistency among judges in how they rated the arrangements. More specifically, whether the judges were licensed or not, they gave the arrangements similar scores; thus, it appears that Louisiana’s licensing law does not produce more discriminating florists.

TABLE 3 INTRACLASS CORRELATION COEFFICIENTS

ICC Coefficient F p All Judges, All Arrangements 0.91 11.72 0.00 All Judges, Louisiana Arrangements 0.88 8.45 0.00 All Judges, Texas Arrangements 0.93 14.89 0.00 Louisiana Judges, All Arrangements 0.87 7.80 0.00 Louisiana Judges, Louisiana Arrangements 0.85 6.46 0.00 Louisiana Judges, Texas Arrangements 0.89 9.48 0.00 Texas Judges, All Arrangements 0.83 5.91 0.00 Texas Judges, Louisiana Arrangements 0.75 3.99 0.00 Texas Judges, Texas Arrangements 0.86 7.17 0.00

Focus Group Results

In the focus groups, florists were first asked a few questions about the specific arrangements they rated, with the intention of “warming up” the participants and also masking the questions of primary interest—those related to licensure. When the focus group discussion turned to the latter topic, the judges’ comments were consonant with the quantitative findings presented above. As a reminder, the judges had no idea which arrangements came from which state or that the arrangements even came from different states. And before each focus group, I calculated the mean difference in ratings based on state as a way to present the participants with some tentative results and frame a question about licensing.

When I revealed the fact that the arrangements came from different states and the differences in ratings based on state were small, the judges did not appear at all surprised. The first dominant theme from the focus groups was that good florists could be found anywhere, not just in a particular location. The judges overwhelmingly agreed that quality of floral design depends on the person rather than the state or region. This view held even after it was pointed out that Louisiana requires a license and Texas does not. Again, almost all of the judges—including the licensed florists from Louisiana—expected no difference in the quality of arrangements because of Louisiana’s regulation. They noted that quality of work was a function of the standards set by individual businesses rather than a licensing regime, and those standards themselves were a function of consumer demand and market competition. As one florist commented, “If you don’t do good work, you’re not going to have any business.”

The second dominant theme was the low esteem in which the florist license is held among Louisiana florists. Most thought it provided little of value. All of them derided the practical test as outdated and irrelevant, since aspiring florists were tested on skills and knowledge that bear no resemblance to contemporary floristry. Instead, the practical test required test-takers to demonstrate proficiency with techniques that were abandoned decades ago. Moreover, the Louisiana florists dismissed the floral designs required on the test as “old school” and “ugly”—nothing a contemporary consumer would ever demand or want.


A third theme was the purposes florists ascribe to licensure. Many in the focus groups thought that instead of producing quality florists, the licensing requirement served two purposes—raising money for the state and shutting out competition. The license generates revenue from florists or aspirants through testing fees and annual licensure renewal. At the time of this data collection, test-takers paid the state of Louisiana $150 each time they took the full test and $100 each time they re-took the design portion. Annual licensure renewals cost $75.

The second purpose, shutting out competition, was frequently mentioned by focus group participants. Most referred to it negatively, but a few supported the idea of excluding competition from “amateurs”—those who work out of their homes—or “freelancers”—people who buy flowers from the same wholesalers but create and sell arrangements through means other than a traditional florist shop or store. Despite these few, however, most disparaged the anti-competitive effects of licenses. One even shared how a competing floral shop across the street from her store repeatedly called the Louisiana Horticulture Commission, the agency responsible for administering the license, with trumped up complaints as a way to damage her business.

Because participants at no time ascribed any of the purported purposes to the license, I represented the views of licensing proponents by suggesting licensure was a beneficial way to protect the public from poor quality or unsafe florists or floral products. This was an idea representatives of the Louisiana Horticulture Commission and the state’s Agriculture Department repeatedly have asserted in defense of their license (Notes and quotes, 2007; United States District Court, 2004a, 2004b). The florist-judges were not convinced, however. They maintained that licensure served the aforementioned two purposes. In fact, the public health and safety suggestion drew skeptical laughter. As one Louisiana judge concluded, “You really can’t hurt anybody with a flower.” DISCUSSION AND CONCLUSION

This study examined some hypothesized benefits associated with an occupational license in one long-regulated industry in Louisiana—floristry. In so doing, this article adds to the small but growing literature that has considered a number of occupations with licenses as a way to determine to what extent licensing regulations result in theorized benefits that might justify the effects and costs associated with licensure systems. Results indicate the license appears not to result in a statistically significant difference in quality of product, nor do licensed practitioners appear to differ from unlicensed practitioners in how they rate a sample of products in their industry. Instead, florist-judges, whether or not they worked under a license, appeared quite consistent in how they rated 50 floral arrangements from randomly chosen floral retailers.

Such results add further support to earlier studies that find little evidence of purported benefits from licensing (Angrist & Guryan, 2008; Buddin & Zamarro, 2008; Carpenter, 2008; Carroll & Gaston, 1981; Kleiner & Kudrle, 2000; Kleiner & Petree, 1988; Kleiner & Todd, 2007; Paul, 1984; Skarbek, 2008). Given recent estimates that licenses increase wages as much as 15% (Kleiner & Krueger, 2010), which are typically borne by consumers in higher costs, this body of research calls into question whether society truly benefits from licenses in light of real costs. The costs are not limited only to the consuming public; they can also be borne by potential workers. By design, licenses fence out would-be practitioners who fail testing regimes or lack the wherewithal to complete the training, educational, or apprenticeship requirements (Kleiner, 2006b). Although using the power of the state to fence out unqualified aspirants may seem reasonable in some professions (physicians, for example) where important knowledge is abstract and asymmetric (Akerlof, 1970; Elliott, 1972) and in which a consumer’s choices are higher stakes, it seems less so in non-technical occupations, like floristry, where abstract or asymmetric knowledge is not prevalent, where consumers’ choices are lower stakes, and/or where there is little to no evidence of significant public benefit from the license, such as this article demonstrates. As a result, non-technical occupations that would be ideal for those entering or re-entering the employment or entrepreneurial sector are fenced off through regulation. Using the “economic ladder” metaphor discussed by Williams (1982), licenses eliminate the first rungs, making it unnecessarily more difficult to mount the economic ladder.


This is not to say that society receives no benefit or finds no value in “signals” of quality and competence commonly associated with licensing (Spence, 1973). Rather, the question is whether such signaling requires state sanctioned licenses that come with real costs instead of other forms of signaling that come without such costs. The latter is commonly referred to as “market transparency” (Friedman, 1962; Haas, 2008; Stigler, 1971) where consumers receive signals through traditional vehicles likes warranties or brand names (Akerlof, 1970). Other signaling occurs through third party consumer organizations, such as the Better Business Bureau (Klein, 2001; Skarbek, 2008), and more contemporary versions built on new information and communication technologies (Potts, 2009), such as Angies list (www.anglieslist.com).

There is also credentialing through private or nonprofit professional associations that grant credentials to practitioners who successfully demonstrate the requisite knowledge, skills, and/or education (Freidson, 1986). Examples include ASE certification for automobile mechanics, CTC designation for travel agents, or CFP appellation for financial planners. In this scheme, a private non-profit industry body oversees the process and grants the certificate. Certification is not mandatory; therefore a non-certified practitioner also may provide similar services. However, given that certification indicates the achievement of a certain level of skill, some consumers might pay a premium for using a certified practitioner as opposed to a non-certified one as a way to receive some assurance of higher quality products or services.

Specific to florists, multiple designations currently exist. At the national level, the American Institute of Floral Designers offers certification that requires a prescribed educational background, a written test, and a practical test to earn the title of Certified Floral Designer (American Institute of Floral Designers, n. d.). State floral associations also offer similar designations. For example, Texas—one of the states included in the experiment reported herein—has a professional association that offers a “master florist” title. This certification requires the completion of a series of training workshops and a multi-stage testing regime before certification (http://www.tsfa.org/default.aspx?p=TMFInformation). Interestingly, the Louisiana State Florist Association offers its own “master florist” designation. This certification could serve the same signaling function to prospective florist employers and consumers without the aforementioned costs associated with licensure.

In a postscript to this research, that certification program may become more active. Despite the assertion by Potts (2009) that occupational deregulation almost never occurs, in June 2010 (six months after the data in this article were gathered), Governor Jindal signed a law stripping the florist license of its practical test, leaving in place the fees associated with the license and the written test (Scott, 2010a). Louisiana’s legislature passed the bill after a lawsuit, claiming an unconstitutional infringement upon the right to work, was filed against the state (Scott, 2010b). Because of that legislation, the lawsuit was dropped, and Louisiana continues to license florists. REFERENCES Abbott, A. (1988). The system of professions. Chicago: University of Chicago Press. Abel, R. L. (1979). The rise of professionalism. British Journal of Law and Society, 6(1), 82-98. Adams, A. F., Jackson, J. D., & Ekelund, R. B. (2002). Occupational licensing of a credence good: The regulation of midwifery. Southern Economic Journal, 69(3), 659-675. Akerlof, G. A. (1970). The market for lemons: Qualitative uncertainty and the market mechanism. Quarterly Journal of Economics, 84(3), 488-500. American Institute of Floral Designers. (n. d.). Certified floral designer. Retrieved February 5, 2010, from http://www.aifd.org/NewMembership/Accreditation.htm.


Angrist, J. D., & Guryan, J. (2008). Does teacher testing raise teacher quality? Evidence from state certification requirements. Economics of Education Review, 27, 483–503. Belcher, B. (1993). Creative flower arranging: Floral design for home and flower show. Portland, OR: Timber Press. Brain, D. (1991). Practical knowledge and occupational control: The professionalization of architecture in the United States. Sociological Forum, 6, 239-268. Buddin, R., & Zamarro, G. (2008). Teacher qualifications and student achievement in urban elementary schools. Journal of Urban Economics, 66, 103–115. Carpenter, C. G., & Stephenson, E. F. (2006). The 150-hour rule as a barrier to entering public accountancy. Journal of Labor Research, 27(1), 115-126. Carpenter, D. M. (2008). Regulation through titling laws: A case study of occupational regulation. Regulation and Governance, 2(3), 340-359. Carroll, S. L., & Gaston, R. J. (1981). Occupational restrictions and the quality of service received: Some evidence. Southern Economic Journal, 47(4), 959-976. Chevalier, J. A., & Morton, F. M. S. (2008). State casket sales restrictions: A pointless undertaking? Journal of Law and Economics, 51, 1-23. DelPrince, J. M. (2005). AIFD Guide to Floral Design terms, techniques, and traditions. Baltimore, MD: American Institute of Floral Designers. Dorsey, S. (1983). Occupational licensing and minorities. Law and Human Behavior, 7(2-3), 171-181. Elliott, P. (1972). The sociology of the professions. London: Macmillan. Elzinga, A. (1990). The knowledge aspect of professionalization: The case of science-based nursing education in Sweden. In R. Torstendahl & M. Burrage (Eds.), The formation of professions (pp. 151-173). London: Sage. Federman, M. N., Harrington, D. E., & Krynski, K. J. (2006). The impact of state licensing regulations on low-skilled immigrants: The case of Vietnamese manicurists. American Economic Review, 96(2), 237-241. Finch, S. (2004, January 5). Florist licensing under attack; Lawsuit says test is unconstitutional, New Orleans Times-Picayune, p. 1. Freidson, E. (1986). Professional powers. Chicago: University of Chicago Press. Freidson, E. (1994). Professionalism reborn: Theory, prophecy, and policy. Chicago: University of Chicago Press. Freidson, E. (2001). Professionalism: The third logic. Chicago: University of Chicago Press. Friedman, M. (1962). Capitalism and freedom. Chicago: University of Chicago Press.


Haas, J. (2008). Occupational licensing versus company-led training: The controversy over the competence assurance system for European aircraft technicians. European Societies, 10(4), 597-617. Halliday, T. C. (1987). Beyond monopoly. Chicago: University of Chicago Press. Harrington, D. E., & Krynski, K. J. (2002). The effect of state funeral regulations on cremation rates: Testing for demand inducement in funeral markets. Journal of Law & Economics, 45, 199-223. Harrington, D. E., & Treber, J. (2009). Designed to exclude. Arlington, VA: Institute for Justice. Hellberg, I. (1990). The Swedish veterinary profession and the Swedish state. In R. Torstendahl & M. Burrage (Eds.), The formation of professions (pp. 174-185). London: Sage. Hunter, N. T. (2000). Art of Floral Design (2nd ed.). Albany, NY: Delmar. Jackson, R. E. (2006). Post-graduate educational requirements and entry into the CPA profession. Journal of Labor Research, 27(1), 101-114. Jacob, J., & Murray, D. (2006). Supply-side effects of the 150-hour educational requirement to CPA licensure. Journal of Regulatory Economics, 30(2), 159-178. Johnson, L. L., & Loucks, C. (1986). The effects of state licensing regulations on the real estate brokerage industry. Real Estate Economics, 14(4), 567-582. Johnson, T. (1972). Professions and power. London: MacMillan. Karpik, L. (1990). Technical and political knowledge: The relationship of lawyers and other legal professions to the market and the state. In R. Torstendahl & M. Burrage (Eds.), The formation of professions (pp. 186-197). London: Sage. Klein, D. B. (2001). The demand for and supply of assurance. Economic Affairs, 21(1), 4-11. Kleiner, M. M. (2000). Occupational licensing. Journal of Economic Perspectives, 14(4), 189-202. Kleiner, M. M. (2006a). A license for protection. Regulation, 29(3), 17-21. Kleiner, M. M. (2006b). Licensing occupations: Ensuring quality or restricting competition. Kalamazoo, MI: Upjohn Institute. Kleiner, M. M., & Krueger, A. B. (2008). The prevalence and effects of occupational licensing. Cambridge, MA: National Bureau of Economic Research. Kleiner, M. M., & Krueger, A. B. (2010). The prevalence and effects of occupational licensing. British Journal of Industrial Relations, 48(4), 676–687. Kleiner, M. M., & Kudrle, R. T. (2000). Does regulation affect economic outcomes: The case of dentistry. Journal of Law and Economics, 43(2), 547-582. Kleiner, M. M., & Petree, D. L. (1988). Unionizing and licensing of public school teachers: Impact on wages and educational output. In R. B. Freeman & C. Ichniowski (Eds.), When public sector workers unionize (pp. 305-319). Chicago: University of Chicago Press.


Kleiner, M. M., & Todd, R. M. (2007). Mortgage broker regulations that matter: Analyzing earnings, employment, and outcomes for consumers. Cambridge, MA: National Bureau of Economic Research. Lamancusa, K. (1990). Floral design. Blue Ridge Summit, PA: Tab. Larson, M. S. (1977). The rise of professionalism: A sociological analysis. Berkeley, CA: University of California Press. Legislature of the State of Louisiana. (1936). Acts passed by the legislature of the state of Louisiana at the regular session. Baton Rouge, LA: State of Louisiana. Leland, H. E. (1979). Quacks, lemons, and licensing: A theory of minimum quality standards. Journal of Political Economy, 87(6), 1328-1346. Lueck, D., Olsen, R., & Ransom, M. (1995). Market and regulatory forces in the pricing of legal services. Journal of Regulatory Economics, 7(1), 63-83. Mester, J. L., Trepanier, A. M., Harper, C. E., Rozek, L. S., Yashar, B. M., & Uhlmann, W. R. (2009). Perceptions of licensure: A survey of Michigan genetic counselors. Journal of Genetic Counseling, 18, 357–365. Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage. Notes and quotes. (2007, June 6). Notes and quotes from the Louisiana Legislature, Associated Press State & Local Wire, npn. Paul, C. (1984). Physician licensure legislation and the quality of medical care. Atlantic Economic Journal, 12(4), 18-30. Peterson, M. A. (2001). From trust to political power: Interest groups, public choice, and health care. Journal of Health Politics, Policy, and Law, 26(5), 1146-1163. Potts, J. (2009). Open occupations--why work should be free. Economic Affairs 29(1), 71-76. Pryke, P. (2006). Flower school: Mastering the art of floral design. New York: Rizzoli. Raudenbush, S., Bryk, A., Cheong, Y. F., Congdon, R., & Toit, M. (2004). HLM 6. Lincolnwood, IL: Scientific Software International. Raudenbush, S., Bryk, A., & Congdon, R. (2007). HLM for windows (Version 6.04). Lincolnwood, IL: SSI Scientific Software International. Scott, R. T. (2010a, June 16). Governor poised to pluck out florists' flower-arranging test, New Orleans Times-Picayune p. A01. Scott, R. T. (2010b, March 4). Lawsuit attempts to kill state's florist licensing law; La. is only one with the requirement, New Orleans Times-Picayune, p. A03. Senate of the State of Louisiana. (1936). Official journal of the Senate of Louisiana. Baton Rouge, LA: State of Louisiana.


Shapiro, C. (1986). Investment, moral hazard, and occupational licensing. Review of Economic Studies, 53(5), 843-862. Shilling, J. D., & Sirmans, C. F. (1988). The effects of occupational licensing in complaints against real estate agents. Journal of Real Estate Research, 3(2), 1-9. Skarbek, D. (2008). Occupational licensing and asymmetric information: Post-hurricane evidence from Florida. Cato Journal, 28(1), 73-82. Spence, M. (1973). Job market signaling. Quarterly Journal of Economics, 87(3), 355-374. Stigler, G. (1971). The theory of economic regulation. Bell Journal of Economics and Management Science, 2(1), 3-21. Straus, A., & Corbin, J. (1998). Basics of qualitative research. Thousand Oaks, CA: Sage. Summers, A. (2007). Occupational licensing: Ranking the states and exploring alternatives. Los Angeles: Reason Foundation. Svensson, L. G. (1990). Knowledge as a professional resource: Case studies of architects and psychologists at work. In R. Torstendahl & M. Burrage (Eds.), The formation of professions (pp. 51-70). London: Sage. Timmermans, S. (2008). Professions and their work: Do market shelters protect professional interests? Work and Occupations, 35(2), 164-188. Timmons, E. J., & Thornton, R. J. (2008). The effects of licensing on the wages of radiologic technologists. Journal of Labor Research, 29, 333–346. Turner, S. (2001). What is the problem with experts? Social Studies of Science, 31, 123-149. United States District Court. (2004a). Sandy Meadows et al., v. Bob Odom et al.; Answers to Interrogatories. Baton Rouge, LA: Middle District Court of Louisiana. United States District Court. (2004b). Sandy Meadows, Shamille Peters, and Barbara Peacock v. Bob Odom et al.; Testimony of Ben Knight Baton Rouge, LA: Middle District Court of Louisiana. Weisz, G. (1983). The emergence of modern universities in France, 1863-1914. Princeton, NJ: Princeton University Press. White, W. (1978). The impact of occupational licensure of clinical laboratory personnel. Journal of Human Resources, 13, 91-102. White, W. (1980). Mandatory licensing of registered nurses: Introduction and impact. In S. Rottenberg (Ed.), Occupational licensure and regulation (pp. 47-72). Washington, DC: American Enterprise Institute. Williams, W. E. (1982). The state against blacks. New York: McGraw-Hill. Zukin, S. (1998). Urban lifestlyes: Diversity and standardization of spaces of consumption. Urban Studies, 35(5-6), 825-840.


Human and Economic Development in China and India: A Comparative MDG Assessment

Bruce A. Forster

University of Nebraska at Kearney Arizona State University

This paper assesses economic and human development in China and India using the United Nations set of Millennium Development Goals (MDGs) as the framework for analyzing the relative performance of the two countries. The MDG framework was used by Rausch & Kostyshak (2009) to assess the development of Arab countries in the Middle East and Africa, and by Forster (2010) to examine development in Sub-Saharan Africa. In summary, the paper shows that China has been more successful in improving its positions relative to the set of MDGs than has India. China also leads several comparator groups while India lags them. INTRODUCTION

This paper assesses China’s and India’s relative progress in economic and human development in. The current conceptual framework for promoting human and economic development in emerging and developing countries is provided by the United Nations Millennium Declaration (UNGA, 2000) which was signed by 189 countries at the September 2000 UN Millennium Summit. A set of eight Millennium Development Goals (MDGs) set out the specific objectives of the Declaration. The MDGs echo goals that had emerged from other meetings such as the 1991 World Health Assembly sponsored by the World Health Organization (WHO), the 1994 International Conference on Population and Development (also known as the Cairo Consensus), the 1996 World Food Summit, and the 2000 World Education Forum (Education for All). The spirit of the MDGs has also been endorsed by the 2001 “Doha Development Agenda” of the World Trade Organization (WTO, 2001), the 2002 “Monterrey Consensus” (United Nations, 2002). The Millennium Declaration, as the conceptual framework for international development, was reaffirmed by the UN at the 2005 and 2010 World Summits (UNGA, 2005; 2010), and affirmed at the 2005 G8 Gleneagles Summit, and also at the 2009 G20 London (UK) Summit.

In the new millennium, China and India have emerged as economic power houses driving the overall growth of Developing Asia. This paper seeks to determine if the well-being of their populations is commensurate with their growth in economic output. The United Nations Millennium Development Goals (MDGs) provide the vehicle for assessing the comparative success of China and India in improving the well-being of their respective populations. Information is distilled from several sources in order to provide the necessary data inputs for the MDG analysis. The MDG framework was used by Rausch & Kostyshak (2009) to assess relative development across three sets of Arab countries (the Middle East, North Africa and in Sub-Saharan Africa) and by Forster (2010) to compare development in Sub-Saharan Africa with other developing regions of the world.


In order to set the stage for the MDG assessment, it is useful to become familiar with some background information on relative macroeconomic performance of, and the population profiles for, China and India, and some other comparative regions of the world. RELATIVE MACROECONOMIC PERFORMANCE

First consider the relatively recent macroeconomic performances of China and India and selected regions of the world shown in table 1.

TABLE 1 REAL GDP GROWTH IN CHINA, INDIA, DEVELOPING ASIA,

ADVANCED AND WORLD ECONOMIES

REGION AVG 1993-2002

2003 2004 2005 2006 2007 2008 2009 2010 2011* (Est.)

CHINA 9.8 10.0 10.1 11.3 12.7 14.2 9.6 9.2 10.3 9.6 INDIA 5.8 6.9 8.1 9.2 9.8 9.9 6.2 6.8 10.4 8.2 DAa 7.1 8.1 8.6 9.5 10.4 11.4 7.7 7.2 9.5 8.4 AEb 2.8 1.9 3.1 2.7 3.0 2.7 0.2 - 3.2 3.0 2.2 World 3.3 3.6 4.9 4.6 5.2 5.4 2.9 -0.5 5.0 4.3 Source: IMF (2011) a) DA: Developing Asia, and b) AE: Advanced Economies

All of the regions/countries in table 1 experienced positive growth in real output from 2002 to 2008 with DA dominating AE and the World performances in each year. The economic problems of 2008-09 hit the World and AE regions relatively hard causing contractions in output in 2009. DA had reduced growth rates in 2008-09 but was spared contractions in real GDP. DA’s strong growth reflects those of India and (especially) China, each of which avoided actual contractions in real GDP.

According to one succinct assessment, “China’s strong and sustained growth over the past several years has served as a linchpin for global trade…..”(IMF, 2011) In early 2011, China surpassed Japan to become the world’s second largest economy. A report from PriceWaterhouseCooper predicts that by 2030, China will pass the United States and occupy the number one position in global trade (BBC, 2011).

China’s growth over the last two decades has frequently featured double-digit performance. Both China and India had slight drops in growth in 2008 and in 2009 due to the global crisis. India’s lowest growth rate occurred in 2008, and China’s occurred in 2009. While India’s growth has been very strong, it has lagged China’s performance until 2010, when India’s growth rate was marginally higher than China’s. Bosworth and Collins (2008) provide an excellent comparative analysis of the relative sources of economic growth in China and in India. Their paper and the current paper are useful complements in understanding human and economic development in China and India. POPULATION PROFILES

From 1930 to mid-2011, the world’s population grew from 2 billion people to slightly less than7 billion people. Not only has world population grown dramatically over that 80 year period, its distribution has been altered in terms of location and composition. In particular, commencing about 1950, the growth in world population shifted towards developing countries. In 1950, the world population was about 2.5 billion people of which about 1.7 billion people were in developing countries. Thus, developing countries represented roughly 68% of the world’s population.


TABLE 2 CHINESE, INDIAN, LESS DEVELOPED COUNTRIES, AND WORLD POPULATION

INFORMATION, MID-2011

Countries and Benchmarks

Population (millions)

Share of World Population

(%)

Total Fertility Rate

Children per female China 1,345.9 19.3 1.5 India 1,241.3 17.8 2.6 Less Developed Countries

5,745 82.2 2.6

World 6,987 100.0 2.5 Source: PRB (2011a)

The world’s population in mid-2011 was roughly 6.99 billion, and the population of developing

countries was 5.66 billion. Developing countries accounted for about 82% of the world’s mid-2011 population. China and India are the world’s two most populous countries. China’s mid-2011 population was 1.35 billion and India’s was 1.24 billion. Together, in mid-2011, these two countries have 2.6 billion people – amounting to about 47% of the population of the developing countries and about 37% of the world’s population.

China’s total fertility rate is 1.5 while India’s is 2.6. Since China’s fertility rate is below the replacement value of about 2.1-2.3 children per female, China’s population will ultimately stop growing and then start declining. India’s fertility rate of 2.6 is above the replacement value, and as a result India’s population will continue growing and it will ultimately pass China to become the world’s largest country (barring major changes in fertility rates of course).

Some western business interests look at China and India and see huge market possibilities, and others see huge competitors penetrating their markets. Both groups are correct. Western export-oriented businesses see the market potential available as China and India become wealthier, while import-competing enterprises will feel threatened by the competition. THE MILLENNIUM DEVELOPMENT GOALS

This paper discusses the first seven of the eight MDGs (MDG 8 is largely beyond the domain of individual countries). The eight MDGs are: MDG 1: Eradicate Extreme Poverty and Hunger MDG 2: Achieve Universal Primary Education (UPE) MDG 3: Promote Gender Equality and Empower Women MDG 4: Reduce Child Mortality MDG 5: Improve Maternal Health MDG 6: Combat HIV/AIDS, Malaria, and Other Diseases MDG 7: Ensure Environmental Sustainability MDG 8: Develop Global Partnerships for Development Each MDG has at least one target. Not all of the targets will be discussed here. The targets are numbered in their order of presentation in this paper. As a rough rule of thumb, the performance measures are benchmarked by a 1990 level of performance, and the target is to improve by 50% the performance from its 1990 level to its 2015 level.

In terms of population size, both China and India dominate their respective regional sub-group(i.e. East Asia and South Asia respectively); hence, these sub-groups may not be appropriate benchmarks for


their performance. Accordingly, in the following discussions broader categories of countries are selected as benchmarks. For example, the developing countries and the world, or different human development levels (defined below), may be more appropriate benchmarks to use in assessing Chinese and Indian performance. MDG 1: ERADICATE EXTREME POVERTY AND HUNGER Target 1: Halve, between 1990 and 2015, the proportion of people whose income is less than “one dollar a day”

In early 2009, the threshold used for extreme poverty was revised from $1.00 per day to $1.25 at 2005 PPP (World Bank, 2009a, b); however, some people still refer to the “$1.00 per day” criterion. The revision in threshold is incorporated in Tables 3 and 4.

TABLE 3 PROPORTION OF PEOPLE IN CHINA, INDIA AND DEVELOPING COUNTRIES LIVING ON

LESS THAN $1.25 A DAY

Countries and Benchmark

Proportion of people living on less than $1.25 per day (%)

1990

2005

2015*

2015**

2015***

China 60.2 15.9 5.0 5.0 6.0 India 51.3 41.6 22.7 23.6 29.4 Developing countries

41.7 25.2 14.1 15.0 18.5

Source: World Bank (2010) 2015* - The pre-crisis trend based upon growth 2000-2007 2015** - The post-crisis trend assumes a relatively rapid economic recovery in 2010 2015*** - The low growth scenario assumes little or no growth for roughly 5 years.

In 1990, both China and India had larger proportions of their people living on less than $1.25 than the

average of developing countries with China’s proportion exceeding India’s. By 2005, China’s proportion had plummeted to 15.9 % -- a quarter of the 1990 rate—and thus China fulfilled the MDG 1 goal and target 10 years early! India is predicted to meet the goal by 2015 in two scenarios but it falls short in the low growth scenario.

A stark perspective of the implications of the above results is provided by the number of people living on less than $1.25 per day in the various years shown in table 4.

In 1990, China and India combined had about 1,118 million people living on less than $1.25 per day which amounts to 62% of the people in developing countries living on less than $1.25 per day. By 2015 these countries are predicted to have less than one-half billion, thanks mainly to the dramatic reduction in the number of Chinese living on less than $1.25 per day.


TABLE 4 THE NUMBER OF PEOPLE IN CHINA, INDIA AND DEVELOPING COUNTRIES LIVING ON

LESS THAN $1.25 A DAY


Number of people living on less than $1.25 per day (millions)

1990

2005

2015*

2015**

2015***

China 683 208 69 70 82 India 435 456 283 295 367

Developing countries

1,817 1,371 865 918 1,132

Source: World Bank (2010). 2015* - The pre-crisis trend based upon growth during 2000-2007 2015** - The post-crisis trend assumes a relatively rapid economic recovery in 2010 2015*** - The low growth scenario assumes little or no growth for roughly 5 years.

An alternative assessment of “poverty” is given by the country’s Human Development Index (HDI).

Commencing with the 2010 Human Development Report (HDR), the HDI is based upon life expectancy at birth, the mean years of schooling and the expected years of schooling, and Gross National Income (GNI) per capita. The 1990-2010 HDI indices are given in table 5.

TABLE 5 HDI VALUES 1990-2010, AND RANK IN 2010 FOR CHINA AND INDIA

Countries and Benchmarks 1990 1995 2000 2005 2010 2010-Rank in

169 countries China 0.460 0.518 0.567 0.616 0.663 89 India 0.389 0.415 0.440 0.482 0.519 119 High Human Development

0.633 0.634 0.659 0.692 0.717

Medium Human Development

0.440 0.480 0.510 0.555 0.586

World 0.526 0.554 0.570 0.598 0.624 Source: UNDP (2010)

Starting with HDR 2010, the setting of development categories will no longer use pre-determined cut-

off values for HDIs. Instead, the categories will be set according to quartiles. The top quartile will be the countries classified as having Very High Development levels, and these also receive designation as developed countries. Countries in the other three quartiles are termed developing countries. In 2010, both China and India have HDIs in the Medium Development category (UNDP, 2010). China has outperformed the average for the Medium Development group for each year in table 6, and has outperformed the world average for 2005 and 2010. India has underperformed relative to the average of the Medium Development group and the world average in each year in table 5.

HDR 2010 introduced an Inequality-adjusted HDI (IHDI) to account for losses in human well-being due to unequal levels of the components of the HDIs within countries. IHDIs are presented in table 6.

IHDIs are lower than the HDIs for China and India and the benchmarks. The differences between HDI and IHDI measures reflect the losses associated with inequality in each group. While inequality results in a loss in welfare for all groups, it has not altered the rank ordering of the groups in table 5 determined by the basic HDIs.


TABLE 6 2010 IHDI AND HDI VALUES FOR CHINA AND INDIA AND BENCHMARKS


2010 IHDI

2010 HDI

China 0.511 0.663 India 0.365 0.519 High Human Development

0.575 0.717

Medium Human Development

0.449 0.586

World 0.489 0.624 Source: UNDP (2010)

Target 2: Halve, between 1990 and 2015, the proportion of people who suffer from hunger

TABLE 7 PROPORTION OF POPULATION UNDERNOURISHED 1990-92 TO 2005-2007


1990 – 1992 %

2000– 2002 %

2005-2007 %

China 18 10 10 India 20 19 21 Developing countries 20 17 16 World 16 14 13

Source: FAO (2010).

The 1996 World Food Summit’s related hunger goal is to halve the number of undernourished people by 2015 (see table 8 below). China has shown a substantial decline in the proportion, and the number, of undernourished. As of 2005-07, China had not quite reached the milestone for MDG undernourishment target nor had it reached the World Food Summit goal; however, it appears to be within striking range of achieving both goals. India had shown a very slight drop in the proportion of undernourished in 2000-2002 followed by more than an offsetting increase in 2005-2007. With population growth, this produced an increase in the number of undernourished in India. India is unlikely to satisfy either goal by 2015. Developing countries collectively are unlikely to achieve either goal.

TABLE 8 NUMBER OF PEOPLE UNDERNOURISHED (MILLIONS), 1990-92 TO 2005-2007

Countries and Benchmarks 1990-1992 2000-2002 2005-2007

China 210.1 133.1 130.4 India 172.4 200.6 237.7 Developing Countries 862.6 816.0 835.2 World 843.4 833.0 847.5

Source: FAO (2010). UNDP (2010) estimates that the 2008 food price increases may have increased the number of

undernourished people by about 63 million, and the economic crisis of 2008-2009 may have added another 41 million in 2009 than would otherwise have been the case. The World Bank’s Food Price


Watch (2011) announced that the global price increase between June 2010 and January 2011 added about 44 million more people to the number living in poverty.

Undernourishment causes health problems for children. In particular, it affects their height and weight.

TABLE 9 RATES OF LOW BIRTH WEIGHT AND STUNTING, 2003-2008


% Infants with Low birth weights

% Under age-5 stunting

China 4 15 India 28 48 Developing Countries 11 29 World 9 26 Source: UNESCO (2011a, Statistical Tables).

Consistent with the undernourishment rates, the Chinese and Indian rates for low birth weight and

stunting bracket the rates for the benchmarks with China having the lowest rates and India the highest. MDG 2: ACHIEVE UNIVERSAL PRIMARY EDUCATION (UPE) Target 3: Ensure that, by 2015, children everywhere, boys and girls alike, will be able to complete a full course of primary schooling.

TABLE 10 PRIMARY EDUCATION NET ENROLLMENT RATIOS (NERS) AND SECONDARY

EDUCATION GROSS ENROLLMENT RATIOS (GERS), 1999 AND 2008


Primary Education

NER %

Secondary Education GER %

1999 2008 1999 2008 China --- --- 61 76 India --- 90 44 57 Developing Countries 80 87 51 62 World 82 88 59 67

Source: UNESCO (2011 b, c)

No primary education figures are available from the UNESCO Report for China and only 2008 figures for India are available. For India the 2008 NER exceeds the rates for both benchmarks, and at 90% it indicates that India is close to achieving UPE. For secondary enrollments, China’s enrollment figures are higher than India’s and the benchmarks. India falls short of the benchmark performances in each of the years. The flip-side of school enrollments is the number of primary age children not enrolled in school.


TABLE 11 NUMBER OF CHILDREN OF PRIMARY AGE NOT ENROLLED IN PRIMARY

SCHOOL, 1999 AND 2008


1999 (000s)

2008 (000s)

China --- --- India --- 5,564 Developing Countries 103,180 64,117 World 106,269 67,483 Source: UNESCO (2010)

Again, no data are presented for China, and only 2008 figures are available for India, so it is not

possible to comment much on this feature.

TABLE 12 SCHOOL-AGE POPULATION, TOTAL ENROLLMENT IN TOTAL

SECONDARY EDUCATION AND ENROLLMENT TECHNICAL AND VOCATIONAL EDUCATION (TVE), 2008


School-age Population

(000)

Total Secondary Enrollment

(000) % F

TVE Enrollments

(000) % F

China 133,331 101,448 48 18,906 50 India 169,593 96,049 44 742 --- Developing Countries 673,720 416,945 47 39.960 47 World 783,711 525,146 46 56,777 46

Source: UNESCO (2011a)

About 19% of Chinese secondary enrollment is in TVE programs, and Chinese TVE enrollment represents about 47% of developing countries’ TVE enrollments. Indian TVE enrollments are considerably smaller than China’s. The gender make-up of TVE enrollment is equally balanced for China and slightly male dominant for the benchmarks. Gender information for India’s TVE enrollment is not available.

Achieving a 50% improvement in the levels of adult literacy is part of Goal 4 of the 2000 World Education Summit’s “Dakar Framework for Action” (UNESCO, 2010).


TABLE 13 ADULT LITERACY RATES AND THE NUMBER OF ADULT ILLITERATES, 1985-1994 AND

2005-2008


Adult literacy rates (% aged 15 and over)

Adult illiterates (# aged 15 and over) (000)

% F

Adult illiterates (# aged 15 and over) (000)

% F

1985-1994 2005-2008 1985-1994 2005-2008 China

78 94 181,415 70 67,239 73

India 48 63 284,027

61 283,105 65

Developing Countries

67 79 872,565 63 786,386 64

World 76 83 886,508 63 795,805 64 Source: UNESCO (2011a).

MDG 3: PROMOTE GENDER EQUALITY AND EMPOWER WOMEN

The 2010 Human Development Report (UNDP, 2010) presented a new index, the Gender Inequality Index (GII), to account for the welfare impacts associated with gender inequality within countries. The GII includes measures associated with labor markets, empowerment, and reproductive health. A GII of 0 indicates perfect Equality, and a GII of 1 indicates absolute inequality. Table 14 presents GII figures and components for China, India and the benchmarks.

TABLE 14 GENDER INEQUALITY INDICES AND THEIR RESPECTIVE COMPONENTS

Countries And Benchmarks

Gender Inequality Index 2008

Labor market

Empowerment

Reproductive health

Labor Force Participation % 2008

Population with at least Secondary Education % 2010

Parliamentary Seats % 2008

Adolescent Fertility Rate 1990-2008

Maternal Mortality Ratio 2003-2008

F M F M F China 0.405 74.5 84.3 54.8 70.4 21.3 9.7 45 India 0.748 35.7 84.5 26.6 50.4 9.2 68.1 450 Medium Human Development

0.591 54.7 84.1 40.9 57.4 16.0 41.8 242

World 0.560 56.8 82.4 51.6 61.7 16.2 53.7 273 Source: UNDP (2010)


China exhibits less gender inequality than India or the benchmarks while India exhibits more gender inequality than the benchmarks. China’s out performs the comparators in table 14 on all components while India under-performs on all components. Target 4: Eliminate gender disparity in primary and secondary education preferably by 2005, and in all levels of education no later than 2015

Gender Parity Indices (GPIs) can be calculated for several activities or situations. GPIs are the ratios of the female participation rate to the male participation rate for the specified activity. GPIs for primary and secondary school enrollments (in terms of the Gross Enrollment Ratios [GERs]) and for literacy are listed in table 15.

TABLE 15 GENDER PARITY INDICES (GPIS), FOR PRIMARY AND SECONDARY GERS, 1999 AND

2008 AND FOR LITERACY, 1985-1994 AND 2005-2008

GPI (F/M) for Primary Education GER

GPI (F/M) for Secondary Education GER

GPI (F/M) for Literacy


1999 2008 1999 2008 1985-1994 2005-2008

China --- 1.04 --- 1.05 0.78 0.94 India 0.84 0.97 0.70 0.86 0.55 0.68 Developing Countries

0.91 0.96 0.88 0.95 0.76 0.86

World 0.92 0.97 0.91 0.96 0.84 0.90 Source: UNESCO (2011b, c)

In school enrollments the participation of females in China in 2008 exceeds males by a very slim

margin. Males dominate enrollment in India and the benchmarks in each of the years. In 2008, India’s GPIs for primary enrollment is in line with the benchmarks; however, for secondary education, India’s enrollments feature considerably more males than females and India is more male dominant than the benchmarks in both years. Since India did not achieve parity in education by 2008, it did not satisfy the first part of target 4.

China, India and the benchmarks show improvements in gender parity for literacy. In 2005-2008, China was the closest to parity and is in position to achieve parity by 2015. India lags behind China and the benchmarks considerably, and it is unlikely to achieve parity by 2015. MDG 4: REDUCE CHILD MORTALITY Target 5: Reduce, by two-thirds, between 1990 and 2015, the under-five Mortality Rate (U5MR).

According to WHO (2005), there were about 136 million births each year, of which 3.3 million babies were stillborn, another 4.0 million or more died within the first 28 days after birth, and 6.6 million children died before their fifth birthday. 1990 and 2009 figures for the number of deaths per 1000 live births for children under age 5 (U5MR), and the number of deaths, for China and India and the benchmarks are presented in table 16.

The U5MRs fell for China, India and the benchmarks between 1990 and 2009. China has the lowest rates in both years by a considerable margin, and is close to satisfying target 5. India’s rate was higher than the benchmarks in 1990, but roughly the same in 2009. Neither India nor the developing countries is likely to satisfy target 5.


TABLE 16 REGIONAL U5MRS (UNDER 5 DEATHS PER 1000 LIVE BIRTHS) AND THE

NUMBER OF U5 DEATHS, 1990 AND 2009

Source: UNICEF (2010)

In 2009, there were almost 8.1 million under-5 deaths globally. China and India accounted for almost 2.1 million of the global total, or 26% of the total. Globally, pneumonia is the largest single cause of child mortality accounting for 18% of cases. The next highest cause is diarrhoel disease (15%) followed by pre-term births (12%).

Table 17 provides information on infant mortality rates (the number of infant deaths per 1000 live births) and the number of infant deaths.

TABLE 17 INFANT MORTALITY RATES (IMRS), AND INFANT DEATHS, 1990 AND 2009


IMR 1990

IMR 2009

Infant deaths 1990 (000s)

Infant deaths 2009 (000s)

China 37 17 1,031 302 India 84 50 2,223 1,316 Developing regions 68 47 8,371 5,613 World 62 42 8,688 5,751

Source: UNICEF (2010)

Infant mortality rates in China, India and the benchmarks decreased between 1990 and 2009. China has the lowest IMR in both years by a considerable margin. India’s IMR was higher than the benchmarks in both years. In 1990, China and India accounted for about 38% of global infant deaths. By 2009, they accounted for only 28% of global infant deaths.

The causes of infant mortality are different from those of child mortality with prematurity, severe infection and asphyxia being the three largest causes of infant mortality. MDG 5: IMPROVE MATERNAL HEALTH Target 6: Reduce, by three-quarters, between 1990 and 2015, the maternal mortality ratio (MMR)

According to WHO (2010), about 358,000 women died due to maternal causes in 2008. The WHO definition of a maternal death is:

the death of a woman while pregnant, or within 42 days of termination of pregnancy, irrespective of the duration and site of the pregnancy, from any cause related to, or aggravated by, the pregnancy or its management but not from accidental or incidental causes (WHO, 2010).


U5MR 1990

U5MR 2009

U5 deaths 1990 (000s)

U5 deaths 2009 (000s)

China 46 19 1,255 347 India 118 66 3,133 1,726 Developing regions 99 66 12,012 7,929 World 89 60 12,393 8,087


The Maternal Mortality Ratio (MMR) is defined as the number of maternal deaths per 100,000 live births during a specified time period. The lifetime risk of maternal death is the probability of dying from a maternal cause during a woman’s reproductive lifespan. MMRs for 1990-2008 and the percentage change in MMRs between 1990 and 2008 are given in table 18.

TABLE 18 MMRS 1990-2008 AND THE PERCENTAGE CHANGE IN MMRS


1990 1995 2000 2005 2008 % change 1990 to 2008

China 110 82 60 44 38 -66 India 570 470 390 280 230 -59 Developing countries

440 410 370 320 290 -34

World 400 370 340 290 260 -34 Source: WHO (2010)

MMR’s fell between 1990 and 2008 in China, India and the benchmarks with China and India

dropping the furthest. China’s drop of 66% between 1990 and 2008 puts it close to satisfying MDG 5 target of reducing its MMR by three quarters. It will be more of a stretch for India, and perhaps impossible stretches for the benchmarks.

TABLE 19 MMRS, THE NUMBER OF MATERNAL DEATHS, AND THE LIFETIME

RISKS OF MATERNAL DEATH, 2008


MMR

Number of maternal deaths

Lifetime risk of maternal death; 1 in:

China 38 6900 1500 India 230 63,000 140 Developing countries

290 356,000 120

World 260 358,000 140 Source: WHO (2010)

In 2008, the number of maternal deaths in India was almost 10 times as many as the number in China.

China had the lowest lifetime risk of maternal death. While India’s lifetime risk of maternal death is significantly higher than China’s, it matches the world’s average risk. MDG 5: Target 7: Achieve by 2015, universal access to reproductive health

Figures for contraceptive use and attendance of skilled personnel at births are given in table 20. Once again, China’s and India’s experiences bracket those of the developing countries and the world

with China having the highest percentages and India with the lowest. In China, attendance of skilled professional at birth is virtually universal. In India, less than half of births are attended by skilled personnel. Of the two countries, China appears to be making better use of family planning methods and maternal and infant health.


TABLE 20 THE PROPORTION OF MARRIED WOMEN USING CONTRACEPTIVE METHODS, AND

THE PROPORTION OF BIRTHS ATTENDED BY A SKILLED HEALTH CARE PROFESSIONAL


% of married women using contraceptive method

% of births attended by skilled health personnel

Any method Modern method China 87 86 98 India 56 49 47 Developing Countries

60 54 63

World 62 55 67 Source: PRB (2011b)

MDG 6: COMBAT HIV/AIDS, MALARIA, AND OTHER DISEASES Target 8: Have halted by 2015 and begun to reverse the spread of HIV/AIDS

“We have halted and begun to reverse the epidemic. Fewer people are becoming infected with HIV and fewer people are dying from AIDS.” UNAIDS (2010)

Globally, new HIV infections peaked at 3.2 million in 1997, and AIDS deaths peaked at 2.1 million in

2004 (UNAIDS, 2010). Over the known history of the AIDS/HIV epidemic up until 2007, HIV has caused a total of 25 million deaths (UNAIDS, 2009).

China and India are two of the 25 countries with the most people living with HIV in 2009. All but eight of the 25 countries are from Sub-Saharan Africa, the region hit hardest by HIV. Sub-Saharan Africa has 68% of the HIV- infected individuals globally.

TABLE 21 THE NUMBER OF PEOPLE LIVING WITH HIV, THE NUMBER OF NEWLY INFECTED

ADULTS, THE NUMBER OF PEOPLE DYING FROM AIDS-RELATED CAUSES

Adults and children living with HIV

Adults and children newly infected

AIDS-related deaths of adults and children


2001 (millions)

2009 (millions)

2009 (millions)

2001 (millions)

2009 (millions)

China 0.24* 0.74 0.05* 0.09* 0.026 India 2.5 2.4 0.14 0.14 0.17 Global 28.6 33.3 2.6 1.8 1.8 Source: UNAIDS (2010). * denotes lower bound estimate.

In 2009, globally there were 33.3 million people living with HIV (up from 28.6 million in 2001). The number of people in China living with HIV in 2009 was 740,000 (up from 240,000 in 2001). India shows a drop in the number of people living with HIV from 2.5 million in 2001 to 2.4 million in 2009. The


newly HIV-infected individuals figure for India is almost three times higher than the number of newly infected individuals in China. India’s antiretroviral therapy coverage is less than 40%.

Globally, 1.8 million people suffered AIDS-related deaths in 2001and 2009 (recall that in 2004 the number of AIDS-related deaths globally was 2.1 million). China showed a drop in AIDS deaths while India showed a slight increase in AIDS deaths. MDG 6: Target 9: Have halted by 2015 and begun to reverse the incidence of malaria and other major diseases.

WHO (2010) summarizes information from106 malaria-endemic countries and other partners. It is estimated that there were 225 million episodes of malaria world-wide in 2009. Malaria is believed to have caused more than three-quarters of a million deaths globally in 2009 with 85% being children under 5 years of age. SSA, by far, has the highest risk of malaria. SSA had 78% of the malaria cases, and 91% of malaria deaths. South-East Asia followed SSA with 15% of malaria cases and 6% of the malaria deaths.

TABLE 22 THE NUMBER OF CONFIRMED MALARIA CASES AND MALARIA

DEATHS IN CHINA AND INDIA

Countries Number of confirmed cases (000)

Number of deaths

China (2008)

17 23

India (2009)

1,560 1,133

Source: WHO (2010b)

India’s burden in terms of the number of confirmed cases of malaria and the number of malaria deaths is several orders of magnitude higher than those for China.

Table 23 shows that for infant immunizations, China has uniformly higher immunization rates for five important diseases than India or the benchmarks. For India, the tuberculosis immunization rate is considerably higher than for its other diseases.

TABLE 23 THE PERCENT OF INFANTS (1-YEAR OLDS) IMMUNIZED AGAINST

SELECTED DISEASES, 2008


Tuberculosis

Diptheria, Pertussis, tetanus

Polio

Measles

Hepatitis B

China 97 97 99 94 95 India 87 66 67 70 21 Developing countries

95 91 91 90 91

World 96 93 93 92 92 Source: UNESCO (2011a)


MDG 7: ENSURE ENVIRONMENTAL SUSTAINABILITY

Table 24 provides estimates of Carbon Dioxide (CO2) from 1990 to 2007 showing increases in all groups. China’s estimate increases by more than double between 1999 and 2007. It is important to notice that the data for 2007 comes from a different source than the earlier years. The CO2 information is of dubious usefulness. For global climate change it is total emissions that matter not per capita emissions.

TABLE 24 CO2 EMISSIONS PER CAPITA

(METRIC TONS)

1990* 1999* 2007** China 2.1 2.3 5.0 India 0.8 1.1 1.4 Lower Income 0.7 1.0 0.3 World 3.4 3.8 4.6

Source: *UNDP (2003); **World Bank (2011)

Table 25 indicates the proportion of people having improved sanitation.

TABLE 25 URBAN POPULATION WITH ACCESS TO IMPROVED SANITATION FACILITIES

(PERCENT)

1990* 2000* 2008** China 56 69 55 India 44 61 31 Lower Income 58 72 35 World -- 85 61

Source: *UNDP (2003); **World Bank (2011)

Population access to improved sanitation facilities expanded from 1990 to 2000 for China and India and for the set of lower income countries. The access drops off dramatically between 2000 and 2008 for all groups in the table; however, this may reflect different data sources. SUMMARY

China’s performance on the MDGs is stronger than India’s, and it is stronger than the average of the Medium Human Development countries. India generally lags behind the average of the Medium Human Development benchmark. In summary, China has been more successful in improving its status relative to the MDGs than has India. The ability of these countries to achieve the MDG targets by 2015, like other developing countries, will depend not only on their own dedication to improving their performance from now until 2015, but also upon unforeseen shocks to their economies, to those of countries with whom they have very close trading relationships or the world economy generally. REFERENCES BBC. (2011). “China To Overtake Us And Dominate Trade By 2030”. BBC News. March 24 2011. www.BBC.Co.Uk


Bosworth, B., and Collins, S.M. (2008). “Accounting For Growth: Comparing China and India,” The Journal Of Economic Perspectives, 22 (1) 45-66. Food And Agriculture Organization (FAO) Of The United Nations. (2010). The State Of Food Insecurity In The World, 2010. Rome, Italy. Forster, B.A. (2010). “A Survey Of Economic And Human Development Performance Indicators For Sub-Saharan Africa,” International Journal Of Business And Economics, 10 (1) 82-93. International Monetary Fund, (IMF). (2011). World Economic Outlook, April 2011. Population Reference Bureau, (PRB). (2011a). 2011 World Population Data Sheet, Washington, D.C. Population Reference Bureau, (PRB). (2011b). The World’s Women And Girls: 2011 Data Sheet. Washington, D.C. Rauch, J.E. & Kostyshak, S. (2009). The Three Arab Worlds. The Journal Of Economic Perspectives, 23 (3) 165-188. United Nations. (2002). Report On “The International Conference On Financing For Development” (Also Known As The Monterrey Consensus). New York, NY. United Nations. (2010). The Millennium Development Goals Report 2010. New York, NY. UNAIDS. (2010). UNAIDS Report On The Global Aids Epidemic 2010. United Nations Development Program (UNDP). (2003). Human Millennium Development Goals: A Compact Among Nations To End Human Poverty. Development Report (HDR) 2003. New York, Ny. United Nations Development Program (UNDP). (2010). The Real Wealth Of Nations. Human Development Report (HDR) 2010. New York, NY. UNESCO. (2010). Reaching The Marginalized. Education For All: Global Monitoring Report 2010. Paris, France. UNESCO. (2011a). The Hidden Crisis: Armed Conflict And Education. Education For All: EFA Global Monitoring Report 2011. Paris, France. UNESCO. (2011b). The Hidden Crisis: Armed Conflict And Education. Education For All: EFA Global Monitoring Report 2011., Regional Overview: South And West Asia. Paris, France. UNESCO. (2011c).The Hidden Crisis: Armed Conflict And Education. Education For All: EFA Global Monitoring Report 2011., Regional Overview: East Asia And The Pacific. Paris, France. United Nations General Assembly (UNGA). (2000). United Nations Millennium Declaration. New York, NY. United Nations General Assembly (UNGA). (2005). United Nations 2005 World Summit Outcome. New York, NY. UNICEF. (2010). Levels and Trends In Child Mortality. Report 2010. New York, NY.


United Nations Population Fund (UNFPA). (2004). The State Of World Population, 2004 (The Cairo Consensus At Ten: Population, Reproductive Health And The Global Effort To End Poverty), New York, NY. World Bank. (2009a). A Development Emergency. Global Monitoring Report 2009. Washington, D.C. World Bank. (2009b). Global Economic Prospects—Commodities At The Crossroads. Washington, D.C. World Bank. (2010). The MDGS After The Crisis. Global Monitoring Report 2010. Washington, D.C. World Bank. (2011a). Food Price Watch. February 2011. Washington, DC. World Bank. (2011b). Conflict, Security, And Development. World Development Report (2011). Washington, D.C. World Health Organization (WHO). (2005). World Health Report 2005. Geneva, Switzerland. World Health Organization (WHO). (2010a).Trends In Maternal Mortality: 1990-2008. Geneva, Switzerland. World Health Organization (Who). (2010b). World Malaria Report. Geneva, Switzerland. World Trade Organization (WTO). (2001). Ministerial Declaration. (Known As The Doha Development Agenda), Geneva , Switzerland.


Leadership Styles and Organizational Citizenship Behavior: The Mediating Effect of Subordinates’ Competence and Downward Influence Tactics

Lee Kim Lian

UCSI University

Low Guan Tui Vesseltech Engineering Sdn Bhd

The objective of this study is to test a theory-based model predicting the relationships between leadership styles, subordinates’ competence, downward influence tactics and outcome of organizational citizenship behavior in Malaysian-based organizations. Data was collected from 347 respondents that represent major industries like services, manufacturing, mining and construction companies. Path analysis technique was used to test the model developed. The results show that the transformational leadership style has significant positive relationship with subordinates’ organizational citizenship behavior, whereas the transactional leader style is negatively related to organizational citizenship behavior. This result illustrates the direct effects of leadership styles on the subordinates’ outcome. In addition, inspirational appeals and consultation tactics, as downward influence tactics, were found to mediate the relationship between transformational leadership and organizational citizenship behavior. Likewise, subordinates’ competence mediates the relationship between transformational leadership and consultation tactics. These results only partially support the efficacy of the influence theory, and therefore lend support to contingency theories of leadership. Implications for research and direction for future research are also discussed. INTRODUCTION

This study explores how superior leadership styles may impact subordinates’ organizational citizenship behavior (OCB). The importance of leadership style as predictor of OCB has been well established in Western settings (Bass, 1985; Organ, 1988; Podsakoff, MacKenzie, Morrman & Fetter, 1990; Howell & Avolio, 1993; Lowe, Kroeck & Sivasubramaniam, 1996; Podsakoff, MacKenzie & Bommer, 1996; MacKenzie, Podsakoff & Rich, 2001; Geyer & Steyrer, 1998; Wang, Law, Hackett, Wang, Chen, 2005; Schlechter & Engelbrecht, 2006; Boerner, Eisenbeiss, Griesser, 2007). However, there is scant research explore the indirect effects between this two variables. Hence, the inclusion of subordinates’ competence and downward influence tactics served to investigate the role of intervening effect between leadership styles and OCB.

Several researchers have suggested that leadership research needs to focus more on the “fundamental” issues, such as influence processes that characterize leader-follower interaction (Bass, 1990; Hollander & Offermann, 1990; Yukl, 1989). Research has also shown that effective leaders must have the ability to recognize when to use different tactics of influence as well as the skill necessary to effectively carry out


these influence attempts (Kipnis, Schmidt & Wilkinson, 1980; Yukl & Falbe, 1990; Yukl, 1998; Bolino & Turnley, 2003). Moreover, in terms of using downward influence tactics effectively, several empirical studies offer strong support for the idea that the most effective leaders in organizations should understand the nature of influence, “what” influence tactics are available to them, and “how” and “when” to use those tactics (Case, Dosier, Murkison & Keys, 1988; Kaplan, 1986; Kipnis & Schmidt, 1988; Mowday, 1978; Schilit & Locke, 1982; Yukl & Falbe, 1990). These works seems to infer that influence is important in all human relationships.

On the other hand, studies on OCB around the issue of interpersonal relationships have been driven by the conviction that sound superior-subordinate relationship is crucial to organizational success. Positive interpersonal relationship at workplace should enhance positive OCB among the employees. Subordinates with high levels of OCB are more likely to be committed to the organization (William & Anderson, 1991; Smith, Organ & Near, 1983). Therefore, it is worthwhile for the superior to be aware of his/her leadership style in work situations and how it promotes subordinates’ OCB. Graham (1988) and Podsakoff, MacKenzie, Moorman and Fetter (1990) have indicated that superior’s leadership style and subordinates’ OCB are inter-related. Inappropriate leadership styles may trigger negative consequences, which might further increase the sensitivity and susceptibility to misunderstanding that may lead to organizational dysfunction such as decline in work performances, absenteeism and high turnover (Lamude, 1994; Motowidlo, 2003). Thus, prevention of subordinates’ negative outcome is important vis-a-vis the use of different leadership styles. The mismatch might precipitate an unending and potentially disruptive vicious cycle that many organizational leaders want to avoid and therefore, they might want to address their styles and the attendant consequences more rigorously. Objectives of the Study

So far, no studies have been carried out to investigate the superior’s downward influence tactics and subordinates’ competence as mediators between leadership style and OCB in Malaysian work settings. Thus, this research is carried out with the intention of achieving greater understanding of the appropriate downward influence tactics that allow the superiors to better achieve their objectives of maintaining subordinates’ OCB. Secondly, there are quite a substantial amount of research focused on upward influence and little attention has been given to link the issues arises around downward influence. Knowing how downward influence tactics relate to leadership styles and their consequences would enable a superior to consider changing or maintaining his/her styles and influence tactics in order to achieve certain desirable outcomes. Although several studies have explored the relationship between leadership styles and citizenship behavior, hitherto there has yet a study carried out to examine the mediating effect of subordinates’ competence and downward influence tactics on such relationships. Research Questions

The major motivation of this research is to examine how leadership styles affect subordinates’ OCB in Malaysian companies and how subordinates’ competence and downward influence tactics mediate these associations. More specifically, it seeks to answer the following research questions:

• Can leadership style predict downward influence tactics and subordinates’ organizational citizenship behaviour?

• Can downward influence tactics mediate the relationship between leadership style and organizational citizenship behaviour?

• To what extent the subordinates’ competence mediates the relationship between the leadership style and downward influence tactics?


THEORETICAL FRAMEWORK AND LITERATURE REVIEW

This section reviews the relevant constructs and variables as well as their interactions involving: (1) Leadership styles; (2) Subordinates’ Competence; (3) Downward influence; and (4) OCB as shown in Figure 1.

FIGURE 1 PROPOSED MODEL OF DOWNWARD INFLUENCE TACTICS AND INTERACTIONS

Leadership Styles Leadership is defined as the ability to influence others to get things done. It reflects an influence

relationship behavior between leaders and followers in a particular situation with the common intention to accomplish the organization end results (Stogdill, 1948; Bass, 1981). Generally, leadership researchers suggest that an effective leader should be able to articulate vision, instill trust, belief, loyalty and lead employees’ talents directly towards achieving the organizational goals (Kirkpatrick & Locke, 1996; Strange & Mumford, 2002; Levin, 1999; Bennis, 2002; DePree, 2002).

There are several well established dichotomy approaches to the classification of leadership styles. Stogdill (1963, 1974) proposes a leadership dichotomy as “consideration leadership” and “structure leadership”, likewise Fiedler (1967) suggests “task orientation” versus relationship orientation” and Hersey and Blanchard (1977) recommend “concern for people” and “concern for task”. However, this study focused on the transactional and transformational leadership style. Past investigation proposed the dichotomy methods of transactional-transformational leadership may be applicable in the study of phenomenological-based leadership styles (Misumi, 1985; Misumi & Peterson, 1985), in addition to the insights exploration of leaders-subordinates communication patterns (Penley & Hawkins, 1985) that shape both parties influence behaviors. The following section specifically discussed the transactional and transformational leadership styles Transactional Leadership

Past researchers have studied on transactional leadership as the core component of effective leadership behavior in organizations prior to the introduction of transformational leadership theory (Bass, 1985; Burns, 1978; House, 1977). Exchange relationship is the key element reflected by the transactional leadership. Transactional leaders demand their subordinates to agree with, accepted or complied with their request if the subordinates hope for rewards and resources or avoidance of punitive action (Burns, 1978; Podsakoff, Todor & Skov, 1982; Podsakoff, MacKenzie, Moorman & Fetter, 1990). This dyadic exchange process of leadership style has been linked with contingent reward and punishment behavior and termed as transactional leader behavior by Bryman (1992). The typical manager who is a transactional leader tends to identify employees lower level needs by determining the goals that subordinates need to achieve and communicate to them on how successful execution of those tasks will lead to receive of desirable job rewards (Avolio & Bass, 1988; Avolio, Waldman & Yammarino, 1991; Bass, 1985, 1990; Zaleznik, 1983). In fact, this process only helps employees to meet their basic work

LEADERSHIP STYLES

• Transformational • Transactional

Subordinates’ Competence

DOWNWARD INFLEUNCE

TACTICS • Inspirational Appeals • Consultation Tactics • Ingratiation Tactics • Exchange Tactics • Pressure Tactics • Legitimating Tactics

OUTCOME • Organizational

Citizenship Behavior


requirements and maintain the organizational status quo. Moreover, the transactional leader also limits the employees’ effort toward goals, job satisfaction and effectiveness (Bass 1985). Bass (1986) suggests that transactional leadership is acceptable as far as it goes, but fundamentally is a prescription for organizational mediocrity. Transformational Leadership

Transformational leadership refers to leader transformation process involving individuals, group and organization. It involves creating substantive change in the attitude of employees, moral elevation and organization direction. Kuhnert and Lewis (1987) highlighted that transformational leadership “is made possible when a leader’s end values (internal standards) are adopted by followers thereby producing changes in the attitudes, beliefs and goals of followers” (p.653). Similarly, Bryman (1992) has stated that “transforming leadership entails both leaders and followers raising each other’s motivation and sense of purpose. This higher purpose is one in which the aims and aspirations of leaders and followers congeal into one. Both leaders and followers are changed in pursuit of goals which express aspirations in which they can identify themselves” (p.95). It is no doubt that transformational leadership is of great interest of study due to its popularity and attractiveness of this leadership style found to be consistently associated with superior performance (Barling, Weber & Kelloway, 1996; Bass, Avolio, Jung & Berson, 2003; Dvir, Eden, Avolio & Shamir, 2002; Yammarino & Bass, 1990), increased morale-related outcomes such as self efficacy (Kirkpartick & Locke, 1996), affective commitment (Barling et al, 1996), intrinsic motivation (Charbonneau, Barling & Kelloway, 2001) and trust in the leader (Podsakoff et al., 1990). Positive relationships have also been consistently reported between individual, group and organizational performance. Typically, these findings have been explained as showing that leader behaviors cause basic values, beliefs and attitudes of followers to align with organizational collective interests (Podsakoff, MacKenzie, Moorman & Fetter, 1990). Downward Influence Tactics

Kipnis, Schmidt and Wilkinson (1980) and Yukl and Tracey (1992) have developed a simplified definition of influence. According to them, influence occurred when an influence agent is able to alter the target’s perceptions by getting the target to see the advantages of the intended behavior. Changes of behavior can be in the form of beliefs, attitudes and values. Yukl and his colleagues (Yukl, 1998; Yukl & Falbe, 1990; Yukl & Tracey, 1992) have further examined a variety of downward influence tactics available to leaders (Table 1). The ability to exert influence on the decisions made by a superior is an important objective. How superiors persuasively frame their downward influence tactics has been shown to impact performance ratings (Kipnis & Vanderveer, 1971), organizational influence (Floyd & Wooldridge, 1997), promotability (Thacker & Wayne, 1995), job effectiveness (Yukl & Tracey, 1992) and supervisors liking of the employee (Wayne & Ferris, 1990).

According to Porter, Allen and Angle (1981), downward influence tactics has received less conceptual and empirical attention across the various behavioral literatures than have upward influence in management and leadership discipline (Ansari & Kapoor, 1987; Deluga, 1991; Dutton & Ashford, 1993). However, the recent increasing interest in studying downward influence tactics mirrors the shifts in power distributions in many organizations. As organizations have downsized and flattened to meet the demands of competitive environments, employees in some firms have been “empowered”, with more decision-making authority vested in lower level employees (Cotton, 1993). Coupled with increased competitive pressure required employees’ involvement and empowerment to meet the need for more innovation and more productivity (Gustavsen, 1986), it seems likely that managers will have to acquire effective influence skills to convince their subordinates to perform job beyond duty. Moreover, people today are better-educated and more articulate. They can no longer be commanded in the same way as before. There need to be much more involvement and participation at work (Stewart, 1994). Thus, a better understanding of downward influence tactics will ultimately benefit many organizations

The focus on downward influence tactics by superior directed towards their subordinates is essential for effective management. In other words, to be effective, a manager must influence others to carry out


requests, support proposals and implement decisions. The success of an attempt by the superior to influence the subordinates depends on a great extent on the downward influence tactics used by the superior. The proactive downward influence is used to convince subordinates to carry out an immediate request, especially, in situations where the superior has little authority over subordinates.

TABLE 1 DEFINITION OF INFLUENCE TACTICS

Inspirational Appeals The agent makes a request or proposal that arouses target enthusiasm by appealing to target values, ideals, and aspiration, or by increasing target self-confidence.

Consultation

The agent seeks target participation in planning a strategy, activity, or change for which target support and assistance are desired, or the agent is willing to modify a proposal to deal with target concerns and suggestions.

Ingratiation The agent uses praise, flattery, friendly behaviour, or helpful behaviour to get the target in a good mood or to think favourably of him or her before asking for something.

Exchange The agent offers an exchange of favours, indicates willingness to reciprocate at a later time, or promises a share of the benefits if the target helps accomplish a task.

Legitimating The agent seeks to persuade others that the request is something they should comply with given their situation or position.

Pressure The agent uses demands, threats, frequent checking, or persistent reminders to influence the target to do what he or she wants.

Adapted from Yukl & Falbe (1990) and Yukl & Tracey (1992) Subordinates’ Competence

Boyatzis (1982) interprets competency as “an underlying characteristic of an individual which is casually related to effective or superior performance”. A related perspective here is the notion that competencies related to the willingness and ability of the employee to use his/her capacities in specific situations (Spencer, 1983). Competencies are factors contributing to high levels of individual performance and therefore, organizational effectiveness (Armstrong, 1999). McClelland (1973) who saw competencies as components of performance associated with important life outcomes and as an alternative to the traditional trait and intelligence approaches to predicting human performance. Perceived competence, which refers to the experience of feeling that one is effective in dealing with the environment (Skinner and Wellborn, 1997). Competencies used in this way refer to broad psychological or behavioral attributes that are related to successful outcomes, be it on the job or in life in general. Competencies are operationalized in the current studies as those behavioral characteristics that significantly differentiate exemplary subordinates from others. It is also important to point out that competence refers not to how competent employees actually are but rather to their internal feelings about how competent they seem to themselves from engaging in a work and solving problems in it.


Organizational Citizenship Behaviour (OCB)

Organizational citizenship behavior represents a human conduct of voluntary action and mutual aid without request for pay or formal rewards in return and now become quite a relatively new concept in performance analysis. According to George and Brief (1992), OCB is an important element of employees’ productivity as organizations cannot foresee the entire job scope required for goals attainment except the contractually stated minimum job descriptions. The construct of OCB was introduced by Bateman and Organ (1983) by drawing upon the concept of super role behaviors as articulated by Katz and Kahn (1966). Examples of employees OCB include: accepting extra duties and responsibilities at work, working overtime when needed and helping subordinates with their work (Organ, 1988; Masterson, Lewis, Goldman & Taylor, 1996). Determining why individuals engage in OCB has occupied a substantial amount of research attention in both organizational behavior and social psychology (Brief & Motowidlo, 1986; McNeely & Meglino, 1994). Past researches have suggested that there is a relationship between OCB and a host of outcomes, such as satisfaction (Bateman & Organ, 1983); commitment (O’Reilly & Chatman, 1986); perceptions of fairness (Folger, 1993; Martin & Bies, 1991; Moorman, Rohit & Zaltman, 1993; Tepper & Taylor, 2003) and perceptions of pay equity (Organ, 1988). HYPOTHESIZED RELATIONSHIP Leadership Styles and Downward Influence Tactics

Burns (1978) and Bass (1985) conceptualize leadership styles in terms of transactional and transformational characteristics. Burns (1978) views transformational leadership as a process of activating followers’ higher level needs by inspiring higher ideals and raising moral consciousness. He posits that transformational leader heightens subordinates’ motivation to accomplish goals that exceed expectations through inspiration, and by instilling pride and confidence. It also argued that transformational leader can motivate and inspire employees to perform beyond expectations, which is the criteria for success (Bass, 1985). It may be expected that transformational leaders would employ a more personal and soft influence tactics such as inspirational appeals, consultation and ingratiation (Falbe & Yukl, 1992; Yukl, 1998). There are several reasons for suspecting an association between certain influence tactics and transformational leadership. Leaders’ behaviors that inspire others to change their beliefs and values (Bass, 1997) are reminiscent of inspirational appeals. Inspirational appeals refer to the use of values and ideals to arouse an emotional response in the subordinates (Yukl, 2002; Yukl & Seifert, 2002). The request is presented in such a way that it resonates with the subordinate’s needs, values and ideals. Inspirational appeals are known to be an effective tool to raise subordinate’s enthusiasm towards the request (Yukl et al., 1996). Thus, inspirational appeals tactic is expected to be associated with transformational leaders who often communicate with vivid imagery and symbols in a way that generates enthusiasm (Yukl, 2002; Cable & Judge, 2003).

Transformational leader should also be more likely to influence subordinates by getting them personally involved and committed to a project through consultation tactic, such as encouraging them to contribute and suggest ways to improve a proposal, or help plan an activity (Falbe & Yukl, 1992; Yukl, 2002; Yukl et al., 1996; Yukl & Seifert, 2002; Yukl & Tracey, 1992; Cable & Judge, 2003). Ingratiation involves flattery and doing favor that enhance managerial liking of the subordinate (Higgins, Judge & Ferris, 2003). Downward influence tactics such as inspirational appeals, consultation and ingratiation are said to be used by transformational leaders to induce employees’ commitment through the transformation of employees’ value systems – the value systems that align with the organizational goals (Emans, Munduate, Klaver & Van de Vliert, 2003). It is thus hypothesized that:

Hypothesis 1a: Transformational leader attempts to influence subordinates will be more likely to adopt downward influence tactics that emphasize on inspirational appeals, consultation and ingratiation.


Burns (1978) contrasts transformational leadership from the transactional leader – the type of leader who invokes exchange processes in order to satisfy subordinates’ self-interests by exchanging pay and other benefits for subordinates’ effort. He suggests that transactional leadership is a style based on bureaucratic authority and legitimacy within the organization, and that transactional leaders emphasize work standards, assignments and task-oriented goals. It is also proposed that transactional leaders tend to focus on tasks completion and employee compliance, and that these leaders rely quite heavily on organizational rewards and punishments to influence employee performance. Therefore, it may be predicted that transactional leaders frequently exert influence by offering to reciprocate or exchange favors. Transactional leader may employ exchange tactics including promises of future commitments and personal incentives to gain subordinates’ help. Previous research suggests that when transactional leaders believe that softer tactics are unlikely to be effective, they resort to pressure tactic or legitimating tactic. Transactional leader may view pressure tactic as the most effective strategy for influencing subordinates using demands, persistence and repeated requests in cases where subordinates tend to watch and wait for others to do assigned tasks (Avolio, 1999). Legitimating tactic may also be efficacious for influencing subordinates to comply with requests mandated by organizational policies, rules or procedures (Kipnis, 1984). Finally, a study by Tepper (2000) provides support to the notion that transactional leaders employ exchange and pressure tactics more frequently than transformational leaders. Therefore, it is hypothesized that:

Hypothesis 1b: Transactional leadership is positively associated with downward influence tactics that emphasize on exchange, pressure and legitimating.

Leadership Styles and Subordinates’ Competence

According to leadership theorists, the performance of leader is dependent on his or her leadership style to influence subordinates with vary competency level to carry out the tasks successfully. Today, leaders are aware that they deal with diverse background of subordinates reporting to them. This has allowed them to respond differently especially with different subordinates’ competence. The importance of subordinates’ competence affecting leadership style has not been stressed and discussed extensively in the theoretical and management literature. Past research found that supervisors reacted more warmly, permissively and collegially to a subordinate when the latter performed efficiently (Lowin & Craig, 1968), while initiated more structure and showed less consideration for poor performers (Greene, 1975). Dockery and Steiner’s (1990) research findings suggest that subordinates’ ability has effect on leadership styles. The rationale behind this is that transformational leader would want to give more latitude and support to subordinates who have high ability and perform efficiently and effectively. The study of “subordinates’ ability” implied that superior exercises of leadership styles can be affected by subordinates’ competence. It can be conjectured then that if the subordinates’ competence is high, the superior may use transformational leadership, and that when subordinates’ competence is low, the superior may be expected to adopt transactional leadership style. Thus, the following hypotheses are put forward:

Hypothesis 2a: A superior exercises of transformational leadership is positively correlated with subordinates’ competence Hypothesis 2b: A superior exercises of transactional leadership is negatively correlated with subordinates’ competence.

Subordinates’ Competence and Downward Influence Tactics Based on the study by Dockery and Steiner (1990), any particular influence tactics leaders used may affect the subordinates’ competency level. Subordinates’ competence may raise the question as to whether feelings of confidence affect the influence tactics employed. One may expect that competence interact with available influence tactics in a fashion so that, low competent subordinates will be influenced differently than high competent subordinates. Hence, a potential influencing leader cannot be sure


whether or not his or her judgment will be superior or inferior to the judgment of the subordinates’ competency level. Keeping this in mind, one could formulate the following argument: when working on a task, subordinates will probably feel obligated to contribute more whenever they think that they can contribute positively. When subordinates have greater competence in their own task solution, they will expect to be able to contribute more successfully to task performance and will therefore have a stronger tendency to offer task contributions and to wield influence than when they are less competence about their own judgment, and may expect their judgment to be wrong (Littlepage, Schmidt, Whistler & Frost, 1995). However, their willingness to participate in the task will probably be greater when their superiors’ influence styles are more of consultation approach and less controlling influence tactics than by using pressure or legitimate influence tactics. Therefore, subordinates’ competence that is the extent to which subordinates effectiveness in doing their work is suspected to be associated with consultation tactics. Thus, the following hypothesis is postulated:

Hypothesis 3a: When subordinate exhibits higher competence, superior tends to use consultation tactics in his or her exercises of influence.

If leader uses pressure tactic to force low competence subordinate to comply, this may result in

negative outcome. On the other hand, it may be easier for a leader to use exchange and pressure tactics to handle less competent subordinates, because these tactics will allow the influence subordinates to decide if, and to what extent, the influence will be accepted. Thus, the following hypothesis is proposed:

Hypothesis 3b: In the case of subordinate who exhibits lower competence, superior tends to use exchange and pressure tactics in his or her influence attempts.

Subordinates’ Competence as Mediator in the Relationship Between Transformational Leadership and Consultation Tactics.

The direct relationship between leadership and influence was well supported by numerous studies (Tepper, 2000, Charbonneau, 2004; Warren, 1998; Lamude & Scudder, 1995). Some even asserted that these two concepts are inextricably linked (Burns, 1978; Gardner, 1990; Hinkin & Schriesheim, 1989). On the other link, it was also empirically generalized that leaders react differently to different subordinates’ competence (Lowin & Craig, 1968; Greene, 1975; Dansereau, Graen & Hage, 1975). It can be also surmised that the reciprocal relationship may also exist in that; the subordinate perception of own competency is related to how he or she would perceived leadership style was imposed upon him or her. Evidence also exists although limited, on the direct relationship between subordinates’ competence and influence tactics (Knippenberg, et al., 1999; Tepper, et al. 1998). These studies posited that subordinates’ competence affects the use of particular influence tactics used in their attempt to achieve desirable outcome or leader-member relations. The evidence of these multi-interaction relationships between leadership style, subordinates’ competence and influence tactics in their logical causal flow suggest that one of the variables may act as mediating variables in these interactions. Taking a cue from the study of Locke and Schweiger (1979) and Locke, Feren, McCaleb, Shaw and Denny (1980) which view subordinates’ competence as a moderating variable in the participative decision making and work performance relationship, it can be put forward that subordinates’ competence can be a mediating variable in the relationship between transformational leadership style and consultation tactics. Empirically, this can be substantiated if the existence of the third variable in this case, subordinates’ competence, can decrease or increase the total effect of transformational leadership style on the consultation tactics. Based on the implication of the previous findings on the nature of the multi-interaction relationships, it is predicted that:

Hypothesis 4: Subordinates’ competence will mediate the relationship between transformational leadership and consultation tactics.


Leadership Styles and Organizational Citizenship Behavior It is suggested that the most important effects of transformational leadership behavior should be on

extra-role behaviors that exceed the requirements of in-role expectations (Graham, 1988). Furthermore, these extra-role behaviors are best articulated by the OCB construct (Organ, 1988; Deluga, 1995; Organ & Konovsky, 1989; Podsakoff et al. 1990). OCB is a behavior, largely discretionary and seldom included in formal job descriptions. This behavior is said able to promote efficient and effective functioning of the organization (Organ, 1988). Transformational leaders motivate followers by getting them to internalize and prioritize a larger collective cause over individual interests. Individuals make contributions because in performing these acts their senses of self-worth and self-concepts are enhanced. Individuals for whom this link between the interests of self and others has not been established are less likely to make these largely discretionary, non-tangibly rewarded contributions. Results of past researches show that transformational leadership has been consistently linked to followers’ higher level of OCB (Bass, 1985; Organ, 1988; Podsakoff et al., 1990; Howell & Avolio, 1993; Wang, Law, Hackett, Wang, Chen, 2005; Schlechter & Engelbrecht, 2006; Boerner, Eisenbeiss, Griesser, 2007). Considering these past findings, the following hypothesis is suggested:

Hypothesis 5: Transformational leadership style is positively correlated with OCB. Downward Influence Tactics and Outcome

Since past investigations of influence tactics and OCB have been carried out separately, little is known about their level of distinctiveness. This omission represents a research need since both of these categories of behavior are common within organizations and both have been found to be associated with supervisor-subordinate relationship quality and important organizational outcomes. Some researchers have discovered that influence tactics are often used by superiors as a means of obtaining personal goals, promoting their own self interest, exercising social control and changing the behavior of others (Ferris & Judge, 1991; Ferris, Russ & Fandt, 1989; Kipnis et al., 1980; Barry & Watson, 1996); and successful use of influence tactics tends to reduce resistance by subordinates (Pfeffer, 1981; Tedeschi & Melburg, 1984). According to Blau (1964) and Organ (1988), employment relationship engenders feelings of personal obligation when subordinates (treated well by superiors) feel obligated to discharge it by engaging in extra-role behavior directed at helping others and the organization. Initial conceptual and theoretical work in influence tactics research and extra-role behavior suggest that inspirational appeals, consultation and ingratiation tactic used would enhance supervisor-subordinate relationship (Tedeschi & Melburg, 1984). Inspirational appeals (using emotional language to emphasize the importance of a new task and arouse enthusiasm), consultation (involving employees in the decision-making process) and ingratiation (engaging in friendly behavior toward the target to ensure the subordinate is well disposed toward the leader’s request) have been demonstrated to be effective in generating subordinates’ OCB (Yukl & Tracey, 1992; Yukl & Falbe, 1990; Kipnis et al., 1980; Tedeschi & Melburg, 1984; Wayne & Liden, 1995). Likewise, other studies recorded that superior uses of pressure, exchange and legitimating tactic are likely to be negatively linked to subordinates’ OCB (Kipnis & Schmidt, 1988; Schriesheim & Hinkin 1990; Falbe and Yukl, 1992; Sparrowe, Soetjipto, Kraimer, 2006). Thus, the following are expected:

Hypothesis 6a: Superior’s exercise of influence tactics of inspirational appeals, consultation and ingratiation will have a direct and positive effect on organization citizenship behavior. Hypothesis 6b: Superior’s exchange, pressure and legitimating tactic will have a negative effect on organization citizenship behavior.


Mediating Effects of Downward Influence Tactics on the Relationship Between Leadership Styles and OCB

There are substantial empirical supports for the direct relationship between transformational leadership styles and OCB (Chen & Farh, 1999; Ferres, Travaglione & Connell, 2002; MacKenzie et al., 2001). Same amount of support are found between transformational leadership and influence tactics (Charbonneau, 2004; Warren, 19982). There is also a widely demonstrated theoretical link between influence tactics and OCB/extra role behavior (Dulebohn, Shore, Kunze & Dookeran, 2005; Soetjipto, 2002; Sparrowe, Soetjipto & Kraimer, 2006). These multi interactions may give rise to the mediation effect with downward influence tactics provide the most cogent reason as mediator.

Transformational leader aspires, challenges and raising the subordinates’ self-confidence and enthusiasm towards goals accomplishment that is exceeding their own self-expectations (Bass, 1997, 1998; Cable & Judge, 2003; Yukl, 2002; Yukl et al., 1996). In addition, transformational leader is more inclined to influence subordinates by personally involving them in performing task assignment (Cable & Judge, 2003; Falbe & Yukl, 1992; Yukl, 2002; Yukl et al., 1996; Yukl & Seifert, 2002; Yukl & Tracey, 1992). Inspiration and involvement, in essence, represent the exercise of downward influence tactics of inspirational appeals and consultation tactic (Yukl & Tracey, 1992). Moreover, when an individual is a transformational leader and his or her influence style is perceived as inspirational and consultation, the leader should be particularly likely to employ inspirational or consultation influence tactic with subordinates to inspire and get their personal involvement in the project. Thus, subordinates would be likely to respond positively to a transformational leader when a proper use of downward influence tactic is employed. A consequence of transformational leadership is employees’ OCB. This effect is consistent with the notion that transformational leader recognizes the effectiveness of downward influence tactics of inspirational appeals and consultation tactic to attain the employees’ OCB. Based on the above discussion, the following hypothesis is proposed:

Hypothesis 7: Downward influence tactics of inspirational appeals and consultation will increase the positive relationship between transformational leadership style and organizational citizenship behavior.

Mediating Effects of Subordinates’ Competence on the Relationship Between Leadership Styles and OCB

A study by MacKenzie et al. (2001) examined the effect of transformational and transactional leadership on marketing personnel’s performance at an insurance company suggest that transformational leadership has higher influence on performance than transactional leadership. This finding supports the assumption that the transformational leadership, as compared to transactional leadership style, has a stronger relationship with in-role performance and with OCB. Locke and Schweiger (1979) and Locke et al. (1980) studied group member knowledge and competence in the context of participative decision making (PDM) and performance view competence as a potential moderator variable. Their position would be strengthened if it could be shown that participation enhances the performance of more competent employees but fails to accentuate the performance of less competent personnel. There is an alternative view of the influence of competence on this relation that is supervisors may permit their more competent (and more productive) employees to participate in decisions that affect them. In this scenario, competence (and performance) would determine the level of PDM for each subordinate. Thus, the following hypotheses are advanced.

Hypothesis 8: Transformational leadership style is positively correlated with OCB. This relationship is mediated by the subordinates’ competence.


METHODOLOGY AND RESEARCH DESIGN Sampling Design

The sample for this study includes 2000 firms. Respondent from each company comprises of the executives, managers and professionals in services, manufacturing, mining and construction companies located mainly in Klang Valley. This sample was selected from companies with a number of employees more than 35 where a more formalized structure and system of supervision are likely to exist and function (Blau & Schoenherr, 1971; Hall, 1977; Pugh, Hickson, Hinnings, MacDonald, Turner & Lupton, 1963). The companies that fulfil the above criteria were selected from the master list of the Federation of Manufacturers Malaysia (FMM), Service Directory, Construction Industry Development Board (CIDB) and Malaysian Trade and Commerce Directory, published in 2008. In order to decrease the pitfalls of inexact sampling, no more than two questionnaires were sent to the same company. Research Instruments

Each of the measurements of relevant constructs was discussed here. Downward Influence Tactics

Yukl’s Influence Behaviour Questionnaire-2000 (IBQ-2000) was used to measure downward influence tactics. Leadership Styles

The leadership style scale consists of the Transformational Leadership Behavior Inventory (TLI: Podsakoff, MacKenzie, Morrman & Fetter, 1990) that measures six dimensions including articulating a vision, providing an appropriate model, fostering the acceptance of group goals, having high performance expectations, providing individualized support and providing intellectual stimulation. A 7-item Likert scale was used to assess the transactional leadership from Leader Reward and Punishment Behavior Questionnaire (LRPQ: Podsakoff, Todor, Grover & Huber, 1984; Podsakoff, Todor & Skov, 1982). In this study, the leadership scale is treated as unidimensional. Some researchers have treated transformational and transactional scale as unidimensional by combining the scores of all dimensions belonging to the respective key styles (Podsakoff & Organ, 1986; Podsakoff, MacKenzie & Fetter, 1993; Podsakoff, Niehoff, MacKenzie & Williams, 1993; Podsakoff, MacKenzie & Bommer, 1996). The reason leadership is treated as unidimensional is to achieve construct parsimony that best differentiate the leadership style. Organizational Citizenship Behavior (OCB)

OCB scale was measured using a 7-item scale developed by Smith, Organ and Near (1983). The scale measures the altruism and compliance of OCB. Incumbent rated these items on a 7-point Likert scale ranging from 1 strongly disagree to 7 strongly agree. Subordinates’ Competence

Wagner and Morse’s (1975) self-reported measure of individual sense of competence was used to measure the employee’s task competence in lieu of a more direct measure of competency level. The instrument is made up of 23 items. All items are scaled on 5-point agree-disagree rating scales. Data Analysis Procedure

The main statistical technique used was Path Analysis. Other statistical analysis employed is the correlational analysis.


RESEARCH RESULTS AND DISCUSSIONS Sample Characteristics

From the total of 2000 questionnaires mailed, a total of 374 responses were received and only 347 were usable and 27 rejected due to incomplete answer. The response rate was approximately 17%. The highest number of respondents was from the Chinese ethnic group. By gender, 46% were male and 54% were female. More than 70% of the respondents were from companies located in the Klang Valley. The highest proportion of respondents falls into the 30-39 years age group. On the whole, the education level of the respondents was high. This was reflected in the position or the type of occupation held by the majority of the respondents. The average salary of the respondents was higher than the population’s average. The population average salary was RM2215.50/month (Source: Malaysian Economy Report, 2008). The survey also revealed the following information about the respondent’s superior. 64% superiors reported in the survey were males. A majority of them were holding medium to high management positions. Racial composition of the superiors was: 50% Chinese, 32% Malay, 11% Indian and 7% from other races. Most of the superiors were holding high positions in the company with 24% of them in the first hierarchical level. Their educational level was also predictably high with 94% of them having had tertiary education. Validating the Scales

The standardized Cronbach Alpha for each subscale is provided in Table 2. The internal consistency reliability coefficients for all the scales were satisfactory (Nunnally, 1978). All the scales had coefficient Cronbach Alpha greater than .78.


TABLE 2 MEANS, STANDARD DEVIATIONS AND PEARSON CORRELATIONS AMONG KEY VARIABLES

Variables

Means

S.D. 1 2 3 4 5 6 7 8 9 10

1 Transformational leadership 4.56

1.14 .92

2 Transactional leadership 3.71

1.47 -.63** .91

3 Subordinates’ competence 3.35

.41 .31**

-.15** .79

4 Inspirational appeals 3.33

.88

.66**

-.45**

.23** .90

5 Consultation 3.50

.82 .69**

-.41** .36**

.58** .83

6 Ingratiation 3.11

.82 .54**

-.33** .23**

.63** .48** .81

7 Exchange 2.83

.83

-.19**

.31**

-.14*

-.08

-.15**

.13* .84

8 Pressure 2.90

.88

-.35**

.40**

-.24**

-.24**

-.31*

-.16**

.43** .78

9 Legitimate

3.15

.94

-.10

.22**

.05

-.04

-.08

-.08

.31**

.46**

.82

10

Organizational citizenship behavior

4.63

1.33

.63**

-.40**

.31**

.53**

.58**

.42**

-.10

-.33**

-.16**

.89

* Correlation is significant at the .05 level (2 tailed) ** Correlation is significant at the .01 level (2 tailed) Figure in diagonal represent coefficients alpha

TABLE 3 RESULTS OF MULTIPLE REGRESSION ANALYSIS

Dependent and

independent variables

Regression coefficients

for a full model

Path coefficients

for a full model

t values for a full model


for a trimmed model

Path coefficients

for a trimmed model

t values for a trimmed model

Subordinate’s Competence

Transformational .130(.024) .356 5.410*** .112(.019) .308 6.023*** Transactional .021(.019) .075 1.143 (Constant) 2.681(.163) 16.448*** 2.838(.088) 32.310***

R2 .093 .093 F 18.809*** 36.281***

Df 2,344 1,345 Inspirational Transformational .476(.042) .615 11.409*** .512(.031) .663 16.437*** Transactional -.037(.031) -.061 -1.184 Subordinate’s Competence

.060(.090) .028 .665

(Constant) 1.092(.364) 3.001*** .990(.147) 6.753***

R2 .437 .438 F 90.575*** 270.162***

Df 3,343 1,345 Consultation Transformational .468(.037) .646 12.552*** .461(.029) .636 15.739*** Transactional .009(.028) .016 .320 Subordinate’s Competence

.316(.081) .159 3.923*** .317 (.080) .160 3.956***

(Constant) .274(.325) .844 .336(.263) 1.278

R2 .488 .489 F 110.914*** 166.756***

Df 3,343 2,344


TABLE 3 (Cont) RESULTS OF MULTIPLE REGRESSION ANALYSIS

Dependent and



for a full model

Path coefficients

for a full model



for a trimmed model

Path coefficients

for a trimmed model


Ingratiation Transformational .378(.044) .523 8.633*** .390(.033) .540 11.929*** Transactional .004(.033) .007 .118 Subordinate’s Competence

.140(.094) .071 1.483

(Constant) .902(.382) 2.363* 1.328(.154) 8.630***

R2 .290 .290 F 48.214*** 142.297***

Df 3,343 1,345 Exchange Transformational .037(.050) .051 .750 Transactional .184(.037) .327 4.984*** 1.75(.029) .311 6.068*** Subordinate’s Competence

-.207(.107) -.104

(Constant) 2.676(.433) 6.182*** 2.187(.115) 19.073***

R2 .098 .094 F 13.595*** 36.816***

Df 3,343 1,345 Pressure Transformational -.076(.050) -.098 -1.518 Transactional .190(.037) .317

.225(.029) .375 7.663***

Subordinate’s Competence

-.338(.108) -.159 -3.123*** -.384(.104) -.181 -3.689***

(Constant) 3.680(.438) 8.409*** 3.357(.383) 8.762***

R2 .192 ---- F 28.372*** 41.251***

Df 3,343 2,344


TABLE 3 (Cont) RESULTS OF MULTIPLE REGRESSION ANALYSIS

Dependent and



for a full model

Path coefficients

for a full model



for a trimmed model

Path coefficients

for a trimmed model


Legitimating Transformational .030(.058) .036 .520 Transactional .163(.043) .253 3.752*** .141(.034) .220 4.185*** Subordinate’s Competence

.164 (.126) .072 1.301

(Constant) 1.864(.509) 3.665*** 2.630(.134) 19.561***

R2 .047 .046 F 6.693*** 17.512***

Df 3,343 1,345 OCB Transformational .425(.081) .363 5.277*** .408(.071) .349 5.741*** Transactional .023(.049) .026 .479 Inspirational .233(.090) .154 2.580** .232(.082) .153 2.829*** Consultation .330(.092) .204 3.565*** .339(.092) .210 3.685*** Ingratiation

-.023(.089) -.014 -.251 Exchange

.139(.076) .086 1.830 Pressure - .137(.077) -.091 -1.778 Legitimating - .132(.065) -.094 -2.040* -.146(.056) -.104 -2.603** Subordinate’s Competence

.291(.140) .091 2.078* .309(.136) .096 2.266*

(Constant) .192(.604) .234(.470) .498

R2 .466 .463 F 34.527*** 60.752***

Df 9,337 5,341

* p < .05, ** p < .01, *** p < .005 Numbers in parentheses are the standard errors


TABLE 4 TEST OF SIGNIFICANCE OF INDIRECT EFFECT THROUGH A MEDIATOR

Degrees of freedom = N-1 = 346 * p <0.05, ** p <0.01, *** p <0.001

FIGURE 2 INDIRECT EFFECTS THROUGH MEDIATORS

X1. Transformational

Leadership

X2. Transactional Leadership

X4. Inspirational appeals

X5. Consultation

X6. Ingratiation

X7. Exchange

X8. Pressure

X9. Legitimating

X10. Organizational CitizenshipBehavior

X3. Subordinates’ Competence

(.663)

(.636)

(.540)

(.375)

(.220)

(.311)

(.153)

(.210)

(-.104)

(.308)

(.160)

(-.181)

(.096)

(.349)

Testing of Hypotheses H1a & H1b: Leadership and Downward Influence Tactic

Hypothesis H1a predicts that transformational leader attempts to influence subordinates will be more likely to adopt downward influence tactics that emphasize on inspirational appeals, consultation and ingratiation. The correlational analysis in Table 2 provides good support for H1a. In the relationship of transformational leadership to downward influence tactics, consultation tactic ranked highest among other

Measurement path

Before mediator After mediator t-statistic t=(ab)/√ (b2sa2+a2sb2)

Regression coefficient ( a)

Standard errors (sa)

Regression coefficient (b)

Standard errors (sb)

X1 → X3 → X5 0.130 0.024 0.316 0.081 3.166*** X1 → X3 → X10 0.130 0.024 0.291 0.140 1.941* X1 → X4 → X10 0.476 0.042 0.233 0.090 2.524** X1 → X5 → X10 0.468 0.037 0.330 0.092 3.451***


tactics (r = 0.69, p < .01). This was followed by inspirational appeals and ingratiation (r = 0.66, p < .01 and r = 0.54, p < .01 respectively). Moreover, the result of path analysis which partial out other effects in Table 3 indicates that this three influence tactics have positive direct effect on transformational leadership i.e. relationship between inspirational appeals (β= 0.663, p < 0.005), consultation (β= 0.636, p < 0.005) and ingratiation (β= 0.540, p < 0.005) are significantly related with transformational leadership. It was expected that consultation, inspirational appeals and ingratiation tactics represent a higher level of inner acceptance between superior and subordinate relationships. Evidence from extant literature on organizational influence and correlational results also shows that consultation, inspirational appeals and ingratiation tactics are positively associated with transformational leadership. For instance, Yukl and Seifert (2002) found that consultation, inspirational appeals and ingratiation tactics exercised by transformational leader tend to foster a more satisfied, cooperative and prolonged relationships among superiors and subordinates.

Result of a study by Charnonneau (2004) supports the transformational leader’s uses of rational persuasion, inspirational appeals and consultation as effective in generating subordinate commitment to perform a task. This finding sheds some light on the underlying influence processes at work in transformational leadership. Indeed, results suggest that leaders who use more influence methods that result in targets’ internalization of a request or task are perceived as more transformational. In turn, transformational leadership has been associated with organizational commitment (Barling, Weber & Kelloway, 1996) and team commitment (Arnold, Barling & Kelloway, 2001). Hence, the manner in which leaders make request is important because it may ultimately lead to followers’ broader commitment to the organization’s goals and values.

Among subordinates, consultation tactics emerge as a very important cue for acceptance and recognition of the superior’s influence management style as reflected in the present result. It most likely gains their compliance and least likely to provoke their resistance (Gross & Guerrero, 2000; Tepper, 2000). The present results support the general view that consultation, inspirational appeal and ingratiation tactics have a positive effect on the superior-subordinate relationship. The high degree of intercorrelations among the consultation, inspirational appeals and ingratiation tactics serves to temper the previous discussions and tends to suggest that while consultation emerges as the dominant explanation for the downward influence tactic, its effective utilization might be tied to some extent, to the superior’s exercise of a combination of other styles such as inspirational appeals and ingratiation tactics.

Hypothesis H1b predicts that transactional leadership is positively associated with downward influence tactics that emphasize on exchange, pressure and legitimating tactic. This hypothesis was supported by the data. Both the correlational and path analysis indicate that a positive and significant relationship between transactional leadership and downward influence variable of exchange (r = 0.31, p < 0.01; β = 0.311, p < 0.005), pressure (r = 0.40, p < 0.01; β = 0.375, p < 0.005),) and legitimating (r = 0.22, p < 0.01; β = 0.220, p < 0.005). The results support the general contention that transactional leader exerts influence by offering to reciprocate or exchange favours (i.e. exchange tactics) as reported in the study conducted by Tepper (2000). Transactional leaders are reward-sensitive (Stewart, 1994), making them especially likely to use tactic that is linked to exchange tactics, which is the purpose of exchange behaviors (Tedeschi & Melburg, 1984). When an individual is perceived as transactional leader by subordinates, he/she should be more likely to employ exchange tactics with their subordinates because this approach propelled the leader to action while still abiding by the formal rules of achievement in the organization.

Pressure tactics may be the most effective strategy for influencing subordinates by using demands, persistence and repeated requests when subordinates “sit and wait for others to take the necessary initiatives imposed by the tasks” (Avolio, 1999, p. 38). Study by Tepper (2000) has provided support that transactional leader employed pressure tactics more frequently than transformational leader. Finally, legitimating also may be efficacious for influencing subordinates to comply with the requests mandated by organizational policies, rules or procedures (Kipnis, 1984). Study by Vroom and Jago (1988) showed a link between legitimating to authoritarian leadership, thus implied that legitimating influence tactics is associated with transactional leadership. The present result is generally consistent with the literature,


which suggests that transactional leader employed more pressure, exchange and legitimating influence tactic to obtain organisational objectives (Kipnis, 1984). H2a & H2b: Leadership Styles and Subordinates’ Competence

Hypothesis H2a states that a superior exercises of transformational leadership is positively correlated with subordinates’ competence. The correlational analysis in Table 2 indicates that there was a positive and significant relationship between transformational leadership and subordinates’ competence (r = 0.31, p < 0.01). The result of path analysis which partial out other effects indicated the relationship between transformational leadership and subordinates’ competence to be significant (β = 0.308, p < 0.005). The result provided full support for hypothesis H2a. The result seems to show that leader would use more transformational leadership style when dealing with subordinates who are more competent as evidenced in Dockery and Steiner (1990) study. According to Dockery and Steiner (1990), subordinates’ ability affects leadership style. Their reasoning is that transformational leader would want to give more support to subordinates who have high ability so that they can perform well. Their findings which is supported in this study, implies that superior’s exercise of transformational leadership style can be affected by subordinates’ competence. This attitude can be explained from situational leadership approach in which the leader analyzes the situation and then decide the appropriate approach. The first situational model of leadership was proposed by Tannenbaum and Schmidth (1958). They described how manager should consider three factors before deciding on how to lead: (i) forces in the manager, (ii) forces in the subordinate and (iii) forces in the situation. Forces in the manager include the manager’s personal values, inclinations, feeling of security and confidence in subordinates. Forces in the subordinate include his or her knowledge and experience (thus competence), readiness to assume responsibility for decision making, interest in the task or problem and understanding and acceptance of the organization’s goal. Forces in the situation include the type of leadership style the organization values, the degree to which the group work effectively as a unit, the problem itself and the type of information needed to solve it and the amount of time the leader has to make the decision. On the other hand, the leader should refrain from using “unnecessary” approach that may be counter-productive such as using a transactional top-down exchange approach on highly competence employee.

Hypothesis H2b states that a superior exercises of transactional leadership is negatively correlated with subordinates’ competence. The result in Table 2 indicates that there was a modest correlation between transactional leadership and subordinates’ competence (r = - 0.15, p < 0.05). This relationship was however, not confirmed by the path analysis result in Table 3. The divergent results imply that the observed association of transactional leadership and subordinates’ competence is probably contributed largely by the spurious effects of other correlated variable. In this case, the positive strong correlation between transformational leadership and subordinates’ competence. Thus, the observed significant correlation between transactional leadership and subordinates’ competence could be due to the reason that they share common negative association with transformational leadership. The result suggests that the transactional style will not be necessarily being adopted if subordinate is incompetence, since true to its intention, incompetence subordinate could not be fully entrusted with work standard, assignments and task-oriented goals. With the lack of direct effect, Hypothesis H2b was not supported. H3a & H3b: Subordinates’ Competence and Downward Influence Tactics

Hypothesis H3a posits that when subordinate exhibits higher competence, superior tends to use consultation tactics in his or her exercises of influence. The correlational and path analysis results show a significant and positive relationship between the subordinates’ competence and consultation tactics (r = 0.36, p < 0.01; β = 0.160, p < 0.05). More specifically, increased subordinates’ competence was tied to the consultation tactic and thus implying positive outcome. This result implicitly shows that superior will be cautious in employing influence tactics to highly competence subordinate – leading them to choose a less risky, more participative and ego enhancing approach. This way, the superior will avoid jeopardizing his or her integrity by asserting a hard approach when his or her judgment is inferior to the highly competence subordinate. Dockery and Steiner (1990) on the other hand, attribute this superior’s behavior


to their innate intention to give more latitude and support to well performing subordinates. This pattern of superior’s response toward the subordinate’s competence was collaborated by the study of Lowin and Craig (1968) and Greene (1975). This finding has an important implication in that while certain downward influence tactics generally tied with certain leadership styles, it does not occur in isolation of superior’s predisposition or characteristic alone, but also influence by the subordinates’ characteristics, in this case, the subordinates’ competence. Thus, the adoption of downward influence tactics is a deliberate and conscious action of the superior.

Hypothesis H3b predicts that in the case of subordinate who exhibits lower competence, superior tends to use exchange and pressure tactics in their influence attempts. The result of the correlational and path analysis only shows a significant negative relationship between subordinates’ competence and pressure tactics p (r = -0.24, p < 0.01; β = -.181, p < 0.005). Thus, hypothesis H3b is partially supported. The superior will specifically avoid the use of pressure tactics when the subordinates’ is highly competence. This can be explained from the perspective of conventional wisdom itself, that is, inappropriate application of pressure or hard tactics can be counter-productive in itself and especially so when the workgroup is either high performance or of high competency level. This result also lends credibility to the previous finding by Lowin and Graig (1968) and Greene (1975) as well as explanation by Dockery and Steiner (1990). H4: Subordinates’ Competence as Mediator in the Relationship Between Leadership Styles and Downward Influence Tactics

Hypothesis H4 predicts that subordinates’ competence will mediate the relationship between transformational leadership and consultation tactics. The correlational result in Table 2 shows that there is a positive and significant association between the transformational leadership and consultation tactics (r = 0.69, p < 0.01). The effect of transformational leadership on consultation tactics in a separate path was further confirmed by the path analysis result in Table 3 (β = 0.636, p < 0.005). Related to this link, transformational leadership style is related positively to the subordinates’ competence (r = 0.31, p < 0.01; β = 0.308, p < 0.005). Also, in the next link, the correlational and path analysis results show also a significant and positive relationship between the subordinates’ competence and consultation tactics (r = 0.36, p < 0.01; β = 0.160, p < 0.005). This all positive and significant paths in the triangular relationship between the transformational leadership, subordinates’ competence and consultation tactics give rise to the speculation that the intermediate variable that is subordinates’ competence could be a mediation variable.

Further test was conducted to access the significance of an indirect effect of transformational leadership on consultation tactics through a mediator that is subordinates’ competence by Baron and Kenny (1986) and Sobel (1982) method. Result of this analysis in Table 4 shows that the subordinates’ competence contributed significantly to the increased association between transformational leadership and consultation tactics. This mediation effect is significant at 0.05 level. The evidence of the relationship between transformational leadership style and subordinates’ competence was discussed in the testing of hypotheses H2a and well supported by the extent literature. Likewise, the relationship between subordinates’ competence and superior exercises of consultation tactics was supported in the testing of hypotheses H3a with sufficient past findings (Littlepage, Schmidt, Whistler & Frost, 1995; Dockery & Steiner, 1990). The empirical support for hypothesis H4 brings out an important qualification to the intrinsically plausible explanation between the transformational leadership and consultation tactics. This is in the form of the magnification of the strength of the relationship between transformational leadership and consultation tactics in the case of highly competent subordinates. Transformational leader tends to adopt an even more submissive consultation tactics when trying to influence more competence subordinates. This was deliberately adopted to bring positive outcomes to such relationships. The nature of this interaction is quite similar to the result of the study by Locker and Schweiger (1979) and Locke, Feren, McCaleb, Shaw and Denny (1980) in which subordinates’ competence is found to moderate the relationship between participative decision making and work performance. The result gives additional support to the assertion that leaders consciously and continuously evaluate the level of competence of the


subordinates and select the appropriate influence tactics to bring desired outcomes. In overall, support for Hypothesis H4 was found. H5a & H5b: Leadership Styles and Organizational Citizenship Behavior

Hypothesis H5a predicts that transformational leadership style is positively correlated with organizational citizenship behavior. The positive relationship between transformational leadership and organizational citizenship behavior is in the hypothesized direction. The correlational result in Table 2 indicates that transformational leadership was highly related to organizational citizenship behavior (r = .63, p < 0.01). This relationship was also further affirmed by the path analysis result in Table 3 (β = 0.349, p < 0.005). This result expectedly, is quite similar to past studies (Schlechter & Engelbrecht, 2006; Ferres, Travaglione & Connell, 2002; Mackenzie, Podsakoff & Rich, 2001; Chen & Farh, 1999; Gerstner & Day, 1997) that unambiguously indicate the existence of a positive relationship between transforma-tional leadership and OCB. Moreover, the relationships between leadership and OCB have been empirically studied with the conclusion that transformational leadership was consistently linked to followers’ higher level of OCB (Goodwin, Wofford & Whittington, 2001; Wang, Law, Hackett, Wang & Chen, 2005; Mackenzie, Podsakoff & Rich, 2001). Hence, there is a strong conceptual support for the notion that transformational leaders motivate their followers to exhibit extra-role behaviors. Study by MacKenzie, Podsakoff and Rich (2001) on transformational and salesperson performance concluded that transformational leadership influences salesperson to perform “above and beyond the call of duty” and that transformational leader behaviors actually have stronger direct and indirect relationships with sales performance and OCB. Bass (1985) asserted that employees choose to perform tasks out of identification with transformational leader in the organization. He further stresses that transformational leadership can create identification with and internalization of desirable values as opposed to the limited goal of transactional leadership to produce a compliant workforce. This way, the subordinates are more encouraged to go beyond self-interest and the effect will be that they are more enthusiastic, productive, hardworking and more committed to the organization.

Hypothesis H5b predicts that transactional leadership style is negatively correlated with OCB. The correlational result in Table 2 provides evidence that transactional leadership was negatively correlated with OCB (r = -.40, p < 0.01). Although seemingly logical, this relationship was however, not affirmed by the path analysis results in Table 3. Cumulatively, a transactional leadership style did seem to influence OCB in a negative way. The direct effect of the transactional leadership on OCB was too weak and insignificant to lend unqualified support for hypothesis H5b. The logical explanation is that transactional leader uses hard approach which is viewed as ineffective in engaging subordinates’ commitment. The present result however, implies a less deterministic (expected) negative relationship between the transactional leadership on OCB relationships. Although the application of transactional leadership can be effective in certain situation, for example, Yammarino and Bass’s (1990) investigation found that transactional leadership can have a favorable influence on attitudinal and behavioral responses of employees but it generally fails to evoke a volunteeristic initiative beyond the normal call of duty. This however, does not necessarily degenerate to the extent that it creates a negative response to the OCB. H6a & H6b: Downward Influence Tactics and OCB

Hypothesis H6a states that superior exercises of downward influence tactics of inspirational appeals, consultation and ingratiation will have a significant positive association with subordinates’ organization citizenship behavior. In linking the downward influence to OCB, correlational result indicates a significant association between inspirational appeals and OCB (r = 0.53, p < 0.01), consultation tactics and OCB (r = 0.58, p < 0.01) and also ingratiation tactics and OCB (r = 0.42, p < 0.01). However, only inspirational appeals and consultation tactics with the OCB were further affirmed by the path analysis in which the path coefficient for inspirational appeals β = 0.153 was significant at the 0.005 level and consultation tactics β = 0.210 was significant at the 0.005 level. These results provide partial support for hypothesis H6a and could well indicate that the social exchange theory has prominence in explaining linkage between downward influence tactics and OCB. Thus, it would be expected that manager’s uses of


inspirational appeals and consultation tactics would encourage employees to increase their OCB (Sparrowe, Soetjipto & Kraimer, 2006; Dulebohn, Shore, Kunze & Dookeran, 2005). However, there was also an assertion that OCB was exhibited for self-interest purposes that coincides with the self-interest explanation for influence tactics used (Dulebohn et al., 2005). Irrespective of the attribution of this behavior, the superiors consciously use interpersonal or inspirational influence strategies to elicit OCB from their colleagues (Barbuto, 2000; Bass, 1985; Katz & Kahn 1978). The present result, however also collaborate the Falbe & Yukl (1992) findings in which influence tactic that rely on personal power and power sharing such as consultation tactics was more effective in motivating employees’ OCB than tactic that rely on authority and position power. On the other hand, Organ (1988) explains that the employees’ OCB is encouraged by the positive impression they perceived in their supervisors and co-workers. Some other researchers (Folger, 1993; Niehoff, 2000; Penner, Midili & Kegelmeyer, 1997; Rioux & Penner, 2001) have advanced the proposition of three distinct motives of employees engaging in OCB that is pro-social values, organizational concern and impression management. Unfortunately, the relationship between ingratiation tactics and OCB is not supported in the current study. This could be due to this tactic emphasizing on strong relationship between superior and subordinates and this does not necessarily mean leading to subordinates’ OCB.

Hypothesis H6b proposes that superior exercises of downward influence tactics of exchange, pressure and legitimating will have a significant negative association with subordinates’ organization citizenship behavior. As shown in Table 2, the correlation coefficients are significant in the case of between pressure tactics and organization citizenship behavior (r = -0.33, p < 0.01) and also between legitimating tactics and organization citizenship behavior (r = -0.16, p < 0.01) but not with exchange tactics. The path analysis result in Table 3, however failed to achieve statistical significant level. Therefore, on the strength of both the correlational and path analysis result, hypothesis H6b was not supported. The present result seems to be in consistent with the result of research finding by Sparrow et al (2006) which suggested that pressure tactics and legitimating tactics are not related to the employee’s helping behavior (OCB) although some other researchers concludes that “forcing” influence tactics is counterproductive in engaging employee commitment and motivation (Emans et al, 2003, Falbe & Yukl, 1992; Yukl and Tracey, 1992, Yukl et al., 1996). While the present result is not providing an equivocal support for the earlier proposition, it offers an interesting counter-argument that held promise of the supremacy of OCB. If OCB is not exactly affected by the ‘hard” influence attempt, this will provide empirical evidence that OCB is a valuable characteristics of the individual that is resilient and enduring and thus not easily affected by the nature of relationship between the subordinate and superior. In a way, OCB can be a highly sought after characteristic of a subordinate, since this extra-role behavior represent an intrinsic characteristics of the individuals, the propensity for this behavior remains even in the environment that is less than ideal. H7: Downward Influence Tactics as a Mediator on the Relationship Between Leadership Style and OCB

Hypothesis H7 suggests that downward influence tactics of inspirational appeals and consultation will increase the positive association between transformational leadership style and organizational citizenship behavior. There is a strong direct relationship between transformational leadership and OCB as shown in Table 3 with path coefficient of 0.349 (p < 0.005). The mediation effects of inspirational appeals is significant at the 0.01 level for the path X1 (transformational leadership) → X4 (inspirational appeals) → X10 (OCB) and the mediation effects of consultation tactics is significant at the 0.001 level for the path X1 (transformational leadership) → X5 (consultation tactics) → X10 (OCB) as shown in Table 4 or diagram in Figure 2. The result confirms that transformational leadership has significant direct relation to the subordinates’ OCB and this relationship is mediated by downward influence tactics of inspirational appeals and consultation tactics. When transformational leadership is exercised, the use of inspirational appeals and consultation tactics further increases the tendency of the subordinates to involve in OCB behavior. This finding seemed to support the result of a study conducted by Soetjipto (2002) who concludes that only inspirational appeals and consultation tactics mediate the relationship between leaders-members’ perceptions of LMX quality. A leader’s exercise of inspirational appeals and


consultation tactics may be perceived by his or her subordinates as reflecting the leader’s active support for the subordinates to get the task completed. Thus, the use of such tactics may foster a high subordinates’ OCB. The present result seems to support the assertions of some scholars who think that the concept of leadership and influence are extricably linked (Burns, 1978; Gardner, 1990; Hinkin & Schriesheim, 1989). The link is established by the appropriate selective of influence tactics that are related to a particular leadership style. The present result clearly indicate that the use of the suitable influence tactics tend to make the usage of leadership style more effective. H8: Mediating Effect of Subordinates’ Competence on the Relationship Between Transformational leadership and OCB

Hypothesis H8 suggests that transformational leadership style is positively correlated with OCB. This relationship is mediated by the subordinates’ competence. The strong direct relationship between transformational leadership and OCB is shown in Table 3 with path coefficient of 0.349 (p<0.005). From Table 4, the mediation effect of subordinates’ competence is significant at the 0.05 level for the path X1 (transformational leadership) → X3 (subordinates’ competence) → X10 (OCB). The result confirms that transformational leadership has significant direct relation to the subordinates’ OCB and this relationship is mediated by subordinates’ competence (Figure 2). When transformational leadership is exercised, the subordinates’ competence would further increase the tendency of the subordinates to involve in OCB behavior. Transformational leader may be perceived by his/her subordinates as reflecting the leader’s active support for the subordinates to get the task completed. Thus, the use of such leadership style may foster a high subordinates’ OCB. CONCLUSION

One of the most pertinent aspects of the findings is the efficacy of the links among organizational constructs that have almost never been studied simultaneously in the present scale. In terms of theoretic perspectives, seldom does one come across a comprehensive study incorporating leadership, subordinates’ competence, downward influence tactics and OCB combined in a single study, particularly in empirical research. Still, the relationships among these relationships appeared to be both plausible and significant. A combination of the leadership theory and influence theory not only showed that they are mutually reinforcing and afford a more comprehensive understanding of the organizational conducts than any perspective by itself. The benefit of the integration was also illustrated in the findings of myriad of contingent variables in this study. The present study concludes that leadership style, downward influence tactics and subordinates’ competence are the strong predictors of subordinates’ outcome. These represent a more complete configuration of variables. The present study confirms a significant linkage between leadership styles and the influence tactics and provides justification for integrating leadership theory and influence theory in the organizational behavioral studies. This study proposes that the leadership styles takes effects through the appropriate predisposed influence tactics. This study revealed that transforma-tional manager tends to use inspirational appeals, consultation and ingratiation approach to gain subordinates’ OCB. Such behavior should be promoted in organization and it should offer great practical significance. On the other hand, transactional leader tends to use influence tactics that emphasize on exchange, pressure and legitimating tactics. This offers a suggestion that transactional leader tends to influence by reciprocating and exchange of favor, with tendency to be reward sensitive and has inclination to use organizational formal and legitimate channel to achieve the goals. The mediating effect of subordinates’ competence was investigated and shed some light on how this variable strengthen or weaken the interaction between leadership styles and downward influence tactics. This study also found support for the mediation effect of subordinates’ competence on the relationship between transformational leader and consultation tactics. Specifically, transformational leader tends to use more consultation tactics to deal with subordinates who exhibit higher competence level. The study also lends support for the mediation effect of the leaders’ exercise of downward influence by the way of inspirational appeals and consultation tactics on the relationship between transformational leader and OCB. Although, most of the


previous researchers tend to attribute differences in subordinates’ OCB to leadership styles, while confirming the same, this study provide an additional insight to this evidence, that is, apart from the direct effect of leadership styles on outcomes, the outcomes implication were also largely influenced by the appropriate choice and successful use of downward influence tactics of inspirational appeals and consultation tactics. CONTRIBUTION OF THIS STUDY

This research has made several contributions. First, it is an empirical test of a more comprehensive model that comprises of leadership styles, downward influence tactics, subordinates’ competence and OCB. This comprehensive model was developed to reconcile and explain some unequivocal results in the past. Although the empirical contributions of this research are modest, they are nevertheless believed to be important. This is the first study to examine the variables in such a wide scale that involves combination of various perspectives. Moreover, a more complex model developed here is to allow for exploration of multi-interaction hypothesis. One of the contributions of this study is in the investigation of the mediating role of downward influence tactics between leadership styles and subordinates’ OCB. This mediating role has largely been overlooked and little attention has been given to empirically examine the extent of this mediation effect. For example, prior research in leadership styles has demonstrated that members’ perceptions of their leaders’ styles may differentiate their performance. In other words, previous research has only investigated the input and output components of the exchange process. Consequently, little, if any, explanation is offered on why different leaders’ style tends to generate different members’ OCB. By incorporating leader’s influence attempts and members’ responses to such attempts, the present study constitutes a contribution of influence literature in terms of providing a plausible explanation on the connection between leadership styles and subordinates’ outcome because leaders with different leadership styles exercise varied influence tactics and members respond to such tactics in various ways reflected in their OCB. In addition, subordinate’s competence and role ambiguity were found to also mediate the transformational leadership styles and downward influence tactics. MANAGERIAL IMPLICATIONS

There are several specific managerial implications that can be derived from the present study. From a practical standpoint, the research findings suggest that when the superior has a choice in the leadership styles, he/she should emphasize more on the transformational leadership in order to achieve greater OCB. Transformational leadership style seems to alter destructive influencing network created by fluctuating superior-subordinate power differences. Implication for fostering transformationally oriented organizational cultures through training and development, job and organizational design as well as human capital decisions seem important. Training in mentoring and recognizing the varying development needs of employees can nurture the transformational leader behavior. The intellectual stimulation of transformational leadership in integrative problem solving relationship should be promoted rather than the win-lose relationships. The transformational leadership style can be acquired through the learning of scenarios, role play and videotapes of actual case in organization. With suitable feedback, work productivity would increase. Similarly, organizations facing rapid environmental change would benefit from the flexibility cultivated by transformational leadership at all levels. Similarly, understanding downward influence tactics has implications for the managerial behavior. It appears that managers can elicit favorable outcomes using inspirational appeals and consultation tactics. On the other hand, while the use of pressure and exchange tactics may be effective to achieve pre-specified goal under certain situations (Soetjipto, 2002; Sparrowe, Soetjipto & Kraimer, 2006), it fails to encourage employees to engage in OCB behavior. Secondly, although it may be premature to suggest a strict guideline as to how managers should combine influence tactics, it appears that managers are more likely to be effective by invoking inspirational appeals and consultation tactics. Moreover, it seems that managers can reduce the deleterious effects of the outcome by using a combination of leadership styles and influence tactics.


However, further study is warranted to ascertain the effects of various tactical combinations and to determine how their effects vary across contexts and tasks. DIRECTION FOR FUTURE RESEARCH

Future studies of leadership styles should focus on identifying other important respondent characteristics and more importantly, on understanding the processes by which such variables impact person perceptions. In future studies, attempts should be made to incorporate additional endogenous variables such as motivation, compliance and performance of subordinate which are more indicative of organizational outcomes. In addition it may be useful to investigate whether the perception of superior leadership styles carry the same attitudinal and behavioral implication across the demographic variables such as gender, role status, race, and etc. This will complicate the study but may offer richer explanation of the organizational behavior. This study should also provide further encouragement to researchers to use elaborate models in the management and organizational theory research. Where scholarly contributions were littered with many simplistic approaches using under-represented models, researchers should make an effort in constructing elaborate and real world models that can help in building a cohesive theory. Researchers should realize that their research model should dictate the methodological approach to be applied to entangle the theoretical mysteries among the variables and not the other way round. The advent of the multivariate analysis tools is seen as the main driver in the advancement of this cause. Objective ratings of context would have been desirable. However, using objective ratings would have introduced difficult issues and required the sampling of additional organizations. The use of additional organizations, however, would have allowed for a comparison of across different industries and geographic regions. This would have made the findings of this study more generalizable and would have avoided some common method bias problems. As has often been highlighted, the strength of a particular theory is as good as its ability to consistently explain a certain phenomenon and is not expected to perform well in all of the phenomena. Thus, it would appear that complex models incorporating many interactions would call for diverse perspectives for credible explanations. Theory integration is still not wide spread and would take many research replications in different environmental conditions to confirm these findings. Thus, future researchers are encouraged to explore the organizational phenomena by trying to employ various theories to advance the knowledge and understanding of the causes and effects of the downward influence tactics. REFERENCES Aguinis, H., Nesler, M.S., Hosoda, M. & Tedeschi, J.T. (1994). The use of influence tactics in persuasion. The Journal of Social Psychology, 134, 429-38. Ansari, M. A. & Kapoor, A. (1987). Organizational context and upward influence tactics. Organizational Behaviour and Human Decision Processes, 40, 39-49. Arnold, K.A., Barling, J. & Kelloway, E.K. (2001). Transformational leadership or the iron cage: Which predicts trust, commitment and team efficacy?” Leadership & Organizational Development Journal, 22, 315-320. Avolio, B. J. & Bass, B. M. (1988). Transformational leadership, charisma, and beyond. In J. G. Hunt, B. R. Baliga, H. P. Dachler, & C. A. Schriesheim (Eds.), Emerging leadership vistas (pp.29-49). Lexington, MA: Lexington Books. Avolio, B.J. & Waldman, D.A. & Yammarino, F.J. (1991). Leading in the 1990’s: The four I’s of transformational leaderhsip. Journal of European Industrial Training, 15¸ 9-16.


Barbuto, J.E. (2000). Influence triggers: A framework for understanding follower compliance. Leadership Quarterly, 11, 365-387. Barling, J., Weber, T. & Kelloway, E.K. (1996). Effects of transformational leadership training on attitudinal and financial outcomes: A field experiment”. Journal of Applied Psychology, 81, 827-932. Baron, R. M. & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Barry, B. Watson, M. R. (1996). Communication aspects of dyadic social influence in organizations: A review and integration of conceptual and empirical developments. Communication Yearbook, 19, 269-318. Bass, B. M. (1981). Stogdill’s Handbook of Leadership (2nd Ed.). New York: Free Press. Bass, B. M. (1985). Leadership and Performance Beyond Expectations. Free Press, New York. Bass, B. M. (1990). From transactional to transformational leadership: Learning to share the vision. Organizational Dynamics, 18(3), 19-6. Bass, B. M. (1986). Implications of a New Leadership Paradigm. Binghamton: School of Management, State University of New York. Bass, B. M. (1997). Does the transactional-transformational leadership paradigm transcend organizational and national boundaries? American Psychologist, 52. Bass, B. M. (1998). Transformational leadership: Industry, military, and educational impact. Mahwah, NJ: Erlbaum. Bass, B. M. & Avolio, B. J. (1990). Transformational leadership: A response to critiques. In Chimers, M. M. & R. Ayman, (Eds.), Leadership Theory and Research: Perspectives and Directives (pp 49-80). New York: Academic Press. Bass, B. M. & Avolio, B. J. (1995). Multifactor leadership questionnaire for research. Pa Alto, CA: Mind Garden, Inc. Bass, B. M. & Avolio, B. J. (2000). MLQ-Multifactor leadership questionnaire. Redwood City, CA: Mindgarden. Bass, B. M. & Avolio, B. J., Jung, D.I. & Berson, Y. (2003). Predicting unit performance by assessing transformational and transactional leadership. Journal of Applied Psychology, 88, 207-218. Bateman, T. S. & Organ, D. W. (1983). Job satisfaction and the good soldier: The relationship between affect and employee “citizenship”. Academy of Management Journal, 26, 587-595. Bennis, W. (2002). Become a tomorrow leader. In L.C. Spears (Ed.), Focus on Leadership: Servant-Leadership for the 21st Century, pp. 101-9. Wiley, New York, NY. Berson, Y. & Linton, J. D. (2005). An examination of the relationships between leadership style, quality, and employee satisfaction in R&D versus administrative environments. R&D Management, 35(1), 51–60.


Blau, P.M. (1964). Exchange and Power in Social Life. New York: John Wiley & Sons. Blau, P.M. & Schoenherr, R.A. (1971). The Structure or Organizations. New York: Basic Books. Boerner, S, Eisenbesiss, S.A. & Griesser, D. (2007). Follower behavior and organizational performance: The impact of transformational leaders. Journal of Leadership and Organizational Studies, 13(3), 15-26. Bolino, M.C. (1999). Citizenship and impression management: Good soldiers or good actors? The Academy of Management Review, 24(1), 82-98. Bolino, M. C. & Turnley, W. H. (2003). “Going the extra mile: Cultivating and managing employee citizenship behavior”. Academy of Management Executive, 17(4), 60-71. Boyatzis, R.E. (1982). The Competent Manager: A Model for Effective Performance. John Wiley & Sons, New York, NY. Brief, A. P. & Motowidlo, S. J. (1986). Prosocial organizational behaviours. Academy of Management Review, 11, 710-725. Bryman, A. (1992). Charisma and Leadership in Organizations. London: Sage. Burns, J. M. (1978). Leadership. Harper and Row, New York, NY. Burns, T. & Stalker, G. M. (1961). The management of innovation. London: Tavistock. Bycio, P., Hackett, R. D. & Allen, J. S. (1995). Further assessments of Bass’s (1985) conceptualization of transactional and transformational leadership. Journal of Applied Psychology, 80, 468-478. Cable, D.M. & Judge, T.A. (2003). Managers' upward influence tactic strategies: The role of manager personality and supervisor leadership style. Journal of Organizational Behavior, 24(2), 197-212. Caldwell, D.R. & Burger, J.M. (1997). Personality and social influence strategies in the workplace. Personality and Social Psychology Bulletin, 23, 1003-12. Case, T., Dosier, L., Murkison, C. & Keys, B. (1988). How managers influence superiors: A study of upward influence tactics. Leadership and Organizational Development Journal, l 9(4), 25-31. Charbonneau, D. (2004). Influence tactics and perceptions of transformational leadership. Leadership and Organization Development Journal, 25(7/8), 565-577. Charbonneau, D., Barling, J. & Kelloway, E.K. (2001). Transformational leadership and sports performance: The mediating role of intrinsic motivation. Journal of Applied Social Psychology, 1, 1521-1534. Chen, X.P. & Farh, J.l. (1999). The effectiveness of transactional ad transformational leader behaviors in Chinese organizations: Evidence from Taiwan. Paper presented at the annual meeting of the Academy of Management. Chicago. Chiok, Foong, Loke. (2001). Leadership behaviours: effects on job satisfaction, productivity and organizational commitment. Journal of Nursing Management, 9(4), 191-204.


Cotton, J.L. (1993). Employee Involvement. Sage, Newbury Park, CA. Covin, J.G. & Slevin, D.P. (1989). Strategic management of small firms in hostile and benign environments. Strategic Management Journal. 10, 75-87. Crozier, M (1964). The Bureaucratic Phenomenon. University of Chicago Press. Dansereau, F., Jr., Graen, G. & Haga, W.A. (1975). A vertical dyad linkage approach to leadership within formal organizations: A longitudinal investigation of the role making process. Organizational Behavior and Human Performance, 13, 46-78. Deluga, R.J. (1991). The relationship of upward-influencing behavior with subordinate-impression management characteristics. Journal of Applied Social Psychology, 21, 1145-1160. Deluga, R. J. (1995). The relationship between trust in the supervisor and subordinate organizational citizenship behavior. Military Psychology, 7, 1-16. DePree, M. (2002). Servant leadership: three things necessary. In L.C. Spears (Ed.), Focus on Leadership: Servant-Leadership for the 21st Century, pp. 89-97. Wiley, New York, NY. Dockery, T.M. & Steiner, D.D. (1990). The role of the initial interaction in Leader-Member Exchange. Group & Organization Studies, 15(4), 395-413. Dulebohn, J.H., Shore, L.M., Kunze, M. & Dookeran, D. (2005). The differential impact of OCBs and influence tactics on leader reward behavior and performance ratings over time. Organizational Analysis, 12(1), 73-90. Dunham, J. & Klafehn, K.A. (1990). Transformational leadership and the nurse executive. Journal of Nursing Administration, 20(4), 28-34. Dutton, J.E. & Ashford, S.J. (1993). Selling issues to top management. Academy of Management Review, 18, 397-428. Dvir, T., Eden, D., Avolio, B.J. & Shamir, B. (2002). Impact of transformational leadership on follower development and performance: A field experiment. Academy of Management Journal, 45, 735-744. Eastman, K.K. (1994). In the eyes of the beholder: An attributional approach to ingratiation and organizational citizenship behavior. The Academy of Management Journal, 37(5), 1379-1391. Emans, B. J. M., Munduate, L., Klaver, E., & Van de Vliert, E. (2003). Constructive consequences of leaders’ forcing influence styles. Applied Psychology: An International Review, 52, 36–54. Erez, M. & Rim, Y. (1982). The relationship between goals, influence tactics, and personal and organizational variables. Human Relations, 35, 871-878. Erez, M., Rim, Y. & Keider, I. (1986). The two sides of the tactics of influence. Agent vs. Target. Journal of Occupational Psychology, 59, 25-39. Falbe, C. M. & Yukl, G. (1992). Consequences for managers of using single influence tactics and combinations of tactics. Academy of Management Journal, 35, 638-652.


Felfe, J. (2003). Transformationale und charismatische Fuhrung und Commitment im organisationalen Wandel. University of Halle-Wittenberg: Unpublished professional dissertation. Felfe, J. & Schyns, B (2006). Personality and the Perception of Transformational Leadership: The Impact of Extraversion, Neuroticism, Personal Need for Structure, and Occupational Self-Efficacy. Journal of Applied Social Psychology, 36 (3), 708–739. Ferres, N., Travaglione, T. & Connell, J. (2002). Trust: A precursor to the potential mediating effect of transformational leadership. International Journal of Organizational Behavior, 5, 242-263. Ferris, G. R., Dulebohn, J. H., Frink, D. D., George-Falvy, J., Mitchell, T. R. & Matthews, L. M. (1997). Job and organizational characteristics, accountability, and employee influence. Journal of Managerial Issues, 20(2), 59-71. Ferris, G.R. & Judge, T.A. (1991). Personnel/Human Resources Management: A Political Influence Perspective. Journal of Management, 17(2), 447-488. Ferris, G. R., Russ, G. S. & Fandt, P. M. (1989). Politics in organizations. In R. A. Giacalone & P. Rosenfeld (Eds.), Impression management in organization (pp. 143-170). Hillsdale, NJ: Erlbaum. Fiedler, F. (1967). A Theory of Leadership Effectiveness. McGraw-Hill, New York, NY. Floyd, S.W. & Wooldridge, B. (1997). Middle management’s strategic influence and organizational performance. Journal of Management Studies, 34(3), 465-485. Folger, G. (1993). Justice, motivation, and performance beyond role requirements. Employee Responsibilities and Rights Journal, 6(3), 239-248. George, J.M. & Brief, A.P. (1992). Feeling good–doing good: A conceptual analysis of the mood at work - organizational spontaneity relationship. Psychological Bulletin, 112, 310-329. Gerstner, C.R. & Day, D.V. (1997). Meta-analytic review of leader member exchange theory: Correlates and construct issues. Journal of Applied Psychology, 82(6), 827-871. Geyer, L. & Steyrer, J. M. (1998). Transformational leadership and objective performance in banks. Applied Psychology: An International Review, 47, 97-420. Goihl, K., Tartler, K. & Kroger, M. (2001). Mitarbeiterbefragung. Freie Universitat Berlin: unpublished report. Goodwin, V. L., Wofford, J. C. & Whittington, (2001). A theoretical and empirical extension to the transformational leadership construct. Journal of Organizational Behaviour, 22, 759-774. Graham, J. (1988). Transformational leadership: Fostering follower autonomy, not automatic followership. In J. Hunt, B. Baglia, H. Dachler, C. Schriesheim (Eds.), Emerging Leadership Vistas. Lexington Books: Lexington, MA; pp 73-79. Greene, C. N. (1975). The reciprocal nature of influence between leader and subordinate”. Journal of Applied Psychology, 60, 187-193.


Gustavsen, B. (1986). Evolving patterns of enterprise organization: The move towards greater flexibility. International Labour Review, 125, 367-382. Hambley, L., Kline, T.J.B. & O’Neill, T.A. (2005). The effects of leadership style and communication medium on team interaction styles and outcomes. Academy of Management Annual Meeting. Hater, J. J. & Bass, B. M. (1988). Superiors’ evaluations and subordinates’ perceptions of transformational and transactional leadership. Journal of Applied Psychology, 73, 695-702. Heller, F. & Yukl, G. (1969). Participation, Managerial Decision-Making and Situational Variables. Organization Behavior and Human Performance. 4, 227-241. Herman, J. B. & Hulin, C. L. (1972). Studying organizational attitudes from individual and organizational frames of reference. Organizational Behavior and Human Performance, 8, 84-108. Hersey, P. & Blanchard, K. (1977). Management of Organizational Behavior. Prentice-Hall, New York, NY. Hespanhol, A., Pereira, A.C., Pinto, A.S. (1999). Job satisfaction in Portuguese physicians in general medicine. Aten Primaria, 24, 456-61. Hickson, D. J., Hinings, C. R., Lee, C. A., Schoeck, R. E., & Pennings, J. M. (1971). A strategic contingency theory of intraorganizational power. Administrative Science Quarterly, 6, 216-229. Higgins, C.A., Judge, T.A. & Ferris, G.R. (2003). Influence tactics and work outcomes: a meta-analysis. Journal of Organizational Behavior, 24, 89-106. Hinkin, T.R. & Schriesheim, C.A. (1990). Relationships between subordinate perceptions of supervisor influence tactics and attributed bases of supervisory power. Human Relations, 43(3), 221-237. Hollander, E. & Offermann, L. R. (1990). Power and leadership in organizations: Relationship in transition. American Psychologist, 45, 179-189. House, R. (1977). A 1976 theory of charismatic leadership. In J.G. Hunt & L.L. Larson (Eds.), Leadership on the cutting edge (pp. 189-207). Carbondale, IL: Southern Illinois University Press. Howell, J. M. & Avolio, B. J. (1993). Transformational leadership, transactional leadership, locus of control, and support for innovation: Key predictors of consolidated-business-unit performance. Journal of Applied Psychology, 78, 891-902. Hunt. J. G. & Schuler, R. S. (1976). Leader reward and sanctions behavior in a public utility: What difference does it make? Working paper, Southern Illinois University. Jackson, S.E. & Schuler, R.S. (1985). A meta-analysis and conceptual critique of research on role ambiguity and role conflict in work settings. Organizational Behavior and Human Decision Processes, 36 (1), 16-78. Kahn, R. L., Wolfe, D. M., Quinn, R. P., Snoeck, J. D. & Rosenthal, R. A. (1964). Organizational stress. Studies in role conflict and ambiguity, New York: Wiley.


Kaplan, R. E. (1986). Trade routes: “The manager’s network of relationship. Organizational Dynamics, 37-52. Katz, D. & Kahn, R. L. (1966). The social psychology of organizations. New York: Wiley. Katz, .D. & Kahn, R.L. (1978). The Social Psychology of Organizations. New York: Wiley. Kelloway, E.K. & Barling, J. (2000). What we have learned about developing transformational leaders. Leadership & Organization Development Journal, 21, 355-62. Kelman, H. C. (1958). Compliance, identification and internalization: Three processes of opinion change. Journal of Conflict Resolution, 2, 51-60. Khandwalla, P.N. (1977a). The Design of Organizations. New York: Harcourt-Brace-Jovanovich. Kipnis, D. (1984). Technology, power, and control. In S.B. Bacharach and E.J. Lawler (Eds.), Research in the Sociology or Organizations, pp.125-156. JAI Press, Greenwich CT. 3. Kipnis, D. & Cosentino, J. (1969). Use of leadership powers in industry. Journal of Applied Psychology, 53, 460-466. Kipnis, D. & Lane, W.P. (1962). Self confidence and leadership. Journal of Applied Psychology, 46, 291-295. Kipnis, D. & Schmidt, S.M. (1988). Upward influence styles: Relationship with performance evaluations salary and stress. Administrative Science Quarterly, 33, 528-542. Kipnis, D., Schmidt, S. M. & Wilkinson, I. (1980). Intraorganizational influence tactics: Explorations in getting one’s way. Journal of Applied Psychology, 65, 440-452. Kipnis, D. & Vanderveer, R. (1971). Ingratiation and the use of power. Journal of Personality and Social Psychology, 17, 280-286. Kirkpartick, S.A. & Locke, E.A. (1996). Direct and indirect effects of three core charismatic leadership components on performance and attitudes. Journal of Applied Psychology, 81, 36-51. Krone, K. J. (1992). A comparison of organizational, structural and relationship effects on subordinates’ upward influence choice. Communication Quarterly, 40, 1-15. Kuhnert, K. W., & Lewis, P. (1987). Transactional and Transformational Leadership: A Constructive/ Developmental Analysis. The Academy of Management Review, 12, 648-656. Lamude, K. G. (1994). Supervisors’ influence tactics for handling managers’ resistance. Psychological Reports, 75, 371-374. Lamude, K. G. & Scudder, J. (1995). Relationship of managerial work roles to tactics used to influence subordinates. The Journal of Business Communication, 32(2), 163-173. Levin, S.L. (1999). Development of an instrument to measure organizational trust. Dissertation Abstracts International, UMI No. 9920326.


Liepmann, D. & Goihl, K. (2001). Mitarbeiterbefragung in der Verwaltung. Freie Universitat Berlin: unpublished report. Likert, R. & Likert, J. G. (1976). New ways of managing conflict. New York: McGraw-Hill. Locke, E. A. (1976). The nature and causes of job satisfaction. In M. Dunnette (Ed.), Handbook of Industrial and Organizational Psychology, Chicago, IL, pp 1297-1350. Locke, E. A., Feren, D.B., McCaleb, V.M., Shaw, K.N. & Denny, A.T. (1980). The relative effectiveness of four methods of motivating employee performance. In K. Duncan, M. Gruneberg & D Wallis (Eds.). Changes in Working Life, pp. 362-388. Chichester, England: John Wiley. Locke, E. A. & Schweiger, D.M. (1979). Participation in decision-making: One more look. In B.M. Staw (Ed.), Research in Organizational Behavior, pp. 265-339. Greenwich, CT: JAI Press. Low, G.T. (2005). Strategic Persistence and Reorientation of the Firms Following an Environmental Shift: An Empirical Study of Malaysian Manufacturing Companies, Unpublished PhD Thesis, University Malaya. Lowin, K. B. & Craig, J. (1968). The influence of level of performance on managerial style: An experimental object-lesson in the ambiguity of correlational data. Organizational Behaviour and Human Performance, 3, 440-458. Lowe, K. B., Kroeck, K. G. & Sivasubramaniam, N. (1996). Effectiveness correlates of transformational and transactional leadership: A meta-analytic review of the MLQ literature. Leadership Quarterly, l 7, 385-42. Madison, D. L., Allen, R. W., Porter, L. W., Renwick, P. A. & Mayes, B. T. (1980). Organizational politics: An exploration of angers’ perception. Human Relations, 33, 79-100. Martin, B.J. (1990). A successful approach to absenteeism. Nursing Management, 21, 45-8. Martin, C. L. & Bies, R. J. (1991). Just laid off, but still a good citizen? Only if the process is fair”. Annual Meeting of the Academy of Management, 6(3), 227-238. Masterson, S. S., Lewis, K., Goldman, B. & Taylor, M. S. (1996). Integrating justice and social exchange: The differing effects of fair procedures and treatment on work relationships. Academy of Management Journal, 43, 738-748. MacKenzie, S.B., Podsakoff, P.M. & Rich, G.R. (2001). Transformational and transactional leadership and salesperson performance. Academy of Marketing Science Journal, 29(2), 115-134. McClelland, D.C. (1973). Testing for competence rather than for intelligence. American Psychologist, 28(1), 1-14. McNeely, B.L. & Meglino, B.M. (1994). The role of dispositional and situational antecedents in prosocial organizational behavior: An examination of the intended beneficiaries of prosocial behavior. Journal of Applied Psychology, 79(6), 836-844. Meadows, D.H. (1980). The unavoidable a priori. In J. Randers (Ed.). Elements of the System Dynamics.


Mintzberg, H. (1973). The nature of managerial work. New York: Harper & Row. Misumi, J. (1985). The Behavioral Science of Leadership. University of Michigan Press, Ann Arbor. Misumi, J. and Peterson, M.F. (1985). The performance-maintenance (PM) theory of leadership: Review of a Japanese research program. Administrative Science Quarterly, 30, 198-223. Moorman, C., Rohit, D. & Zaltman, G. (1993). Factors affecting trust in marketing research relationships. Journal of Marketing, 57, 81-101. Morrison, E. W. (1994). Role definitions and organizational citizenship behavior: The importance of the employee’s perspective. Academy of Management Journal, 37, 1543-1567. Mosadeghrad, A.M. (2003a). The role of participative management (suggestion system) in hospital effectiveness and efficiency. Research in Medical Sciences, Isfahan, 8(3), 85-9. Motowidlo, S. M. (2003). Job performance, in W. C. Borman, D. R. Ilgen & R. Klimoski, (Eds.), Comprehensive Handbook of Psychology, Vol. 12: Industrial and Organizational Psychology. New York: Wiley. Mowday, R. T. (1978). The exercise of upward influence in organizations”. Administrative Science Quarterly, 23, 137-156. Niehoff, B. P. (2000). A motive-based view of organizational citizenship behaviours: Applying an old les to a new class or organizational behaviours. Proceedings, Midwest Academy of Management. Nunnally, J.C. (1978). Psychometric Theory, New York: McGraw-Hill. O’Reilly, C. & Chatman, J. (1986). Organizational commitment and psychological attachment: the effects of compliance, identification, and internalization on prosocial behavior. Journal of Applied Psychology, 71(3), 492-499. Organ, D. W. (1988). Organizational citizenship behavior: The good soldier syndrome. Lexington, MA: Lexington Books. Organ, D. W. & Konovsky, M. (1989). Cognitive versus affective determinants of organizational citizenship behavior. Journal of Applied Psychology, 74, 157-164. Parker, C.P., Dipboye, R.L. & Jackson, S.L. (1995). Perceptions of Organizational Politics: An Investigation of Antecedents and Consequences. Journal of Management, 21(5), 891-912. Payne, R.L. & Pugh, D.S. (1976). Handbook of Industrial and Organizational Psychology. Rand McNally, Chicago. Penley, L.E. & Hawkins, B. (1985). Studying interpersonal communication in organizations: A leadership application. Academy of Pedhazur (1997Journal, 28(2), 309-326. Penner, L.A., Midili, A.R. & Kegelmeyer, J. (1997). Beyond job attitudes: A personality and social psychological perspective on the causes of organizational citizenship behavior. Human Performance, 10, 111-132.


Pfeffer, J. (1981). Power in organizations. Boston: Pitman. Podsakoff, P.M., MacKexzie, S.B. & Bommer, W.H. (1996). A meta-analysis of the relationships between Kerr and Jermier’s substitutes for leadership and employee job attitudes, role perceptions and performance. Journal of Applied Psychology, 81, 380-399. Podsakoff, P.M., MacKenzie, S.B. & Fetter, R. (1993). Substitutes for leadership and the management of professionals. Leadership Quarterly, 4(1), 1-44. Podsakoff, P. M., MacKenzie, S. B. Morrman, R. H. & Fetter, R. (1990). Transformational Leader Behaviors and their Effects on Follower’s Trust in Leader Satisfaction, and Organizational Citizenship Behaviors. Leadership Quarterly, 1, 107-142. Podsakoff, P.M., Niehoff, B.P., MacKenzie, S.B. & Williams, M.L. (1993). Do substitutes for leadership really substitutes for leadership? An empirical examination of Kerr and Jermier’s situational leadership model. Organizational Behavior and Human Decision Processes, 54, 1-44. Podsakoff, P.M. & Organ, D.W. (1986). Self reports in organizational research: Problems and prospects. Journal of Management, 12(4), 531-544. Podsakoff, P. M., Todor, W. D., Grover, R. A. & Huber, V.L. (1984). Situational moderators of leader reward and punishment behaviors: Fact or fiction? Organizational Behavior and Human Performance, 34(1), 1-63. Podsakoff, P. M., Todor, W. D. & Skov, R. (1982). Effects of Leader Contingent and Non Contingent Reward and Punishment Behaviours on Subordinate Performance and Satisfaction. Academy of Management Journal, 22, 810-821. Porter, L.W., Allen, R.W. & Angle, H.L. (1981). The politics of upward influence in organizations. In L.L. Cummings and B.M. Staw (Eds.), Research in Organizational Behavior. Greenwich, CT: JAI Press, 3, 109-149. Pugh, D.S., Hickson, D.J., Hinings, C.R., MacDonald, K.M., Turner, C. & Lupton, T. (1963). A conceptual scheme for organizational analysis. Administrative Science Quarterly, 8, 289-315. Rahman, M. & Zanzi, A. (1995). A Comparison of Organizational Structure, Job Stress, and Satisfaction in Audit and Management Advisory Services in CPA Firms. Journal of Managerial Issues, l 7(3), 290. Rioux, S.M. & Penner, L.A. (2001). The causes of organizational citizenship behavior: A motivational analysis. Journal of Applied Psychology, 86, 1306-1314. Rizzo, J.R., House, R.J. & Lirtzman, S.I. (1970). Role Conflict and Ambiguity in Complex Organizations. Administrative Science Quarterly, 15(2), 150-163. Rousseau, D. M. (1978). Characteristics of departments, positions, and individuals: Contexts for attitudes and behavior. Administrative Science Quarterly, 23, 521-540. Roznowski, M. (1989). Examination of the Measurement Properties of the Job Descriptive Index with Experimental Items. Journal of Applied Psychology, 74(5), 805-814.


Ruello, S. H. (1973). Transferring managerial concepts and techniques to operating management. Advanced Management Journal. Schilit, W. K. & Locke, B. (1982). A study of upward influence in organizations. Administrative Science Quarterly, 27, 304-316. Schlechter, A.F. & Engelbrecht, A.S. (2006). The relationship between transformational leadership, meaning and organizational citizenship behavior. Management Dynamics, 15(4), 2-16. Schmidt, S. M. & Kipnis, D. (1984). Manager’s Pursuit of Individual and Organizational Goals. Human Relations, 37, 781-794. Schriesheim, C. A., & Hinkin, T. R. (1990). Influence tactics used by subordinates: A theoretical and empirical analysis and refinement of the Kipnis, Schmidt, and Wilkinson Subscales. Journal of Applied Psychology, 759, 246-257. Seltzer, J. & Bass, B. M. (1990). Transformational leadership: Beyond initiation and consideration. Journal of Management, 16(4), 693-703. Seo, Y., Ko, J., Price, J.L. (2004). The determinants of job satisfaction among hospital nurses: model estimation in Korea. International Journal of Nursing Studies, 41, 437-46. Soetjipto, B.W. (2002). Downward influence in leader-member relationships. Unpublished DBA Thesis. Cleveland State University. Skansi, D. (2000). Relation of managerial efficiency and leadership styles – Empirical study in Hrvatska Elektroprivreda D.D. Management, 5(2), 51-67. Skinner, A. & Wellborn, J.G. (1997). Children’s coping in the academic domain. In S. Wolchik, and I. Sandler (Eds.), Handbook of Children’s Coping: Linking Theory and Intervention, pp. 387-422. Plenum Press, New York. Smith, P.C., Kendall L.M. and Hulin C.L. (1969). The Measurement of Satisfaction in Work and Retirement. A Strategy for the Study of Attitudes. Chicago: Rand McNally. Smith, C. A., Organ, D. W., & Near, J. P. (1983). Organizational citizenship behavior: Its nature and antecedents. Journal of Applied Psychology, 68, 653-663. Snyder, R. A. & Bruning, N. S. (1985). Quality of vertical dyad linkages: Congruence of supervisor and subordinate competence and role stress as explanatory variables. Group & Organization Studies, 10(1), 81. Sobel, M. E., (1982). Asymptotic confidence intervals for indirect effects in structural equation models. In S. Leinhardt (Ed.), Sociological Methodology (pp. 290-312). Washington, DC: American Sociological Association. Sparrowe, R.T., Soetjipto, B.W. & Kraimer, M.L. (2006). Do leaders’ influence tactics relate to members’ helping behavior? It depends on the quality of the relationship. Academy of Management Journal, 49(6), 1194-1208. Spencer, L.M. (1983). Soft Skills Competencies. Scottish Council for Research in Education, Edinburgh.


Steward, D.M. (1994). Handbook of Management Skills, 2nd Edition. Gower Publishing Co., Aldershot. Stogdill, R.M. (1948). Personal factors associated with leadership: A survey of the literature. Journal of Psychology, 25, 35-71. Stogdill, R.M. (1963). Manual for the Leader Behavior Description Questionnaire – From XI. Bureau of Business Research. Ohio State University, Columbus, OH. Stogdill, R.M. (1974). Handbook of Leadership: A Survey of Theory and Research. Free Press, New York. Stordeur, S., Vandenberghe, C. & D’hoore, W. (2000). “Leadership styles across hierarchical levels in nursing department”. Nursing Research, 49(1), 37-43. Strange, J.M. & Mumford, M.D. (2002). The origins of vision: charismatic versus ideological leadership. Leadership Quarterly, 13, 343-77. Tannenbaum, A. (1968). Control in organizations. McGraw-Hill, New York. Thacker, R.A. & Wayne, S.J. (1995). An examination of the relationship between upward influence tactics and assessments of promotability. Journal of Management, 1(4), 739-756. Tharenou, P. & Harker, P. (1984). Moderating influence of self-esteem on relationships between job complexity, performance, and satisfaction. Journal of Applied Psychology, 69(4), 623-632. Tedeschi, J. T. & Melburg, V. (1984). Impression management and influence in the organization. In S.B. Bacharach & E.J. Lawler (Eds.), Research in the sociology of organizations (Vol. 3, pp. 31-58). Greenwich, CT: JAI Press. Tepper, B.J. (2000). Consequences of abusive supervision. Academy of Management Journal, 43, 178-190. Tepper, B.J., Eisenbach, R.J., Kirby, S.L. & Potter, P.W. (1998). Test of a justice-based model of subordinates’ resistance to downward influence attempts. Group and Organization Management, 23, 144-160. Tepper, B. J. & Taylor, E. C. (2003). Relationships among supervisors' and subordinates' procedural justice perceptions and organizational citizenship behaviors. Academy of Management Journal. 46(1), 97-05. Thomas, K. W. (1976). Conflict and Conflict Management. In M.D. Dunnette (Ed.), Handbook of Industrial and Organizational Psychology (pp. 889-935). Chicago: Rand McNally. Tirchy, N. M. & Devanna, M. A. (1986). The transformational leader. New York: Wiley. Uyterhoeven, H. (1972). General managers in the middle. Harvard Business Review, 50(92), 75-85. Vance, C. & Larson, E. (2002). Leadership research in business and health care. Journal of Nursing Scholarship, 34 (2), 165171


Wagner, F.R. & Morse, J.J. (1975). A measure of individual sense of competence. Psychological Reports, 36, 451–459. Walumba, F. O., Wang, P., Lawler, J.J. & Shi, K. (2004). The role of collective efficacy in the relations between transformational leadership and work outcomes. Journal of Occupational and Organizational Psychology, 77 (4), 515-530. Wang, H., Law, K.S., Hackett, R.D., Wang, D. & Chen, Z.H. (2005). Leader-member exchange as a mediator of the relationship between transformational leadership and followers’ performance and organizational citizenship behavior. Academy of Management Journal, 48(3), 420-432. Wayne, S.J. & Ferris, G.R. (1990). Influence tactics, affect and exchange quality in supervisor-subordinate interactions: A laboratory experiment and field study. Journal of Applied Psychology, 75, 487-499. Wayne, S. J. & Liden, R. C. (1995). Effects of impression management on performance ratings: A longitudinal study. Academy of Management Journal, 38, 232-260. Weber, M. (1946 translation). From Max Weber: Essays in Sociology. In H. H. Gerth & C. W. Mills (Eds.). New York: Oxford University Press. Wells, G. T. (1990). Influence of organizational structure on nurse manager job satisfaction. Nursing Administration Quarterly, 14, 1-8. Williams, L. J. & Anderson, S. E. (1991). Job satisfaction and organizational commitment as predictors of organizational citizenship and in-role behavior. Journal of Management, 17, 601-617. Wren, J.T. (1995). Leader’s companion: Insights on leadership through the ages. New York: Free Press. Yammarino, F.J. & Bass, B.M. (1990). Long term forecasting of transformational leadership and its effects among naval officers: Some preliminary findings. In K.E. Clark and M.B. Clark (Eds.), In Measures of Leadership, pp.151-169. Greensboro, NC: Center for Creative Leadership. Yukl, G. (1989). Managerial leadership. A review of theory and research. Yearly Review of Management, 15, 251-289. Yukl, G. A. (1998). Leadership in organizations. Englewood Cliffs, NJ: Prentice Hall. Yukl, G. (2002). Leadership in Organizations, 5th ed., Prentice Hall, Upper Saddle River, NJ. Yukl, G. A. & Falbe, C. M. (1990). Influence tactics in upward, downward, and lateral influence attempts. Journal of Applied Psychology, 75, 132-140. Yukl, G. A., Falbe, C. M. & Youn, J. Y. (1993). Patterns of influence behavior for managers. Group and Organizational Management, 18, 5-28. Yukl, G., Kim, H. & Falbe, C.M. (1996). Antecedents of influence outcomes. Journal of Applied Psychology, 81(3), 309-317.


Yukl, G. & Seifert, C.F. (2002). Preliminary validation research on the extended version of the influence behaviours questionnaire. Toronto, paper presented at the Society for Industrial and Organizational Psychology annual conference. Yukl, G. A. & Tracey, B. (1992). Consequences of influence tactics used with subordinates, peers, and the boss. Journal of Applied Psychology, 77, 525-535. Yukl, G., Lepsinger, R. & Lucia, T. (1993). Preliminary Report on the Development and Validation of the Influence Behavior Questionnaire. In K.E. Clark & B. Clark (Eds.), The impact leadership (pp. 417-427). Greensboro, NC: Center for creative leadership. Zaleznik, A. (1983). The leadership gap. Washington Quarterly, 6(1), 32-39.


A New Normal for Portfolios: Construction, Assessment and Evaluation

Jeffry Haber Iona College

The financial crisis found many investors questioning what went wrong and how they might structure their portfolio going forward to avoid similar catastrophe. Such re-examination has led to new buzz-words and what seems like new paradigms. An example is the new normal. Other frequently seen headlines deals with whether the endowment model is broken. This paper looks at the investing issues in portfolio construction, assessment and evaluation in terms of the “new normal” and the “broken endowment model” and finds that the new normal is no different than the old normal and the endowment model is actually the blind following and replication of a few industry leaders. Despite the ample rhetoric, nothing much has changed or will change. INTRODUCTION

When the financial crisis began, it began slowly. At first it was sub-prime lending that collapsed, then prime credit, then the equity markets, the international markets and culminated in a world-wide liquidity crisis where there were ready sellers and no buyers. The crisis lasted roughly 16 months, from December 2007 through March 2009. During this period virtually every asset class, save government bonds, had periods of severe impact. Once the dust had cleared institutional investors began assessing their portfolios and developed a strategy for changes to better incorporate the previously unanticipated risk elements.

At the same time, consultants, fund managers and others began developing a new set of paradigms to capitalize on the institutional investors’ travails. This includes “the new normal” and the “broken endowment model.” Well how new is the new normal and how broken is the endowment model? WHY INVEST

Every institutional investor that has a portfolio has to ask themselves “why should we invest?” The answer, most likely, will be to enable the organization to carry it outs mission. It is therefore incumbent on the organization to develop a set of policy statements surrounding how it will invest its endowment (the investment policy) and how it will spend its endowment (the spending policy). Both of these statements are required by most enacted versions of UPMIFA (Uniform Prudent Management of Institutional Funds Act).

Most investment policies boil down to two statements: - Goal of perpetuity - Goal of a real, X% return


The “X%” will be determined by the organization – private foundations have an annual 5% payout requirement, so they will usually add some element for expenses and taxes above the 5%. Spend-down organizations will not have a goal of perpetuity, so their investment policy statement might look very different.

Some investment policy statements contain additional language, some about risk (which usually does not help the investment process since it is typically a vague statement) and some organizations have a socially responsible investing policy (“we won’t invest in companies that produce, manufacture or distribute cigarettes” for example) that screens out certain companies and funds, while others have a policy that screens in certain companies and funds (“we will invest in companies that produce clean energy”).

So the short answer to “why invest” is contained in the investment goal – if you need a real return above some percentage and your goal is perpetuity, holding just cash won’t cut it. RISK PARAMETERS

Many organizations and funds talk about risk, but few understand what risk is or how it will affect them. Risk is a multi-attribute factor that takes on many different forms. Is risk:

- Quantitative or qualitative - A noun, verb or adjective - Subject to a normal distribution or some other distribution

In some cases risk is quantitative (VaR, or value at risk, the measure of how much of the investment

can be lost in a single day) or qualitative (headline risk, or the risk of adverse publicity). Often, individuals discussing risk are talking about different types simultaneously. That’s also what makes quantification of risk impossible1 – risk is too encompassing a word. Individual elements of risk can be quantified, but not the totality of what risk is. STRESS TESTING

Funds often tout that they stress test their fund, which implies a quantification of risk and the ability of the fund to handle risk. But stress testing is really adjusting existing parameters, whereas risk is more comprehensive. Donald Rumsfeld once said (and I paraphrase):

There are known knowns, Known unknowns and Unknown unknowns.

There are the things we know that we know, and there are the things that we know that we do not

know. However, there are things that we do not know that we do not know, and therein lays the elements of risk. The “unknown unknowns” by definition cannot be incorporated into a model except by a surrogate, such as a random or noise factor. What amount should be incorporated? A good start would be to take the stress tested funds and look to see that actual performance – the difference between the stress test NAV (net asset value) and the value at the bottom would be the noise factor. Going forward, incorporating this amount and publishing it may make the fund unsalable, but intellectually honest. CORRELATION

Often, investment opportunities are described in ways that appeal to investors, rather than in ways that are factually correct. There are an incredible number of funds that “do not correlate” with other investments. Non-correlation is a desired feature for certain investments, and funds take advantage of this by billing themselves as uncorrelated, even if the period of non-correlation is an extended duration.


Investors are sometimes careless in describing non-correlation, confusing it with negative correlation. Correlation is a statistical measure of the predictability of the future movement of one stream of data given the movement of a second stream. The greater the correlation, the better the predictive accuracy. Correlation ranges from -1.0 to +1.0. A correlation of -1.0 is exactly as strong as a correlation of +1.0, but is an inverse relationship (as one stream goes down, the other goes, up, and vice versa). A positive correlation is where one stream goes up the other stream will go up as well2.

Non-correlation (or uncorrelated) is some specified amount around zero. Investors and fund managers are usually not clear about what constitutes non-correlation. My personal preference is to consider high correlation as being +/- .70, low correlation as +/- .20, with the range in-between being whatever comes between low and high3. THE MARKET GOES TO 1

In times of financial stress there can be heard (time and time again) that the “market went to 1.” This is used as an excuse for why portfolios lost value; the inherent meaning is that everything went down so there was no place to put funds. Of course, no one ever says that the market went to 1 on the upside, which would imply that the investor had no acumen, they merely followed the trend.

Well does the market go to 1? Theoretically, no4 and empirically, no5 as well. It may be convenient to claim the down markets had highly correlated investment options, but even in times of extreme distress there were investment options that were non-correlated. Naturally, no one would suggest that the market literally went to 1, but that phrase is used to mean that asset allocation choices were highly correlated. I considered any correlation at or above +/- .70 as being at 1. During the tenure of the financial crisis (I used the 16 months from December 2007 through March 2009) all developed equity markets (domestic and international) were highly correlated. Emerging equity markets were just under the cutoff (.68) for high correlation (or “going to 1”). Debt markets, both government and commercial, along with commodities were non-correlated, dispelling the well-held notion that the “market went to 1.” PURPOSE OF ASSET ALLOCATION

What is the purpose of asset allocation in an institutional portfolio? Since I hold the belief that risk, on the whole, cannot be quantified (except by looking at very specific characteristics in a singular fashion) there has to be some way for an organization to establish a framework to decide on what constitutes proper investments. Asset allocation is that framework. Used correctly, each allocation grouping (or “bucket”) should represent those investments that have similar risk characteristics and are highly correlated. Unfortunately, this is often not the case.

Contemporary asset allocation is done according to the security type and where the investment is headquartered. So the stock of a United States (US) company would be part of the domestic equity allocation. And the stock of a European company would be part of the international equity allocation. If the US company made 100% of their sales (assume it is a manufacturer for the retail market) to Europe, would their risk characteristics be closer aligned with other US companies or European companies? Now assume that the raw materials they use in the production process come from the commodity markets. Are the risk characteristics still closely aligned with US companies?

Now suppose that the European company makes 100% of its sales to the US homebuilding market. Are the risk characteristics of this stock similar to other international stocks or more closely tied to the US and the real estate market? The point being that asset allocation needs to be considered on the basis of how the stock will react, not on where the headquarters of the company is.

Additional considerations need to be factored in as well, such as liquidity. Certain investment types (such as limited partnerships, mutual funds with lock ups, any fund with redemption limitations, etc) may not provide the cash flow needed by the organization when the organization needs it. Allocation has to be done a number of ways on a variety of bases.


BENCHMARKING

Benchmarks are an important part of investment management – it allows the portfolio manager to ascertain whether they made a good hiring choice. The benchmark needs to be relevant to the type of investing the fund manager is hired to do, and the benchmark usually represents the low cost option within that type of investing (i.e., an index fund).

Multi-strategy funds represent a more complex situation and usually require a benchmark that is a combination of the various universes that the manager can invest in. Overall, my experience is that most investors have a good understanding of benchmarking and apply it to fund managers appropriately.

The one area that where I find that portfolio managers leave much to be desired in the benchmarking of the portfolio as a whole, which essentially answers the question of whether the Chief Investment Officer (CIO, or anyone who is charged with the fiduciary duty of managing the portfolio) did a good job. Most overall portfolios are benchmarked against a composite return using a combination of US and international indices, such proportions determined by how the organization allocates its portfolio. I view this as incorrect, because the organization has an investment policy statement that contains a return goal – I believe that this should be the portfolio benchmark. Other benchmarks can be used as well, the evaluation need not be limited to one metric, but the return goal should be one facet of the evaluation.

A CIO who delivers exactly what was asked of him/her by the investment committee should be congratulated. However, with the prevalence of peer comparison, often times an organization that has perpetuity and low risk as guidelines and gets the appropriate return will not feel their performance was adequate when they wind up in the lowest quartile. Organizations need to be resolved in understanding their long-term goals and what that translates to on annual return basis. MEAN REVERSION

No words have been more overlooked than “past performance is not indicative of future performance.” Investors tend to flock to the managers with the current highest returns, thinking that this will continue into the future. I am beginning to think (not supported by any research I have done) that managers with good processes (low turnover, succession planning, etc) who were second quartile in return and the best choices, and those funds that are top quartile are candidates to be sold while at the top. Mean reversion seems to be getting traction as an investing philosophy.

To some extent the basic tenet of mean reversion theory is that with the spate of investment managers in the field there is very little alpha that can be imparted, and that most of the excess return is generated by luck. Luck, whether good or bad, does not last forever, and managers in the top quartile cannot be expected to stay there. CONCLUSION

If, by the “new normal,” we mean that organizations will no longer blindly follow industry leaders with goals and aims that may not be aligned, then there is a new normal. But this new normal is nothing more than having organizations ask what the long-term goals are and what should the portfolio look like in order to reach these goals. Too often investors made decisions based on what others were doing without considering whether it was appropriate for them. REFERENCES Haber, Jeffry, “The Problems in Quantifying Risk,” Journal of Applied Business and Economics,” accepted September, 2010. Haber, Jeffry, “Correlation of Uncorrelated Asset Classes,” The Journal of International Business and Economy, Volume 9, Issue 2, December 2008, pp1-12.


Haber, Jeffry, “Examining the Role of Short-Term Correlation in Portfolio Diversification,” Graziado Business Report, 2009, Volume 12, Issue 3. Haber, Jeffry, “Does the Market Go To 1?” The Journal of Business and Economics Research, Volume 8, Number 3, March 2010, pp99-102. Haber, Jeffry, “The Financial Crisis: Did the Market Go to 1? and Implications for Asset Allocation,” Economics & Business Journal: Inquiries & Perspectives, Accepted February 2011.


The Impact of the Sarbanes-Oxley Act (SOX) on the Cost of Equity Capital of S&P Firms

Sheryl-Ann K. Stephen

Butler University

Pieter J. de Jong University of North Florida

This study examines the impact of SOX on the cost of equity capital for small and large S&P firms. The provisions of SOX aim to improve internal control systems and reduce information asymmetry by improving corporate governance systems and increasing transparency. Using a fixed-effects regression model, our findings suggest that the cost of equity capital has decreased post-SOX for the overall sample of firms, but more specifically for the small firms, which are usually associated with poor internal control systems and high information asymmetry. Collectively, our results provide evidence that SOX has had a positive impact on firms. INTRODUCTION

The Sarbanes-Oxley Act (hereafter, SOX or the Act) has been in effect since July 2002 but its proposed benefits, in light of the exorbitant costs that surrounded its implementation, remain a controversial issue. A 2004 survey conducted by the Financial Executives International (FEI) found that the initial cost of compliance for the largest firms was about $4.6 million. Supporters of the legislation argue that the Act was necessary to prevent the types of financial meltdowns that were witnessed over the 2000-2002 period. However, what remains unclear is whether SOX is actually having a positive impact on public firms given the recent global financial meltdown. In this paper, the impact of the adoption of SOX on the cost of equity capital is empirically investigated. This must be of particular interest and importance to corporate insiders, investors, and indeed all market players as the cost of equity represents the compensation the market demands for owning a stock and assuming the risk of ownership. It is essential to establishing the hurdle rate for a firm’s investment projects, which in turn affects the firm’s profitability. Also, management uses the cost of equity capital in trying to determine a firm’s optimal capital structure, which remains one of the primary determinants of the health and success of any firm.

Easley and O’Hara (2004) show that the quality of information, as well as a firm’s information structure, has an effect on its cost of equity capital. The authors report that investors require a higher return to acquire stocks of firms with more private information. This means that the better the information environment, the lower the information asymmetry, and the lower the return demanded by investors because stocks will be viewed as being less risky. Therefore, to the extent that the emphasis of SOX is on improved corporate governance systems, transparency, and information quality (including the accurate


dissemination of information), the effect of the Act on the cost of equity capital is pertinent and warrants investigation.

Merton (1987) puts forth the investor recognition hypothesis. In essence, this hypothesis states that firms that receive minimal publicity and are not covered by many security analysts have a higher cost of capital. According to Merton’s hypothesis, it is expected that large firms would have a lower cost of equity capital because these firms are usually well known and garner more media attention. On the other hand, because small firms are less publicized and are usually characterized by high levels of information asymmetry, it is expected that these firms would be subjected to a higher cost of equity capital. Therefore, the impact of SOX on the cost of equity capital of small firms must also be an important consideration, especially if the benefits of SOX to the financial markets as a whole are to be adequately assessed. To our knowledge, we are the first to address this issue.

The main provisions of SOX deal with internal control, financial reporting, board and auditor independence, and insider trading. Essentially, they deal with virtually every single aspect of a firm’s governance and operations that has a financial outcome. To the extent that corporate insiders and firms that are found to be non compliant with the mandates of SOX face severe financial penalties and/or imprisonment, it can be assumed that the risk associated with disclosing inaccurate information and fraudulent behavior by firms has increased significantly post-SOX. However, scholars and practitioners are still trying to assess whether the provisions of SOX are having the desired effect on investors, corporate insiders, and the operations of public firms.

This study examines the impact of SOX on the cost of equity capital. Specifically, the following question is investigated: How has the cost of equity capital of firms changed since the implementation of the Act? Given that the cost of equity capital is fundamental to many corporate decisions, and is in part related to the very objectives that SOX is intended to achieve (e.g., improved information quality and transparency of firms’ operations), this research makes an important contribution to the entire body of work dealing with corporate governance, corporate disclosure, information quality, and the effect on the cost of equity capital. Our study also contributes to the literature that examines how government intervention via legislation affects various aspects of the financial markets.

In this research, we use the most recent and widespread piece of legislation (SOX) to examine the effect on firms’ cost of equity capital, and find that there has been a reduction in the cost of equity capital, specifically in our sample of small firms. PREVIOUS RESEARCH

Given the importance of the cost of equity capital in the decision-making activities of firms, an extensive line of research has developed over the years dealing with information, corporate disclosure, internal control systems and the cost of equity capital.

Diamond (1985) investigates the optimal release of information by firms. Using a general equilibrium model with endogenous information collection, the author shows that the release of information by firms is advantageous to shareholders. He states, “The welfare improvement occurs because of explicit information cost savings and improved risk sharing.” This supports the notion that more information is actually better than less information because of reduced information asymmetry and subsequently, reduced risk. Moreover, this leads to a lower cost of equity capital for firms.

Diamond and Verrecchia (1991) investigate the effect of corporate disclosure on liquidity and the cost of capital. The authors conjecture that under certain conditions reducing information asymmetry reduces the cost of capital. Their findings suggest that by increasing public information (corporate disclosure), there is a reduction in the level of information asymmetry, which in turn leads to a reduction in the cost of capital. They also conclude that corporate disclosure “reduces the risk-bearing capacity available through market makers.”

Botosan (1997) investigates the effect of the corporate disclosure level on the cost of equity capital by regressing firm-specific estimates of cost of equity capital on firm size, market beta, and a self-constructed measure of disclosure level. The author reports that for firms characterized by low analyst


following, increased disclosure leads to a lower cost of equity capital. However, for firms with a high level of security analyst following, the author finds no relationship between the disclosure level and the cost of equity capital. The author attributes this to her measure of disclosure, which she acknowledges may not be a powerful proxy for overall corporate disclosure especially when security analysts have a major role in the communication process.

Healy, Hutton and Palepu (1999) examine stock performance and intermediation changes surrounding increases in corporate disclosure. Their findings indicate that increased voluntary corporate disclosure is associated with better stock performance, increased analyst following, a higher level of institutional ownership, and higher liquidity. In addition, the authors surmise that increases in corporate disclosure correspond to the increased use of debt and equity financing.

Leuz and Verrecchia (2000), in exploring the economic consequences of increased corporate disclosure, surmise that a firm’s commitment to greater disclosure should lower the cost of capital that results from information asymmetries. The authors note that greater corporate disclosure can mean an increase in the quantity of disclosure as well as an increase in the quality of disclosure, or both. A commitment by corporate insiders to increase the disclosure level is reflected in a reduced level of information asymmetry between firms and their shareholders, or among buyers and sellers of the firm’s stock. Moreover, the upside is an increase in liquidity, a lesser discount at which the stock is sold, and a lower cost of equity capital. Using a cross-sectional analysis, they find that greater disclosure levels lead to benefits that are economically and statistically significant.

Botosan and Plumlee (2002) investigate the relationship between the expected cost of equity capital and three types of corporate disclosure – annual report, quarterly and other published reports, and investor relations. The authors find a negative association between the cost of equity capital and the annual report disclosure level. They find no relationship between the cost of equity capital and investor relations activities. However, they find a positive relationship between the cost of equity capital and the quarterly report disclosure level. According to the authors, while this result may be contrary to what the theory predicts, it supports managers’ claims that more timely disclosure increases the cost of capital because of greater stock price volatility.

Gomes, Gorton and Madureira (2007) investigate Regulation Fair Disclosure (Reg FD) and its impact on information and the cost of capital. They find that the implementation of Reg FD affected normal channels of information, making them more complicated, which in turn adversely affected the cost of equity capital. Interestingly, they find that small firms are more affected than large firms because security analysts stop following these small firms, leading to an increase in their cost of equity capital.

Ogneva, Subramanyam and Raghunandan (2007) examine the relationship between the cost of equity capital and internal control weaknesses (ICW) using a sample of firms that files first-time Section 404 reports with the SEC. They find a higher implied cost of equity associated with ICW firms than for a control sample of firms that did not disclose any ICW. The authors conclude that, on average, ICWs are not directly associated with a higher cost of equity capital.

Zhang (2007) finds that SOX has had a negative impact on firm value. Conversely, Li, Pincus and Rego (2008) and Jain and Rezaee (2006) document a positive effect of SOX on firm value. If the objectives of SOX are being achieved and the legislation is having a positive impact on firm value, financial markets are enjoying less information asymmetry, and investors are facing less risk as a result, it is reasonable to hypothesize a decrease in the cost of equity capital.

The special case of small firms in relation to SOX has been previously documented. Poor internal control systems and a high level of information asymmetry usually characterize small firms, and they normally attract less security analysts as well as less publicity. Doyle, Ge and McVay (2007) point out that increased transparency can be especially beneficial to small firms that are not normally subjected to close monitoring. Therefore, if SOX is having a positive influence on these small firms, it will be reflected in a decrease in the risk normally associated with ineffective internal control systems and a poor information environment. Subsequently, these small firms should enjoy a reduction in the cost of equity capital.


Ashbaugh-Skaife, et. al., (2009) examine the effect of SOX internal control deficiencies on the risk of the firm and cost of equity. They find that firms with internal control deficiencies have significantly higher idiosyncratic risk, systematic risk and cost of equity. The authors also find that any changes in the internal control effectiveness of a firm lead to significant changes in the cost of equity ranging from 50 to 150 basis points. As a result of these findings, they conclude that internal control reports have a direct impact on investors’ risk analyses and firms’ cost of equity capital. HYPOTHESES

For examining the effect of the Sarbanes-Oxley Act on the cost of equity capital in both large and small firms, we test the following two hypotheses:

H1: The cost of equity capital is lower post-SOX. H2: Changes in the cost of equity capital are more significant for small firms.

RESEARCH METHOD Data Description and Sample Selection

The data for our study come from multiple sources. Specifically, earnings forecast data are obtained from I/B/E/S, stock price data are obtained from CRSP, company financial data come from COMPUSTAT, and the macro-economic data come from the Federal Reserve Bank of St. Louis. The sample of firms comes from the S&P 500 Index and the S&P Small Cap 600 Index, and includes only those firms that have a December fiscal year end. Firms with missing data from the three databases were dropped from the sample. Also, firms with an analyst following of less than 2 were dropped from the sample since it would not be possible to calculate the dispersion. The time period under review is from January 1996 to December 2006. The pre-SOX period includes the years1996-2002, while the post-SOX period includes the years 2003 to 2006. This time period is chosen in order to minimize the sensitivity of the results to other macro-economic shocks in the economy, especially the most recent financial crisis which experts believe began in 2007.

The final sample consists of 4,642 observations, representing 422 firms. Of these firms, 230 are from the S&P 500 Index and 192 are from the S&P Small Cap Index. Approximately 37 percent of the observations are during the post-SOX period, and 46 percent of the observations represent small firms. The specific number of observations in each test may be lower depending on the omission of outliers and the cost of equity capital measure used.

In estimating the cost of equity capital, many different methods were considered. Previous studies (Gebhardt, et. al., (2001)) have used average realized returns to estimate the cost of equity capital even though expected returns may be the more appropriate measure. This is because expected returns are not directly observable. In defense of the widespread use of average realized returns, Gebhardt, et. al., (2001) state, “…in an efficient market where risk is appropriately priced, the average ex post realized returns should be an unbiased estimator of the unobservable ex ante expected returns.” Fama and French (1997) used the CAPM and a three-factor model to estimate the cost of equity capital, but found that the estimates were not precise. In fact, other estimates of the cost of equity capital, including earnings to price ratio (E/P), Gordon Growth Model, and average return on equity (ROE), all have disadvantages and imperfections (Botosan, 1997).

Against this backdrop, the literature has moved to using the positive earnings growth (PEG) ratio as a proxy for the cost of equity capital. The PEG ratio is defined as the price-earnings (PE) ratio divided by the short-term earnings growth rate. Easton (2004) surmises that the main advantage of using the PEG ratio is that it considers differences in short-run growth, thereby producing a better ranking than simply the PE ratio. The author further acknowledges that the use of the PEG ratio provides a more parsimonious method of ranking firms even though expected rates of return estimates based on the PEG ratio are biased downward. Also, Botosan and Plumlee (2005) examine the reliability of five popular methods to


estimating the risk premium, and find that the PEG ratio estimates are robust and consistently related to market risk, leverage risk, information risk, residual risk and growth. Following Botosan and Plumlee (2005), we estimate the cost of equity capital in our study using the PEG ratio as shown in Figure 1 below:

FIGURE 1 ESTIMATION OF THE COST OF CAPITAL EQUATION

PEG Ratio (Rit) = (P0/eps1) / ((100*( eps2 - eps1)/ eps1))

NOTE: P0 = stock price

epst = earnings per share P0/eps1 = PE ratio

Consistent with Gebhardt, et. al., (2001), we use risk and firm characteristics in the panel regression

analysis. For each firm, the standard deviation of returns is calculated from the previous year’s monthly returns (from CRSP) providing a measure of the market volatility. Firm size is measured as the natural log of the market value of equity, according to Ashbaugh-Skaife, et. al., (2009). Firm size, along with the number of security analysts following a firm are used to proxy for the information environment and liquidity. The rationale is that the information environment is more efficient for larger firms than for smaller firms, and also, larger firms are normally characterized by more liquidity than smaller firms (Leuz and Verrecchia, 2000); (Easley and O’Hara, 2004). In terms of earnings variability, and consistent with Fama and French (1993), forecast error, dispersion of analysts’ forecasts, and earnings volatility are all used to capture risk in the model. Methodology

Gebhardt, et. al., (2001) highlight many factors that can have an impact on the cost of equity capital. In our study, however, the primary concern is in investigating the impact of SOX on the cost of equity capital. Since panel data is being used, a model with firm fixed effects is employed to isolate the effect of SOX on the cost of equity capital, while controlling for the firm characteristics discussed above. This model is appropriate because the variables are homogenous across the firms, and the firms are fixed throughout the entire sample period. In order to get the most efficient estimates, robust t-statistics from a heteroskedasticity-autocorrelation consistent estimator are reported. We estimate the fixed-effects regression in Figure 2.

FIGURE 2 FIXED EFFECTS REGRESSION EQUATION

PEGit = a0 + a1 (SOXt) + a2 (Debt Ratioit) + a3 ln (NANit) + a4 (ANFEit) + a5 ln (Dispit) +

a6 ln (Evolit) + a7 ln (Ret Volit) + a8 ln (BE/ME Ratioit) + a9 ln (Firm Sizeit) + a10 ln (GDPit) + a11 ln (CPIit) + a12 ln (PPIit) + a13 ln (RETAILit) + uit

SOX is a dummy variable that equals one for the post-SOX period, and zero otherwise. Given that the

SOX dummy variable could reflect alternative constructs other than ‘how SOX affected firms’, we introduce into the model macroeconomic control variables that are unrelated to SOX, which may have an impact on the cost of equity capital. These variables include gross domestic product (GDP), consumer price index (CPI), producer price index (PPI) and real retail and food services sales (RETAIL).

The debt ratio (Debt Ratioit) is calculated as long-term debt divided by the year-end total common equity and is used to proxy for financial leverage. Modigliani and Miller (1958) document a positive relationship between the debt ratio and cost of equity capital. Thus, we expect a positive sign on the coefficient of the debt ratio. Security analyst following, ln (NANit), is equal to the number of analysts making a forecast for each firm. A negative sign is expected on this variable as previous research has shown that more security analyst coverage is associated with a lower cost of capital (Gomes et al., 2007).


The absolute normalized forecast error (ANFEit) is measured as the absolute value of the difference between the consensus forecast earnings per share (EPS) for firm i for year t and the actual EPS for firm i for quarter t, divided by the absolute value of the actual EPS. Dispersion of security analysts’ forecasts, ln (Dispit), is equal to the natural log of the coefficient of variation of the consensus forecasts (i.e., the standard deviation of the consensus forecast earnings over the year normalized by the mean consensus forecast). Earnings volatility, ln (Evolit), is equal to the natural log of the coefficient of variation of annual pre-tax income over the past two years (i.e., the standard deviation of the pre-tax income over the past two years divided by the mean). The return volatility, ln (Ret Volit), is calculated as the natural log of the standard deviation of the previous year’s monthly returns. Following Fama and French (1995), the book to market ratio, ln (BE/ME Ratioit), is calculated as the natural log of the book value of equity divided by the market value of equity and is used as a proxy for risk. A positive association is expected between the cost of equity capital and the forecast error, dispersion, earnings volatility and the book to market ratio.

In order to examine the impact of SOX on large and small firms, the sample is categorized into firm size quartiles with quartile 1 representing the smallest firms and quartile 4 representing the largest firms, based on the market value of equity of each firm at the end of the first year of the sample period. Consistent with previous research (e.g., Brennan, et. al., 1993), it is expected that smaller firms with less security analyst following will have a higher cost of equity capital.

Table 1 summarizes the descriptive statistics separately for the overall sample and for the four size groups. Panel A displays descriptive statistics for the overall sample, while Panels B and C display descriptive statistics for the size groups. For each variable, the mean, median and standard deviation (SD) are reported. Panel A shows that the average cost of equity capital (PEG) for the entire sample is 7%. As expected, in Panels B and C, we note that for the smallest firm size quartile (Q1), the average cost of equity capital is 10%, while for the largest firm size quartile (Q4), the average cost of equity capital is 6%. In terms of security analyst coverage, the mean number of analysts for the smallest firms is 5.29 while for the largest firms, the mean number of analysts is 20.47. These findings are consistent with Brennan, et. al., (1993), Brenann, et. al., (1998), and Gebhardt, et. al., (2001) who found that smaller firms with less security analyst following have a higher cost of equity capital.

TABLE 1 DESCRIPTIVE STATISTICS (1996-2006)

Panel A: All Firms Variable Mean Median SD Cost of capital (PEG) 0.07 0.03 0.13 SOX 0.36 0 0.48 Debt Ratio 0.57 0.45 0.50 Analysts (NAN) 12.46 11.00 8.15 Forecast Error (ANFE) 0.13 0.05 0.16 Dispersion -3.15 -3.52 1.42 Earnings Volatility -1.09 -1.09 0.97 Return Volatility -0.03 0.07 0.99 BE/ME Ratio 0.42 0.39 0.22 Firm Size 8.02 7.97 1.48 GDP 9.19 9.20 0.09 CPI 5.17 5.18 0.08 PPI 4.92 4.89 0.09 RETAIL 11.97 11.98 0.08


Panel B: Firms by Size (Quartiles 1 and 2) Variable Mean Median SD

Q1 Q2 Q1 Q2 Q1 Q2 Cost of capital (PEG) 0.10 0.07 0.04 0.03 0.18 0.11 Debt Ratio 0.54 0.52 0.38 0.42 0.52 0.46 Analysts (NAN) 5.29 9.41 4.00 8.00 3.32 5.28 Forecast Error (ANFE) 0.19 0.13 0.10 0.06 0.19 0.15 Dispersion -2.39 -3.09 -2.77 -3.42 1.56 1.38 Earnings Volatility -0.72 -1.08 -0.66 -1.15 0.92 0.97 Return Volatility -0.47 -0.04 -0.46 0.06 0.92 0.93 BE/ME Ratio 0.56 0.44 0.58 0.41 0.22 0.21 Firm Size 6.12 7.35 5.99 7.37 0.28 0.37 GDP 9.16 9.20 9.19 9.22 0.09 0.09 CPI 5.15 5.18 5.15 5.19 0.07 0.08 PPI 4.89 4.93 4.88 4.89 0.07 0.09 RETAIL 11.95 11.98 11.98 11.99 0.07 0.07 Panel C: Firms by Size (Quartiles 3 and 4)

Variable Mean Median SD Q3 Q4 Q3 Q4 Q3 Q4 Cost of capital (PEG) 0.07 0.06 0.03 0.03 0.12 0.10 Debt Ratio 0.60 0.63 0.49 0.48 0.47 0.51 Analysts (NAN) 14.46 20.47 14.00 19.00 6.00 7.78 Forecast Error (ANFE) 0.10 0.08 0.04 0.03 0.14 0.13 Dispersion -3.40 -3.70 -3.69 -3.98 1.24 1.12 Earnings Volatility -1.15 -1.33 -1.12 -1.34 0.99 0.91 Return Volatility 0.10 0.27 0.26 0.46 0.96 0.99 BE/ME Ratio 0.40 0.30 0.37 0.25 0.20 0.17 Firm Size 8.64 9.98 8.67 10.07 0.39 0.40 GDP 9.19 9.21 9.20 9.22 0.09 0.09 CPI 5.17 5.19 5.18 5.19 0.08 0.08 PPI 4.92 4.93 4.89 4.90 0.09 0.10 RETAIL 11.97 11.98 11.98 11.99 0.08 0.08 NOTE: Bold text in panels B and C indicates that the firms in quartiles 1 and 2 (small firms) are significantly different from the firms in quartiles 3 and 4 (large firms) at p = 0.05 (two-tailed). Differences in means and medians are examined using a Wilcoxon rank-sum t-test. EMPIRICAL RESULTS Cost of Equity Capital Post-SOX

Table 2 presents the results on the impact of SOX on cost of equity capital for the overall sample. It is hypothesized that cost of equity capital has decreased post-SOX. This hypothesis is motivated by the


mandates of the Act, which require that public firms increase the transparency and accuracy of all of their financial operations, improve their internal control systems as well as their corporate governance systems. If the legislation is having the desired effect, and the information environment as well as investor confidence have improved, public firms will be viewed as being less risky which will subsequently impact the cost of equity capital.

Table 2 (Panel A) shows a negative and statistically significant relationship between the cost of equity capital and SOX in the absence of the macroeconomic control variables (t-statistic = -2.76). When the macroeconomic control variables are introduced in the model, the results confirm the negative and statistically significant relationship between the cost of equity capital and SOX (t-statistic = -2.64). This means that the cost of equity capital has decreased post-SOX providing support for the notion that the implementation of SOX has been beneficial to public firms. These findings are also consistent with the research of Ashbaugh-Skaife, et. al., (2009) who find that firms with less internal control deficiencies have a lower cost of equity capital. Given that one of the primary objectives of SOX is to improve firms’ internal control systems, our results point to a possible achievement of that objective. Also consistent with prior studies (Brennan, et. al., 1998), the findings show a negative and significant association between the number of security analysts, firm size and the cost of equity capital.

TABLE 2

REGRESSION OF THE COST OF EQUITY CAPITAL

Panel A: Model without macroeconomic variables. Variable Coefficient t-Statistics SOX -0.019 -2.76*** Debt Ratio -0.003 -0.20 Analysts (LNAN) 0.006 0.55 Forecast Error (ANFE) 0.017 0.49 Dispersion -0.004 -1.66* Earnings Volatility 0.005 0.97 Return Volatility -0.005 1.53 BE/ME Ratio 0.051 3.41*** Firm Size -0.028 2.59*** Constant 0.345 4.02*** Adjusted R2 0.05 Panel B: Model with the macroeconomic variables. Variable Coefficient t-Statistics SOX -0.033 -2.64*** Debt Ratio 0.006 0.36 Analysts (LNAN) -0.001 -0.11 Forecast Error (ANFE) 0.028 0.84 Dispersion -0.004 -1.97** Earnings Volatility 0.005 0.96


Return Volatility -0.005 1.59 BE/ME Ratio 0.064 4.18*** Firm Size -0.020 -1.92* GDP 1.777 1.57 CPI -0.794 -1.56 PPI 0.428 3.36*** Retail -1.755 -1.89* Constant 6.982 2.53** Adjusted R2 0.07 *** Denotes statistical significance at the 1% level in two-tailed tests. ** Denotes statistical significance at the 5% level in two-tailed tests. * Denotes statistical significance at the 10% level in two-tailed tests. Cost of Equity Capital Post-SOX by Firm Size

Table 3 presents the results on the effect of SOX on the cost of equity capital by firm size. Quartile 1 represents the smallest firms while quartile 4 represents the largest firms. The hypothesis is that the effect of SOX on the cost of equity capital post-SOX for small firms will be more significant than the effect on large firms. Panel A displays the results for quartile 1, which represents the smallest firms based on the market value of equity. A negative and significant sign on the SOX dummy variable provides evidence that the cost of equity capital for small firms have decreased post-SOX (t-statistic = -1.73). The results lend empirical support to the view that there has indeed been an improvement in the overall operations of firms post-SOX. This is being reflected in a decrease in the perceived riskiness of the smallest public firms and in turn, a decrease in the cost of equity capital. Panel B presents the results for quartile 2. The results are consistent with the findings in quartile 1, as well as with the findings in the overall sample. For this group of firms, the cost of equity capital has decreased post-SOX (t-statistic = -1.78).

The results for quartile 3 and 4, which represent the larger firms are presented in Panels C and D. Interestingly, the negative relationship between the implementation of SOX and the cost of equity capital is not significant in these groups of larger firms (t-statistic = -0.14 and -1.46, respectively). We conclude that the reduction in the cost of equity capital post-SOX differs across the quartiles, with the statistically significant results being found in the sample of smaller firms. These results are not surprising for the following reasons: As one would expect, the internal control systems, financial operations, and the information environment of the largest firms were already very efficient before SOX was implemented. In addition, one would associate less information asymmetry with the largest firms. Therefore, we can surmise that any decrease in the cost of equity capital post-SOX for these large firms are not as significant as the reductions in the cost of equity capital for the smaller firms in quartiles 1 and 2. These small firms may have benefitted more from the provisions of SOX in terms of less systematic risk, firm-specific risk, information asymmetry, and improved internal control systems.

TABLE 3

REGRESSIONS OF THE COST OF EQUITY CAPITAL BY FIRM SIZE

Panel A: Quartile 1; N = 367

Variable Coefficient t-Statistics SOX -0.0655 -1.73* Debt Ratio -0.023 -0.61 Analysts (LNAN) -0.002 -0.06


Forecast Error (ANFE) 0.145 1.47 Dispersion -0.012 -1.68* Earnings Volatility -0.037 -1.66* Return Volatility -0.006 -0.53 BE/ME Ratio 0.125 2.31** Firm Size 0.034 0.46 GDP 4.037 1.08 CPI -2.184 -1.02 PPI 1.104 1.85* Retail -3.583 -1.28 Constant 11.579 1.46 Adjusted R2 0.08 Panel B: Quartile 2; N = 529

Variable Coefficient t-Statistics SOX -0.029 -1.78* Debt Ratio 0.012 0.78 Analysts (LNAN) 0.002 0.15 Forecast Error (ANFE) -0.005 -0.08 Dispersion -0.007 -1.88* Earnings Volatility 0.008 0.64 Return Volatility -0.008 -1.39 BE/ME Ratio 0.048 2.01** Firm Size -0.012 -0.39 GDP 0.392 0.34 CPI -0.197 -0.38 PPI 0.405 2.21** Retail -0.566 -0.51 Constant 2.259 0.59 Adjusted R2 0.06 Panel C: Quartile 3; N = 538

Variable Coefficient t-Statistics SOX -0.003 -0.14 Debt Ratio -0.014 -0.57 Analysts (LNAN) -0.009 -0.35 Forecast Error (ANFE) 0.117 1.30


Dispersion -0.008 -1.38 Earnings Volatility 0.013 1.73* Return Volatility -0.003 -0.45 BE/ME Ratio 0.074 2.68*** Firm Size -0.012 -0.39 GDP 2.386 1.21 CPI -0.936 -1.04 PPI 0.125 0.55 Retail -2.200 -1.36 Constant 8.894 1.85*

Adjusted R2 0.08

Panel D: Quartile 4; N = 672

Variable Coefficient t-Statistics SOX -0.033 -1.46 Debt Ratio 0.024 0.67 Analysts (LNAN) 0.010 0.38 Forecast Error (ANFE) 0.033 0.64 Dispersion 0.000 0.05 Earnings Volatility 0.010 2.01** Return Volatility -0.004 -0.71 BE/ME Ratio 0.053 2.25** Firm Size -0.055 -2.98*** GDP -0.117 -0.11 CPI 0.015 0.02 PPI 0.274 1.60 Retail -0.190 -0.22 Constant 2.602 0.89 Adjusted R2 0.07 *** Denotes statistical significance at the 1% level in two-tailed tests. ** Denotes statistical significance at the 5% level in two-tailed tests. * Denotes statistical significance at the 10% level in two-tailed tests. CONCLUSIONS

Investors, and other proponents of the Act, stress the point that SOX’s main goal of improving the accuracy and quality of financial reporting benefits internal control systems, the information environment, corporate governance and management. This, in turn, leads to more investor confidence and efficient capital markets. In a further defense of SOX, they point out that misconduct by insiders should be reduced


because of increased transparency, leading to growth, wise risk-taking, increased confidence in the capitalist system, higher rates of investment and economic progress.

The effect of SOX on the cost of equity capital is a matter of considerable importance and interest to corporate insiders, investors, security analysts, and indeed all market participants. The benefits and costs of the Act are still being heavily debated primarily because they have not been properly established and quantified. Our paper presents empirical evidence of the effect of SOX on the cost of equity capital. For a sample of 422 firms, the findings suggest that the cost of equity capital has decreased post-SOX for the overall sample. However, when small and large firms are examined separately, the results only hold for the sample of smaller firms. The reduction in the cost of equity capital disappears for the larger firms.

Our study sheds some light on the ongoing debate about the impact of SOX on the welfare of small firms. Opponents of SOX believe that the costs of compliance are too burdensome for these small firms to realize any benefits. Proponents, on the other hand, hold the view that SOX will improve the efficiency of these small firms. The findings presented here give credence to the view that SOX has had a positive impact on small firms. The decrease in the cost of equity capital experienced by the sample of small firms points to a reduction in the overall riskiness of these small firms, and supports the notion that small firms do indeed stand to benefit from the provisions of SOX.

Further, the evidence presented here indicates that smaller firms that initially are characterized by poor internal control systems and low disclosure levels (high information asymmetry) experience a decrease in their cost of equity capital after the implementation of SOX. This reduction in the cost of equity capital is not seen in larger firms that traditionally enjoy good internal control systems and low information asymmetry. We, therefore, surmise that for small firms, SOX has been effective in decreasing the information risk to investors and has led to a lower cost of equity capital. REFERENCES Ashbaugh-Skaife, H. Collins, D.W., Kinney Jr., W.R. & Lafond, R. (2009). The Effect of SOX Internal Control Deficiencies on Firm Risk and Cost of Equity. Journal of Accounting Research, 47 (1), 1-43. Botosan, C. A. (1997). Disclosure Level and the Cost of Equity Capital. Accounting Review, 72 (3), 323-349. Botosan, C. A., & Plumlee, M. A. (2002). A Re-examination of Disclosure Level and the Expected Cost of Equity Capital. Journal of Accounting Research, 40 (1), 21-40. Botosan, C. A., & Plumlee, M. A. (2005). Assessing Alternative Proxies for the Expected Risk Premium. Accounting Review, 80 (1), 21-53. Brennan, M. & Hughes, P. (1991). Stock Prices and the Supply of Information. Journal of Finance, 46 (5), 1665-1691. Brennan, M., Jegadeesh, N., & Swaminathan, B. (1993). Investment Analysis and the Adjustment of Stock Prices to Common Information. Review of Financial Studies, 6 (4), 799-824. Brennan, M., Chordia, T., & Subrahmanyam, A. (1998). Alternative Factor Specifications, Security Characteristics, and the Cross-section of Expected Returns. Journal of Financial Economics, 49 (3), 345-373. Diamond, D. W. (1985). Optimal Release of Information by Firms. Journal of Finance, 40 (4), 1071–1094.


Diamond, D. W. & Verrecchia, R. E. (1991). Disclosure, Liquidity, and the Cost of Capital. Journal of Finance, 46 (4), 1325–1359. Doyle, J., Ge, W., & McVay, S. E. (2007). Accruals Quality and Internal Control Over Financial Reporting. Accounting Review, 82 (1), 1141-1170. Easley, D., & O’Hara, M. (2004). Information and the Costs of Capital. Journal of Finance, 59 (4), 1553–1583. Easton, P. D., (2004). PE Ratios, PEG Ratios, and Estimating the Implied Expected Rate of Return on Equity Capital. The Accounting Review, 79 (1), 73-95. Fama, E., & French, K. R. (1993). Common Risk Factors in the Returns of Stocks and Bonds. Journal of Financial Economics, 33 (2), 3-56. Fama, E., & French, K. R. (1995). Size and Book-to-Market Factors in Earnings and Returns. Journal of Finance, 50 (1): 131-155. Fama, E., & French, K. R. (1997). Industry Costs of Equity. Journal of Financial Economics, 43 (2), 153-193. Financial Executives International Study, (2004). Gebhardt, W. R., Lee, C., & B. Swaminathan (2001). Toward an Implied Cost of Capital. Journal of Accounting Research, 39 (1), 135-176. Gitman, L., & Mercurio, V. (1982). Cost of Capital Techniques Used by Major U.S. Firms: Survey and Analysis of Fortune’s 1000. Financial Management, 11 (4), 21-29. Gomes, A., Gorton, G., & Madureira, L. (2007). SEC Regulation Fair Disclosure, Information, and the Cost of Capital. Journal of Corporate Finance, 13 (2-3), 300-334. Healy, P. M., Hutton, A., & Palepu, K. (1999). Stock Performance and Intermediation Changes Surrounding Sustained Increases in Disclosure. Contemporary Accounting Research, 16 (3), 485-520. Jain, P. K., & Rezaee, Z. (2006). The Sarbanes-Oxley Act of 2002 and Security Market Behavior: Early Evidence. Contemporary Accounting Research, 23 (3), 629-654. Lang, M. H., & Lundholm, R. J. (1996). Corporate Disclosure Policy and Analyst Behavior. Accounting Review, 71 (4), 467-492. Leuz, C., & Verrecchia, R. E. (2000). The Economic Consequences of Increased Disclosure. Journal of Accounting Research, 38 (Supplement), 91–124. Li, H., Pincus, M., & Rego, S. (2008). Market Reaction to Events Surrounding the Sarbanes-Oxley Act of 2002 and Earnings Management. Journal of Law and Economics, 51 (1), 111-134. Merton R. C. (1987). A Simple Model of Capital Market Equilibrium with Incomplete Information. Journal of Finance, 42 (3), 483–510.


Modigliani, F., & Miller, M. H. (1958). The Cost of Capital, Corporate Finance, and the Theory of Investment, American Economic Review, 48 (2), 261-297. Ogneva, M., Subramanyam, K.R., & Raghunandan, K. (2007). Internal Control Weakness and Cost of Equity: Evidence from SOX Section 404 Disclosures. Accounting Review, 82 (5), 1255-1297. Wooldridge, J. M., 2002. Econometric Analysis of Cross Section and Panel Data, Cambridge, MA: MIT Press. Zhang, I. (2007). Economic Consequences of the Sarbanes-Oxley Act of 2002. Journal of Accounting and Economics, 44 (1), 74-115.


Comparing GDP Indexed Bonds to Standard Government Bonds

Lillian Kamal University of Hartford

Malek Lashgari

University of Hartford

This paper discusses the construction of a new financial instrument whose payoffs are linked to the overall performance of the U.S. economy. The idea behind this financial instrument, known as a GDP bond, has been explored by financial economists for several years. Since there is no historical data for GDP bonds in the U.S., we construct a simulation process through which the payoff function for a GDP bond could be compared with other high quality bonds during the 1947-2010 period. It is observed that the GDP bond has a superior performance to other high quality bonds in a risk return framework. INTRODUCTION

Today’s macroeconomic climate is one where there is weak economic growth, burgeoning budget deficits and higher than average risk in financial instruments. At the time of writing, although the U.S. economy is showing some signs of recovery, the Federal Reserve has downgraded its growth forecasts for the United States through 2013, and the stock market continues to be volatile. The U.S. government budget deficit is at $15.25 trillion, which is more than 100% of the 2010 value of U.S. GDP. The federal funds rate is still at a range between 0% and 0.25%, and thus market interest rates in general are rather low. Many firms have had their credit rating downgraded, and debate continues in Congress regarding the debt ceiling for the United States government. In fact, the U.S. economy itself has had its credit rating downgraded slightly. Thus the investment climate is one where there are greater risks while at the same time returns are comparatively low. The financial climate is thus ripe for an instrument that could offer a multi-faceted solution to these problems. This paper discusses a financial instrument that could mitigate the rise in U.S. government debt, during down-turns in economic activity, provide a new variety of low default risk instrument for risk-averse investors, and mitigate inflation risk.

Rational individuals in the pursuit of wealth accumulation seek to mitigate risks faced when investing in different financial instruments. One primary risk associated with financial instruments, especially in economies that are on up-swings of the real business cycle, is inflation risk1. A rational investor would therefore contemplate earning the prevailing real rate of interest while being compensated for inflation. For a bond seeker, one way of doing this is purchasing a bond that is indexed to inflation. An existing example of such an instrument is TIPS (U.S. Treasury Inflation Protected Securities) whose par value and annual cash flows change with the Consumer Price Index (CPI).

Another option, not currently offered in the United States, would be a GDP indexed bond (GDP bond). A GDP bond is a financial instrument that provides investors with periodic realized real returns


together with observed inflation as measured by the GDP deflator. This paper shows one way of constructing these bonds using a comparatively more intuitive model as compared to the existing literature – the result is a bond that mitigates inflation risk, whose coupon payments maintain ex ante purchasing power expectation, and whose returns compare favorably with U.S. government intermediate and long term securities. The GDP bonds, as formulated in this paper, possess the aforementioned properties.

Argentina has achieved great success with its GDP indexed bond, which works by way of GDP warrants. The bonds are designed to pay investors when economic growth rates exceed government projections. Argentina has been experiencing a growth spurt since 2010 and holders of GDP warrants are expected to realize returns of about 24%2. These bonds became a very useful way for Argentina to restructure debt from the economic crisis she experienced earlier, and now that economic growth has been re-established, investors are realizing returns that keep at pace with this economic growth. Thus countries that are seeking to restructure debt and encourage investors (both domestic and international) may seek to follow Argentina’s example and issue nominal GDP indexed bonds.

The remainder of the paper is divided into an overview of related literature, the bond pricing model, simulations, and supporting econometric analysis, and the conclusion. REVIEW OF THE LITERATURE

Baker and Wurgler (2009) show that during the 1962-2005 period, U. S. government bonds and common stock of large, well established, dividend paying companies tend to move together over time. The reasons appear to be the business cycle, changes in investor sentiments and the required risk premium. In a business cycle contraction investors tend to concentrate in government bonds and larger company common stock due to their cash flow predictability. Furthermore, perceptions of increased risk appear to have a lesser impact on the required risk premiums for larger companies as compared to the smaller ones.

The potential benefits of GDP bonds are analyzed by scholars in the past. Li (2002) reviews data during 1958-2001 and finds that the correlation between common stock and bond returns is stronger during rising uncertainty regarding long term expected inflation. According to this observation, GDP bonds should perform better than either common stock or regular bonds during rising inflation due to their inflation hedging properties.

Schroeder, Heinemann, Kruse and Meitner (2004), show by way of simulations that GDP bonds tend to outperform regular bonds in the presence of unexpected rise in GDP, and under-perform in cases of unanticipated slowdown in the economy. Ruban, Poon and Vonatsos (2008) analyze GDP bonds using Monte Carlo simulation and conclude that indexation to the nominal growth in GDP appears to be the proper design. This conclusion is based on an evaluation of cash flows, default risk, investors risk aversion, and economic fluctuations.

Griffith-Jones and R. Sharma (2006) provide various benefits of GDP bonds for the issuing institutions and express that the idea goes back to 1980s citing Lessard and Williamson (1985) and Shiller (1993). The benefit of GDP bonds for the issuing countries is expected to be stabilization of government expenditures during the business cycles since the bond’s cash flow expenses for the issuer would decline during recessions, as the government revenue also diminishes. Investors meanwhile may benefit from a GDP bond as they share in the growth of the economy. The authors further review experiences of countries including Bulgaria, Bosnia and Costa Rica in the 1990s as well as Argentina in 2006. The problems with those bonds appear to be their complexity and pricing issues. Griffith-Jones and Sharma further explain that GDP bonds appear to have the risk return characteristics of common stock.

Kamstra and Shiller (2009) express the idea for a bond to be issued by U.S. Treasury “with a coupon tied to the United States’ current dollar GDP,” with a coupon of one-trillionth of the GDP. Such a bond is expected to provide growth of income as well as a compensation for inflation. The authors continue to state that the GDP bond would provide a financial security that would reflect the growth in labor income


that constitutes about two-thirds of the national income. They further state that GDP bonds would be expected to provide a return close to common stock with half the risk.

An important benefit of any inflation protected security is highlighted by Kamstra and Shiller (2009) in the discussion on the composition of retirement funds. On average about one-third of retirement funds assets are made up of inflation protected assets and fixed income assets. This would seem to indicate that a significant factor in investment decision in such funds is low risk-load. GDP indexed bonds fit in perfectly in such a scenario, especially for the savvy pension fund managers. Such managers can hedge against risk in down-turns by moving money from risky sectors of the market into GDP indexed bonds (recessionary periods are also when governments would typically want to issue these instruments in order to raise funds for expansionary fiscal policies) and thus avoid credit risk and other forms of risk that arise with corporate bonds and the stock market. GDP BOND PRICING MODEL

One way to initiate a GDP bond is to set the issue price at one trillionth of GDP with a total return tied to the growth of GDP. This structure for the GDP bond is expected to provide price transparency and competitive yield for it in the capital markets. As shown in this paper the price of such a bond is expected to closely track the reported GDP. In addition, random fluctuations over the short time horizon should provide opportunities for trading in the derivative markets. By providing simulations comparing accumulated wealth resulting from the GDP bond and other high quality bonds we show its superior performance in a risk return framework.

In a GDP bond an investor will receive periodic incomes that are based on the growth in the economy as measured by the real growth in GDP as well as the inflation embedded in the GDP deflator. TABLE 1 provides information regarding price and total return for the bond. As shown in TABLE 1, the GDP bond will have a book value in line with concurrent GDP while earning the prevailing real return.

As shown in TABLE 1, the principal value of the bond, as the investor’s wealth, will rise (or fall) in line with changes in GDP and fair value of the bond meanwhile should be the same as GDP at the time of transactions. Discrepancies however may arise due to the uncertainty regarding the true value of GDP at the time of transactions. An investor in a GDP bond would thereby earn a real return on investment as well as being fully compensated for inflation as measured by the GDP deflator.

Let ri = r + ei (1) where ei ~( o , constant variance); error terms possess zero mean and constant variance over time, and r being a persistent component of real return. Then the year-end principal value of GDP bond would be GDPo (geometric average of real return)t

= GDPt , for t= 1,2,3 years. In this manner, the year-end principal value of GDP bond in any given year would be in line with the observed value of GDP. Furthermore, as shown in TABLE 1, the periodic cash flows resulting from a GDP bond would be GPD0 (geometric average of real return) t-1 * rt .

This implies a GDP bondholder would expect to earn observed real returns as well as inflation prevailing during the holding period time horizon. These features for the GDP bond are similar to cash flow patterns in common stock. In addition, the possibility of earning the real return, adjusted for inflation, for GDP bonds, tends to help in the convergence of interests of bondholders and common stockholders.


TABLE 1 GDP BOND PRICING MODEL

GDPo denotes nominal GDP at the initial issue price divided by $1 Trillion, rt= real growth rate (%) in GDP during year = t. ROI denotes return on investment. Hedging Against Inflation

The inflation hedging properties of a GDP bond would make it a desirable investment with other financial instruments. Past studies have shown that common stock can have inflation hedging properties (see for example, Bodie (1976), Ely and Robinson (1997)), especially in the long run, and regardless of whether the inflation has real or monetary causes. However, Ely and Robinson (1997) while showing that on average stocks hedge against inflation in the long run, have also highlighted the U.S. as an exception to this case. Common stockholders hoping for a real return in line with growth in the economy thus may or may not be compensated for expected inflation. This strengthens the case for the introduction of a GDP indexed bond market in the United States. The introduction of U.S. Treasury Inflation Protection Bonds (TIPS) in January 1997 appears to help in protecting bondholders against expected and unanticipated inflation while earning a real return. The real return on TIPS, however, is constant for its tenure. For example, a buyer of a 10-year TIPS in 2010 will earn the prevailing real return in 2010 during the next 10 years together with the later observed inflation in the years ahead.

One perceived problem with TIPS is that real return changes over time since the real return tends to rise during the recovery phase of the business cycle and falls towards the trough. Assuming an efficient capital market, an investor in TIPS would more likely buy TIPS toward the peak of the business cycle in order to lock into a higher real rate. While TIPS theoretically resolve the inflation issue, regular bondholders will have to forecast a reasonable real return commensurate with the later prevailing state of economy for as long as 30 years.

In spite of the fact that TIPS are supposed to perfectly protect an investor against inflation risk, evidence has emerged that suggests that 10 year expected inflation rates calculated from TIPS data averages about 50 basis points below those of professional inflation forecasters (Carlstrom and Fuerst, 2004). The illiquid nature of the TIPS market may be one reason why inflation forecasts from these markets tend to under-estimate actual expected inflation. All of these reasons combined make the case ever stronger for the introduction of a new instrument that better matches both inflation rates and movements in the real business cycle.

The biggest selling point for a GDP indexed bond is that the investor has a large essence of risk removed. Rational investors seek to maximize return at any time while controlling for their risk tolerance levels. A typical yield curve (term structure of interest rates) displays this behavior on the part of investors who seek higher returns for longer term securities. In fact, past research has shown that the term

Yr ROI Year-End Principal Value Income Earned or Received

Real Interest Earned

1 r1 GDPo (1+r1) = GDP1 GDPo * r1 r1 2 r2 GDPo (1+r1) (1+r2) = GDP2 GDPo (1+r1) r2 r2 3 r3 GDPo (1+r1) (1+r2) (1+r3 )= GDP3 GDPo (1+r1) (1+r2)r3 r3

T rt

GDPo

t

avggeometric

returnreal of = GDPt

for t= 1,2,3 years

GDPo

1

returnreal of

−

t

avggeometric

*rt

rt


structure (slope of the yield curve) is an important predictor of economic activity (see for example, Estrella and Hardouvelis, 1991). The advantage of GDP indexed bonds is that the problem of an inverted yield curve is largely mitigated. The return on the bond moves in tandem with the growth rate of real GDP and so has co-movement with the real business cycle. Therefore coupon rates are continually adjusting with the rate of economic growth. Fisher’s hypothesis regarding the relationship between the nominal and real return is stated as shown in eq. (2).

NR = RR + I + I * RR (2) where NR denotes nominal return, RR denotes real return and I denotes inflation. Thereby, the nominal return desired by regular bondholders is the combination of forecasted real return, inflation and interaction between the two. At the time of purchasing a bond, a rational investor is expected to require the expected real return and inflation during the planning time horizon. Since the periodic cash flows of the bond are fixed, regular bond holders would suffer from unanticipated inflation. However, regular bondholders tend to benefit from a subsequent decline in inflation, real return or slowdown in economic activity since their return is fixed. Such a stable return is not, however, beneficial to an existing bondholder in the case of unpleasant surprises during rising inflation. Meanwhile, bond holders as well as common stockholders tend to suffer from unanticipated inflation.

A GDP bond, as formulated in this paper, resolves both issues of observed real return and inflation. Its property of providing year-by-year observed real return appears to make it superior to TIPS. Meanwhile, the combined properties of providing real return and inflation for GDP bonds would make its return profile dominant to regular bonds. Regular bondholders would have to assess both the real return and inflation in the upcoming years during the time remaining to maturity of the bond.

In a GDP bond both the real return and inflation can vary. If investors are likely to assume that business recoveries are accompanied by both higher real returns and inflation, they would more likely buy GDP bonds during the trough of the business cycles or during recessionary times, whereas TIPS would more likely be purchased at the peak. These securities would thereby provide reasonably good returns for active investors. Given that the time horizon for recoveries is longer than the downturns, the payoff for a buy and hold investment in GDP bonds should dominate those of TIPS. Inflation Indices

The question arises as to whether nominal GDP adequately captures true inflation. If so, then nominal GDP must have some correlation with consumer prices. Inflation is usually gauged using either the Consumer Price Index (CPI) or the GDP Deflator. These indexes provide information regarding changes in prices for representative bundles of goods and services. Eq.(3) shows the regression of changes in the CPI onto changes in the GDP Deflator.

%ΔGDP deflatort = a + β*%Δ CPIt-1 + εt (3)

The results for eq.(3) are shown in TABLE 2 and indicate that the lagged inflation rate (measured using the CPI) is an important explanatory variable for the current inflation rate measured according to the GDP deflator. This result is logical given that the CPI tracks prices of goods and services consumed by consumers whereas the GDP deflator tracks prices of all final goods and services in the economy and is therefore based on a large bundle of goods and services.


TABLE 2 REGRESSION OF LAGGED CPI ONTO THE GDP DEFLATOR

Intercept Lagged CPI Coefficient value 0.374698 0.475886 Std. Error 0.053423 0.040087 t-statistic 7.013816** 11.87124** Prob. 0.0000 0.0000 Sample: 1947 to 2010 Adjusted R-squared: 0.50 **Significant regressors.

Services including wages and salaries have a large share of GDP. In addition, corporate profits and proprietors’ income comprise a reasonably large portion of GDP. A sample component of GDP for 2007 is shown in TABLE 3.

TABLE 3 COMPONENTS OF US GROSS DOMESTOC PRODUCT

Sector % of GDP a. services, including wages & salaries 60% b. corporate profits 12% c. proprietors’ income 8% d. other? 20% Total 100%

As shown in TABLE 3, services including wages and salaries account for 60 percent while corporate profits and proprietors’ income for 20 percent of GDP in 2007. These observations appear to show that since CPI does not properly account for these components it may not provide adequate information regarding inflation. In particular, final value of financial wealth should be in line with the growth in the wages, salaries and profits of business enterprises.

That is, the GDP deflator appears to play an important role in testing for inflation hedging properties of financial assets. A GDP linked bondholder would benefit from the growth in the service sector and the associated rise in profit. Such variables may not be reflected in CPI and as such TIPS (inflation linked bonds) may not be as good as GDP bonds in compensating investors for inflation. In addition, an investor would like to benefit from the growth in the economy and the current structure of U.S. economy is heavily concentrated in the service sector.

The sociological evolution in the capital markets appears to help in explaining factors affecting accumulation of wealth as Piketty and Saez (2003) state that the share of wage income has outpaced the share of financial and real assets in such a way that the wealthy are now due to “the work income.” Higher wages appear to follow the growth in the economy and thereby investors in GDP bonds will share in the prosperity of business enterprises. Thereby, lenders, in line with wage earners and owners of the business enterprises, will all share in the profitability of the corporation. Selected Issues

The fair value of a GDP bond may differ from its current price. This is because GDP is estimated on a quarterly basis, subject to periodic revisions. Thereby, the true price of a GDP bond at any point in time will not be known until several months later. Such a mispricing, however, can be resolved in the


derivatives markets. For example, a swap may be purported for an exchange of a fixed price (or total return) at the end of each quarter with the later observed values (or total return).

Meanwhile, a GDP bond is subject to random price fluctuations during each quarter. This drawback is dealt with financial contracts such as futures, options and swaps. Through these instruments, a counter-party will agree to pay the reported price at the end of the quarter, earning a profit for assuming the risk involved. The pricing models developed by Kruse, Meitner and Schroder (2005) can mitigate these issues.

An investor in a GDP bond may be provided with a choice of receiving periodic total returns comprising of the observed real return as well as inflation, or a lump sum at a later time. At any rate, the total payoffs will be the same and in line with the growth of the economy. The income generated from a GDP bond, either in a lump sum or periodic, would account for as regular income and since GDP includes tax revenues, it would be automatically indexed to inflation in taxes. GDP BOND SIMULATION RESULTS Data

Total returns data are sourced from Ibbotson’s Yearbook (Morningstar), for both long term and intermediate U.S. government bonds3. Nominal GDP data and inflation indexes are sourced from the Federal Reserve Bank of St. Louis Database (FRED)4. The period of focus for all simulations and estimations of real returns on actual instruments in this paper is 1947-2010. Methodology

As Kamstra and Shiller (2009) propose, a GDP bond would generally be expected to be one of long maturity. If such a bond were issued in the United Kingdom, it would probably be a perpetuity. Since no perpetuities of the bond variety are offered in the United States, a perpetual term GDP indexed bond would widen the U.S. bond market.

A bond that attempts to follow the real business cycle of an economy would need to reflect the changes in the price of real variables in an economy. The question then arises as to whether an aggregate price index such as the GDP deflator or the CPI does indeed correlate with changes in the prices of real variables such as employment. If so, then an aggregate price index should be highly correlated with a cost index of real variables. The cost index of choice here is the Employment Cost Index (ECI), sourced from the Bureau of Labor Statistics. TABLE 4 shows the results of regressing the inflation rate (measured using the GDP deflator) on to the rate of change in the ECI as shown in eq. (4).

%ΔECIt = a + β * %Δ GDP Deflatort + εt (4)

TABLE 4 REGRESSION OF GDP DEFLATION ONTO EMPLOYMENT COST INDEX (ECI)

Intercept GDP deflator Coefficient value 0.380613 0.779100 Std. Error 0.044042 0.042676 t-statistic 8.642030** 18.25622** Prob. 0.0000 0.0000

Sample: 1947 to 2010 Adjusted R-squared: 0.70 **Significant regressors.

RESULTS AND ANALYSIS

The principal of the bond is taken to be $1000. The simulation assumes annual returns on GDP bonds (these returns are annualized nominal GDP growth rates). The continuous compounding method is then


used to compare the bond returns with those of competing instruments. This section discusses the results of the simulation and comparison exercises. The simulated nominal GDP indexed bond is compared to the following instruments:

a. Treasury bill b. Intermediate term government bond c. Long term government bond d. Large company bond e. Large company common stock

Thus the comparisons are done over instruments of varying risk and maturity. A priori it is not expected that the simulated bond will out-perform the corporate bonds or common stock, but these checks are carried out for robustness purposes.

The results indicate that the simulated GDP bond performs very well in comparison to typical competing instruments that are issued by the government. Since this is a long term bond and its returns are based on long run nominal GDP, it easily out-performs short term Treasury Bills. The gap between the GDP indexed bond and actual returns on instruments lessens as the term to maturity of the instrument increases. Thus the simulated bond still out-performs the intermediate term and long term government bonds, and thus would be a good option for a cautious investor who would like to hedge against inflation risk and enjoy a return that matches the growth rate of the real economy.

It is rather startling that the simulated bond out-performs long term corporate bonds. Typically even well rated corporations offer returns that beat government instruments since there is an inherent credit risk even with AAA rated corporations. However, the simulated GDP indexed bond allows the long term investor to enjoy higher returns with a lower risk level. This is a significant selling point for the simulated bond as this indicates that medium risk tolerant consumers who typically seek AAA rates corporate bonds or mutual funds that are primarily composed of such bonds would achieve higher returns with a government instrument.

It is of course not surprising that the simulated bond cannot outperform long term common stock. An investor with a higher tolerance for risk would therefore earn much more with long term stocks than with the simulated bond. The simulated instrument would thus primarily appeal to investors with a lower risk tolerance.


CHART A SIMULATED GDP BOND VS. TREASURY BILL

FVGB: Simulated GDP Bond FVIGB: Actual returns based on Treasury Bills.

CHART B SIMULATED GDP BOND VS. INTERMEDIATE TERM GOVERNMENT BONDS

FVGB: Simulated GDP Bond FVIGB: Actual returns based on intermediate term government bonds.

0.0000

10000.0000

20000.0000

30000.0000

40000.0000

50000.0000

60000.0000

70000.0000

1946

-01-

01

1949

-01-

01

1952

-01-

01

1955

-01-

01

1958

-01-

01

1961

-01-

01

1964

-01-

01

1967

-01-

01

1970

-01-

01

1973

-01-

01

1976

-01-

01

1979

-01-

01

1982

-01-

01

1985

-01-

01

1988

-01-

01

1991

-01-

01

1994

-01-

01

1997

-01-

01

2000

-01-

01

2003

-01-

01

2006

-01-

01

2009

-01-

01

FVGB

FVTB

0.0000

10000.0000

20000.0000

30000.0000

40000.0000

50000.0000

60000.0000

70000.0000

1946

-01-

01

1949

-01-

01

1952

-01-

01

1955

-01-

01

1958

-01-

01

1961

-01-

01

1964

-01-

01

1967

-01-

01

1970

-01-

01

1973

-01-

01

1976

-01-

01

1979

-01-

01

1982

-01-

01

1985

-01-

01

1988

-01-

01

1991

-01-

01

1994

-01-

01

1997

-01-

01

2000

-01-

01

2003

-01-

01

2006

-01-

01

2009

-01-

01

FVGB

FVIGB


CHART C SIMULATED GDP BOND VS. LONG TERM GOVERNMENT BONDS

FVGB: Simulated GDP Bond FVIGB: Actual returns based on long term government bonds.

CHART D SIMULATED GDP BOND VS. LARGE CORPORATE BONDS

FVGB: Simulated GDP Bond FVIGB: Actual returns based on the long term corporate bonds.

0.0000

10000.0000

20000.0000

30000.0000

40000.0000

50000.0000

60000.0000

70000.0000

1946

-01-

01

1949

-01-

01

1952

-01-

01

1955

-01-

01

1958

-01-

01

1961

-01-

01

1964

-01-

01

1967

-01-

01

1970

-01-

01

1973

-01-

01

1976

-01-

01

1979

-01-

01

1982

-01-

01

1985

-01-

01

1988

-01-

01

1991

-01-

01

1994

-01-

01

1997

-01-

01

2000

-01-

01

2003

-01-

01

2006

-01-

01

2009

-01-

01

FVGB

FVLGB

0.0000

10000.0000

20000.0000

30000.0000

40000.0000

50000.0000

60000.0000

70000.0000

1946

-01-

01

1949

-01-

01

1952

-01-

01

1955

-01-

01

1958

-01-

01

1961

-01-

01

1964

-01-

01

1967

-01-

01

1970

-01-

01

1973

-01-

01

1976

-01-

01

1979

-01-

01

1982

-01-

01

1985

-01-

01

1988

-01-

01

1991

-01-

01

1994

-01-

01

1997

-01-

01

2000

-01-

01

2003

-01-

01

2006

-01-

01

2009

-01-

01

FVGB

FVLCB


CHART E SIMULATED GDP BOND VS. LARGE COMMON STOCK

FVGB: Simulated GDP Bond. FVIGB: Actual returns based on long term corporate stocks. CONCLUSION

A market exists for investors to enjoy returns based on real variables in the economy. A bond that has its returns based on nominal GDP growth rates mitigates inflation risk, but also allows the cautious investor to partake in returns associated with the upswing of the real business cycle. This bond would be very advantageous for government trying to raise money during recessionary periods as investors are typically wary of the stock market and seek to maintain their funds in lower risk instruments. In the case of a country like the United States that has budget deficit issues, the widening of the bond market using an instrument like this would be very useful for deficit control. Although governments that issue such bonds do have to pay greater returns during growth periods of the real business cycle, clauses can be inserted such that payment is issued only when growth rates exceed certain levels. This is the methodology used by Argentina with its GDP warrants.

Simulations show that the instrument compares very well against both short and medium term bonds, and even out-performs long term corporate bonds. Issues that need to be addressed are the issue price for such bonds, and the fact that speculative investors may choose to purchase them only towards the upswings of the business cycle and sell them towards the downswings in the business cycle. Thus while these bonds are cheaper for a government to institute during downswings, they are more expensive during upswings in the business cycle as returns rise with the business cycle. Thus a further issue to contend with is whether a payment clause needs to be inserted into the instruments such that interest coupons are only issued above certain GDP growth rates. Regardless of these issues, the simulations indicate that there is a very lucrative missing market that government can tap to mitigate budget issues during downturns in the business cycle.

0.0000

100000.0000

200000.0000

300000.0000

400000.0000

500000.0000

600000.0000

700000.0000

800000.0000

900000.0000

1900

-01-

01

1900

-01-

04

1900

-01-

07

1900

-01-

10

1900

-01-

13

1900

-01-

16

1900

-01-

19

1900

-01-

22

1900

-01-

25

1900

-01-

28

1900

-01-

31

1900

-02-

03

1900

-02-

06

1900

-02-

09

1900

-02-

12

1900

-02-

15

1900

-02-

18

1900

-02-

21

1900

-02-

24

1900

-02-

27

1900

-03-

01

1900

-03-

04

FVGB

FVLCS


END NOTES

1. Inflation risk is defined as the potential loss of purchasing power when price indexes rise in an economy.

2. Bloomberg News. 3. Table A-6 from Ibbotson’s Yearbook provided total returns data for the Long-Term Government

Bonds, and Table A-10 from Ibbotson’s Yearbook provided total returns data for the Intermediate-Term Government Bonds.

4. http://research.stlouisfed.org/fred2/. REFERENCES Baker, Malcolm and Wurgler, Jeffrey. (2009). Government Bonds and the Cross-Section of Stock Returns. Working paper, Harvard Business School, NYU Stern School of Business and NBER. March 18. Bodie, Zvi. (1976). Common Stocks as a Hedge Against Inflation, The Journal of Finance, Vol. 31, No. 2, May. Carlstrom, Charles T., and Timothy S. Fuerst (2004). Expected Inflation and TIPS. Federal Reserve Bank of Cleveland. Economic Commentary, November, pp. 1-4. Ely, David P., and Robinson, Kenneth J. (1997). Are stocks a hedge against inflation? International Evidence Using a long-Run Approach, Journal of International Money and Finance, Vol. 16, Issue 1, February. Estrella, Arturo and Hardouvelis, Gikas A. (1991). The Term Structure as a Predictor of Real Economic Activity, The Journal of Finance, Vol. 66, No. 2, June. Griffith-Jones, Stephany and Sharma, Krishnan. (2006). GDP-Indexed Bonds: Making it Happen. UN/DESA Working Paper No. 21, April. Kamstra, Mark and, Shiller, Robert J. (2009). The Case for Trills: Giving the People and Their Pension Funds a Stake in the Wealth of the Nation. Cowles Foundation Discussion Paper Series, August. Kruse, Susanne, Meitner, Matthias and Schroder, Michael. (2005). On the Pricing of GDP-Linked Financial Products. Applied Financial Economics, Vol. 15, pp. 1125-1133. Lessard, Donald, and John Williamson (1985). Financial Intermediations Beyond the Debt Crisis. Institute of Intermediation Economics, No. 12. Li, Lingfeng. (2002). Macroeconomic Factors and the Correlation of Stock and Bond Returns. Working Paper, Yale University, November. Piketty, T. and E. Saez.(2003). “Income Inequality in the United States: 1913-1998.” Quarterly Journal of Economics, Vol. 118, pp. 1-39. Ruban, Oleg, Poon, Ser-Huang and Vonatsos, Konstantinos. (2008). “GDP Linked Bonds: Contract Design and Pricing.” Working Paper, Manchester Business School.


Schroder, Michael, Heinemann, Fredrich, Kruse, Susanne and Meitner, Matthias. (2004). GDP Linked Bonds as a Financing Tool For Developing Countries and Emerging Markets. Centre for European Economic Research, Working Paper, September. Shiller, Rober J. (1993). Macro Markets: Creating Institutions for Managing Society’s Largest Economic Risk. New York: Oxford University Press.


The Influence of Winning on Mid-Major College Football Attendance

Tim Padgett Arkansas State University

C. Shane Hunt

Arkansas State University

A strong positive correlation exists between winning percentage and attendance within a given season. Understanding this correlation is important to sports marketers and sports business managers. Considering that marketers have no impact on win-loss record, this impact can be extracted before evaluating an organization’s marketing performance. Sports business managers have to consider the uncertainty of a team’s performance during an upcoming season when they attempt to accurately project revenues. BACKGROUND

The relationship between higher winning percentages and higher attendance in sports has long been a given. The naked eye can clearly see that the New York Yankees draw more than the Pittsburgh Pirates, the Los Angeles Lakers draw more than the Los Angeles Clippers, and the University of Alabama football program draws more than the University of Alabama at Birmingham football program. A complex analysis is not necessary to realize that this higher attendance is due in large part to the winning tradition of the teams with the greater attendance.

What is not as clear is the specific impact that higher winning percentages have on attendance. If we are able to quantify this impact then we would be able to extract this impact on the front end of marketing evaluations and have a much clearer picture of the effectiveness of an organization’s marketing efforts.

For example, an increase in attendance may be easily attributed to a jump in a team’s winning percentage. In reality the increased winning percentage may be a big part of that increase, but another important part may have been the hard work of an organization’s marketing department. Without being able to accurately extract the impact of the increased winning percentage, the hard work of the marketing department would be unrecognized and unrewarded. On the other hand, attributing a drop in attendance entirely to a drop in winning percentage may be overlooking an ineffective marketing department. PURPOSE

In this paper we will examine the impact that winning percentage has on college football at the mid-major level. The mid-major level is chosen because most of these programs rarely play before capacity crowds. The table below shows the average attendance by conference between 2005 and 2009 and how that attendance compares to the average capacity of the conference’s stadiums.


TABLE 1

Conference 2009 2008 2007 2006 2005 Average Avg Cap. %CapacitySEC 76,288 76,844 75,139 75,706 74,579 75,711 77,401 97.82%Big Ten 71,769 70,125 71,158 69,643 72,566 71,052 74,560 95.30%Big 12 62,875 62,956 60,941 59,968 58,397 61,027 66,374 91.94%ACC 51,249 52,737 53,787 52,936 52,242 52,590 58,690 89.61%Pac 10 54,186 57,350 57,956 56,314 57,479 56,657 64,547 87.78%Big East 44,804 43,145 41,456 39,043 39,400 41,570 48,984 84.86%Moutain West 33,202 32,021 33,937 32,766 35,275 33,440 46,905 71.29%WAC 22,749 24,475 23,014 17,696 21,595 21,906 31,675 69.16%Sun Belt 16,463 18,584 16,374 17,097 16,881 17,080 28,093 60.80%C-USA 26,600 27,118 27,666 26,645 25,543 26,714 46,252 57.76%MAC 15,317 16,727 17,334 17,696 14,489 16,313 31,452 51.86%Sources: NCAA.org, collegegridirons.com

Table 1

It is most probably no coincidence that those schools in the bottom conferences of the above chart are also those with the most limited resources (Eichelberger, 2009). Therefore it is even more crucial to programs at this level to understand what drives their attendance. A better understanding of what drives their attendance means a better understanding of what drives their revenue, which is crucial to any business but especially to those experiencing economic hardship.

By focusing on the mid-major level we are able to see large fluctuations in attendance that allows us to more accurately examine the impact of winning percentage. Examining a 5% increase in attendance of a Conference USA team will tell us more than a 0.2% increase in a SEC school’s attendance. Such an increase would also mean significantly more to the bottom line of the Conference USA team’s budget than the SEC team’s budget. Even a 5% fluctuation in attendance at the SEC level would mean little to a SEC team’s bottom line given that revenues at these schools are also driven significantly by television contracts and bowl payouts. LITERATURE REVIEW

A study done by David C. Funk, Kevin Filo, Anthony A. Beaton, and Mark Pritchard focused on several factors that motivate fans to attend sport events (Funk, Filo, Beaton, & Pritchard, 2009). Among these factors was “esteem,” which was measured by whether or not fans feel as though they win when their team wins and whether fans get a sense of accomplishment when the team wins. While this factor is very closely related to the win/loss factor we are examining, it is probable that there are still fans out there that choose to attend or not to attend games based on the quality of the team, even if the result does not have a significant impact on how they feel about themselves.

Emotional attachment has been found to be a factor that fans consider when attending games (Koo & Hardin, 2008). This emotional attachment is built over several years and often includes going to games as a young child. This emotional attachment is often even stronger when the team represents the spectator’s alma-mater. Memories of one’s college years and reunion with old college friends are likely to keep a significant number of people coming to games even when the team’s performance may be subpar.

Several findings in Michael Davis’ research on the correlation between winning and attendance in Major League Baseball will be applicable to our study as well. Davis first solved the “Chicken or the Egg Conundrum” of whether winning leads to increased attendance or increased attendance leads to winning.


A (positive) correlation was only found with the former. Davis also found that a successful season will lead to increased attendance during the following years as momentum of the previous success draws in more through a bandwagon effect (Davis, 2008 ).

Research done by Glen Knowles, Keith Sherony, and Mike Haupert focused upon how the uncertainty of the outcome impacted attendance in Major League Baseball. They found that attendance was maximized when the home team had a 60% chance of winning. This supports the hypothesis that sports fans want to see a winning a product (Knowles, Sherony, & Haupert, 1992).

One leading article on college football attendance factors was done by Donald Price and Kabir Sen. This article is extremely helpful in examining the many underlying factors that affect college football attendance (Price & Sen, 2003). However, most of these factors are factors that administrators have little to no control over. Our piece will hopefully prove a strong compliment to Price and Sen in future works that examine factors affecting college football attendance.

Those who know the most about college football, the coaches themselves, also agree that winning games are important for attendance. In fact, when asked to rate factors affecting attendance on a 1-7 scale, winning was found to be the most important with a 6.68 rating (Hay & Rao, 1984). METHODS AND HYPOTHESIS

We have already established that we will focus on mid-major schools who often struggle filling their stadiums. Furthermore, since inclement weather in the late fall plays a huge role in determining attendance at cold weather schools, we will focus strictly on Conference USA and the Sun-Belt Conference. As seen in Table 1, these conferences on average only had 57.76% (CUSA) and 60.80% (SBC) capacity over the last five years.

Our first study (appendix 1) will look at how a team’s performance during a season (measured by increase or decrease in winning percentage from the previous season) impacts attendance during the next season. It would be a reasonable assumption for a team to believe that an increase in winning percentage should lead to an increase in season ticket sales the next season. By the same logic, a drop-off in winning percentage should lead to a drop-off in season ticket sales.

While season ticket sales are affected by the previous season, tickets bought on a game-by-game basis should be affected minimally by the previous season. While consumers may decide to purchase tickets to the first game or two based upon the previous season, eventually the current season will become the driving factor in determining whether or not they attend the game.

The second study (appendix 2) examines the impact winning percentage has on the current season’s attendance. We expect a strong correlation between these two factors. A team that is winning more than the previous year should expect to see higher attendance accordingly. More importantly, this analysis will lead us one step closer to being able to quantify the impact. RESULTS AND ANALYSIS

The first study showed a weak correlation between improvement or decline in winning percentage and attendance the following season. This is probably due to only a small fraction of ticket sales at this level being season tickets. The bigger conferences that have a much larger season ticket as part of overall tickets sold percentage would probably show a much greater correlation in this study.

The second study that examined the effect of winning percentage on attendance during the same season showed a much stronger correlation (R Square of .189). This strongly indicates that a significant percentage of the fan base of CUSA and SBC teams decide to attend games on a game by game basis based upon how the team is performing during that season.


MANAGERIAL IMPLICATIONS

From a marketing perspective, we now have a better understanding of the impact winning percentage fluctuation has on season attendance. With this knowledge, marketing evaluators are able to extract the significant portion of the fluctuation of attendance attributed to on-field performance, which is out of the control of marketing managers.

Since attendance is directly related to revenue, it also illustrates the difficulty in accurately projecting revenue in budget forecasts. Considering that accurately projecting win-loss records is an art-form that nobody has shown any evidence of mastering, these programs are going into each fiscal year with a tremendous amount of revenue uncertainty.

This evidence also strongly indicates that discretionary resources should be invested in strengthening the on-field product given the impact illustrated in this study on attendance. This can be done by increasing recruiting budgets, allocating more resources towards attracting top-tier coaching talent, or investing in new weight room equipment. LIMITATIONS AND FUTURE RESEARCH

This study would benefit from specific information from schools that separates season ticket sales and walk-up ticket sales. This would allow a better understanding of the effect previous success or failure has on season ticket sales, most of which are bought prior to the next season. If the success or failure of the previous season is shown to have a significant impact on season ticket sales than the increase of season ticket sales would be one avenue towards achieving more revenue certainty. REFERENCES Davis, M. C. (2008 ). The Interaction Between Baseball Attendance and Winning Percentage. International Journal of Sport Finance , 58-73. Eichelberger, C. (2009, January 6). Bloomberg. Retrieved May 15, 2010, from http://www.bloomberg.com/apps/news?pid=20601103&sid=aYYY_mDwYMkY Funk, D. C., Filo, K., Beaton, A. A., & Pritchard, M. (2009). Measuring the Motives of Sport Event Attendance: Bridging the Academic-Practitioner Divide to Understanding Behavior. Sport Marketing Quarterly , 126-138. Hay, R., & Rao, C. (1984). Football Coaches Rank Factors Impacting Attendance at Games. Marketing News , 5. Knowles, G., Sherony, K., & Haupert, M. (1992). The Demand for Major League Baseball: A Test of the Uncertainty of Outcome Hypothesis. American Economist , 72-81. Koo, G.-Y., & Hardin, R. (2008). Difference In Interrelationship Between Spectator Motives and Behavioral Intentions Based On Emotional Attachment. Sport Marketing Quarterly , 30-43. Price, D., & Sen, K. (2003). The Demand for Game Day Attendance in College Football: An analysis of the 1997 Division 1-A Season. Managerial and Decision Economics , 35-46.


Risk Behavior and Performance in Chinese Private Firms: From Regulatory Impact to Owner-Managers

Chuanyin Xie

The University of Tampa

Given the background that private firms in China were risk taking in once an unfavorable regulatory environment, this study examines their risk behavior and performance implications in a new era when the “invisible hand” begins to exert its influence. In the new era, private firms have choices as to whether or not to take risk, so risk behavior needs to be examined beyond the regulatory impact. I shift the research attention to owner-managers who make risk decisions. I argue that owner-managers’ characteristics, including who they are, why they run their own business, and to what degree they control their business, have direct implications for firm risk behavior. These characteristics can also affect the relationship between risk behavior and performance. INTRODUCTION

Risk is an important concept in organizations (Miller and Leiblein, 1996; Sitkin and Pablo, 1992). Risk taking has important implications for firm survival and growth (Shapira, 1995). Though risk behavior has been extensively studied in the West, it is still relatively unexplored in a context of transitional economies such as China. Probably, risk research is not important when an economy is dominated by the “iron fist” control. State-owned enterprises (SOEs), a dominant form of business organization, largely serve as manufacturing machines taking orders from the state, so risk almost does not exist from a market perspective. Private firms, often treated as a “fringe group”, have little choice but to take risk because of poorly protected property rights and hostile regulatory attitudes toward them. When the economic transition reaches a point at which the “invisible hand” begins to exert its influence, risk research would become meaningful. Risk behavior such as risk taking can be based on managerial choice, instead of either being unnecessary (the case of SOEs) or indispensible (the case of private firms).

China is the world’s largest transitional economy. Private business has been one main force driving its economic growth since 1980s. China’s economic transition has been subject to two opposing forces: the “iron fist” control and the “invisible hand” control which “co-exist, compete, and counteract” (Tan, 2005).The early phase of economic transition, dominated by the “iron fist” control, presented an unfavorable environment for private firms. In response to regulatory hostility, they were risk taking (Tan, 1996). The struggle between the two forces has finally led to a more market-based, the “invisible hand” control model, though “uniquely Chinese” (Tan, 2005), because of undisputed inefficiency of the “iron fist” control. The regulatory regime has become more favorable to private business (IFC, 2000; Tan, 2005). Laws and regulations have been established to protect property rights. China’s entry into WTO in 2002 has strengthened its obligation of respecting ownership of private properties. As a result, the private sector has seen a fast growth in recent years (He, 2009).


In a market-based economy, the “invisible hand” guides business activities through the combination of self-interest, private ownership, and competition. The self-interest nature of private ownership motivates individuals to allocate resources in pursuit of profitable opportunities (Gibb and Li, 2003). The competitive nature of market makes resource allocation risky due to uncertainty. As China’s economy becomes more market-based (I label as the “competitive phase” of its economic transition), private firms, with their ownership rights being recognized, face new opportunities and challenges. Are they willing to commit resources toward activities with uncertain outcomes? No research has directly addressed this question. This study attempts to fill the gap.

In China’s transitional economy, risk behavior in private firms has been studied from a regulatory perspective (Tan, 1996, 2001) when the “iron fist” was relatively strong. In the competitive phase characterized by the relatively strong “invisible hand”, the impact from the regulatory regime would decrease. In this study, I shift the focus from the regulatory impact to own-managers’ influence on firm risk behavior. In China, three questions related to owner-managers have received research attention: who they are (social groups), why they run their own business (ownership motivation), and to what degree they control their business (business control). I investigate how the characteristics of owner-managers, including their social groups, ownership motivation, and business control, might affect firm risk behavior and how risk behavior might affect performance. I put owner-managers in a context of economic transition to examine their impact. Figure 1 depicts the research model.

FIGURE 1 RESEARCH MODEL

Major Sources of Information

This study uses several sources of information. A major source of information is research reports made by He (2009). Based on seven large scale multiple-industry surveys conducted by five Chinese institutions during 1993-2006, He reported information on owners’ backgrounds, ownership motivation, and governance structures in Chinese private firms. Other sources of information include two nation-wide surveys on Chinese private firms, conducted by Asian Development Bank (ADB, 2003) and International Finance Corporation (IFC, 2000) respectively. I also use information provided by Global Entrepreneur-ship Monitor Reports (GEM, 2002, 2007, 2009) on a global basis. ECONOMIC TRANSITION AND BUSINESS OWNERSHIP

China’s economic transition makes private business ownership possible. A private firm is defined as “a for-profit organization that is owned by individuals and employs more than eight people” (IFC, 2000). In this section, I explore the two phases of economic transition and how they affect owner-managers’

The Owner-Manager

Social Group Performance Risk Behavior P1 P4

Ownership Motivation P2

Risk Behavior P5

Performance

Desire for Control Risk Behavior P3 P6

Performance


characteristics. In the entrepreneurship literature, owner-managers are often referred to as entrepreneurs, so I use “owner-managers” and “entrepreneurs” interchangeably. Economic Transition

As a transitional economy, China shares the following basic features: low-income, rapid growth, underdeveloped legal systems, inadequate protection of private property rights, and market imperfections (Hoskisson et al, 2000). During the early phase of economic transition, all these features were salient. SOEs dominated the whole economy. Their inefficiency, along with the inefficiency of the centrally planned system, created serious shortages in most markets, information and resource asymmetries (Peng, 2001), and “institutional holes” (Yang, 2004). Private firms were presented with both opportunities and challenges. On the one hand, they could take advantage of the inefficiencies of SOEs and the planned system and exploit numerous market niches or fill institutional gaps. On the other hand, they had to deal with adverse situations brought about by the regulatory regime, including political threats and restricted access to resources. These opportunities and challenges shaped private firms’ behaviors: using guanxi (relationships) or “red hat” (a strategy used to disguise their private ownership by registering as a collectively-owned enterprise) as substitutes for formal institutional support ( Xin and Pearce, 1996), pursuing short-term profitability and reluctant to make long-term investments (Nee, 1992), and acting proactively to identify and occupy market niches (Peng, 2001).

China’s economy has now become more market-oriented. Accompanying the establishment of market-based rules is intensified competition. For private firms, they not only compete against each other, but also confront more formidable competitors: domestic SOEs and foreign firms. SOEs have gradually been weaned from their dependence upon state budget allocations during the economic transition. With a separation from the government and clarified rights and responsibilities, SOEs have greatly transformed themselves. Endowed with abundant resources and oriented toward the market, they are now globally competitive (Ralston et al, 2006). Foreign firms have rushed to China in response to the “opening-up to the outside world” policy and preferential treatments. Because of its consistency in implementing the policy and its huge internal market, China has now become the largest foreign direct investment recipient (Chang and Xu, 2008) and the most appealing host for investments (UNCTAD, 2009) in the world. As Stuttard (2000) commented, “everyone is here” in China – Americans, Europeans, Japanese, and others – “whether it be automotive, consumer products, electricity generation, electronics, or new technology.” Foreign firms are often equipped with advanced technologies, know-how, and management and marketing skills, as well as plentiful financial resources (Ralston et al, 2006).

Domestic SOEs and foreign firms have put private firms at a competitive disadvantage. Reports by IFC (2000) and ADB (2003) indicate that Chinese private firms are resource-constrained. They have limited access to formal sources of finance, poor availability of information, and difficulties in hiring highly qualified employees. In addition, many private firms have management-related problems such as nontransparent and inconsistent implementation of policies. Business Ownership

Why do people own and run a private business? According to GEM Report (2002), individuals perform entrepreneurial activities for two major reasons: pursuing an opportunity (opportunity motivated) or seeing entrepreneurship as their “last resort” (necessity motivated). More than 97% of entrepreneurs can be identified as either opportunity or necessity motivated. Opportunity entrepreneurs run a business “as one of several possible career options”. Necessity entrepreneurs are forced to run their own business because “all other options for work are either absent or unsatisfactory”. In this study, I use the “opportunity vs. necessity” categorization to distinguish individuals’ motivation for business ownership. He’s (2009) survey reports suggest that in the early phase of China’s economic transition, a majority of entrepreneurs were necessity-based. Continued economic transition has seen more opportunity-motivated entrepreneurs.

Who run private firms in China? Based on He’s (2009) survey reports, private business owners come from five social groups: individual household business owners, industrial workers, peasants, former


cadres, and professionals. I regroup these individuals into three broad categories: blue-collars, former cadres, and professionals. The blue-collar group represents private entrepreneurs from a lower social class. Many of them lived in poor rural areas or lost their job from poorly-run SOEs before starting their own business. Their entrepreneurial career is more often motivated by necessity than opportunity. Blue-collars are close to craftsman entrepreneurs described by Smith and Miner (1983). They are characterized by “narrowness in education and training”, “low social awareness” and “a limited time orientation”. Former cadres were government officials or SOE managers. They gained social and political capital and owned business networks during their employment. Compared with blue-collars, former cadres are less necessity-driven when running their own business. Professionals largely use their own specialized skills and knowledge to develop new products or services and fill unserved markets. They tend to be well-educated and have options for an employment career. Like cadre entrepreneurs, professional entrepreneurs are often driven by opportunity. Ownership Form and Control

In Chinese private firms, ownership takes different forms, including proprietorship, partnership, and limited liability. He’s (2009) survey reports show that proprietorship was a dominant form of business ownership during early 1990s, but it has been declining since then. In contrast, the number of private firms adopting a limited liability form was low in early times, but has gradually been increasing since then. The proportion of partnership among the three ownership forms has been remaining low. The increased adoption of the limited liability form during the economic transition reflects to some extent the improvement of the regulatory regime and the progress of marketization. This ownership form is aimed at achieving professionalism in business management. The three ownership forms by themselves are clear, but ownership in private firms was not defined well in the early phase of market transition. Because of inadequate protection of private property rights, private ownership had high property risk (IFC, 2000). In order to keep their private firms safe, owner-managers put on a “red hat” to cover up the private nature of their firms (Chen, 2007). “Red hat” existed throughout 1980s and 1990s.

Ownership and control can be aligned or separated. Survey reports (ADB, 2003; IFC, 2000) indicate that most private firms in China are managed by owners, suggesting an alignment between ownership and control. In addition, ownership concentration is often high. The major decision makers tend to be owner-managers. However, there is a small percentage of private firms whose board of directors has begun to exercise influence.

In summary, China is transitioning from a more planned economy to a more market-based economy. The two phases of economic transition present different characteristics, which are summarized in Table 1. RISK BEHAVIOR IN A COMPETITIVE ENVIRONMENT

As argued earlier, risk behavior in Chinese private firms needs to be examined beyond the regulatory impact in the new competitive era. Firm risk behavior is managerial choice of projects with uncertain outcomes (Desai, 2008; Li and Tang, 2010). Given that most private firms in China are small and managed by owners (ADB, 2003; IFC, 2000), it is reasonable to assume that owner-managers play a key role in their firms’ risk behavior. In this section, I investigate how owner-managers’ characteristics such as their social groups, ownership motivation, and business control might affect firm risk behavior. Social Groups and Risk Behavior

In the early phase of economic transition, most private firms in China were started by people from lower social classes such as farmers and industrial workers (He, 2009). The former planned economy left numerous unfilled market niches yet to be exploited. Those who were bold enough ventured into an uncertain world with a hope of “earning money” or “getting rich”. They were often “pushed” into self-employment due to poor living conditions or job losses. Prospects for making money also attracted cadres such as government officials and SOE managers. A few of them were motivated to quit their job and “pulled” into the uncertain but exciting private sector. With privileges in accessing information and


mobilizing resources, former cadres were in a good position to exploit arbitrage opportunities generated by the old system. Compared with the blue-collar and cadre groups, the group of professional entrepreneurs was relatively small in the early phase.

TABLE 1 ECONOMIC TRANSITION, BUSINESS OWNERSHIP AND CONTROL

Economic Transition

The Early Phase The Competitive Phase Environment

Characteristics

Implications

Market imperfection: High Competition: Low Private property protection: Low Opportunity: Unfilled market niches Threat: Regulatory

Market imperfection: Decreasing Competition: Increasing Private property protection: Increasing Opportunity: Fewer market niches Threat: Competition

Business Ownership Who (Social groups)

Blue-collars: High Former cadres: Medium Professionals: Low

Blue-collars: Decreasing Former cadres: Decreasing Professionals: Increasing

Why (Ownership Motivation)

Opportunity-driven: Low Necessity-driven: High

Opportunity-driven: Increasing Necessity-driven: Decreasing

Business Control Ownership Forms

Control

Proprietorship: High Partnership: Low Limited liability: Low “Red hat” Owner-manager: High Board of directors: Low

Proprietorship: Decreasing Partnership: Low Limited liability: Increasing Owner-manager: High, but decreasing Board of directors: Low, but increasing

After more than two decades of economic transition, China has become more market-based. First,

competition has heated up in most industries. Compared with SOEs and foreign firms, private firms are often poor in resources. Previously, they were able to take advantage of former SOEs’ inefficiency, but now SOEs have become formidable competitors (Ralston et al, 2006). Second, a large number of industries have become more or less mature during many years of development. Unfilled market niches do not exist as widely as before. Recent GEM Report (2009) shows that entrepreneurs in China have perceived fewer opportunities in the future.

The economic transition has also rendered private firms’ strategies less effective. In the early phase, private firms largely used guanxi, quick response, and flexibility to compete and survive. Guthrie (1998) found that the significance of guanxi began to decline in China’s economic activities during 1990s. Li and colleagues’ study (2008) provides further support for the declining role of guanxi: increased competition reduced the role of managerial networks in China. Probably, private firms will continue to rely on quick response and flexibility for survival when facing powerful SOEs and foreign firms (Ralston et al, 2006). I argue that quick response and flexibility are less effective in generating profits in the new competitive era, given that profitable opportunities are becoming fewer and are more difficult to exploit.

Among the three groups of entrepreneurs, the blue-collars could be in the most disadvantageous position in the competitive environment. They were able to exploit shortages in many markets previously due to their courage, not necessarily their competitive skills. In the competitive phase, however, skills are


becoming important. Because of their limited education and short-term orientation, blue-collars would be “rigid in nature” (Smith and Miner, 1983). As a result, they are less likely to develop competitive skills for the new era. He’s (2009) survey reports show that the proportion of blue-collar entrepreneurs has already decreased recently. Cadre entrepreneurs succeeded because they were able to exploit arbitrage opportunities during the early phase. Equipped with social and political capital, they are likely to be more adaptive than blue-collars during the economic transition. However, cadre entrepreneurs compete on relational and informational advantages. This “dominant logic” (Prahalad and Bettis, 1986) may hardly change. Based on He’s (2009) survey reports, the number of cadre entrepreneurs would decline as marketization continues.

Compared with blue-collars and former cadres, professionals are more likely to have the right skills to deal with competition. Their strengths are less based on exploiting failures of the old economic system, but more on their professional knowledge and skills related to technologies, products, or markets. According to He’s (2009) survey reports, the proportion of professional entrepreneurs has been increasing recently, suggesting that professionals may be more willing to take business risk in the new era. The current environment in China also seems to favor professional entrepreneurs. For example, the Chinese government has established large incubation programs (Lalkaka, 2002) and science and technology parks (Watkins-Mathys and Foster, 2006) to facilitate technology entrepreneurship and innovation. Clearly, professionals would benefit more from those programs than non-professionals.

The competitive era seems to pose more challenges to non-professional entrepreneurs because of their less effective skills. Skills contribute to perceived control over behavioral activities. Low perceived control is less likely to initiate a behavior (Ajzen, 1991). Scholars in the field of entrepreneurship have similar arguments. According to McClelland (1961), entrepreneurs do not take risk deliberately. Instead, they pursue initiatives that are achievable and controllable, using their own skills to realize a profit (Cunningham and Lischero1991). Thus, I make the following proposition:

Proposition 1: Private firms whose owner-managers are professionals are more likely to commit resources to risky projects than those whose owner-managers are not professionals.

Ownership Motivation and Risk Behavior

Entrepreneurs are often opportunity or necessity motivated to own and run a business. Ownership motivation may affect the direction and nature of their existing and future businesses (GEM, 2002). Opportunity entrepreneurs choose to run their own business out of “a burning desire”, not because they have no other choices (Bailey, 2002). GEM Report (2002) shows that opportunity entrepreneurs are more likely to expect business growth and new market creation than necessity entrepreneurs. GEM Report (2007) reveals that high expectation and high growth entrepreneurs are more likely to display entrepreneurial attitude and behavior. The different expectations between opportunity and necessity entrepreneurs may also be explained by different opportunity costs they face when choosing an entrepreneurial career. Compared with necessity entrepreneurs, opportunity entrepreneurs incur higher opportunity cost when they switch from salaried employment to more uncertain self-employment. Therefore, they are likely to pursue a “higher upside potential” (GEM, 2007). High aspirations were found to have positive impact on risk taking (Bromiley, 1991).

In the Chinese context, cadre and professional entrepreneurs are more likely to be opportunity-driven than their counterparts in the blue-collar group (He, 2009). They incurred high opportunity costs when leaving their salaried positions, so they have a reason to set high goals, which would prompt them to commit resources to projects with “higher upside potential”. In addition, cadre and professional entrepreneurs tend to be better educated than blue-collar entrepreneurs. GEM Report (2007) shows that education has positive impact on entrepreneurs’ aspirations in both high and low income economies. From the education point of view, the two groups of entrepreneurs would set higher aspirational goals than the blue-collar group.


Necessity may also encourage people to take risk. Based on prospect theory (Kahneman and Tversky, 1979), when people are in a loss situation, they tend to be risk taking. Blue-collar entrepreneurs in China seem to fit this situation. Given their limited education and skills, however, they can be more “rigid” (Smith and Miner, 1983) than adaptive in a competitive environment, so innovative behaviors are less likely. Cadres and professionals may also be driven by necessity, such as unemployment or dissatisfaction with current jobs, to run a business. If they are forced to be self-employed, they would hope “a job comes along in the meantime” (Bailey, 2002). This hope may become strong in a competitive environment where opportunities are hard to identify and exploit. As a result, they would be less interested in pursuing risky activities.

Proposition 2: Private firms whose owner-managers are motivated by opportunity are more likely to commit resources to risky projects than those whose owner-managers are motivated by necessity.

Business Control and Risk Behavior

As China’s economic transition reaches the competitive phase, ownership rights are being recognized. Ownership makes it possible for owner-managers to control their business. The relationship between ownership and control has been extensively studied, particularly from an agency perspective (Fama and Jensen, 1983). Scholars have emphasized the role of ownership and control alignment in innovative activities with uncertain outcomes (Francis and Smith, 1995; Wright et al, 2000), but Chandler (1990) advocated professional management of the firm. Survey reports (ADB, 2003; IFC, 2000) suggest that many owner-managers in Chinese private firms are not willing to cede control over their business.

Owner-managers’ desire for business control often leads to concentrated ownership. Though ownership can provide strong incentives that may encourage risk taking and innovation (Wright et al, 2000), its effect on risk taking may be limited to the extent that owner-managers do not bear too much risk. Risk bearing occurs when managers’ wealth is tied to firm performance (Wiseman & Gomez-Mejia, 1998). Research suggests that ownership, as an incentive device, makes managers risk averse because they bear risk resulting from their stake in the firm (Beatty and Zajac, 1994; Ortega-Argiles et al, 2005). In a Chinese context, Kao (1993) argued that Chinese family firms, whose owner-managers often have a high desire for control, tend to be risk averse. Carney and Gedajlovic (2002) found that Hong Kong-based firms with coupled ownership and control were more likely to pursue short-term profitability and less likely to invest in capital-intensive projects.

If owner-managers’ desire for control is low, ownership is likely to be less concentrated and risk bearing can thus decrease. The reduction of risk bearing may be particularly important for entrepreneurial activities in a competitive environment. Competition increases uncertainty of investment projects. Risk sharing could increase owner-managers’ willingness to launch risky projects. A diffuse ownership structure also help ease resource constraints through resource pooling from different owners, providing conditions for risk taking.

Proposition 3: Private firms whose owner-managers have high desire for control are less likely to commit resources to risky projects than those whose owner-managers have low desire for control.

PERFORMANCE IMPLICATIONS

Risk taking has direct implications for firm performance, causing performance variations (Palmer and Wiseman, 1999) or negative outcomes (Naldi et al, 2007). In the Chinese context, research on the relationship between risk taking and firm performance is limited. Two studies, conducted in China’s transitional economy characterized by uncertainty, provide implications for this relationship. Tan and Litschert’s (1994) study found that defensive strategies led to higher overall performance than proactive, future-oriented, and risky strategies. When using Chinese private firms and SOEs as a sample, Tang and


colleagues (2007) found that entrepreneurial orientation, reflected by innovativeness, proactiveness, and risk taking, had positive impact on performance. The results of the two studies appear to be conflicting, indicating that risk taking may have both positive and negative relationships with firm performance.

In private firms, owner-managers’ industry and technical skills were found to have positive impact on firm growth (Baum et al, 2001). GEM Report (2007) suggests that right skills are important for entrepreneurial success. It may be argued that firm risk taking is more likely to have positive performance implications when owner-managers have right skills. The early phase of economic transition in China was characterized by market shortages and imperfections. If individuals were willing to take risk to be private entrepreneurs, they were always able to find profitable opportunities and had chance to earn more money than those in SOEs. “It’s financially better to wield a barber’s razor than a surgeon’s scalpel” and “It’s financially better to sell tea-boiled eggs than to make guided missiles” (Young, 1991). Risk taking paid off because of their courage, not necessarily their sophisticated skills.

Those situations do not seem to exist anymore in the competitive phase. It’s not likely that blue-collars are able to use courage and gut feelings to make money; it’s also less likely that cadre entrepreneurs can make their business successful only based on their social and political connections. In the competitive environment, competitive resources and capabilities could be more effective than networks and relationships (Peng, 2003). As argued earlier, blue-collar and cadre entrepreneurs may not have right skills for the competitive environment, so their ability to choose and control risky projects would be questionable. If they do take the risk, they would not be as lucky as they were in the early phase.

Proposition 4: Committing resources to risky projects will be more likely to lead to upside outcomes in private firms whose owner-managers are professionals than in those whose owner-managers are not professionals.

GEM Report (2002) suggests that opportunity entrepreneurs can be more optimistic than necessity

entrepreneurs when running their business. Based on Brown and Marshall (2001), there is a curvilinear relationship between optimism and performance across many different activities. When individuals are low in optimism, they often achieve low performance because they tend to “lack motivation” and “focus on negative information” (Hmieleski and Baron, 2009). Performance increases with optimism but decreases when optimism passes a certain point. The reason is that overoptimistic individuals often set unrealistic goals. Optimism leads to best performance when it is moderate. Opportunity entrepreneurs tend to have high expectations (GEM, 2002), which may lead to overoptimism when initiating a new business, but they would become more realistic when running their business (GEM, 2007). I argue that in a context of China, opportunity entrepreneurs are not likely to be highly unrealistic because of their resource constraints.

When necessity-based people choose to be self-employed, they have no other choices. Self employment may be “against their will”, so they are less likely to have growth intentions and more likely to use business as an income substitute and employ only themselves (Bailey, 2002), indicating low optimism. They are less prepared to take risky actions than opportunity entrepreneurs. If they do take risk, however, their risk taking can be more like gambling. They already lost their job, so they would feel that they have nothing else to lose. They could be tempted to pursue high risky projects with expectations of quick returns. For gamblers, success can only be left to chance.

Proposition 5: Committing resources to risky projects will be more likely to lead to upside outcomes in private firms whose owner-managers are motivated by opportunity than in those whose owner-managers are motivated by necessity.

When owner-managers’ desire for business control is high, they often make all important decisions by

themselves. Internal and external monitoring is less likely and less effective. Opportunistic investments can be made on the basis of “animal spirits” or “gut feel” and “without regard to internal and external processes of accountability” (Carney, 2005). Such a decision making process is flawed and less


successful. Naldi and colleagues (2007) found that risk taking was negatively related to performance in family-controlled firms.

If owner-managers have a low desire to control their business, their personal influence would decrease. Firm risk behavior can be less likely to reflect individual owners’ will and value and monitoring from both internal and external environments is more possible. The improved decision making process helps the firm to commit resources to right projects. Recently, an increasing number of Chinese private firms have adopted a limited liability form of ownership (He, 2009). In a limited liability company, ownership is spread and a board of directors is installed. As a result, the company could avoid “animal spirits” and “gut feel” in risky decisions. Cooke (2008) studied 30 top performing Chinese private firms and found that family-owned firms have moved away from family control to professional management. Therefore, I propose the following:

Proposition 6: Committing resources to risky projects will be less likely to lead to upside outcomes in private firms whose owner-managers have high desire for control than in those whose owner-managers have low desire for control.

DISCUSSION

China’s economic transition has made it possible to own a private business. Given the background that private firms were risk taking in the early phase of economic transition, this study investigates their risk behavior and performance implications in the competitive era. I argue that different phases of economic transition may have different impacts on private firms’ risk behavior. In the early phase, private firms did not seem to have choice but to take risk because of regulatory hostility. In the competitive phase, however, they have choices as to whether or not to take risk, so decision makers, that is, owner-managers, would become important. I focus on three basic issues related to owner-managers: who they are, why they run their own business, and to what degree they control their business, and examine how these issues may affect firm risk behavior.

I also explore performance implications of risk behavior. In a competitive environment, committing resources to new projects is important for gaining or sustaining competitive advantage, but resource commitments may not bring positive results. Private firms in China are now facing a different environment. It is not likely that they can make profits on the basis of courage, as they could in the early phase of economic transition. I argue that risk taking is more likely to lead to upside outcomes if owner-managers are equipped with right skills for the competitive era, are driven by opportunity rather than necessity, and are willing to share control of their business.

This study contributes to our understanding of risk behavior in private firms. Private ownership is a key feature of private business. On the one hand, private ownership provides “high powered” incentives that may encourage risk taking (Wright et al, 2000); on the other, private ownership makes owner-managers risk averse due to risk bearing (Beatty and Zajac, 1994; Wiseman and Gomez-Mejia, 1998). It seems that risk behavior is a result of two opposing forces: ownership incentives and risk bearing. Based on this study, the relative strength of the two forces may be affected by ownership motivation and business control. Opportunity motivated owner-managers could be more likely to take advantage of ownership incentives than necessity-motivated ones; owner-managers who control their business tightly may be more influenced by risk bearing than ownership incentives.

Firm risk behavior can also be affected by organizational factors (Rajgopal and Shevlin, 2002; Singh, 1986; Wright et al, 2007), but I only focus on the impact of owner-managers. I believe this focus is appropriate for Chinese private firms, given that most of them are small and managed by owners. Scholars have emphasized the importance of managerial perception of organizational factors on organizational activities (Jauch and Kraft, 1986). Future research may focus on how decision makers mediate the relationship between organizational factors and firm risk behavior. For example, slack resources have been found to have both positive and negative impact on risk taking (Palmer and Wiseman, 1999; Singh, 1986). It is likely that the decision maker perceives the role of slack resources in


different ways. Slack may be viewed as a “buffer” absorbing shocks (Palmer and Wiseman, 1999), as a resource for innovation (March and Shapira, 1992), or as a sign of organizational success (Bromiley, 1991). I suspect that managerial perception of slack affects the relationship between slack and risk behavior. CONCLUSION

As the largest and fastest growing transitional economy in the world, China provides a research context in which almost all entrepreneurial activities can be studied as “natural experiments” (Phan et al, 2010). In this study, I explore private firms’ risk behavior and performance implications. My central argument is that in a competitive environment, risk behavior in Chinese private firms needs to be examined beyond the regulatory impact. Since private firms can choose to take risk in the new era, the owner-managers would play a key role in making that decision. REFERENCES ADB (Asian Development Bank). (2003). The Development of Private Enterprise In the People’s Republic of China. Manila, Philippines. Ajzen, I. (1991). The Theory of Planned Behavior. Organizational behavior and human decision processes, 50, 179-211. Bailey, J. (2002). Enterprise: Desire - More than Need - Builds a Business - Entrepreneurs by Choice, rather than Necessity, Create Firms that Grow. Wall Street Journal - Eastern Edition, 239, 99: B4. Baum, R., Locke, E., & Smith, K. (2001). A Multidimensional Model of Venture Growth. Academy of Management Journal, 44(2), 292-303. Beatty, R. P., & Zajac, E. J. (1994). Managerial Incentives, Monitoring, and Risk Bearing: A Study of Executive Compensation, Ownership, and Board Structure in Initial Public Offerings. Administrative Science Quarterly, 39(2), 313-335. Bromiley, P. (1991). Testing a Causal Model of Corporate Risk Taking and Performance. Academy of Management Journal, 34(1), 37-59. Brown, J. D., & Marshall, M. A. (2001). Great Expectations: Optimism and Pessimism in Achievement Settings, in Optimism & Pessimism: Implications for Theory, Research, and Practice. Ed. E. C. Chang. Washington, DC: American Psychological Association, 239-255. Carney, M. (2005). Corporate Governance and Competitive Advantage in Family-Controlled Firms. Entrepreneurship Theory and Practice, 29(3), 249-265. Carney, M., & Gedajlovic, E. (2002). The Coupling of Ownership and Control and the Allocation of Financial Resources: Evidence from Hong Kong. Journal of Management Studies, 39(1), 123-146. Chandler, A. D. (1990). Scale and Scope: The Dynamics of Industrial Capitalism, Cambridge, MA: Harvard University Press. Chang, S. J., & Xu, D. (2008). Spillovers and Competition among Foreign and Local Firms in China. Strategic Management Journal, 29(5), 495-518.


Chen, W. (2007). Does the Color of the Cat Matter? The Red Hat Strategy in China's Private Enterprises. Management & Organization Review, 3(1), 55-80. Cooke, F. L. (2008). Competition and Strategy of Chinese Firms: An Analysis of Top Performing Chinese Private Enterprises. Competitiveness Review, 18(1/2), 29-56. Cunningham, J. B., & Lischeron, J. (1991). Defining Entrepreneurship. Journal of Small Business Management, 29(1), 45-61. Desai, V. M. (2008). Constrained Growth: How Experience, Legitimacy, and Age Influence Risk Taking in Organizations. Organization Science, 19(4), 594-608. Fama, E. F., & Jensen, M. C. (1983). Separation of Ownership and Control. Journal of Law and Economics, 26(2), 301-325. Francis, J., & Smith, A. (1995). Agency Costs and Innovation: Some Empirical Evidence. Journal of Accounting and Economics, 19(2-3), 383-409. GEM (Global Entrepreneurship Monitor). (2002). Executive Report. Babson College; Ewing Marion Kauffman Foundation; London Business School. GEM (Global Entrepreneurship Monitor). (2007). Global Report on High-Growth Entrepreneurship. Babson College, Babson Park, MA, US; London Business School, London, UK. GEM (Global Entrepreneurship Monitor). 2009. Global Report. Babson College, Babson Park, MA, United States; Universidad del Desarrollo, Santiago, Chile; Reykjavík University, Háskólinn Reykjavík, Iceland; London Business School, London, United Kingdom. Gibb, A., & Li, J. (2003). Organizing for Enterprise in China: What can We Learn from the Chinese Micro, Small, and Medium Enterprise Development Experience. Futures, 35(4), 403-421. Guthrie, D. (1998). The Declining Significance of Guanxi in China’s Economic Transition. The China Quarterly, 154, 254-282. He, X. (2009). The Development of Entrepreneurship and Private Enterprise in the People's Republic of China and its Relevance to Transitional Economies. Journal of Developmental Entrepreneurship, 14(1), 39-58. Hmieleski, K. M., & Baron, R. A. (2009). Entrepreneurs' Optimism and New Venture Performance: A Social Cognitive Perspective. Academy of Management Journal, 52(3), 473-488. Hoskisson, R. E., Eden, L., Lau, C. M., & Wright, M. (2000). Strategy in Emerging Economies. Academy of Management Journal, 43(3), 249-267. IFC (International Finance Corporation). (2000). China’s Emerging Private Enterprises: Prospects for the New Century. IFC: Washington, DC. Jauch, L. K., & Kraft, K. L. (1986). Strategic Management of Uncertainty. Academy of Management Review, 11(4), 777-790. Kao, J. (1993). The Worldwide Web of Chinese Business. Harvard Business Review, 71(2), 24-36.


Kahneman, D., & Tversky, A. (1979). Prospect Theory: An Analysis of Decisions under Risk. Econometrica, 47, 262-291. Lalkaka, R. (2002). Technology Business Incubators to Help Build an Innovation-based Economy. Journal of Change Management, 3(2), 167-176. Li, J. J., Poppo, L., & Zhou, K. Z. (2008). Do Managerial Ties in China always Produce Value? Competi-tion, Uncertainty, and Domestic vs. Foreign Firms. Strategic Management Journal, 29(4), 383-400. Li, J., & Tang, Y. (2010). CEO Hubris and Firm Risk Taking in China: The Moderating Role of Managerial Discretion. Academy of Management Journal, 53(1), 45-68. March, J., & Shapira, Z. (1992). Variable Risk Preferences and the Focus of Attention. Psychological Review, 99(1), 172-183. McClelland, D. C. (1961). The Achieving Society. D. Van Nostrand, Princeton, NJ. Miller, K. D., & Leiblein, M. J. (1996). Corporate Risk-Return Relations: Returns Variability versus Downside Risk. Academy of Management Journal, 39(1), 91-122. Naldi, L., Nordqvist, M., Sjöberg, K., & Wiklund, J. (2007). Entrepreneurial Orientation, Risk Taking, and Performance in Family Firms. Family Business Review, 20(1), 33-47. Nee, V. (1992). Organizational Dynamics of Market Transition: Hybrid Forms, Property Rights, and Mixed Economy in China. Administrative Science Quarterly, 37(1), 1-27. Ortega-Argiles, R., Moreno, R., & Caralt, J. (2005). Ownership Structure and Innovation: Is There a Real Link? Annals of Regional Science, 39, 637-662. Palmer, T. B., & Wiseman, R. M. (1999). Decoupling Risk Taking from Income Stream Uncertainty: A Holistic Model of Risk. Strategic Management Journal, 20(11), 1037-1062. Peng, M. W. (2001). How Entrepreneurs Create Wealth in Transition Economies, Academy of Management Executive, 15(1), 95-110. Peng, M. W. (2003). Institutional Transitions and Strategic Choices. Academy of Management Review, 28(2), 275-296. Phan, P., Zhou, J., & Abrahamson, E. (2010). “Creativity, Innovation, and Entrepreneurship in China,” Management & Organization Review, 6(2), 175-194. Prahalad, C. K., & Bettis, R. A. (1986). The Dominant Logic: A New Linkage between Diversity and Performance. Strategic Management Journal, 7(6), 485-501. Rajgopal, S., & Shevlin, T. (2002). Empirical Evidence on the Relation between Stock Option Compensation and Risk Taking. Journal of Accounting and Economics, 33(2), 145-171. Ralston, D. A., Terpstra-Tong, J., Terpstra, R. H., Wang, X., & Egri, C. (2006). Today's State-Owned Enterprises of China: Are They Dying Dinosaurs or Dynamic Dynamos? Strategic Management Journal, 27(9), 825-843.


Singh, J. V. (1986). Performance, Slack, and Risk Taking in Organizational Decision Making. Academy of Management Journal, 29(3), 562-585. Sitkin, S. B., & Pablo, A. L. (1992). Reconceptualizing the Determinants of Risk Behavior. Academy of Management Review, 17(1), 9-38. Smith, N. R., & Miner, J. B. (1983). Type of Entrepreneur, Type of Firm, and Managerial Motivation: Implications for Organizational Life Cycle Theory. Strategic Management Journal, 4(4), 325-340. Stuttard, J. (2000). The New Silk Road: Secrets of Business Success in China Today. New York, John Wiley & Sons, Inc. Tan, J. (1996). Regulatory Environment and Strategic Orientations in a Transitional Economy: A Study of Chinese Private Enterprise. Entrepreneurship Theory and Practice, 21(1), 31-46. Tan, J. (2001). Innovation and Risk-Taking in a Transitional Economy. Journal of Business Venturing, 16(4), 359-376. Tan, J. (2005). Venturing in Turbulent Water: A Historical Perspective of Economic Reform and Entrepreneurial Transformation. Journal of Business Venturing, 20(5), 689-704. Tan, J., & Litschert, R. J. (1994). Environment-Strategy Relationship and its Performance Implications: An Empirical Study of Chinese Electronics Industry. Strategic Management Journal, 15(1), 1-20. Tang, J., Tang, Z., Zhang, Y., & Li, Q. (2007). The Impact of Entrepreneurial Orientation and Ownership Type on Firm Performance in the Emerging Region of China. Journal of Developmental Entrepreneurship, 12(4), 383-397. UNCTAD (United Nations Conference on Trade and Development). (2009). World Investment Prospects Survey 2009-2011. New York and Geneva. Watkins-Mathys, L., & Foster, M. J. (2006). Entrepreneurship: The Missing Ingredient in China's STIPs? Entrepreneurship & Regional Development, 18(3), 249-274. Wiseman, R. M., & Gomez-Mejia, L. R. (1998). A behavioral Agency Model of Managerial Risk Taking. Academy of Management Review, 23(1), 133-153. Wright, M., Hoskisson, R. E., Busenitz, L. W., & Dial, J. (2000). Entrepreneurial Growth through Privatization: The Upside of Management Buyouts. Academy of Management Review, 25(3), 591-601. Wright, P., Kroll, M., Krug, J., & Pettus, M. (2007). Influence of Top Management Team Incentives on Firm Risk Taking. Strategic Management Journal, 28(1), 81-89. Xin, K., & Pearce, J. L. (1996). Guanxi: Connections as Substitutes for Formal Institutional Support. Academy of Management Journal, 39(6), 1641-1658. Yang, K. (2004). Institutional Holes and Entrepreneurship in China. Sociological Review, 52(3), 371-389. Young, S. (1991). Wealth but not Security: Attitudes towards Private Business in China in the 1980s. The Australian Journal of Chinese Affairs, 25, 115-135.


An Analysis of the Behavior of Alternative Employment Indicators:

Was the Great Recession Different?

Bill Seyfried Rollins College

From 2007-2009, the United States experienced its worst economic downturn since the Great Depression. Though economic growth experienced a significant decline, the labor market was particularly hard hit. Some economists expressed surprise as to the severity of job losses. Unlike traditional models, this study develops a model that estimates the responsiveness of the labor market to economic growth as well as a risk premium to account for the impact of the availability of credit on the labor market. Alternative employment indicators as well as economic growth are considered. Empirical findings indicate that the behavior of employment during the Great Recession fit the historical experience based on the model developed. INTRODUCTION

There are many ways to gauge the performance of the job market. Each month, the government releases an assessment of the labor market based on two surveys – the household survey and the establishment survey. The household survey provides various estimates of the job market, with the two measures receiving the most attention being employment and the unemployment rate. The establishment survey also provides a variety of measures including nonfarm payrolls, private payrolls, and aggregate hours worked. Though they tend to move together over time, each measure provides a slightly different perspective as to the state of the labor market. When considering the number of jobs created, the most closely watched indicator has traditionally been nonfarm payrolls. In recent years, more attention has begun to be paid to the private sector job component of nonfarm payrolls, in order to assess the strength of the private sector in terms of economic recovery. Though the amount of net new jobs reported each month typically refers to the number from the establishment survey, the household survey also reports an estimate for net job creation. Though the results of the two reports tend to move together over time, they tend to differ somewhat in the short run. Some economists think the household survey is a more accurate gauge of a recovery in the job market (for example, see Meltzer 2003). However, most economists think the establishment survey is more reliable due in part to its larger sample size (440,000 establishments as opposed to 60,000 households). Just tallying the number of people employed may not be the most precise gauge of the labor market since to be considered employed one just needs to have a job, whether it’s full time or part time. The index of aggregate hours worked, obtained from the establishment survey, measures employment in terms of hours instead of jobs, which makes it a more precise assessment of the job market, particularly when companies reduce hours, but hold onto workers as seen during the Great Recession.


Regardless of which measure of the job market is used, its performance during the Great Recession was the worst since the Great Depression. Many post-Depression records were set, including a 6.2% decline in nonfarm payroll employment (8.7 million jobs) as well as a 9.5% decline in aggregate hours worked.

Much has been written about the downturn in the job market relative to the decline in the overall economy. According to Elsby, et.. al. (2010), the labor market performed similar to past recessions until early 2009, when the labor market dynamics began to change relative to past recessions. Some have considered the significant rise in unemployment while others focused on declines in employment. Uchitelle (2010) makes use of a version of Okun’s Law to find that, if historical relationships had held true, the unemployment rate in 2009 should have risen from 7.4% in the beginning of the year to 9% by year end. Various economists have explored the weakness of the labor market during the last few years. Given the decline in GDP during the recession, Brad Delong (2009) concludes that the unemployment rate should have risen to about 8% instead of 10%. Among other findings, Arpaia and Curci (2010) detect a difference in the behavior of jobs from hours worked as some economies tend to rely on reducing the number of hours that employees work rather than laying off workers. Burda and Hunt (2011) also find a distinction between the behavior of hours worked compared to jobs when comparing the performance of the US and German labor markets during the Great Recession.

Some other economists have proposed that the relationship between employment and economic growth has changed or that augmentations to existing models may help to explain the behavior of the labor market during the Great Recession. Gordon (2010) suggests that Okun’s Law broke down in 1986 as the cyclical behavior of productivity changed. Businesses became quicker to lay off workers than in prior times, resulting in a closer relationship between economic growth and employment. Instead of using the percent change in real GDP to measure economic growth, Wolfers (2010) uses the growth rate of real Gross Domestic Income (GDI). Results of his model indicate that it more accurately explains the steep rise in unemployment that took place in 2008-2009. Fatas and Mihov (2010) suggest that when one considers employment growth instead of the unemployment rate, 2009 stands out as an outlier, though 2007 and 2008 fit the historical pattern. They suggest that structural changes and/or access to credit may help explain the aberration. Nalewaik (2010) explores the differences in GDP and GDI and concludes that GDI better captures the business cycle fluctuations in true output growth. Furthermore, he finds that it is more highly correlated with other business cycle measures such as the unemployment rate and employment growth as measured by the household survey.

There’s a body of literature that has explored employment elasticities – how much employment responds to a percent change in economic growth. High levels of employment elasticity suggest that the labor market is very sensitive to changes in economic growth. For example, given an equal decline in economic growth, a country with a relatively high employment elasticity would experience a more significant downturn in its labor market than one with a relatively low employment elasiticity. Padalino and Vivarelli (1997) found employment elasticities close to zero for most G7 countries other than the US, which had an employment elasticity of 0.5. Using data from 1970-1998 for the EU and US, Walterskirchen (1999) found elasticities ranging from about 0.25 in Austria to 0.75 in Spain with the elasticity being about 0.5 in the US. Together this suggests that given the same decline in GDP, nations like Spain and the US would be expected to experience steeper declines in employment relative to other countries in the EU such as Austria. The experience of the Great Recession supports this finding as US unemployment peaked at just over 10% while Spanish unemployment rose past 20%. Meanwhile, unemployment in Austria did not rise past 5.5% despite it suffering a larger decline in GDP than the US. However, employment elasticities would not necessarily be static. For example, Pini (1997) found that the elasticity for the US didn’t change from the period of 1960-79 compared to 1979-1995 while it did change in several other nations – rising in some while declining in others.


DESCRIPTIVE STATISTICS

Quarterly data for all of the economic variables were obtained for the period from 1970 to 2009. Data for employment were obtained from the Bureau of Labor Statistics while the data for the growth rates of GDP, GDI, and Final Sales were obtained from the Bureau of Economic Analysis. All variables are reported in terms of seasonally-adjusted annualized rates. Descriptive statistics for the variables can be found in tables 1a-b.

TABLE 1A

DESCRIPTIVE STATISTICS OF THE GROWTH OF EMPLOYMENT INDICATORS

Payroll Employment

Private Employment

Household Employment

Aggregate Hours Worked

Mean 1.53% 1.56% 1.44% 1.27% Median 1.92% 2.15% 1.62% 1.68%

Minimum -6.56% -8.86% -6.36% -12.78% Maximum 7.16% 8.26% 6.87% 9.22%

Standard Deviation 2.36 2.81 2.35 3.79

TABLE 1B DESCRIPTIVE STATISTICS OF MEASURES OF GROWTH AND RISK PREMIUM

Growth Rate of GDP Growth Rate of GDI Final Sales Risk Premium

Mean 2.89% 2.87% 2.88% 2.11% Median 3.05% 3.20% 3.05% 1.95%

Minimum -7.9% -7.7% -7.6% 0.98% Maximum 16.7% 12.8% 16.8% 5.68%

Stand. Dev. 3.55 3.55 2.96 0.73

Nonfarm payrolls, estimated from the establishment survey, are the most widely followed measure of net job creation. Quarterly growth averaged 1.53% for the period, with a median growth rate of 1.92%. The largest decline, 6.56%, occurred in the first quarter of 2009 while the biggest increase in jobs, 7.16%, took place in the second quarter of 1978. Nonfarm payrolls grew about 78% of the time. Similarly, private employment rose by 1.56% on average, with a median growth rate of 2.15% per quarter. The steepest decline in employment, 8.86%, occurred in the fourth quarter of 1974 while the most rapid increase, 8.26%, was in the first quarter of 1978. Employment as measured by the household survey displayed mean and median growth rates of 1.44% and 1.62%, respectively. As with nonfarm payrolls, the largest drop occurred in the first quarter of 2009, a decline of 6.36%, with the biggest increase taking place in the second quarter of 1978 (6.87%). Household employment rose about 76% of the time.

Aggregate hours worked exhibited the lowest mean and median growth rates of 1.27% and 1.68%, respectively while growing about 71% of the time with zero growth 4% of the time. As expected, it experienced the most volatility with its sharpest decline of 12.8% in 1975Q1 and its largest gain was 9.2% in 1973Q1. This is to be expected since firms are more likely to adjust hours before adding or eliminating jobs. For example, firms may keep workers during the initial part of an economic downturn, but reduce their hours in order to maintain the worker once the economy begins to recover. Similarly, companies respond to the initial phase of an economic recovery by working existing workers longer hours before bringing on new workers. Hiring new workers involves the expense of a job search, incurring fringe benefits and job training while having existing employees work more avoids each of these expenses.


Economic growth, as measured by the percent change in real GDP, grew by 2.89% on average, with a median growth rate of 3.05%. The highest economic growth rate was 16.7% in the second quarter of 1978 while the economy plunged by 7.9% in the second quarter of 1980 (see figure 3 in appendix). The economy grew about 85% of the time. An alternative measure of economic growth, the percent increase in real gross domestic income, increased by 2.87% on average, with a median of 3.20%. The weakest quarter was a decline of 7.7% in the first quarter of 2009 while the strongest quarter was 12.8% in the second quarter of 1978 (see figure 4 in appendix). Real final sales grew at an average rate of 2.88% with a median growth rate of 3.05%. Growth in final sales reached a maximum of 16.8% in the second quarter of 1978 and experienced its steepest decline of 7.6% in the second quarter of 1980.

Estimates of confidence and risk can be subjectively measured using surveys, for example surveys of consumer and business confidence. However, a market-based estimate of risk exists in the form of credit spreads or risk premium. The risk premium is measured as the difference between the yield on a bond and a US Treasury bond of the same maturity. The US government is generally considered to be the least risky borrower (as evidenced by funds flowing to US Treasuries when there are “flights to safety”). The higher the yield on a bond relative to a comparable Treasury bond, the more return an investor is demanding to compensate for perceived risk. A common measure of the average risk premium for corporations is the yield on Baa corporate bonds, which are bonds rated as minimum investment grade by Moody’s. Yields on both corporate and Treasury bonds were obtained from the Alfred database maintained by the Saint Louis Federal Reserve. The risk premium averaged about 2% (a mean of 2.11% and median of 1.95%), reaching a low of 0.98% in the first quarter of 1979 and a high of 5.68% in the fourth quarter of 2008 (note: the risk premium exceeded 6% briefly in late 2008, but the quarterly average peaked at 5.68%). The risk premium was significantly higher during the financial crisis of 2008-2009 than any time since the early 1930s. EMPIRICAL MODEL

We seek to develop a model that explores the relationship between employment and economic growth in order to try to determine whether the relationship experienced during the Great Recession was consistent with prior periods. Three major measures of net job creation are considered: nonfarm payrolls, private sector employment (both from the establishment survey) and employment as estimated by the household survey. In addition, the index of aggregate hours worked is also considered. The first factor incorporated into the model involves the persistence of job creation. In other words, does job growth in one period tend to lead to job growth in the subsequent period? If so, how much persistence exists? Thus, lagged employment growth is included in the model.

When considering factors that affect job creation, growth in the overall economy clearly is an important factor. If the economy is growing, companies will hire more workers in order to increase production. Thus a positive relationship is expected, but the question is about the magnitude of the relationship. Does economic growth lead to slightly more jobs or a disproportionate increase in jobs? In addition to including economic growth, the change in economic growth is also included to account for acceleration or deceleration of economic growth. If economic growth is accelerating, firms are more likely to have confidence in the economic expansion and thus increase hiring. Besides the traditional measure of economic growth, the percent change in real GDP, real Gross Domestic Income (GDI) and real Final Sales are included in separate models to see whether they are superior predictors of growth in labor.

Many times, economists pay too little attention to the effects of financial markets on the overall economy. However, as shown by the financial crisis, credit availability has a direct impact on business. To capture the availability of credit (suggested by Fatas and Mihov, 2010), the risk premium is included - defined as the difference between the average interest rate on the ten-year corporate bond rated Baa by Moody’s and the comparable ten-year US Treasury bond. It should be noted that Baa bonds are investment-grade bonds and thus reflect the cost of credit for firms normally perceived as being relatively safe. Higher risk premiums are indicative of reduced access (or increased cost) of credit. In addition, the


risk premium can also be seen as a market-based measure of fear. When the risk premium is high, companies may become more cautious, resulting in increased layoffs and/or not hiring new workers. Thus, one would expect higher risk premiums to lead to less employment growth both due to tightness of credit and also more cautious business practices in terms of managing payrolls. As with economic growth, the model includes both the lagged risk premium and the change in the risk premium from the previous quarter. A high risk premium is a sign of tight credit and fear, thus having a negative effect on job growth. Meanwhile, an increase in the risk premium indicates deteriorating credit conditions and thus should also hinder job creation.

The model estimated can be described as follows:

FIGURE 1 MODEL OF EMPLOYMENT GROWTH

EMPt = B0 + B1EMPt-1 + B2ECONt-1 + B3∆ECONt + B4RPt-1 + B5∆RPt + et

where: EMP = growth rate of employment ECON = economic growth rate RP = risk premium

Various versions of the model were estimated, using alternative measures of economic growth and employment indicators as described above. In each case, results were tested for standard econometric problems. The only problem detected was the existence of ARCH effects. The existence of ARCH effects is common when analyzing time series data due to the correlation of the variance over time (Kennedy, 2008). As a result, the model was re-estimated to correct for ARCH effects with the results reported in tables 2 to 5. EMPIRICAL RESULTS

In each case, the models displayed high explanatory power in terms of the adjusted R2. Both economic growth, regardless of how it was measured, and the risk premium were highly significant in explaining the behavior of labor, whether in terms of employment or hours worked. Next, we discuss the empirical results for each measure of labor. Nonfarm Payroll Employment

As seen in table 2, when considering nonfarm payrolls, all variables were significant, in most cases at the 1% level. The adjusted R-squared ranged from a low of 0.79 when using GDP to represent economic growth to a high of 0.83 when using GDI. A high degree of persistence in employment growth was detected as the coefficient on lagged employment growth was approximately 0.55 in each case. A one percent increase in economic growth (measured by GDP or GDI) results in about a 0.26% increase in employment in the following period while a one percentage point change in the rate of economic growth has impact of just under 0.25% on job creation. However, when final sales are used to proxy economic strength, the impacts of a one-percent increase in economic growth rises to about 0.34% increase in employment, implying an equivalent change in final sales has a larger impact on nonfarm payroll employment than either GDP or GDI. The size of the risk premium in the prior period has a coefficient ranging from about -0.36 to -0.39, implying that a higher level of perceived risk has a negative impact on job creation. However, the change in the risk premium has a higher coefficient and tends to be more significant, ranging from about -0.7 to -0.94, indicating that changes in the perception of risk have a larger impact on the labor market.


TABLE 2 MODELS OF NONFARM PAYROLLS

GDP GDI Final Sales Constant 0.663**

(2.06) 0.616 (1.51)

0.435 (1.21)

Lagged employment growth 0.564*** (8.33)

0.553*** (7.42)

0.546*** (8.42)

Lagged risk premium -0.393*** (3.12)

-0.359** (2.25)

-0.365*** (2.67)

Change in risk premium -0.861*** (4.80)

-0.708*** (3.16)

-0.939*** (5.60)

Lagged economic growth 0.265*** (6.75)

0.260*** (6.21)

0.339*** (8.47)

Change in economic growth 0.213*** (8.03)

0.244*** (9.82)

0.245*** (8.55)

Adjusted R2 0.79 0.83 0.81 Variance Equation

constant 0.007 (1.48)

0.011* (1.87)

0.005 (1.07)

ARCH -0.038*** (5.58)

-0.050* (1.79)

-0.030*** (4.12)

GARCH 1.017*** (18.85)

1.020*** (44.45)

1.013*** (12.28)

Note: z-statistics in parentheses; *** significant at 1% level, **5% level, *10% level Private Employment

Similar to nonfarm payrolls, private employment displayed persistence, with coefficients close to 0.55 (see table 3). Economic growth, regardless of how it was measured, had a comparable effect on private-sector jobs as it did to nonfarm payrolls, being more sensitive to final sales than either GDP or GDI. Private employment growth was highly related to changes in the risk premium as evidenced by both a higher magnitude and significance of the coefficient on the change in the risk premium relative to that found for nonfarm payrolls, with coefficient ranging from almost -1 to nearly -1.25. Though not as high as for nonfarm payrolls, the R2 was still quite high, ranging from 0.67 to 0.73 with the highest value occurring when economic growth was measured by final sales.

TABLE 3 MODELS OF PRIVATE EMPLOYMENT

GDP GDI Final Sales Constant 0.785*

(1.71) 0.567 (1.40)

0.307 (0.65)

Lagged employment growth 0.526*** (6.61)

0.550*** (7.14)

0.551*** (8.62)

Lagged risk premium -0.456** (2.45)

-0.343** (2.09)

-0.327* (1.80)


-0.966*** (3.72)

-1.232*** (4.65)


0.263*** (4.15)

0.353*** (5.40)



0.245*** (6.08)

0.263*** (7.69)


constant 0.017 (1.32)

0.009 (1.48)

0.019 (1.37)

ARCH 0.033 (1.09)

-0.037*** (23.89)

0.024 (1.04)

GARCH 0.939*** (28.33)

1.018*** (72.25)

0.947*** (31.92)

Note: z-statistics in parentheses; *** significant at 1% level, **5% level, *10% level Household Employment

Employment as estimated from the household survey revealed little persistence, insignificant when using GDP or GDI and less than 0.15 when using final sales. Economic growth (in terms of GDP or GDI) of one percent led to a corresponding increase of about 0.4% employment growth while a one percentage point increase in the rate of economic growth led to about a 0.3% increase in employment. When using final sales, both economic growth and the change in economic growth had slightly larger effects on jobs. Employment from the household survey is quite sensitive to the risk premium with the coefficient on the risk premium from the prior period of between -0.57 and -0.84 and that on the change in the risk premium being between -0.87 and -1.2.

TABLE 4 MODELS OF HOUSEHOLD SURVEY

GDP GDI Final Sales Constant 1.940***

(3.80) 1.253** (2.32)

1.323** (2.37)

Lagged employment growth 0.025 (0.29)

0.050 (0.57)

0.135** (2.00)


-0.572*** (2.72)

-0.699*** (3.33)


-0.867** (2.40)

-1.087*** (3.68)


0.445*** (7.45)

0.479*** (7.84)


0.307*** (7.27)

0.353*** (6.86)


constant 0.169 (0.91)

0.489 (0.73)

4.013 (3.55)

ARCH 0.053 (0.72)

0.101 (0.99)

-0.105** (1.95)

GARCH 0.874*** (7.36)

0.681** (1.96)

-0.581* (1.62)

Note: z-statistics in parentheses; *** significant at 1% level, **5% level, *10% level


Aggregate Hours Worked Though higher than that displayed by the household survey, when significant, aggregate hours

worked displayed a low level of persistence (between 0.17 and 0.27), considerably less than what was found for nonfarm payrolls and private employment. Hours worked was more sensitive to economic growth than the other measures of the labor market, with coefficients on lagged economic growth somewhat above or below 0.7 and those on the change in economic growth of about 0.5. As with economic growth, aggregate hours worked was highly sensitive to the risk premium, with a change in the risk premium of one percentage point leading to just under a two percentage point decline in hours worked in two of the models (and about 1.7 decline in the case of GDI). The risk premium from the prior period also had a sizeable negative effect on hours worked as well, with a one-percent higher risk premium leading to between a 0.8% to 1.1% decline in employment growth, much higher than found for other labor market measures.

TABLE 5 MODELS OF AGGREGATE HOURS WORKED

GDP GDI Final Sales Constant 1.624***

(2.60) 1.060 (1.28)

0.533 (0.76)

Lagged employment growth 0.173** (2.40)

0.153* (1.80)

0.271*** (3.52)


-0.894*** (2.72)

-0.796*** (2.89)


-1.673*** (3.76)

-1.912*** (5.26)


0.670*** (6.93)

0.732*** (7.56)


0.524*** (7.64)

0.532*** (8.14)


constant 0.085* (1.93)

0.098* (1.65)

0.047 (1.04)

ARCH -0.049 (1.53)

-0.056 (1.54)

-0.042 (1.12)

GARCH 1.015*** (37.51)

1.021*** (34.71)

1.022*** (29.26)

Note: z-statistics in parentheses; *** significant at 1% level, **5% level, *10% level SUMMARY AND CONCLUSIONS

Using GDI or final sales marginally improved the explanatory power of the respective models, with adjusted R2 rising regardless of what measure of employment is considered. Coefficients were highly significant, at the 1% level in most cases. A high degree of employment persistence was found when considering nonfarm payrolls or private employment while a lower degree of persistence was detected for aggregate hours worked (no persistence was found for employment based on the household survey). Thus, the labor market as measured by the establishment survey (nonfarm payrolls and private sector employment) display some momentum, whether to the upside or to the downside. Economic growth, from the previous period as well as the change in economic growth positively affected employment growth. Final sales had a marginally higher effect on employment than GDP or GDI.


Despite receiving scant attention from many economists, the risk premium had a very significant effect on the growth in employment, both in terms of the lagged risk premium as well as the change in the risk premium from the preceding period. Whether due to tight credit for investment-grade corporations or as a measure of fear, high risk premiums significantly hindered job growth with the change in the risk premium having a particularly large effect. Given the record-high risk premiums in late 2008 and early 2009, this helps to explain why those who didn’t account for this effect may have missed the subsequent deterioration in the labor market.

Though the Great Recession ended in the summer of 2009, the lingering effects are still being felt years later, particularly in the labor market as it tries to recover from a loss of 8.7 million jobs. Some economists, based on traditional models, expressed surprise as to the dramatic loss of jobs that took place during the recession. In this study, a model was developed to account for the average risk premium on investment-grade (Baa) corporate bonds in addition to more traditional factors such as the effects of economic growth, changes in economic growth, and the persistence in employment growth. Economic growth, had the expected impact on employment growth, but doesn’t fully explain the behavior of the labor market. As suggested by Wolfers (2010), using gross domestic income instead of gross domestic product improved the explanatory power of the model, but only marginally. Employment was even more sensitive to final sales than GDI (or GDP). Though credit is affected somewhat during most recessions, the freezing of credit markets during the financial crisis of 2007-2009 appeared to have a powerful effect on the job market, as suggested by Fatas and Mihov (2010). The empirical results reveal that the dramatic rise in the risk premium during late 2008 and early 2009 helps to explain the huge decline in employment during the depths of the recession. Though economic growth is an important factor in understanding the behavior of employment growth, the risk premium, whether seen as a measure of credit availability or as a market-based measure of fear, is critical in understanding the behavior of the labor market during the Great Recession. REFERENCES Arpaia, Alfonso and Curci, Nicola (2010). EU Labour Market Behavior During the Great Recession. European Commission Economic Papers 405. http://ec.europa.eu/economy_finance/publications/economic_paper/2010/pdf/ecp405_en.pdf Burda, Michael, and Hunt, Jennifer (2011). What Explains the German Labor Market Miracle During the Great Recession? Brookings Panel on Economic Activity, Spring 2011. http://www.brookings.edu/~/media/Files/Programs/ES/BPEA/2011_spring_bpea_papers/2011_spring_bpea_conference_burda.pdf DeLong, Brad (2010). The Life and Strange Death of Okun’s Law. http://delong.typepad.com/sdj/2010/02/the-life-and-strange-death-of-okuns-law.html Elsby, Michael, Hobijn, Bart and Sahin, Aysegul (2010). The Labor Market in the Great Recession. Federal Reserve Bank of San Francisco Working Paper 2010-07. http://www.frbsf.org/publications/economics/papers/2010/wp10-07bk.pdf Fatas, Antonio and Mihov, Ilian (2010). Labor Markets and the Current Cycle (and the Death of Okun’s Law). http://fatasmihov.blogspot.com/2010/02/labor-markets-and-current-cycle-and.html Gordon, Robert (2010). Okun’s Law, Productivity Innovations, and Conundrums in Business Cycle Dating. Presentation at the AEA meeting, Jan 4, 2010. http://faculty-web.at.northwestern.edu/economics/gordon/AEAlong_Combined_100108.pdf


Kennedy, Peter (2008). A Guide to Econometrics. Malden, MA: Blackwell Publishing. Meltzer, Allan (2003). A Jobless Recovery. http://www.mosler.org/wwwboard/messages/518.html Nalewaik, Jeremy (2010). The Income- and Expenditure-Side Estimates of U.S. Output Growth. Brookings Paper on Economic Activity. http://www.brookings.edu/~/media/Files/Programs/ES/BPEA/2010_spring_bpea_papers/spring2010_nalewaik.pdf Padalino, S. and M. Vivarelli (1997). The Employment Intensity of Economic Growth in the G-7 Countries. International Labour Review, 136, 191-213. Pini, P. (1997). Occupazione, Tecnologia e Crescita: Modelli Interpretativi ed Eevidenze Empiriche a Livello Macroeconomico. 1997. Paper Presented to the Conference of the Accademia Nazionale dei Lincei on "Sviluppo tecnologico e disoccupazione: trasformazione della societa", held in Rome, 16-18 January 1997. Uchittelle, Louis (2010). A Broken Economic Law. NY Times, Feb 22, 2010. http://economix.blogs.nytimes.com/2010/02/22/a-broken-economic-law/ Walterskirchen, E. (1999). The Relationship Between Growth, Employment and Unemployment in the EU. European Economist for an Alternative Economic Policy Workshop, Barcelona, Spain. http://www.memo-europe.uni-bremen.de/tser/Walterskirchen_24months.PDF Wolfers, Justin (2010). Is Okun’s Law Really Broken? NY Times, March 1, 2010. http://freakonomics.blogs.nytimes.com/2010/03/01/is-okuns-law-really-broken/ DATA SOURCES Bureau of Labor Statistics

• Aggregate Hours Worked: http://www.bls.gov/webapps/legacy/cesbtab4.htm • Employment (Household survey): http://www.bls.gov/webapps/legacy/cpsatab1.htm • Nonfarm payrolls: http://www.bls.gov/webapps/legacy/cesbtab1.htm • Private Employment: http://www.bls.gov/webapps/legacy/cesbtab1.htm

Bureau of Economic Analysis, National Economic Accounts • GDP (Table 1.1.1): http://www.bea.gov/national/nipaweb/SelectTable.asp?Selected=Y • GDI (Table 1.7.1): http://www.bea.gov/national/nipaweb/SelectTable.asp?Selected=Y • Final Sales (Table 1.2.1): http://www.bea.gov/national/nipaweb/SelectTable.asp?Selected=Y

ArchivaL Federal Reserve Economic Data • Risk Premium (Baa) = ten year Baa corporate bond yield – ten year Treasury bond yield

o Baa corporate bond: http://alfred.stlouisfed.org/series?seid=BAA&cid=119 o U.S. Treasury bond: http://alfred.stlouisfed.org/series?seid=GS10&cid=115


Important theories of Unemployment and Public Policies

Adil H. Mouhammed University of Illinois at Springfield

This paper intends to analyze the most important theories of unemployment. These theories are scientifically developed and confirmed by economists representing various schools of economic thought such as the Keynesian and the classical schools of political economy. These theories are used to develop some essential public policies that can be employed to reduce the unemployment rate. INTRODUCTION

Many countries whether advanced capitalist economies or developing countries have experienced very high rates of unemployment since the Great Recession of December 2007. The American economy faces unemployment rate of 9.2 percent in June 2011, and Egypt has a rate of unemployment of 19 percent. The Saudi economy faces a rate of unemployment of 10 percent. This problem is very costly economically and politically. Economically, unemployment represents a loss in the Gross Domestic Product (GDP). Politically, the world witnesses the Arab revolt in Egypt, Tunisia, Syria, Libya, Iraq, and Bahrain, to mention a few, a revolt that is caused by unemployment, poverty, inequality, and dictatorship.

Economic literature provides many explanations for the unemployment problem. Some causes blame the economic systems, and others blame the unemployed workers. Still, other theories shift the problem to external sources and shocks, or unpredictable events, and others argue that technology and labor market institutions are the causes of the unemployment problem. Other theories think the deficiency in aggregate spending and innovations are the essential factors for explaining the problem.

This paper intends to review analytically these dominant determinants in the next seven sections. Section nine uses some of these theories to develop a set of public policies capable of reducing the rate of unemployment. The last section provides a summary and conclusions of this work. UNEMPLOYMENT IN THE CLASSICAL ECONOMIC THEORY

The classical theory, as analyzed by Pigou (1933) and Solow (1981), argues that the labor market consists of demand and supply of labor. Demand for labor is a derived demand, obtained from the declining portion of the marginal product of labor. The demand curve is a negative function of real wage in that if wages increase the quantity demand for labor will decline and the opposite is correct. The supply of labor is derived from worker's choice whether to spend part of time working or not working (leisure). Supply of hours worked is a positive function of the real wage, because if the real wage rises, workers supply more hours of work. In equilibrium, demand and supply of labor are intersected at a clearing point that determines the equilibrium real wage rate and full employment. Unemployment, Sweezy (1940: 807) explaining Pigou’s Theory of Unemployment, “apart from frictional obstructions…would be nonexistent


if it were not for the fact that wage-earners habitually stipulate for a rate of wages higher than the ‘equilibrium’ level.”

Full employment does not mean that there is no unemployment. Still frictional unemployment does exist at the going real wage rate. For example, if a worker thinks that the disutility of work is greater than the benefit of work or the utility of the real wage, this worker will decide not to work. This type of unemployment is called voluntary unemployment. Frictional unemployment arises because of the dynamic nature of the labor markets, the availability of information, the search for better jobs, and random fluctuations in demand for labor such as closing of a plant and of opening of a new plant. Duration of frictional unemployment is determined by the unemployment insurance benefits and the speed of the information.

Wicksell thinks that if wages are sufficiently flexible downward, then this decline can maintain full employment (Jonung 1989: 28-35). Cheaper credit to businessmen is the most effective measure to fight unemployment. He even thought that the government should support private investment in housing and soils. Government can support the introduction of various inventions as well. Government support should be financed by taxation. Wicksell analyzes technical unemployment due to technological change as well. The introduction of machinery would cause unemployment but the unemployed will search for new jobs, a search that will push wages downward. Hence, full employment is restored again. For the normal (frictional) unemployment, Wicksell thinks that advertisements and employment agencies can reduce the normal rate of unemployment. The cyclical unemployment, as another type of unemployment, is due to the lack of effective demand. He though it would be a good idea to raise wages in order for workers to buy more. But this action may cause workers to lose their jobs as a result of higher wages.

Essentially, for Wicksell the cyclical unemployment was due to the wrong investment of capital. Capital was invested in areas where rates of return were low. He concluded that public works is the best measure to fight cyclical unemployment. After World War I, Wicksell thinks the boom and the rise in prices induced by the war would come to an end. Thus, unemployment would rise. Workers would have to accept lower wages. He also thought that government should provide financial support to the unemployed who could not find jobs. After 1921, Wicksell turns to Malthus. He thought that the causes of the unemployment are the surplus people, shortage of capital brought about by the war, and the disorganized state of the monetary system. For the third cause, after the war prices were falling and producers decided to produce lower amounts of production because they knew they would receive lower prices for their products. Thus, they let their money set idle in banks and workers became unemployed. These causes suggest that emigration became one of the important policies for solving the unemployment problem.

Wage reduction is not a competent policy to increase employment. The increase in wages is most likely due to increased labor productivity and wage reduction will reduce work intensity and productivity. Wage reduction will not force some capital intensive firms to switch to labor intensive techniques in the short run. Higher wages should stimulate the substitution effect by employing more machines for labor. And this substitution will increase labor productivity and employment in the long-run.

Hayek (Nishhiyama and Leube 1984: 7) contends that unemployment is due “to a discrepancy between the distribution of labor…between industries…and the distribution of demand among their producers. This discrepancy is caused by a distortion of the system of relative prices and wages.” In other words, the unemployment is caused by “a deviation from the equilibrium prices and wages which would establish themselves with a free market and stable money.” This is actually a mismatch between demand and supply of labor, which is usually caused by expansionary monetary and fiscal policies and powerful trade unions. These policies create economic dislocation and structural changes in an economy which misdirect labor and other economic resources to other alternatives. Unions are also able to set higher wages compared to market wages, which generate unemployment, particularly in industries that become less profitable. In short, for Hayek the unemployment problem is caused by resources being in the wrong places at the wrong time and can be corrected if wages and prices are determined by the equilibrium of supply and demand.


In line with Hayek theory of unemployment, Trehan (2001) provides an important explanation of the search theory of unemployment. Firms search for the productive workers and workers search for high-paying jobs. So, both agents continue searching until matches are reached. At that point a worker will leave the unemployment pool. But if a worker realizes later on that her productivity is worth higher wages and firms are paying high wages on the average, then the worker’s reservation wage will increase. Consequently, the unemployment rate will start rising gradually, indicating a mismatch has occurred again. UNEMPLOYMENT IN THE THEORY OF INNOVATIONS

Originally, this theory was developed by the German economist Von Mangoldt (Ekelund and Hebert 2007) wrote a book about entrepreneurial profits in 1855 and connected profits to risk. He provided several ways by which the entrepreneur can make profits. These ways are (1) finding particular markets, (2) acquisition of productive agents, (3) skillful combination of factors of production, (4) successful sales policy, and (5) innovations. And it is well understood proposition that entrepreneurial profits will increase employment (Mouhammed 2010).

Schumpeter (1934) does not provide explicitly a theory of unemployment but his theory of the business cycle does demonstrate clearly how unemployment can be reduced. Innovation (see also Vecchi 1995) which creates more jobs relative to job destruction is the basic force beyond the increases in employment and the decreases in unemployment. When entrepreneurs innovate something new such as the production of a new product, the finding of a new market, the finding of a new method of production, and the introduction of new technologies and a new organization they increase investments to materialize those innovations. Domestic investment expenditures will increase demand on economic resources and will increase their prices. Other entrepreneurs will imitate the leaders by adopting the new innovations. Labor and materials will be employed to produce the new items. Consequently, wages will be increasing and unemployment will be declining, assuming that employment creation will outweigh employment destruction due to the new innovations (see also Mortensen and Pissarides 1998 and Manuelli 2000).

Schumpeter started his analysis by explaining economic development. By development, which is the essential part of his endogenous dynamic economics, Schumpeter (1934: 83) means the “changes in economic life as are not forced upon it from without but arise by its own initiative, from within. Should it turn out that there are no such changes arising in the economic sphere itself, and that the phenomenon that we call economic development is in practice simply founded upon the fact that the data change and that the economy continuously adapts itself to them, then we should say that there is no economic development.” Economic development which reflects new changes outlined below is not a phenomenon that can be explained by economic forces only, but it has to be explained by other forces that are external to the ones analyzed by economic theory.

For Schumpeter, economic development generates changes in the socio-economic environment, including the existing equilibrium. As he (1934: 64) puts it: “Development ...is spontaneous and discontinuous change in the channels of the flow, disturbance of equilibrium, which forever alters and displaces the equilibrium state previously existing.” The essential driving force for generating development is innovations introduced by the entrepreneurs whose leadership becomes the triggering device for the discontinuous dynamic changes. Innovations start by “the producer [not consumer] who as a rule initiates economic change, and consumers are educated by him if necessary” (Schumpeter 1934: 65). It follows that economic development is defined “by the carrying out of new combinations” which are triggered by the business entrepreneur and appeared discontinuously (Schumpeter 1934: 66). And the outcomes of these combinations are welcomed by the consumers who are affected by the entrepreneurial leadership. That is, leadership becomes the prime mover to consumers and other imitating producers.

The concept of innovation which creates changes according to Schumpeter (1934: 66) covers the following five areas of development: “(1) the introduction of new good...or of a new quality of a good. (2) The introduction of a new method of production....(3) The opening of a new market....(4) The conquest of a new source of supply of raw materials, or manufactured goods....(5) The carrying out of the new


organization of any industry, like the creation of a monopoly position...or the breaking up of a monopoly position.” The new combinations are usually embodied in new productive enterprises which start by utilizing the unemployed working people, the unsold raw materials, the new technologies, and the unused productive capacity. As Schumpeter (1934: 68) points out, “Development consists primarily in employing existing resources in a different way, in doing new things with them, irrespective of whether those resources increase or not.”

For the continuation of the process of economic development and innovations credit and finance are important requirements: “in carrying out new combinations, financing...is fundamentally necessary” (Schumpeter 1934: 70). Credit is a very important function in economic development because it provides funds for the entrepreneurs to materialize innovations, or to carry out the new combination. Consequently, Schumpeter (1934: 74) argues, the banker who has savings and creates the money (or the purchasing power) for the entrepreneur is “a phenomenon of development.”

Accordingly, entrepreneurial leadership which triggers the process of economic development is “a special kind of function and in contrast to a mere difference in rank, which would exist in every social body” (Schumpeter 1934: 87). Leadership arises “only where new possibilities present,” and innovations require leadership, and the entrepreneur is the leader who responds to reality effectively and creatively. For her efforts, the entrepreneur earns entrepreneurial profits as a surplus over costs. That is, new combinations are carried out if there is development and if total receipts are greater than the total costs. Most importantly, Schumpeter (1934: 154) contends, “without development there is no profit, without profit [there is] no development.” And without profit there is no accumulation of wealth under capitalism. In other words, entrepreneurial leadership becomes the essential driving force for the business enterprises and the backbone of competitive capitalism.

Not only does economic development generate employment, income, and profits, but it also creates the value of land (rent), and without development the value of land does not exist. As the process of development continues the land value will rise due to urban and rural expansion. Moreover, development creates demand for certain goods. This is called repercussion of development, which creates surpluses (Schumpeter 1934: 172). Hence, profits are augmented in the process of the repercussion of development, which will in turn create another price for credit which is called the interest rate. Interests will be paid out of the profit or the surplus value. It is also true that without development there is no interest, but the process of development makes interest act “as a tax upon profit” (Schumpeter 1934: 175). For Schumpeter, supply and demand for credit will determine the interest rate, where the demand for credits is discontinuous because innovations are discontinuous. In short, higher wages and employment, economic profit, interests, and rents are all phenomena generated by innovations which in turn furnished by the entrepreneur.

During the process of economic development the economy is drifted toward a boom which is followed by a downturn, or a recession. Schumpeter contends that during the early period of the prosperity phase of the business cycle, the new innovating firms generate a higher demand for economic resources which must come from other industries. However, an innovative firm means it is able to produce per unit of a product at a smaller cost (Schumpeter 1928: 378). At the same time the innovative firms start selling the new products at reasonable prices, reflecting the economic power of these innovative enterprises. Given the low cost of production, the reasonable prices will generate higher revenues and surpluses which include profit.

The profit, however, is a temporary phenomenon. This is because some older firms become adapted to the new conditions and innovations and will be able to imitate (or copy) the methods and the products of the leading innovative enterprises. On the one hand, demand for economic resources will rise, so will their prices and the cost of production. Cost per unit of output will increase. On the other hand, the large volume of production will lower the prices, as firms lose their economic power for setting higher prices for their products. Consequently, as costs rise and revenues decline, profits will be eliminated, and liquidation will follow. Pessimism emerges and the capitalist economy moves toward a recession or a depression. Revival will start again after new swarms of innovations are initiated by some entrepreneurs.


Business enterprises whose leaders are creative will establish their economic power again for setting higher prices for low cost production. Profits will be rising, so will investments and employment.

Essentially, Schumpeter’s theories of economic development and the business cycle are based on the entrepreneur and her creative leadership and responses. Schumpeter emphasizes the fact that innovation means creative destruction, destroying old products, firms, and markets and creating new products, firms, markets, and technologies that generate secondary waves. Innovation is a matter of entrepreneurial leadership and individual initiatives (Schumpeter 1928: 384). The entrepreneurial creative responses, Schumpeter (1947: 150) argues, are not predictable but are generating significant changes for a long period of time. Hence, entrepreneurship according to Schumpeter (1947: 151) is the mechanism of economic change, and the entrepreneur is the one who gets things done and controls the resistance and difficulties facing her business operations. In fact, the entrepreneur is the force behind economic and institutional changes such as technologies, products, contracts, property, labor relations, regulations, security, and freedom. UNEMPLOYMENT IN THE THEORY OF EFFECTIVE DEMAND

Veblen points out that the volume of output is set to attain a satisfactory profit and is a manifestation of the predatory instinct of the vested interests which aim at domestic and international dominance. But how is this volume of production determined to achieve reasonable profits? Veblen gives a lucid answer. He accurately realizes, and before Lord Keynes reaches a similar conclusion, that vested interests determine the volume of output after taking into consideration the aggregate demand. As Veblen (1904: 195) explains:

In part by actual increase of demand and in part through a lively anticipation of an advanced demand, aggressive business enterprise extends its venture".

And the 'venture', of course, means extending production and operations, assuming the existence of a reasonable level of profits.

The level of aggregate demand will provide the necessary increases in total revenues. On the other side, the cost of production has to decline. If revenue rises and cost declines, then the reasonable level of profits can be found. There are various forces in Veblen’s work that reduce the cost of production. Technology increases production and reduce the cost of inputs used in the production process, and enterprises cut wages and increase productivity in order to cut cost per unit of output. Better technology can reduce the prices of capital goods, and government can cut taxes. Banks can reduce the interest rates as well. Administrative and insurance cost can be declined in order to stimulate business enterprises. The decline in costs, given rising revenues, will increase the profit level for Veblen. Consequently, higher profits will force the business enterprises to expand and employ more workers. Thus, employment will increase and the rate of unemployment will decline.

Keynes (1936) considers unemployment as an involuntary phenomenon. He thinks that employment is cyclical, generated by the deficiency of aggregate demand (Mouhammed 2010). Capitalists hire workers and invest to produce output when the expectations about the economy and profits are favorable. If expectations about the future are supported by reality, investments and employment continue rising until equilibrium is reached. This equilibrium is attained by the intersection of the aggregate demand and supply--the point of the effective demand—which may be less than the full employment equilibrium. If expectations about the future of the economy are not favorable, capitalists invest less and employ less number of workers. Hence, the equilibrium is achieved where cyclical unemployment exists. This unemployment is due to the deficiency of the aggregate demand, particularly investment expenditures.

Consistent with Keynes’s teaching, Davidson (1998), a representative of Post Keynesian economics, argues that involuntary unemployment is explained by insufficiency of effective demand, instability of exchange rates, and international mobility of finances which create uncertainty that weakens entrepreneurial confidence to make investments to reduce unemployment. Similarly, other Keynesians


argue that the unemployment is due to the contractionary nature of the U.S. monetary policy which creates deficiency in aggregate demand. Other Keynesians think that the unexpected increase in price level, or a higher rate of inflation, will reduce the real wage and increase demand for labor. That is, the rate of unemployment will decline. This idea reminds one with the old proposition of Phillips curve suggesting there is a trade-off between the rate of unemployment and the rate of inflation. UNEMPLOYMENT IN THE REAL BUSINESS CYCLE THEORY

It is argued in this theory (Chatterjee 1995 and 1999) that the growth of productivity of input which revolutionizes technology is the main source of employment and unemployment. If the growth of output increases more than the growth of inputs, then total factor productivity or the residual, has increased. If total factor productivity is not growing, then firms and the economy become inefficient. It follows that reallocation of labor and capital cannot be achieved and labor and capital will be used in less profitable opportunities.

There are various causes for the slowdown in total factor productivity. Technology is not improving in the production of goods and services and workers’ skills are not being enhanced. New products are not invented and when the prices of imported materials are increasing. Once total factor productivity is stagnating, the co-movements in other important variables will slowdown. For example, consumption expenditures will not increase above the trend, nor will investment spending. GDP and total hours worked will not be above the trend either. When consumption, investment, GDP, and hours of work decline, the Solow’s residual, which represents the growth in labor productivity and is measured by the difference between actual and predicted productivity growth (or shocks), will decline. That is, there is no improvement in technology and productivity under this condition. Therefore, unemployment will increase.

Technology shocks are brought about by scientific and engineering development, by R&D, management techniques, and by industrial organizations that make inputs more productive. In Schumpeter’s terminology innovations are being introduced and are very effective in making the economy grow. Innovations and favorable technological shocks also reduce inputs and increase total factor productivity. In short, if shocks to productivity brought about by technological shocks do not exit, the unemployment will rise.

Gali and Rabanal (2004) contend that demand and monetary shocks affect the variables of the business cycle, including employment, by about 75 percent, where the technological shocks affect those variables by about 25 percent. Gali (1999) finds that the positive shocks in technology generate a decline in hours of labor and negative comovement between technology shocks and productivity. For him, non technology shocks generated positive comovement between hours and productivity. His results were not consistent with the real business cycle theory. UNEMPLOYMENT AND PRODUCTIVITY

The standard microeconomic theory produces a similar result with different terminology. In this theory the marginal revenue product is MRP which is the multiplication of marginal revenue (MR) by the marginal product of labor (MPL), or productivity. Mathematically, it is MRPL = (MR) (MPL). And the profit-maximizing firm will hire workers until MRPL = W, where W is the given wage rate. It is assumed that the production function is of the form where output (Q) depends on two resources Labor (L) and all other resources combined as O, and is subject to a constant return to scale, where the sum of the exponents of L and O is equal to one. It is also assumed that the production function is affected by the technological level A such that,

Q = ALaOb …1 Differentiate the production function partially with respect to labor yields


∂Q/∂L = aALa-1Ob …2 Use the marginal product of labor in the MRPL equation to obtain

MRPL = MR x MP = MR x (aALa-1Ob) …3

because,

MPL = (aALa-1Ob) The MRPL should equal to the real wage rate W/P, where P is the price of the product. The MRPL is equal to the value of the marginal product of labor if MR = P under perfect competition. Under imperfect competition, the MRPL is smaller than the value of the marginal product, indicating the existence of labor exploitation. In any event, solving for L, we obtain the employment level

L = aPQ/W ...4

where PQ represents the gross domestic product, or GDP. If the numerator and the denominator of the above equation are divided by L, one can obtain

L = aPQ/L/W/L = a (average product of labor)/(average wage)

This equation states that if labor productivity (or the average product of labor) increases, assuming W is constant, the demand for labor, L, will rise, and the unemployment rate will decline. And this shift (or increase) in the demand for labor can occur, for example, if investment or capital formation increases. This is because if labor productivity increases relative to wages, the employer or the producer will increase the firm’s rate of profit by hiring more workers (L).

The previous analysis was adopted by Arthur Lewis (1954). Lewis develops what was called the Lewis model in which he assumes that if there was a surplus of labor and a given demand for labor, then the wage rate is fixed. Lewis points out that under this condition capitalists do make a certain level of profit. The capitalists will reinvest part of the profits in new capitals. This investment will raise labor productivity. Hence, the demand for labor will increase, and these new employed workers can come from low productivity sectors or the rural areas. This increase in employment will provide more profits for the capitalists, and more profits will increase investment, employment, and income. In short, demand for labor will shift to the right when labor productivity rises, indicating an increase in employment and income.

Clearly, the introduction of new innovative marketing techniques will increase the demand for the product and consequently will increase demand for labor. Moreover, if productivity increases due to a greater utilization of capital goods, new technological advances, and better quality of labor (due to education, training, and health), then the demand for labor (or employment) will increase. In other words, successful innovations will increase productivity and employment (Schumpeter 1934). In addition, if the prices of capital goods decline, the quantity demanded for these goods will rise. Consequently, output will increase, so will the employment of labor. If resources are complement, the employment of more capital in the production process will increase the demand for labor or employment.

In fact, technological change or growth will be equal to the growth rate of output minus the growth rate of labor productivity. If productivity increases significantly, it will increase the growth rate of the gross domestic product (GDP) with larger increases than productivity, which forces employers at that point to hire more workers to accommodate expected demand. Wages will rise but if labor productivity increases at a rate faster than the increase in wages, then the rates of inflation and unemployment will decline.


MARKET STRUCTURE AND UNEMPLOYMENT

Microeconomic theory contends that market structure does affect the level of unemployment. Under the perfect competitive market the demand for labor and the supply of labor intersects to determine the wage rate and employment. Under this market condition, the demand for labor reflects labor productivity. Employers will hire workers up to the point where the value of marginal product (VMPL = MPL X P) is equal to the wage rate. Let us call this employment level Ec.

Under imperfect market conditions the outcomes are different. The demand for labor reflects the marginal revenue product (MRPL = MPL X MRX), where MRX is the marginal revenue. Under this condition, MRX is lower than the market price P. It follows that the MRPL is lower than VMPL. That is, there is exploitation of the labor force by employers. The second important outcome under the imperfect market condition is that the marginal labor cost will be above the labor supply. When the MRPL intersects the marginal factor cost, the employment of labor (Em) is determined. But this employment level is lower than the level of employment determined under perfect market condition (Ec).

It follows that an economy dominated by large corporations will generate lower level of employment and a higher rate of unemployment. In some cases such as the NFL and NBA, among others, collective bargaining is used to settle wages. Even John B Clark (1894: 8) points out that monopoly “wants a high price for its own special product and it can get this only by reducing the amount created. This means fewer men in its own shops.” In other words, a restricted output will increase the rate of unemployment. Veblen (1921) calls this intentional restriction of output by large corporations as sabotage. For him, sabotage increases unemployment of plants and workers. GLOBALIZATION AND UNEMPLOYMENT

Recently, the unemployment problem has been attributed to the globalization process. For example, shifting production and outsourcing to other countries have generated a high rate of structural unemployment in the U.S. economy, which has contributed for the increase in the rate of unemployment. Appreciated dollar during the 1980s made American exports expensive in the world market, which reduced the export level, causing unemployment to increase in the exporting industries. But an appreciated dollar increased the imports from foreign countries, which raised the American rate of unemployment. Currently, the opposite has happened. The dollar has been depreciating against the major currencies, which have made the American exports cheaper globally. Hence, exports have increased and imports have become very expensive, which have contributed for increasing the level of employment.

Over the last two decades one can contend that generally the transportation cost has been declining (before the increases in oil prices) and wages and taxes had been increasing before the Great Recession of 2007. These factors provided incentives for corporations to outsource their production tasks to other producers located in foreign countries. It is also true that these forces pushed corporations to relocate to other countries where wages and taxes are low relative to the United States of America. It follows that many American workers lost their jobs due to these corporate decisions. For example, due to outsourcing and relocation of firms a large number of workers lost their jobs from the states of Michigan and Ohio.

Large corporations have also tried to increase their efficiencies in order to make more profits in the long run by reducing the cost of production. Thus, they have been involved in downsizing their operations. This process of downsizing is actually aiming at cutting employment of labor. Consequently, structural unemployment rose due to this process.

Globalization has also played a significant role in the development process of important countries such as India, China, and Brazil, to mention a few. These countries can compete with the United States of America and will be able to control a larger share of the global market. This will affect the American exports and employment negatively. But the process of globalization raises national incomes in many countries, a prosperity that will increase imports from the United States of America. Globalization also creates a competitive environment, forcing many corporations to be innovative. Hence, productivity and


demand for labor will increase in the United States of America, which will reduce the rate of unemployment. IMPORTANT IMPLICATIONS FOR PUBLIC POLICIES

The analysis of these determinants of unemployment provides an excellent picture for the trend of public policies. Governments at all levels should provide accurate information about the job market in order for workers to be able to seize this opportunity for finding employment. Governments should spend for education in order to develop very productive workers and innovative entrepreneurs. Training centers are also extremely useful for training and retraining the unemployed workers. These practices assist workers to develop new skills. In addition, governments can directly employ workers, and the Federal government can provide financial assistance to state governments to achieve this task. State governments can attract foreign direct investments, which will be able to create many jobs for the American workers. And in this context, it must be stated that governments have to cut taxes on domestically investing firms, because this action will provide better profitability for these firms. Furthermore, the Federal government must implement a tax holiday according to which the payroll taxes (taxes for social security) must be cut for a certain period of time, an action that reduces the cost of hiring new workers and provides extra income for the workers, which can be spent on consumption and investment (such as the purchase of a house).

The Federal government needs to reallocate a significant part of its spending from the military to the civilian economy, because the military sector is highly capital-intensive, requiring a lower level of employment. This reallocation is able to reduce uncertainty and foreign tension, which have increased prices of imported materials such as oil. Tax cuts and less costly regulations will increase employment and profits and reduce uncertainty which will increase domestic real investments. Spending for Research and Development and the infrastructure are crucially important tools for better innovations (such as finding new methods of production, new products, and new markets for increasing sales) and productivity which will increase domestic real investments and economic growth.

The Federal government has to use some regulations according to which large corporations can become more competitive. That is, the reduction of monopolistic and oligopolistic corporate power. This is because these large firms, given they employ less number of workers than competitive firms, cut production and employment rather than prices of products when there is a reduction in demand for their products. This behavior worsens the unemployment problem during a recession. If these firms become competitive, then the decline in demand for products will reduce prices rather than output and employment. Once these firms become more competitive they will be able to employ more workers.

The Federal government along with the Federal Reserve Banks can provide more credits at lower interest rates to business people and foreign buyers to purchase American products. This will increase employment in many domestic manufacturing firms. In line with this factor, the Federal government can encourage other foreign government to reduce or eliminate tariffs in order to make American exports cheaper in the global market.

American entrepreneurs with the financial support of the Federal governments and the Federal Reserve must develop new products and new technology. Both will increase domestic investments and employment. In addition, innovative products and technology can increase the export of the country and will enhance the global competitive position of the country. Technology is also a very important factor for increasing productivity which will reduce the cost per unit of output and will increase the profit margin. SUMMARY AND CONCLUSIONS

The important theories of unemployment suggest that there are very important variables for increasing the level of employment and reducing the rate of unemployment. These variables are expectations of high sales (demand) and growth, the provision of cheap loans to business enterprises, the increases in domestic real private and public investments, the improved skills of workers, the reduction in


economic and financial uncertainty such as regulations and higher health and labor costs, the improvement in technological progress and innovations, the transition toward competitive market economy, the reduction in taxes, the availability of competitive entrepreneurs and credits, the continuous increases in productivity, the reduction of prices of physical inputs such as oil, and the provision of job information.

These variables indicate that governments can play a significant role for increasing the level of employment. For example, the Federal government can reallocate half of the wasteful military expenditure to the civilian economy by spending more funds on education, the infrastructure, and training centers for enhancing workers’ skills. The government can spend funds for colleges and universities for innovating new products and technology. The government can provide financial assistance to the state governments to balance their budgets and to increase employment level, and the government can cut taxes on domestic investors in real economic activities such as the real estate and the manufacturing sectors. Cutting payroll taxes such social security tax becomes a very urgent temporary task, because this cutting encourages firms to hire more workers and to stimulate workers to spend for consumption and investment. In addition, the government can cut healthcare and regulation costs in order to reduce business costs and to increase profitability, and the government can make the economy more competitive by reducing monopolistic and oligopolistic economic power in order to increase the employment level. In short, these types of cost cutting and government expenditure will increase productivity and employment, making the economy better off relative to the current status-quo.

The government can really cut the business cost of production by ending the wars which have contributed significantly for increasing oil prices from $25 in 2000 to $140 in 2008. These high oil prices have affected many industries and consumers negatively. Therefore, ending the wars in Iraq and Afghanistan is the simple way towards solving the energy problem in the American economy. Finally, the government can use various ways for solving the trade deficit with China, a trade deficit that has reduced the aggregate demand and the level of employment over the years. REFERENCES Chat Chatterjee, S. (1995). “Productivity Growth and the American Business Cycle”. Business Review, Federal Reserve Bank of Philadelphia, (September/October), 13-23. Chatterjee, S. (1999). “Real Business Cycles: A Legacy of Countercyclical Policies?”. Business Review, Federal Reserve Bank of Philadelphia, (January/February), 17-27. Clark, John Bates. (1904). The Problem of Monopoly: A Study of A Grave Danger And of the Natural Mode of Averting It. New York: The Columbia University Press. Clark, John B. (1894). “The Modern Appeal to Legal Forces In Economic Life,” The Seventh Annual Meeting of the American Economic Association, Columbia College December. Davidson, P. (1998). “Post Keynesian Employment Analysis and the Macroeconomics of OECD Unemployment,” The Economic Journal, 108, 448, 817-831. Ekelund, Jr., Robert and Robert F. Hebert. ( 2007). A History of Economic Theory and Method. Fifth edition, Long Grove, Illinois: The Waveland Press. Gali, Jordi. (1999). “Technology, Employment, and the Business Cycle: Do Technology Explain Aggregate Fluctuations,” The American Economic Review, 1, 249-271. Gali, Jordi and Rabanal, Pau. (2004). “Technology Shocks and Aggregate Fluctuations: How Well Does the Real Business Cycle Model Fit Postwar U.S. Data?,” NBER Macroeconomics Annual, 19, 225-288.


Jonung, Lars. (1989). “Knut Wicksell on unemployment,” History of Political Economy, 21, 1, 27-42. Keynes, J. (1936). The General Theory of Employment, Interest and Money. London: Harcourt. Lewis, A (1954). Economic Development with Unlimited Supplies of Labor. Manchester School, 22, 139-191 Manuelli, R. (2000). “Technological Change, the Labor Market, and the Stock Market,” NBER Working Paper 8022 (November). Morgenstern, Oskar. (1941). “Unemployment: Analysis of Factors,” The American Economic Review, 30, 5, 273-293. Mouhammed, Adil. (2010). “Unemployment and the Entrepreneur,” International Journal of Economics and Research, 1, 1, 1-14. Mortensen, D. and Pissarides, C. (1994). “Job Creation and Job Destruction in the Theory of Unemployment,” The Review of Economic Studies, 61, 3, 397-415. Nishiyama, C. and Leube, K. (1984). The Essence of Hayek. Hoover Institution: Stanford University, Stanford, CA. Pigou, A.C., (1933), The Theory of Unemployment, London: Macmillan. Schumpeter, J. (1934). Theory of Economic Development. Cambridge, MA.: Harvard University Press. [Originally published in 1912]. Schumpeter, J. (1947). “The Creative Response In Economic History,” Journal of Economic History, 7, 2, (November), pp.149-159. Sweezy, P. (1934), “Professor Pigou’s Theory of Unemployment,” The Journal of Political Economy, 42, 6, (December), 800-811. Trehan, B. 2001. Unemployment and Productivity, Economic Letter, Federal Reserve Bank of San Francisco, Number 28, October 12, 1-3. Veblen, T. (1904). The Theory of Business Enterprise. New York: Kelley. Veblen, T. (1921). The Engineers and the Price System. New York: B.W. Huebsch. Vecchi, N. (1995). Entrepreneurs, Institutions and Economic Change: The economics thought of J.A. Schumpeter (1905-1925). London: Edward Elgar.


Estimating the Current Value of Time-Varying Beta

Joseph Cheng Ithaca College

Elia Kacapyr

Ithaca College

This paper proposes a special type of discounted least squares technique and applies it to the Capital Asset Pricing Model. There is evidence that the value of beta, the measure of risk in the model, varies over time. The technique, entropic least squares, detects differences in the past and present standard error of the model. The rate of change in this standard error is referred to as the entropy rate. Unlike discounted least squares where the discount rate must be assumed in an ad hoc manner, entropic least squares estimates the entropy rate simultaneously with the parameters of the model. INTRODUCTION

The idea that beta shifts over time in the capital asset pricing model (CAPM) was first suggested by Blume (1975). He found that there is a consistent tendency for a portfolio with an extremely low or high estimated beta in one period to regress toward a less extreme beta in the following periods. Blume discovered a tendency for betas to regress toward unity, the grand mean of all betas, over time. In other words, beta values are not constant. Some analysts estimate an equity’s beta of by assigning 2/3 weight to the past data and 1/3 weight to unity. Rosenberg and Guy (1976) estimate beta with past data and adjust that estimate according to a set of individual corporate financials. This suggests that beta may change with corporate characteristics over time.

Fabozzi and Francis (1978) suggest that many stocks’ beta coefficients move randomly through time rather than remain stable as the ordinary least squares (OLS) technique presumes. The nonstationarity of beta and the time-varying behavior of equity return co-movements also are suggested by Umstead and Bergstrom (1979), Theobald (1981), McDonald (1985), Lee et al. (1986), Levy (1971), Rosenberg (1985), Kaplanis (1988), and Koch and Koch (1991).

One of the drawbacks of using OLS for time-series estimation is that locally evolving linear trend patterns may be ignored. However, local trends are important when projecting into the immediate future. In this case, it makes more sense to assign greater weight to more recent observations.

We develop a weighted least squares method that amounts to allowing the parameter estimates to change over time. In a scenario where beta values change over time, the objective of financial analysts is to estimate the most current value of beta for making current investment decisions. Recent data should make for better forecasts than older data. Still, the older observations carry some information. As the beta or coefficient value changes over time, the standard error based on the current value of beta would be different for historical periods than for more recent periods. We refer to the rate of change in the standard error over time as the


entropy rate. An iterative procedure is used in this paper to estimate the appropriate entropy rate based on the standard error of the residuals as well as to estimate the most current value of beta. This approach leads to alternate estimates of beta in Capital Asset Pricing Model. In the following section, we derive the entropic least squares (ELS) method for estimating parameters in time-variant standard error cases. DERIVATION OF THE ELS ESTIMATOR

Consider the standard OLS regression equation: Yt = a + b Xt + ut where Yt = value of dependent variable at period t a = constant term Xt = value of independent variable at period t b = beta or coefficient of Xt ut = error at period t

Before deriving the normalizing factor to be used for rendering the variance of the error term constant, we need first to specify the nature of the change in the standard error of the stochastic error term. In this model, the standard error of the stochastic error term is assumed to grow at a constant rate: St = S0 erk where St = standard error of u at period t S0= base standard error, which is the standard error of u for the latest period. e = 2.71828 r = rate of entropy (growth rate for the standard error of u) k = number of time periods retrogressed from the most recent observation, which is zero for the latest time period in the sample.

If we assume the variance grows at a constant rate, the specification changes to

St = S0rke

Applying OLS to Yt = a + b Xt + ut would yield inefficient estimates when the variance of ut is time

variant. To correct this problem, we divide both sides of Yt = a + b Xt + ut by a variance deflator to render the transformed residual homoskedastic. As is typically done to remedy heteroskedasticity, the standard error of the residual term in the original equation is used as the deflator. In this case, the deflator would be the standard error of ut, which is St. Yt/St = a/St + b Xt/St + ut/St

The transformed equation has two independent variables (a/St and Xt/St) and no constant term. The above transformation essentially assigns a weight of 1/St to the residual ut. Since the value of St is positive and increases with the number of periods retrogressed from the present, the residuals (ut/St) in earlier periods have higher standard deviations, or higher St values, and thus carry lower weights (1/ St) than residuals of more recent periods.

The variance of the error term (ut/St) in the transformed equation is now constant with respect to time. The estimated beta (b) derived by minimizing the transformed equation is unbiased and efficient.


Dealing with the Problem of Unknown Base Standard Error and Growth Rate Before OLS can be applied to the transformed equation, the values of St must first be computed so that

the appropriate weights can be assigned. The value of St is dependent on S0 and r. The problem encountered at this stage is that neither S0 nor r is observable. In the method to be proposed, this problem is solved without a priori knowledge of either S0 or r.

Since S0 is unknown at this stage, it is convenient to use it as a divisor for St. And because S0 is a constant, the new ratio (Ft) will be proportionate to St: Ft = St/S0 = erk

Then Ft can be used as the divisor for Yt = a + b Xt + ut since Ft combines the two unknowns (S0 and r) into a single variable. The newly transformed equation is: Yt/Ft = a/Ft + b Xt/Ft + ut/Ft

The newly transformed equation will exhibit homoskedasticity since it is divided through by Ft which is proportionate to St. In addition, b in the newly transformed equation will equal b in Yt = a + b Xt + ut since Yt and Xt have been scaled by the same factor, Ft.

Ft (= erk) is unknown only because r is unknown. Therefore, a reiteration procedure is devised to estimate Yt/Ft = a/Ft + b Xt/Ft + ut/Ft and r simultaneously using a stepwise algorithm.

If there is no change in the standard error of the residuals (St) over time, then the rate of entropy (r) would be zero. In this case, the divisor Ft simply reduces to 1 for all observations and thus Yt/Ft would be same as Yt and Xt/Ft the same as Xt. Consequently, the results of regression on Yt/Ft = a/Ft + b Xt/Ft + ut/Ft are equivalent to OLS, which entails the implicit assumption of r equal to zero. However, if r is positive, then entropy occurs in the data and the magnitude of the error worsens with k. With a positive value for r, Ft would equal unity only for the most recent observation. In such a case, dividing the ut by Ft would mean that the weight assigned to the error terms is 1/Ft. The value of this weight equals 1 for the most recent observation where k = 0. The weight on the error term will decline progressively for observations going back further into the past. The question, however, is what specific value for r, which determines F and thus the weights, should we assume. Since the value of r is unknown, the reiterative procedure begins by setting r to zero, then increasing it by a small increment with each successive iteration. In the first iteration where r is set at zero, the value of F is one and OLS is applied to Yt/Ft = a/Ft + b Xt/Ft + ut/Ft.

For the second iteration, r is increased by .001 (or some other increment deemed appropriate to the case at hand) to generate a new set of Ft’s. Once again, OLS is applied to Yt/Ft = a/Ft + b Xt/Ft + ut/Ft. For the third iteration, r is set at .002. The process continues until the value of r in the particular iteration is essentially equal to the implied value of r. Estimating the Implied Entropy Rate

The implied entropy rate is obtained by taking the natural log of both sides of St = S0 erk: lnSt = lnS0 + r*k where r is replaced with r*, the implied entropy rate Multiply both sides by 2 to obtain: 2 lnSt = 2 lnS0 + 2 r*k or ln (St

2) = ln (S02) + 2 r*k


Since we do not have data on ln(St2), we use each individual observation of estimated ut

2 as a proxy for St

2. In addition, a stochastic term (vt) is added to the deterministic equation above in order to create a stochastic regression equation as follows: ln(ut

2) = ln (S02) + 2 r*k + vt

According to the equation above, the size of the error term squared in log form (ln( ut

2)) is expected to equal the size of the base error term (ln (S0 2)) plus the number of periods away from current period times 2r*. The stochastic error term vt is assumed to be white noise. Note that ln(ut

2) may be viewed as the dependent variable of a simple regression equation and ln(S0

2) as the constant term. The independent variable is 2k with r* as the coefficient. OLS is applied to obtain the estimated value for r*, the implied entropy rate.

To summarize the procedure to this point: An arbitrary value close to zero is selected for r and used to run the following regression: Yt/Ft = a/Ft + b Xt/Ft + ut/Ft Recall that Ft = St/S0 = erk. The error terms (ut/Ft) squared from this regression are used as the ut

2 in ln(ut

2) = ln (S02) + 2 r*k + vt

OLS is applied to the regression above to obtain an estimate for r*, the implied entropy rate. When the

difference between r and r* is essentially zero the process stops. The value of b in this final regression is the entropic least squares estimate. It can be decided beforehand how close r* must be to r before considering them equal. APPLICATION TO THE CAPITAL ASSET PRICING MODEL

As an illustration, the ELS procedure is applied to estimating beta for Apple Inc. Since the structure and product mix of this firm has changed markedly over the sample period (1993-2008) it might be expected that the beta value changed over time. In this situation, the ELS procedure may be particularly effective.

The dependent variable of the CAPM in this case is APPLE RETURN, the monthly returns for Apple, Inc. from February 1993 through July 2008. The independent variable is S&P RETURN, the monthly returns for the Standard and Poor’s 500 Index over the same time period. Table 1 compares the results of the OLS and ELS procedures. We estimate beta for Apple by regressing the monthly return of Apple against that of the S&P, using both ELS and OLS.

Convergence is achieved by the ELS reiteration procedure at r =.0040689. The entropy rate, r, represents the monthly rate of change in the standard error of the ELS model. Higher entropy rates indicate firms whose beta is changing more rapidly over time. The hypothesis that the entropy rate is equal to zero is rejected at the 5 percent critical level in this case.

The ELS estimate of beta (1.65324) suggests that Apple, Inc. is a riskier stock than the OLS estimate (1.47020). However, the two estimates of Apple’s beta are not significantly different at the 5 percent critical level. Given a longer time-series, the estimates would diverge. The fact that the entropy rate (r) is significant suggests that the estimate of Apple’s beta in the CAPM is changing over time. This in turn implies that Apple is a dynamic company in the sense that one of its essential characteristics, riskiness, is changing in relation to the market average.

The intercept of the ELS regression (-6.19427) represents ln(S02). From this we can derive the value of S0

to be 0.045178.


TABLE 1 ESTIMATED RESULTS FOR APPLE, INC.

APPLE RETURNt = a + b S&P RETURNt + ut

OLS ELS

Beta 1.47020 (5.71681)

1.65324 (6.71794)

R-squared 0.150829 0.207383

r N.A. 0.0040689 (2.41958)

Standard Error of (S0) 0.138156 0.045178 (-17.0813)

Standard Error of (St) 0.138156 0.045178*e 0.0040689 k *Figures in parentheses represent t-statistics.

Application to Industry Indexes

In this section, the ELS procedure is used to estimate the CAPM for two industry classifications, pharmaceuticals (DRG) and technology (TXX). The results are contrasted with OLS estimation. Monthly rates of return in each industry are regressed on the S&P rates of return. The estimation period is June 1996 through September 2008. Table 2 gives the results for the pharmaceutical industry.

TABLE 2 ESTIMATED RESULTS FOR PHARMACEUTICALS (DRG)

DRGt = a + b S&P RETURNt + ut

OLS ELS

Beta 0.585990 (7.40749)

0.556135 (7.21914)

R-squared 0.273165 0.26019

r N.A. 0.0055000 (2.58576)

Standard Error of (S0) .040883 0.013372 (-23.6293)

Standard Error of (St) .040883 0.013372*e 0.0055 k *Figures in parentheses represent t-statistics.

The entropy rate for this industry is 0.0055000, which is statistically different from zero at the 5 percent

critical level. Table 3 shows the results for the technology sector (TXX).


TABLE 3 ESTIMATED RESULTS FOR TECHNOLOGY (TXX)

TXXt = a + b S&P RETURNt + ut OLS ELS

Beta 1.81913 (17.4602)

1.75849 (18.8645)

R-squared 0.676173 0.707389

r N.A. 0.0079 (4.007)

Standard Error of (S0) 0.053845 0.013948 (24.9911)

Standard Error of (St) 0.053845 0.013948 *e 0.0079 k *Figures in parentheses represent t-statistics.

The entropy rate for this industry is 0.0079, which is statistically different from zero at the 5 percent

critical level. The above results indicate that pharmaceuticals and technology may be considered dynamic industries because their entropy rates are significantly different than zero.

As a final application, the ELS procedure is used to estimate the CAPM for several industries using the Fama-French classifications and data. The data are monthly from January 1970 through December 2005. First, the return on the market portfolio is calculated as the average return for all 48 Fama-French industries. We then estimated beta for each industry by regressing the monthly return of the respective industry against the market portfolio return using both OLS and ELS. Table 4 presents the results.


TABLE 4 ESTIMATED RESULTS FOR FAMA-FRENCH INDUSTRIES

INDUSTRY RETURNt = a + b 48-INDUSTRY RETURNt + ut

Industry OLS Beta ELS Beta r

Agriculture 0.820065 (18.787)

0.810531 (18.404)

0.00022 (0.5549)

Entertainment 1.31528 (31.8457)

1.321 (31.9167)

0.00026 (0.69100)

Health Services 1.23096 (22.429)

1.14446 (21.421)

0.0009 (2.265)

Medical Equipment 0.83456 (26.052)

0.831657 (25.8968)

0.00022 (0.5848)

Chemicals 0.982735 (37.3815)

0.988259 (37.4076)

0.00006 (0.15298)

Personal Services 1.05628 (26.9408)

1.00585 (25.2758)

0.00065 (1.73293)

Paper 0.955395 (31.6414)

0.959361 (31.6497)

0.00080 (2.05085)

Wholesale Trade 1.03128 (46.8354)

1.0246 (46.2189)

0.00011 (.26938)

Meals 1.06911 (32.8892

1.05456 (32.4307

0.00036 (0.92267

Insurance 0.916294 (28.8947)

0.92031 (28.806)

0.00053 (1.382)

*Figures in parentheses represent t-statistics.

For the Fama-French data set, only two industries have significant entropy rates at the 5 percent critical level - health services and paper. The entropy rate for personal services is significant the 10 percent critical level. These three industries are found to have more significant entropy rates than the others which suggests that they are dynamic industries whose beta values vary over time. In such cases, analysts who use ELS to estimate the beta value of industries may very well outperform analysts who use OLS. CONCLUSIONS

This paper offers an alternative approach for estimating the current value of beta in the Capital Asset Pricing Model (CAPM). Entropic least squares (ELS) is more sophisticated than ordinary least squares (OLS) or discounted least squares (DLS). ELS allows for the standard error of the stochastic disturbance term to vary over time. The rate of growth of the standard error is referred to as the entropy rate, which is to be


estimated rather than assumed. Unlike DLS, ELS estimates of the entropy rate are determined by the data themselves with an iterative process.

In this paper, ELS was used to estimate beta for the Capital Asset Pricing Model (CAPM). We found that Apple, Inc. has an entropy rate that is statistically different from zero. This implies that the CAPM beta for Apple, Inc. varies through time. Furthermore, health services, paper, and perhaps the personal services industries may be considered more dynamic than other industries in the sense that their operations or financial structure are changing through time so that reliance on a fixed beta is not appropriate.

ELS is a regression technique that can be useful in many financial time-series applications. When the entropy rate is found to be statistically significant in a particular model it implies that the regression coefficients are time dependent. In such cases, ELS should outperform more tradition estimation procedures. REFERENCES Blume, M. (1975). Betas and their Regression Tendencies. Journal of Finance, 30, (3), 785-796. Fabozzi, F.J. & Francis, J.C. (1978). Beta as a Random Coefficient. Journal of Financial and Quantitative Analysis, 13, (1), 101-116. Kaplanis, E.C. (1988). Stability and Forecasting of the Co-movement Measures of International Stock Market Returns. Journal of International Money and Finance, 7, 63–75. Koch, P.D. & Koch, T.W. (1991). Evolution in Dynamic Linkages Across National Stock Indexes. Journal of International Money and Finance, 10, 231–251. Lee, C.F., Newbold, P., Finnerty, J.E. & Chu, C.C. (1986). On Accounting-Based, Market-Based and Composite-Based Beta Predictions: Methods and Implications. The Financial Review, 21, 51-68. Levy, R.A. (1971). On the Short Term Stationarity of Beta Coefficient. Financial Analyst Journal, 27, 55-62. McDonald, B. (1985). Making Sense Out of Unstable Alphas and Betas. Journal of Portfolio Management, 11, 19-22. Park, R.E. (1966). Estimation with Heteroscedastic Error Terms. Econometrica, 34, (4), 888. Rosenberg, B. (1985). Prediction of Common Stock Betas. Journal of Portfolio Management, 11, 5-14. Rosenberg, Barr & Guy, J. (1976). Beta and Investment Fundamentals, Part 1. Financial Analyst Journal, 32, (3), 60-72. Rosenberg, Barr & Guy, J. (1976). Beta and Investment Fundamentals, Part 2. Financial Analyst Journal, 32, (4), 62-70. Theobald, M. (1981). Beta Stationarity and Estimation Period: Some Analytical Results. Journal of Financial and Quantitative Analysis, 15, (5), 747-757.


Specialty Funds vs. General Mutual Funds and Socially Responsible Investment (SRI) Funds: An Intriguing Risk/Return Paradigm

Brian D. Fitzpatrick Rockhurst University

Joshua Church

Rockhurst University

Christopher H. Hasse Rockhurst University

This paper examines the benefits and detriments of specialty funds (sector funds) during both bull and bear markets. We created a contrasting analysis between general global mutual funds, specialty funds and Socially Responsible Investment (SRI) funds, incorporating 1 year, 3 year, 5 year and 10 years annualized returns for fifteen popular funds (period ending 5/7/2010), and found that specialty funds exhibited superior risk/return tradeoffs. HISTORY OF SPECIALTY FUNDS

Specialty funds are commonly referred to as sector funds. Sector funds have been and always will be an important segment of the mutual fund industry. Some investors even believe that specialty funds may be the superior business for the mutual fund industry because they charge higher expenses and loads than diversified stock funds. This allows them to generate a greater percentage of the mutual fund revenues. Specialty funds should only be used by investors who believe they can select a sector at the right time or investors who want to be in a particular sector without buying individual stock. Specialty funds are a type of mutual fund that focuses their equity investing within a specific industry or sector of the economy. Some specialty funds cover broad sectors and others direct their investments on an industry group within a sector. The most common sectors include: energy, financial services, health care, precious metals, real estate, technology, and utilities. To be classified as a specialty fund, a fund must invest at least 25 percent of its portfolio into one sector, although the majority of specialty funds invest all of their holdings into a single sector or industry. Specialty funds can be unstable investment vehicles, especially when viewed in isolation. This is why most investors utilize the portfolio approach when investing in specialty funds. Stock prices of companies within a sector or industry move in direct correlation with one another due to casual factors. Some examples of casual factors would be: changes in government policies and regulations, introduction of new technologies or products, business cycle dynamics, shifts in the consumer demand or demographic, transformation of the industry’s structure, or even international events. A specialty fund’s return is dependent on the impact of the sector’s casual


factors. These casual factors that drive the specialty fund’s returns consequently lead to the specialty fund’s level of risk. Many specialty funds are ‘no-load’ funds, which means they do not levy a sales charge. Some mutual fund companies have implemented a short-term redemption fee, incase the investor holds the specialty fund shares for less than a determined minimum period. This discourages short-term trading while protecting the interests of long-term shareholders. Practical investors use specialty funds in ways that “leverage their return potential while managing portfolio volatility. Sector funds can also be used to construct diversified portfolios, enhance returns of diversified portfolios, and stabilize diversified portfolios for consistent profits” (Subramanian). There is limited evidence on the performance of specialty fund managers in the academic finance literature. The primary roots of specialty funds date back to 1981, when Boston-based Fidelity Investments commenced offering them. Around 1985, the flashy performance of some sector funds had caught the attention of many individual investors. In the mid 1980s, “specialty funds were firmly lodged in the top spots of the mutual fund rankings. For example, of the 860 mutual funds whose performances were monitored by Lipper Analytical Services, specialty funds filled the top four spots for the 12-month period ending on August 22, 1985. The top performer was Prudential-Bache Utilities, with a rise of 50.8 percent. Next were Fidelity Select Financial (up 45.4 percent), Fidelity Select Leisure (42.7 percent), and Century Shares Trust, which specializes in insurance companies (39.6 percent). In fact, of the top 10 performers, 7 were specialty funds” (Springer). However, these attractive gains of specialty funds in the mid 1980s only reveal half of the story. “Specialty funds turned in some of the poorest performances as well. Gold funds lost 17.1 percent on average and many technology funds recorded significant losses in the same 12-month period. At the same time, the Standard & Poor’s 500 index rose by 12.1 percent, excluding reinvested dividends” (Springer). Timing is a key element in the equation of specialty funds. Timing is essential to avoid being trapped in a sector that is decaying. Market trends are constantly changing and shifting, which causes different sectors to become more favorable at certain times. Specialty funds in a broad sector can even be difficult to evaluate in regards to timing. Specialty funds have a historical track-record of being consistently inconsistent. EMERGING TRENDS OF SPECIALTY FUNDS In evaluating the current and emerging trends of specialty funds, there is no evidence supporting the claim that specialty fund investors can select the winning funds within a specialty fund category. The performance of a specialty fund that is not rationalized by changes in factor returns can be justified by chance rather than the investment picking skills of the fund’s manager. Specialty funds “neither outperform nor underperform the relevant benchmarks. There is no persistence in specialty fund returns. Specialty fund investors usually do not have the ability to pick the winning funds or the winning sectors. This is all consistent with market efficiency” (Tiwari). On average, specialty funds charge annual expenses of 1.65 percent. The difference in fees and expenses cannot be explained by the variations in the specialty fund size or turnover. “Specialty funds also charge higher loads in the form of rear-end fees. It is not obvious that the higher loads and expenses of a specialty fund arise from higher operating costs” (Tiwari). Fees and expenses are higher among specialty funds that focus on a particular industry. These specialty funds are more costly to monitor and manage. Figure 1 shows how there are significant variations between funds and their respective expense ratios. These higher expenses are partially due to the fact that the cost of information for evaluating specialty is greater than the cost of information for typical stock funds. It becomes an issue of the degree of information asymmetry between insiders and outsiders.


FIGURE 1 EXPENSE RATIOS FOR SELECTED INVESTMENT OBJECTIVES*

Year End 2009

Investment objective 10th Percentile

Median 90th Percentile

Average Asset-weighted

Average Simple

Equity Funds 0.82 1.44 2.28 0.87 1.52 Aggressive growth 0.91 1.49 2.33 1.03 1.58 Growth 0.78 1.33 2.16 0.91 1.43 Sector funds 0.92 1.62 2.50 0.98 1.70 Growth and income .052 1.21 2.00 0.56 1.25 Income equity 0.75 1.24 1.98 0.85 1.32 International equity 0.99 1.60 2.45 1.02 1.67 Hybrid funds 0.63 1.20 2.00 0.84 1.28 Bond funds 0.52 0.96 1.73 0.65 1.08 Taxable bond 0.50 0.99 1.80 0.65 1.09 Municipal bond 0.55 0.92 1.62 0.64 1.07 Money market funds 0.22 0.50 .091 0.34 0.54 *Figures exclude mutual funds available as investment choices in variable annuities and mutual funds that invest primarily in other mutual funds. Sources: Investment Company Institute and Lipper There is a new emerging trend in specialty funds. Due to the vast variation in returns of specific industries, investors have identified the need to invest in the right sector at the right time in order to capture the potential return. Sector rotation is an investment method in which the investor seeks to increase returns by strategically switching from one industry to another. Theoretically, this process will assist specialty fund investors in attempting to exceed the returns generated by buy-and-hold investors. Timely sector rotation increases the value of investment accounts and portfolios. Discreet sector rotation offers significant potential to grow wealth over time. “For example, major corporations recently pruned technology related capital spending, whereas falling interest rates kept consumer spending strong. To profit from such secular trends, an investor may choose to invest in Fidelity funds such as ‘Select Consumer Industries’ (NASDAQ: FSCPX) and ‘Select Leisure’ (NASDAQ: FDLSX) while avoiding ‘Select Technology’ (NASDAQ: FSPTX)” (Subramanian). To employ the sector rotation strategy, investors need to understand and follow the dynamics of each industry. Investors must be able to make informed decisions on which industries to enter and which industries to avoid. Through sector rotation, specialty funds have the potential to outperform the market averages on the foundation of relative returns and risk-adjusted returns. The key word to recognize and remember when entertaining the thought of utilizing specialty funds and sector rotation is ‘potential’. There is also the potential to lose a substantial amount of money. The word ‘consistency’ does not describe specialty funds. There is a great amount of perceived risk involved with specialty funds. For most people, the level of this perceived risk is simply too high when dealing with investing for retirement plans. COMPARISON OF RISKS AND RETURNS It is important to evaluate risk and return when comparing mutual funds to specialty funds. Usually with specialty funds, the potential return on the investment does not off-set the present risk at hand for the investor. “Besides having a lot in a single stock, a sizable weighting in a single sector runs big risks because sometimes everything in an industry goes in the tank at the same time” (Morningstar). On that same note, an industry can greatly exceed expectations during a bull market. Specialty funds vary widely in volatility; it depends on the diversity and volatility of the industry’s businesses as well as the strength


of the correlation between the stock prices of the companies included in the portfolio. There are many high risk tactics associated with specialty fund investing. One high risk strategy is to trade sector funds for short-term returns by taking advantage of opportunities created periodically by market volatility. When evaluating industries and sectors for specialty funds, it is important to understand that each industry is going to possess a different level of risk and return. Each industry is unique in its own aspects. “Sectors like energy, for example, often move with little correlation to the overall stock market. Exposure to such sectors can therefore help dampen portfolio volatility” (Subramanian). You can expect President Obama’s Health Care Reform to have an impact on the healthcare industry; it is too early to determine whether or not it will benefit or weaken the investing in the health care industry sector. On the positive side, the higher risks associated with specialty funds have the possibility to lead to incredible returns. Specialty funds are typically very aggressive with their investing; this can result in big returns or great losses. The unpredictable performance of specialty funds places them at the top or bottom of the mutual fund performance rankings. “A hot fund’s portfolio typically balloons because of big cash inflows from investors eager to climb on the bandwagon. The fund must lower its investment criteria for the companies within its sector to accommodate that new money, making it challenging to continue to beat the market” (Springer). Although mutual funds are not as risky as specialty funds in general, there is still some risk present for those investors. One of the main points to keep in mind when determining the risk associated with mutual funds is the fact that mutual funds are not guaranteed or backed by the FDIC. This immediately puts mutual funds at a higher level of risk. Mutual funds carry multiple types of risk: concentration, sector, price, business, market, credit, interest-rate, liquidity, emerging-markets, and currency. “Funds with a high percentage of assets in their top holdings are not necessarily riskier than other funds, but they can be. Some take on a lot of individual stock risk. When a stock is trading for a high valuation, disappointing news will generally spur much larger losses than one with a low valuation. At the heart of every stock fund is the risk that the businesses of the stocks they own will deteriorate. A fund with losses can slip into a downward spiral if its holdings are so illiquid that its losses spur redemptions and then the redemptions spur more losses because the fund manager has to sell securities at fire-sale prices, as the cycle gains steam. Emerging markets have outsized returns and outsized losses because they are based on rickety economies that work well in some environments but can fall apart in others” (Morningstar). Corporate bonds and emerging-markets-government bonds possess the risk of defaulting. Interest rates weigh on the vulnerability of a fund being impacted by increased rates. Mutual funds should be less risky than specialty funds, predominately due to the level of vulnerability carried by specialty funds. Specialty funds go against the old saying, “Do not put all your eggs in one basket.” Mutual funds possess the quality of diversity and a more balanced level of risk. Many investors incorrectly believe that only high risk will produce great returns. However, brilliant investors such as Warren Buffet and the late John Templeton have proven that low-risk investment can create large returns. In evaluating this notion, it becomes increasingly difficult to defend the position that specialty funds are better investments than general mutual funds. The inconsistency associated with specialty funds tends to make them less attractive. INFLUENCE OF BULL AND BEAR MARKETS “Shareholder sentiment generally moves with stock market performance because of the impact on mutual fund returns. For example, mutual fund companies’ favorability rose in the late 1990s along with stock prices (measured by the S&P 500), then declined between May 2000 and May 2003 as stock prices fell, and increased after 2003 as the stock market gained until the 2008 stock market crash (ICI – Fact Book 2010). Many mutual fund companies are going to advertise and claim that they can outperform the market in both bull and bear markets. These mutual fund companies will focus on specific points in time when the data was in their favor. Companies focusing on specialty funds will comment on how “the bull market in 1999 produced a 132.4 percent gain from technology (FSPTX), which outpaced the S&P 500’s 19.5


percent advance” (Subramanian). Or how “the 2002 bear market produced a 64.3 percent gain from gold (FSAGX), which topped the S&P 500’s 23.4% loss” (Subramanian). Tests have been performed that combined “Jensen’s Alpha (1968) and the Fabozzi and Francis (1979) test for bull and bear market parameters by testing three different asset pricing models: the Capital Asset Pricing Model (CAPM), the Fama and French three-factor model, which adds SMB (small minus big factors to the CAPM), and the Carhart four-factor model, which adds a momentum factor. Of the Ferson and Schadt (1996) mutual funds, only 12 of the funds had positive bear market alphas using the CAPM, 14 positive bear market alphas under the Fama and French three factor model, and 12 positive bear market alphas using the Carhart model. The bull market alphas dominated the bear market alphas, However, the results were statistically insignificant according to their t-tests” (Hamidani). These test results support the claim that markets are mutual-fund efficient and that fund managers generally cannot outperform the market, whether it is a bull or bear market. A mutual fund is going to be more vulnerable to change with a bull or bear market than a specialty fund. This is because a specialty fund has a narrow range of scope in its portfolio. Sometimes the changes in a bull or bear market will not affect the specific industry that the portfolio is invested in. A mutual fund has a much broader portfolio that expands over multiple industries and sectors, thus becoming more susceptible to the changes and shifts in the market and the flow of the stock prices. Both mutual funds and specialty funds rely on the ability of the portfolio manager to identify shifts and changes in the bull or bear market. Reaction time is limited in both cases. INVESTMENTS IN “ETHICAL FUNDS One particularly interesting type of specialty fund is the so-called “ethical” funds. “Ethical” funds are also called “virtue funds” or Socially Responsible Investment (SRI) funds. SRI funds invest only in “virtuous” companies such as those with environmentally friendly practices or objectives, or companies that have outwardly humanitarian upper management. SRI funds exclude companies that profit from “unethical” or “vice” products or services such as tobacco, alcohol, pornography, gambling and the like. They also exclude companies who participate in unethical practices such as child labor or poor working conditions. For years, there has been significant debate in the financial industry about whether it is more or less lucrative to pursue an “ethical” investment strategy by investing solely in SRI funds and socially responsible companies. John Rothchild, a popular investor, called socially responsible investment “a dumb idea” in a May 1996 Fortune magazine article. On the other end of the spectrum, it has also been found that socially responsible mutual funds outperform traditional mutual funds, though slightly. (Guerard, JR.) Ultimately the choice of whether or not to invest in socially responsible funds depends on the investing style and strategy that an investor wishes to pursue. In this case the question becomes, “Is it more lucrative to invest with a sense of ethics and/or morality, or is it better (in terms of returns) to strictly follow a strategy based on other characteristics such as value, growth, or low-risk?” (Guerard, Jr.) In general, SRI funds have been observed to have tendencies towards purchasing the stocks of companies with higher market capitalizations. In addition, SRI funds tend to be more value-oriented than growth-oriented in their investment style. (Guerard, Jr.) The expense ratios for SRI funds tend to be higher than the expense ratios for mutual funds in general. However, the loads and turnover ratios of SRI funds were found to be lower than that of general mutual funds. (Gil-Bazo) Therefore, SRI funds likely balance out with general mutual funds in regards to the costs of investment because their expenses are higher on average, but their average loads and turnover ratios are lower. Only a performance evaluation of SRI funds in comparison to other types of funds can actually provide evidence towards a conclusion for the question of the value in ethical investing. A performance evaluation will also help to explain the one critical aspect of mutual funds that has yet to be addressed: how they react to differing market conditions. So far, the internal aspects and characteristics of general mutual funds and specialty funds have been identified and explained, and the specific differences in the investment styles of conventional mutual funds and specialty funds have been


established. All of that information is useful and helpful in selecting a fund or funds to invest in. However, the final, and perhaps most pertinent, characteristic of a fund is how it performs in varying market conditions. More specifically, it is necessary to analyze how well mutual funds and specialty funds perform in bull markets and bear markets. The best way to analyze that performance is through the statistical analysis of an array of funds or an index. STATISTICAL ANALYSIS To analyze the performance of mutual funds and specialty funds in differing market conditions, we have selected five general mutual funds and five specialty funds to compare and contrast. We will analyze their returns, risk values, fees, as well as their performance against the S&P 500. In addition, we selected five socially responsible funds to be analyzed. We chose the funds on the basis of size, and not return or risk, which could skew the results, but not necessarily. The fifteen funds we chose are among the largest in their respective categories in terms of net asset value. Table 1 lists the fifteen funds, and includes their ticker symbols, fund names, and net asset value as of May 7, 2010.

TABLE 1 FUND SYMBOLS, NAMES AND VVALUES AS OF MAY 7, 2010

Ticker Symbol Company Net Asset Value

(as of 5/7/2010) General Mutual Funds

PTTRX

PIMCO

$128.75 Billion AGTHX American Funds $67.98 Billion VTSMX Vanguard $65.22 Billion FCNTX Fidelity $59.23 Billion CAIBX American Funds $57.63 Billion Specialty Mutual Funds

FKTFX

Franklin

$12.93 Billion

VGHCX Vanguard $11.25 Billion VEURX Vanguard $6.68 Billion FNMIX Fidelity $3.46 Billion FSAGX Fidelity $3.23 Billion SRI Funds PRBLX Parnassus $2.81 Billion ARGFX Ariel $2.31 Billion PAXWX Pax World Funds $1.86 Billion AMAGX Amana $1.72 Billion CAAPX Ariel $1.67 Billion

Data Source: http://www.marketwatch.com/investing/mutual-funds Notice in Table 1 that the Net Asset Value of each fund type decreases as the investment becomes more specific. The general funds are a mixture of growth and value funds, and are not exclusive as to which companies or industries they invest in. The specialty funds are separated as follows:

• FKTFX invests solely in municipal bonds from the state of California • VGHCX is Vanguard’s Health Care industry fund • VEURX is Vanguard’s European investment fund • FNMIX is a Fidelity Emerging Markets fund • FSAGX is a Fidelity Gold (commodities) fund


Though most of the SRI funds simply screen their investments to meet basic socially responsible criteria, the fund AMAGX is a fund that bases its investments on the tenets of Islam. Table 1 simply gives a snapshot of current net asset value of each of the funds used in this analysis, but to analyze accurately, it is important to look at the average returns of each of the fifteen funds throughout different time measures. Table 2 does just that.

TABLE 2 FUND RATES OF RETURN AS OF MAY 7, 2010

RETURNS OF THE FIFTEEN FUNDS

CATEGORY FUND YTD 1 YR 3 YR (Annualized)

5 YR (Annualized)

10 YR (Annualized)

General Mutual Funds

PTTRX

3.77%

13.99%

9.92%

7.43%

7.98%

AGTHX -2.09% 21.04% -6.06% 2.95% 1.03% VTSMX 0.92% 26.97% -7.05% 1.76% 0.09% FCNTX -1.42% 24.78% -3.58% 4.67% 3.27% CAIBX -4.30% 16.23% -607% 2.96% 7.00% AVERAGE -0.62% 20.62% -2.57% 3.95% 3.87% Specialty Mutual Funds

FKTFX

4.31%

12.90%

3.68%

4.15%

5.61%

VGHCX -4.80% 19.79% -3.70% 3.67% 6.70% VEURX -15.65% 12.265 -14.62% 0.47% 0.85% FNMIX 2.42% 23.495 7.335 9.52% 11.99% FSAGX 5.92% 36.815 12.50% 23.99% 19.58% AVERAGE -1.565 21.05% 1.045 8.365 8.95% SRI Funds PRBLX -2.03% 25.60% 0.24% 6.02% 5.52% ARGFX 5.11% 47.94% -7.29% -.014% 8.24% PAXWX -2.07% 15.31% -5.10% 1.41% 1.94% AMAGX -0.42% 23.13% -1.00% 7.58% 3.11% CAAPX 2.575 45.55% -3.54% 2.85% 8.01% AVERAGE 0.635 31.51% -3.34% 3.54% 5.36% Data Source: http://www.marketwatch.com/tools/mutualfunds/fundcomparison.asp Table 2 illustrates that overall specialty funds produce the highest rates of return. It is safe to assume that the three year annualized rate represents primarily a bear market, because 2008, 2009, and some of 2010 were the recent recessionary years. Therefore, specialty funds performed the strongest of the three types with a 1.04% rate of return as compared to a negative 2.57% rate of return for the general mutual funds and a negative 3.34% rate of return for SRI funds. In addition, specialty funds performed the most robustly in both the five year and ten year categories. However, it is possible that the specialty fund sample did so well because of the substantial increase in gold prices over the past three to four years, and that its rate of return was boosted by the returns of FSAGX. To dispute this possibility is the poor performance of the Europe fund (VEURX) over that same period, which should balance out the average return. It is also interesting to note that SRI funds performed more strongly than general mutual funds in the ten year category, and handedly outpaced both general funds and specialty funds with a 31.51% rate of return in the one year category. However, SRI funds also reacted poorly to the bear market, suffering the largest decline in returns of the three for the three year return numbers.


TABLE 3 FIND RISK VALUES AS OF MAY 7, 2010

RISK VALUES OF THE FIFTEEN FUNDS

CATEGORY

FUND

Beta (β)

Std. Deviation (σ) R-squared


PTTRX

0.83

1.39

0.14

AGTHX 0.14 5.70 0.94 VTSMX 0.09 5.95 1.00 FCNTX 0.32 5.39 0.90 CAIBX -0.01 4.39 0.91 AVERAGE 0.52 6.48 0.49 Specialty Mutual Funds

FKTFX

0.36

2.24

0.16

VGHCX 0.13 4.87 0.65 VEURX -0.37 7.57 0.87 FNMIX 0.94 4.50 0.58 FSAGX 1.55 13.20 0.17 AVERAGE 0.52 6.48 0.49 SRI Funds PRBLX 0.62 5.04 0.93 ARGFX 0.36 9.87 0.85 PAXWX 0.05 4.27 0.90 AMAGX 0.52 4.87 0.91 CAAPX 0.58 8.26 0.87 AVERAGE 0.43 6.46 0.89

Data Source:http://www.marketwatch.com/tools/mutualfunds/fundcomparison.asp Table 3 shows that on average, specialty funds have the highest Beta values of all types of mutual funds, illustrating that the risks of investing in specialty funds are the most pronounced overall. Conventional mutual funds, on the other hand, have the lowest Betas of the three types of funds in the sample. This indicates that general mutual funds have the lowest volatility, meaning that they are the least sensitive to market movements (bulls and bears), whereas specialty funds have the highest volatility in returns of the three types of funds and are therefore the most sensitive to market movements. This is no surprise, and relates to the “putting all your eggs in one basket” idea. This table also indicates that all of the funds have low Betas, indicating relatively low volatility, which typically makes returns more dependable and easier to predict, though that is not always the case. The standard deviation values for specialty funds are also the highest of the three types of funds included in the analysis, closely trailed by SRI funds. The conventional mutual funds have the lowest standard deviation values of the three. This indicates that conventional mutual funds do not deviate as much from their mean returns as specialty and SRI funds do. Once again, conventional mutual funds prove to be the least volatile, this time with respect to its rates of return in comparison to past years. Therefore, they are considered the least risky of the three types, and specialty funds are considered the most risky. This would explain why the rates of return of conventional mutual funds in Table 2 are steadier on average than the rates of return of the specialty funds. It is interesting, however, that in this sample specialty funds appear to be the most risky but still performed the most strongly in the bear market.


Finally, the R-squared values for the three categories, which measure the correlation between a fund’s movements and the movements of the market, indicate that the Beta values of SRI funds and conventional funds are measured against the appropriate benchmark, whereas the Beta of specialty funds may not be. R-squared values in this case fall between 0 and 1, which means that values like 0.78 and 0.89 indicate accurate benchmarking for the Beta of their respective funds. However, the R-squared value of specialty funds is at 0.49, which indicates that the Beta of specialty funds may be measured against the wrong benchmark and may therefore be inaccurate. This may be the reason for the conflict between the high Beta of specialty funds and their corresponding high returns in the bear market. It is possible that the actual Beta is a lower value, and therefore could help explain why specialty funds still experienced positive returns despite the bear market. But there are still other factors to analyze.

TABLE 4 FUND FEE RATES AS OF MAY 7, 2010

FEE RATES OF THE FIFTEEN FUNDS

CATEGORY

FUND

Front Load

12b-1 Fees

Expense Ratio

Turnover Ratio


PTTRX

0.00%

0.00%

0.45%

300%

AGTHX 5.75% 0.24% 0.75% 38% VTSMX 0.00% 0.03% 0.18% 5% FCNTX 0.00% 0.005 1.02% 58% CAIBX 5.765 0.23% 0.66% 43% AVERAGE 2.30% 0.10% 0.615 89% Specialty Mutual Funds

FKTFX

4.25%

0.09%

0.57%

11%

VGHCX 0.00% 0.02% 0.36% 6% VEURX 0.00% 0.045 0.27% 18% FNMIX 0.00% 0.00% 0.915 126% FSAGX 0.00% 0.00% 0.87% 42% AVERAGE 0.85% 0.03% 0.60% 41% SRI Funds PRBLX 0.00% 0.00% 0.99% 60% ARGFX 0.00% 0.25% 1.14% 45% PAXWX 0.00% 0.25% 0.98% 43% AMAGX 0.00% 0.25% 1.31% 6% CAAPX 0.00% 0.23% 1.25% 44% AVERAGE 0.00% 0.20% 1.13% 40%

Data Source: http://www.marketwatch.com/tools/mutualfunds/fundcomparison.asp Table 4 is quite surprising in that it illustrates that specialty funds have the lowest expense ratios of the three types of funds. It is usually the case that the specialty funds have higher expense ratios than conventional mutual funds. The highest expense ratios belong to SRI funds, which is consistent with SRI funds throughout the market. The expense ratios of conventional mutual funds are just slightly larger than that of specialty funds at 0.61% to the specialty funds’ 0.60% ratio.


CONCLUSION

Based on the information and data discussed in this paper, we have concluded that in general, specialty funds actually outperform conventional mutual funds. The statistical analysis indicates that the specialty fund category, while carrying the greatest risk, also provide the greatest returns. The risk is simply not significant enough to substantially affect the specialty funds in bear markets. It is very likely that we found the performance of specialty funds in bear markets to exceed the performance of conventional funds because the conventional funds have much higher average turnover ratios, and therefore charge expenses much more frequently. In addition, we must concede to the possibility that our results were slightly skewed in the favor of the specialty fund sample due to our inclusion of Fidelity’s high performing gold commodities fund (FSAGX). However, in reviewing S&P 500 comparison charts, excluding the gold fund, we would still conclude that specialty funds slightly outperform conventional mutual funds, primarily due to their strong performance in bear markets. Much to our dismay, we could not conclude that SRI funds are more lucrative than investment in general mutual funds. The data and statistical analysis simply did not support such a conclusion. However, it is also important to remember that other immeasurable factors influence the performance of any type of mutual fund, such as the fund’s management or expectations for changes in the macroeconomic environment. It is also important to keep in mind that each investor has a different set of investment goals and therefore a different investment strategy that best suits her/him. Investors should determine what strategies best help them achieve their financial and personal goals and invest in funds that employ those strategies. Portfolio customization is the key. We found that, specialty funds have a cyclical nature and investors who utilize them need to be committed to monitoring market conditions and the portfolio’s performance. Investors need to devote the proper time and also possess the expertise to correctly manage these funds in order to avoid devastating losses. Secondly, mutual funds adhere to the common principle of not placing all your eggs in one basket. Mutual funds provide investment portfolios with diversification, variety, daily pricing, and liquidity. Specialty funds carry a higher level of risk because the vast majority (if not all) of the holdings are placed into companies within the same industry. Casual factors cause these companies to fluctuate directly with one another. Drastic changes in casual factors can ruin an industry before managers and investors have time to react. Finally, the costs and fees associated with mutual funds are less than the expenses linked with specialty funds. Both funds can lose their principle, but the costs associated with specialty funds can bury an investor deeper into debt. Specialty funds generally do not out-perform or under-perform mutual funds. In comparing these two funds, mutual funds are considered the stronger investment because they can generate a greater amount of return with less risk. Investments with less risk are very attractive, especially to people who are investing for their future retirement plans. Practically speaking, when we implement a coefficient of variation analysis, and actually show a per unit of risk per unit of return analysis, we witness some interesting phenomena. Combining the risk with the return, we find that coefficients of variation from the ten year period studied, showed that specialty funds only took on .724 units of risk per unit of return, while general mutual funds produced 1.18 units of risk per return unit. SRI funds showed 1.21 units of risk per unit of return, and indicate that the ethical considerations may be important to the investor but the risk return tradeoff does not look very attractive. (See Exhibit 1) Uncharacteristically, for the 10 year returns ending May 7, 2010, we clearly show a lower risk per return relationship for specialty funds over general mutual funds. SRI funds exhibited the highest risk per return relationship during this time period. We are not suggesting that these results would be replicated in future ten year time frames, but we certainly must conclude that once again that the risk/return relationship need not always be symmetrical.


EXHIBIT 1 DATA AS OF MAY 7, 2010

10 Year Returns (Annualized) Standard Deviation Coefficient of Variation

General Mutual Funds PTTRX 7.98% 1.39% 0.174 AGTHX 1.03% 5.7% 5.53 VTSMX 0.09% 5.95% 66.11 FCNTX 3.27% 5.29% 1.65 CAIBX 7% 4.39% 0.63

AVERAGE 3.87% 4.56% 1.18 Specialty Funds

FKTFX 5.61% 2.24% .40 VGHCX 6.70% 4.87% .73 VEURX .85% 7.57% 8.91 FNMIX 11.99% 4.50% .38 FSAGX 19.58% 13.20% .67

AVERAGE 8.95% 6.48% .72 SRI Funds

PRBLX 5.52% 5.04% .91 ARGFX 8.24% 9.87% 1.20 PAXWX 1.94% 4.27% 2.20 AMAGX 3.11% 4.87% 1.57 CAAPX 8.01% 8.26% 1.03

AVERAGE 5.36% 6.46% 1.21

REFERENCES Bogle, J. C. (1994). Bogle on mutual funds: new perspectives for the intelligent investor. New York: Dell Publishing. Carther, S. (no date). Understanding volatility measurements. Retrieved May1, 2010, from http://www.investopedia.com/articles/mutualfund/03/072303.asp. CNN Money. (no date). Different types of stock funds. Retrieved April 28, 2010, from http://money.cnn.com/magazines/moneymag/money101/lesson6/index3.htm. Collins, S. (2010, April). Trends in the fees and expenses of mutual funds, 2009, Investment Company Institute – Research Fundamentals (PDF Printed Version), Washington, D.C., v. 10, n. 2; 1-12. Ferson, W. E. & Schadt, R. W., (1996, June). Measuring fund strategy and performance in changing economic conditions. The Journal of Finance, 51, 2, 425-461. Guerard, Jr., J. B. (2010, April 28). Is there a cost to being socially responsible in investing? Retrieved from http://www.socialinvest.org/pdf/research/Moskowitz/1996%20Winning%20Paper%20-%20Moskowitz.pdf .


Gil-Bazo, J., Ruiz-Verdu, P. & Santos, A. (2008). The Performance of socially responsible mutual funds: the role of fees and management companies. Business Economic Series 09, Working Paper 08-34, 1-29. Hamidani, F. (2004). Mutual fund performance in bull and bear markets: An empirical examination. MBA project in the Global Asset & Wealth Management program at Simon Fraser University. PDF version. Retrieved from: http://ir.lib.sfu.ca/retrieve/221/etd0434.pdf. Investment Company Institute (ICI) (2010). Fact Book 2010, Investment Company Institute (PDF Printed Version), Washington, D.C.; 1-250. Retrieved April 13, 2010, from http://www.ici.org/pdf/2010_factbook.pdf.html. Kalwarski, T. (2010, February 22). Mutual Funds: Funds that rode the bear and bull. Bloomberg’s BusinessWeek Magazine, 22 Feb. 2010; 56. Karceski, J. (2002). Returns-chasing behavior, mutual funds, and beta's death. The Journal of Financial and Quantitative Analysis, 37(4), 559-594. Lynch, A. W., Wachter, J. & Boudry, W. (2003, January). Does mutual fund performance vary over the business cycle? New York University, Stern School of Business. Mahoney, P. G. (2004). Manager-investor conflicts in mutual funds. The Journal of Economic Perspectives, 18(2), 161-182. Morningstar Mutual Fund. (no date). Understanding Mutual Funds Strategies and Fundamental Risk. Fund Spy – Chapter Six (online). Retrieved April 29, 2010, from http://www.morningstar.com/products/PDF/subscription/FundSpy/Chapter6/online.html. Pontiff, J. (1997). Excess volatility and closed-end funds. The American Economic Review, 87(1), 155-169. Rothchild, J. (1996, May 13). Why I invest with sinners. Fortune Magazine, Retrieved May 5, 2010, from http://money.cnn.com/magazines/fortune/fortune_archive/1996/05/13/212407/index.htm. Saxton, J. (Chairman). (2008, February). The Mutual Fund Industry: An Overview and Analysis. Joint Economic Committee – United States Congress. G-01 Dirksen Senate Office Building -Washington, D.C.,; 1-28. Retrieved April 24, 2010, from http://www.house.gov/jec/. Springer, P. (1985, Sept. 8). Personal Finance: The Splashy Returns in Sector Funds. The New York Times Company. New York,1-3. Subramanian, S. (2008, Autumn). Sector Investing: Sector Funds and Sector Rotation. AlphaProfit Investments, LLC, Financial Bridges, 19. Tiwari, A. and Vijh, A. (2001) Sector Fund Performance. University of Iowa – Henry B. Tippie College of Business. Iowa City, IA; 1-38. U.S. Securities and Exchange Commission (MCMXXXIV). (no date). Invest Wisely: An Introduction to Mutual Funds. United States Federal Government – SEC (online). Retrieved April 15, 2010, from http://permanent.access.gpo.gov/LPS113596/LPS113596/www.sec.gov/investor/pubs/inws.html.


U.S. Securities & Exchange Commission. (2008, July 02). Invest wisely: an introduction to mutual funds. Retrieved May 1, 2010, from http://www.sec.gov/investor/pubs/inwsmf.htm. U.S. Securities & Exchange Commission. (2007, August 8). Mutual fund fees and expenses. Retrieved May 1, 2010, from http://www.sec.gov/answers/mffees.htm#management.


PFSA and BaFin - Comparison of Institutional Framework in Dynamic Financial Markets

Agata Kocia

University of Warsaw

Grazyna Szymanska University of Warsaw

At present, theoretical debate in academic communities is ongoing concerning the structures and aims of the centralised supervisory offices. There are some arguments in favour of mergers of individual sector supervisors to create mega institutions and there are also some against it. The article presented below adds to this debate by presenting a more empirical approach aimed at comparing two super-supervisors the Polish PFSA and the German BaFin. The goal of this work is to compare and contrast both institutions and note any differences between, seemingly similar supervisory structures. The article begins with a theoretical explanation of most important terms and concepts related to the subject of financial supervision. Next, we describe the present day structure, aims and economic environment of the Polish PFSA and German BaFin. The following section is dedicated to comparison and contrast of peripheral supervisors’ activities. The final section is dedicated to drawing final conclusions. THE IDEA OF SUPERVISION Definition of Supervision

Supervision from an objective viewpoint it means a group of activities assuring effectiveness of this process, and, from a subject-organisational viewpoint – an organisation, an office, a team of people exercising authority on something, or inspecting somebody or something or indicating aims of performing supervision[2].

Supervision in a public administration consists of potentially imperious intervention in the activity of the supervised institution. Supervisors may call responsible persons to account for lapses and demand to rectify all these errors. They may impose sanctions and fines if recommendations and injunctions are not fulfilled. The law should specify when and in what scope supervision may be conducted as well as methods and possible consequences. It has to be emphasised that the supervisor is responsible for the activity of supervised entity.

Control vs. Supervision

The essence of control is an examination of compatibility of an actual state with a postulated one, an identification of scope and reasons of the discrepancies and presentation of reasons of this situation.[3] That means that a controller observes and analyses a situation and presents outcomes of their observation to an overriding authority but it is not responsible for outcomes of the activity of the controlled entity.


Unlike control, supervision includes also a part which pertains to responsibility. A supervisor makes some observations and assessments but also takes part in management or administration and, in this way, is responsible for outcomes of an activity of the controlled. It must be admitted that this responsibility is limited to an area in which the supervisor disposes an efficient means of pressure. Therefore, supervision always includes control, while performing of control, does not have to be connected with the use of means of supervision. Use of means of supervision results in that the supervisor may imperiously enter into an activity of the entity under supervision. It has to be emphasized that such intervention is allowable only in a mode and a scope prescribed by the law.[4]

Control as an element of supervision consists of four stages. At first obligatory phase, the state prescribed by law must be established. Then, an actual state in a controlled entity must be described. In the third stage, the two above mentioned states are compared and mutual consistencies and inconsistencies are determined. The last stage serves to identify reasons of existing inconsistencies between the actual state and the obligatory state and then drawing of conclusions.[2] Control and supervision consist, accordingly, of a lot of activities which are conducted in accordance with appropriate procedures.

It is very desirable, from a viewpoint of supervised entities and the environment in which they run, that supervisor’s actions do not confine to orthodox (standard) supervision, but they also have educational and supportive elements. The supervisor might help in upgrading knowledge and providing solutions from experience in the best practices. This would be an invaluable contribution in an effective and an efficient work of these entities.

Typically, individual financial institutions are subject to both official supervision and private control. The official supervision of individual entities takes place along two dimensions: prudential and conduct-of-business. The purpose of prudential supervision is to promote the safety and soundness of financial institutions by systemically evaluating the risk-profile and risk-bearing capacity of the supervised entities in order to ensure their solvency. The conduct-of-business supervision, whereas aims at protecting investors and consumers alike by promoting a fair and transparent market process.[5] Apart from the official supervision, financial institutions are also subject to different forms of control, for instance, through audits, ratings, and internal controls.

Some economists have argued that the role of official supervision should be to create an environment conducive to effective private control, generally referred to as a market discipline, by ensuring, in accurate form, information disclosure by financial institutions.[5]

Integration of Supervisions

In the last two decades many countries have reformed the structure of their financial supervision. In increasing number of countries, a trend toward a certain degree of consolidation of powers can be noticed, these forces which in several cases, have resulted in the establishment of unified supervisory authorities. These single authorities supervising the entire financial systems are often different from national central banks.

The single supervisor regime may seem to be a "natural" and best answer to the challenges posed by financial market integration. If, in the long run, the expected financial structure is perfectly integrated and single market results, the best design for the supervisory architecture would seem to be a single / unified authority. But the answer is apparently not that simple.[6] It should be emphasised that changing the institutional structure of supervision cannot guarantee in itself an effective supervision.

Several reasons have been put forward to justify the integration of supervision at the national level. The arguments that have been often advanced in favour of unification are as follows:[7]

1. Supervision of financial conglomerates One of the leading reasons is related to the emergence of financial conglomerates. The rise of

financial conglomerates, which operate diverse groups of financial institutions domestically or internationally, has increased the need for the sectoral supervisors to cooperate and coordinate their actions in an aim to ensure comprehensive supervision. Fragmented supervision may raise concerns about the ability to ensure that supervision is seamless and free of gaps and overall risk


is assessed. There are also group-wide risks1 that may not be adequately addressed by sectoral supervisors who oversight only a part of a conglomerate.

2. Competitive neutrality As financial systems have evolved and matured, the division lines between products and

institutions have blurred. This may lead to a situation, where financial institutions offering similar products or services, are supervised by different sectoral supervisory agencies. Differences in their regulations and associated cost of achieving compliance may give certain institutions a competitive advantage in offering a particular product or service.

The existence of more than one supervisory authority poses the risk of supervisory arbitrage. There is a possibility that a particular service or product is placed in that part of a financial conglomerate where the supervisory oversight is the least intensive or supervisory costs are the lowest.

Unified supervision is better able to iron out differences and inconsistencies and deal with all mentioned above problems.

Complete regulatory neutrality should not be a primary objective of supervision. It is proper to supervise the same operations differently depending on the nature (and thus systemic importance) of the institution in which it is carried out.

3. Regulatory flexibility The unified approach to supervision may allow for development of regulatory arrangements

that are more flexible, especially when a new type of financial product or institution emerges, which was not covered by the original legislation. A status for the unified supervisory agency has to be drafted with adequate flexibility to permit it to rapidly respond to market innovations.

4. Regulatory efficiency There are areas where supervisory unification may lead to cost savings and economies of

scale. The unification may permit cost savings on the basis of shared infrastructure2, administration and support systems. That may diminish the burdens of financial institution under supervision. Unification may help to minimize wasteful overlap and duplication of oversight, research and data collection, thus lay the basis, for a more efficient reporting system3. Unification may also permit the acquisition and taking advantage of information technologies which become cost-effective only beyond a certain scale of operations.

5. Developing a body of professional staff Effective supervision requires a skilled professional staff, so the supervising authority should

be able to attract, retain and develop such staff. Unification can assist this process. Unified supervisory agency would be able to offer its staff a more varied and challenging career, tailored training programs and through these enable its employees to be developed to their greatest potential.

6. Improved accountability The existence of one unified supervisory agency instead of multiple agencies, perhaps with

overlapping responsibilities and areas of jurisdiction, should prevent a moral hazard which can occur in case of sectoral supervisors blaming each other if something goes wrong. The advantage of one supervision is that by creating a single management structure, it should be clear to the market who should be held accountable for particular actions.4

The integration of supervisions has not only its fervent advocates. There are also opponents of this

idea. Among arguments against supervision’s unification are following:[7] 1. Unclear objectives

Integrated supervision may have considerable difficulty in striking an appropriate balance between the different objectives of regulation. The difficulty of designing a single set of objectives may result in a vague or an ill-defined statutory responsibilities, which may cause problems of holding the supervision accountable for its activities.

2. Diseconomies of scale


Unified supervision may suffer from some diseconomies of scale. Some inefficiencies5 are usually associated with monopolies and an integrated supervision is a such monopoly. Another source of diseconomies of scale is a tendency to assign to a unified supervision, the new responsibilities which are beyond its already broad scope.

3. Limited synergies Some economists claim that economies of scope are likely to be much less significant than

economies of scale. The cultures, styles, focus and skills of supervisors from different sectoral supervisory agencies vary markedly which mirrors their distinctive approaches and performances of oversight. The problem is that these differences remain in an unified agency. Even internal organization not assuaged the difficulty as it very often has separate departments reflecting traditional institutional line – i.e. banking, securities, insurance. To some extent the difficulty has been compounded – or at least not assuaged – by the fact that the internal organization of these agencies has tended to mirror traditional institutional lines – e.g. most have been established with separate departments for banking, securities and insurance regulation.

4. Moral hazard The public may have a tendency to assume that all creditors of institutions supervised by a

unified supervision will receive equal protection.6 This is a serious but obviously an informational problem and thus integrated supervision should as soon as possible clarify the rules of the game.

Different review essays have shown advantages and disadvantages of different architectures of

financial supervision and demonstrated that there are not any strong theoretical arguments in favour of any particular model[8]. It is clear that there is no such thing as an optimal supervisory structure.

Regulatory and Supervisory Powers of Unified Agencies

To fulfil their supervisory duties, the supervisors have an access to various supervisory powers. Integrated agencies may have various powers. The basic7 supervisory and regulatory powers that an integrated agency with wide authority in the financial sector is expected to have, are:[9]

1. Conduct of on-site examinations. 2. Conduct of off-site examinations. 3. Imposition of sanctions and fines for non-compliance with rules and regulations. 4. Issuance and change of prudential regulations on credit, market, operational and liquidity risks. 5. Modification of accounting and disclosure rules. 6. Setting of minimum capital requirements and, if deemed appropriate, requiring intermediaries to

comply with higher requirements or grant them temporary suspension (regulatory forbearance). 7. Issuance and change of rules on the composition of capital. 8. Setting of general licensing requirements. 9. Approval / revocation of a license of an intermediary. 10. Resolution of issues related to consumer protection.

The powers of the unified agencies are concentrated around core supervisory functions, such as the

power to conduct on-site and off-site examinations, as well as the power to impose sanctions and fines on financial institutions for non-compliance with existing laws and regulations. If the unified agencies have few regulatory and supervisory powers, it means that other institutions, such as the Ministry of Finance and/or the central bank, continue to have important regulatory and supervisory powers in the country.[9] NATURE OF FINANCIAL SUPERVISORY SYSTEM IN POLAND Origin and Institutional Place of the PFSA

The Polish Financial Supervision Authority (PFSA) initiated its activities on September 19, 2006, i.e. the date when the Act on Financial Market Supervision of July 21, 2006[10] came into force. In the first phase of the merger of financial supervision in Poland, the new supervisory body took over the duties of


the Insurance and Pension Funds Supervision Commission and the Polish Securities and Exchange Commission, which were abolished by the said Act. In the second phase (January 1, 2008), the PFSA took over the tasks of the Commission for Banking Supervision together with its Office – the General Inspectorate of Banking Supervision. The President of the Council of Ministers appointed on September 29, 2006[11] the first Chairperson of the Polish Financial Supervision Authority, Stanisław Kluza. The Chairperson of the PFSA shall perform his functions for five years.

The PFSA is an independent, separated agency functioning outside of the central bank or the ministry of finance. Supervision over the financial market includes:

1) banking supervision, 2) pension supervision, 3) insurance supervision, 4) capital market supervision, 5) supervision over electronic money institutions, 6) complementary supervision.

The Polish Financial Supervision Authority’s activity is supervised by the President of the Council of Ministers. Funding of the PFSA

The Polish Financial Supervision Authority Office is a budgetary entity. The level of expenses on its activities is specified in an annual budget act. According to the legal doctrine, a PFSA Office’s income comes from obligatory fees levied upon the regulated institutions. „In 2009, the PFSA reported a budget revenue of PLN 224,986.6 thousand on an accrual basis and PLN 148,035.7 thousand on a cash basis. The cash revenue derived mainly from fees paid by the industry to cover the costs of supervision of PLN 143,209.4 thousand. The fines incurred and paid by the industry and proceeds other than revenues designated to cover the costs of supervision amounted to PLN 4,826.3 thousand.”[12] The level of these fees is calculated based on separate legal acts governing this specific matter.

Organizational Structure of the PFSA

The PFSA has been headed by Stanislaw Kluza since 2006. The Chairperson should have two Vice-Chairpersons. At present he has only one – Leslaw Gajek. The members of the PFSA are:[13]

1) the minister competent for financial institutions or such minister’s representative (representative of the Minister of Finance - Dariusz Daniluk),

2) the minister competent for social security or such minister’s representative (representative of the Minister of Labour and Social Policy - Marek Bucior),

3) president of the National Bank of Poland or Vice-President delegated by him or her (Vice President of the National Bank of Poland - Witold Koziński),

4) a representative of the President of the Republic of Poland (Danuta Wawrzynkiewicz).

The Commission performs its duties with the assistance of its office – the PFSA Office. The PFSA Office is divided into seven branches[14]. The first three are charged with supervisory

activities: Capital Market Supervision (managed by Director Marek Szuszkiewicz[15]), Insurance and Pension Supervision (headed by Director Dagmara Wieczorek-Bartczak[15]) and Banking Supervision (governed by Director Andrzej Stopczynski[15]). Each of them is charged with supervisory responsibilities in its own area, because the PFSA, although being a single supervision authority regulating the entire financial market, performs its duties based on separate regulations specified in acts pertaining to individual sectors of the financial market[12].

The next very important branch concerns inspection of the entire financial market and is managed by Acting Director Pawel Sawicki[20]. This part emerged in a process of an organisational integration of inspection activities and procedures carried out by the PFSA and unification of the inspection methodologies within the PFSA. A stand-alone Inspection branch, responsible for all the sectors, was


created as legal regulations governing the inspection activities are relatively similar across all regulated sectors of the market.[12]

The fifth branch, headed by Director Adam Plocinski[15], is Financial Market Development and Cross-sector Policy. It performs task with a view to develop financial market and its competitiveness and supervise financial conglomerates[16]. This branch is dedicated i.e. to education, public relation and international cooperation.

Law and Legislation - the sixth branch, governed by Director Marek Wedrychowski[15], coordinates preparation of legislative projects (concerning supervision over the financial market in the PFSA Office) and performs tasks concerning the legal services.[16]

The final branch, Administration, is managed by Stanislaw Soltys[15]. Its basic issues are related to human resources, security and IT, but also administrative and financial matters[14].

The Polish Financial Supervision Authority Office besides above mentioned seven branches also has in its structure an Office of the Commission (a PFSA Cabinet) headed by Director Marzena Borowiec[15], Numerous Plenipotentiaries for Internal Audit and Plenipotentiary for Classified Information Protection.[14]

An organisational chart of the Polish Financial Supervision Authority Office is shown in Figure 1. The head office of the PFSA Office is in Warsaw (the capital of Poland). In addition Banking

Supervision branch has 16 local entities in 13 main Polish cities [16]. To fulfill tasks imposed by law, the Polish Financial Supervision Authority Office makes efforts to

gain highly qualified staff and maintain its high merit quality. At the end of 2008 employment in the PFSA Office amounted to 827 full-time positions and grew fast. Thanks to that the average employment in 2009 reached 891 full-time positions. Most people have worked in Banking Supervision branch – 357.4 full-time positions.[16]

To raise a level of knowledge and cater to the needs resulting from changes in areas connected to supervision, as well as to prepare numerous groups of employees for work in the inspection area, the PFSA Office realizes a lot of trainings. In 2009, subjects of these trainings concentrated on different types of risk, IAS/IRFS8, analysis of financial statements and internal audit. In 2009, per one employee fell an average of 5.2 trainings, which meant 8.7 days of trainings.[16]

All these actions are aimed at supporting present and attracting new employees, who are crucial to meet the challenges appearing in supervision over incessantly changing financial market.


FIGURE 1 THE STRUCTURE OF POLISH FINANCIAL SUPERVISION AUTHORITY OFFICE

Source: based on PFSA webside, Organisational chart of PFSA, http://www.knf.gov.pl/en/Images/organisational_chart_pfsa_tcm81-8078.pdf, accessed 1 September 2010.

Capital Market

Supervision

Insurance and

Pension Supervision

Financial Market Development and Cross-sector Policy

Banking Supervision

Inspection

Law and Legislation

Administration

The Vice-Chairperson

The Chairperson Office of the Commission

Numerous Plenipotentiaries for Internal Audit

Financial Services, Licensing and

Functional Supervision Department

Issuers Department

Trading Supervision Department

System Integirity Supervision Department

Insurence Financial Supervision Department

Risk Monitoring Department

Pension Investments Supervision Department

Financial Intermediaries Department

Customer Protection Department

Education Department

Research Department

International Cooperation Department

Public Relations Department

Bank Licensing Department

Banking Sector Supervision Department

Co-operative Banking Department

Standards and Procedures Inspection

Risk Evaluation Department

Legal Department

Enforcemetnl Department

Human Resources Department

IT Department

Financial and Administrative Department

Security Department

Plenipotentiary for Classified Information Protection

Supervisory Aims, Tasks and Functions The PFSA is responsible for an integrated supervision of Polish financial market. At the end of 2009

Polish supervisory office regulated, inter alia, 643 banks conducting operations (including 49 commercial banks, 18 branches of credit institutions, 576 cooperative banks)[17], 14 open-end pension funds, 1099 employee pension programs, 668 insurance companies, 46 brokerage houses and 13 banks conducting brokerage activities as well as 13 custodian banks.[16]

The primary objective of the Polish Financial Supervision Authority is „to ensure regular operation of the financial market, its stability, security and transparency, a confidence in the financial market, as well as to ensure the protection of interests of these market actors”[10]. To achieve this aim the PFSA fulfill the following tasks:[10]

1) supervising of the financial market (that is banking supervision, capital market supervision, insurance supervision, pension scheme supervision and supervision of electronic money institutions),

2) undertaking measures aimed at ensuring regular operation of the financial market, 3) undertaking measures aimed at development of financial market and its competitiveness, 4) undertaking educational and informational measures related to financial market operation, 5) participating in the drafting of legal acts related to financial market supervision, 6) creating the opportunities for amicable and conciliatory settlement of disputes which may arise

between financial market actors, in particular disputes resulting from contractual relations between entities covered by the PFSA supervision and recipients of services provided by those entities,

7) carrying out other activities provided for by acts of law.

Furthermore, on March 31, 2008 the Conciliatory Court at the PFSA began to function. In February 2009, in response to the situation on the financial market, it extended its scope of operations to include mediation.[12]

Every year the Polish Financial Supervision Authority presents to the President of the Council of Ministers report on its activities.

Interrelations Between Supervisory Office and the Financial System

The Polish Financial Supervision Authority tries not to be introvert and inaccessible, but takes action in aim to make, improve and strengthen contacts with financial market participants. The PFSA carries out i.e. educational initiatives, “Market Meetings” and cooperates with international organisations.

In 2009 the PFSA launched the CEDUR brand (Educational Centre for Market Participants) under which cyclical educational activities are carried out. The PFSA’s educational activities are targeted at:[12]

1) professional financial market participants, 2) the judiciary, prosecutors and law enforcement officers, 3) local and regional consumer advocates, 4) members of the media, 5) scholars and academic communities, 6) consumers of financial services.

The CEDUR initiative conducts activities such as organising training seminars for both professional

and non-professional financial market participants. The seminars offer an opportunity to share unique knowledge and experience in an easily understood form. A relevant advantage of the seminars is the opportunity to exchange views directly with the supervision’s representatives. In 2009, the PFSA has organised 39 CEDUR seminars, which enjoyed high popularity and attracted about 2,600 participants.[12] Included in the scope of CEDUR activities is an operating of the ManyMany.info website9. This educational website, designed and launched in 2009, is addressed to secondary school and academic students. The website, thanks to its user-friendly design, explains in a simple way the mechanisms of household budgeting, pensions, credit cards and other financial products.[12]


The next project is “Market Meetings”, which main purpose is to improve the standards of communication and cooperation with the market organisations. The PFSA’s senior personnel and heads of financial trade organisations meet regularly in four thematic groups: banking sector, pension market, insurance and capital market to discuss the most recent problems of the market. In 2009, 16 rounds of “Market Meetings” were organised.[12]

Apart from national activities, the Polish Financial Supervision Authority cooperates with international organisations. Engagement in international projects enables the PFSA to discuss new regulatory and supervisory concepts, but also to exchange information which then may be used to better perform analytical and supervisory tasks.[12] In 2009, representatives of the PFSA participated in the following initiatives:[18]10

1) European Union a. the Level 3 Committees (CESR11, CEIOPS12, CEBS13), b. public consultations14, c. EU regulations and related documents, d. EU funded projects15,

2) internationals organizations a. IOSCO16, b. IAIS17 c. IOPS18,

3) memoranda of understanding, 4) BSCEE19, 5) TIFS20.

The Polish Financial Supervision Authority also cooperates with the National Bank of Poland – the

Polish central bank. Both institutions signed document concerning mutual cooperation and information exchange[16]. Effective flow of information is important both during calm periods as well as in emergency situations. NATURE OF FINANCIAL SUPERVISORY SYSTEM IN GERMANY Origin and Institutional Place of BaFin

The German Federal Financial Supervisory Office under a German name of Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin), in its current form, was created on May 1, 2002 as a result of merger between Federal Banking Supervisory Office, Federal Securities Supervisory Office and Federal Insurance Supervisory Office. First two of these institutions have their beginnings before World War II while the latter originated much later in the 1990s. The reason for an establishment of comprehensive German supervisory authority is the growing integration of financial intermediaries on one hand, which no longer specialize in single financial product, but these days more often sell an extensive range of them[19]. On the other hand, businesses are growing larger and more powerful and so they demand a variety of products – there are easier to provide by one organization then by a number of individual providers[19]. Additional reasons in favor of a centralized supervisory authority is the move from publicly provided social security systems e.g. pension plans to privately financed ones by investment companies as well as the economies of scale and scope[19]. Thus, as these financial market changes have taken place over last five decades, subsequent changes have began to be implemented by their supervisors.

The legal bases for operation of BaFin are its By-Laws which govern the following: (1) structure and organization, (2) rights and obligations, (3) functions and powers and its Administrative Council and (4) details of budget preparation and control. Although both financial and technical control over BaFin’s activities is performed by the German Federal Ministry of Finance, in terms of budgetary income the supervisory is independent.


Funding of BaFin The supervisory office of BaFin is a non-profit organization. Its funding is regulated by a German act

which established the Federal Financial Supervisory Authority. According to the legal doctrine, BaFin’s income comes from three sources – fees levied upon the regulated institutions[20], separate reimbursements demanded for services provided[20] as well as other contributions[20]. The level of fees and contributions is calculated based on separate legal acts governing this particular matter especially i.e. Ordinance on the Imposition of Fees and Allocation of Costs Pursuant to the FinDAG.

However, expenses of the BaFin come from three main sources: Firstly, the imposition of penalties and fines on supervised institutions results in costs being incurred by the regulator. Secondly, the BaFin as a separate legal entity must maintain accounting documents and disclose all pertinent data which results in administrative costs. Thirdly, expenses are incurred by BaFin due to their regular operational activities.

Organizational Structure BaFin

The main body of BaFin is its Administrative Council whose main role is an oversight of the supervisory authority’s management as well as an advisory role as part of various advisory councils. It is also responsible for acceptance of the authority’s annual budget.

BaFin itself is divided into five branches. The first branch is the office of the President as of 2002 headed by Jochen Sanio [21]. It is divided into two parts: The first one is composed of the President’s office, internal audit and public relations, while the second part consists of a Department INT subdivided into six sections charged with technical cooperation, cross-sectoral multilateral activities, bilateral activities, and supervision of three supervisory branches.

The second branch consists of regulatory services and human resource management and it is headed by Chief Executive Director Michael Sell[21]. This part of BaFin is divided into four departmental units: Department Q1 is dedicated to risk and financial markets analysis, Department Q2 – to consumer and investor protection, Department Q3 – to integrity of the financial system, Department QRM – to cross-sectoral risk modeling and Department Z to central services. This branch also includes GW Group for prevention of money laundering and IT Group for information technology management.

The next three branches are directly charged with supervisory responsibilities. The first one – Banking Supervision, managed by Sabine Lautenschläger-Peiter[21], has five departments: Department BA1 for supervision of major banking and selected commercial banks, Department BA2 – supervision of Landesbanks, savings banks and building societies, Department BA3 – supervision of commercial banks, regional banks and specialist banks, Department BA4 – supervision of credit institutions and housing enterprises and Department BA5 – dedicated to basic issues.

Insurance and Pension Fund Supervision branch with five departments is governed by Thomas Steffen[21]. Department VA1 is primarily charged with occupational retirement provisions, Department VA2 with supervision of life insurance and death benefit funds, Department VA3 with supervision of property/casualty insurers, national insurance groups and quantitative supervision, Department VA4 with supervision of international insurance groups, financial conglomerates and reinsurers as well as their quantitative supervision and internal models and Department VA5 is responsible for basic issues, VA for policy and risk orientation.

The final Securities Supervision and Asset Management branch, managed by Karl Burjhard Caspari[21], includes four departments. That is, Department WA1 – basic issues related to securities supervision, company takeovers, major holdings of voting rights and reporting, Department WA2 for among others, insider surveillance and market surveillance and analysis, Department WA3 basic issues related to supervision of securities and Department WA4 dedicated to investment funds. Below we have presented the onanigram of Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin):


FIGURE 2 THE STRUCTURE OF BUNDESANSTALT FÜR FINANZDIENSTLEISTUNGSAUFSICHT (BaFin)

Source: based on BaFin website, Organisation chart of the Federal Financial Supervisory Authority (pdf/906 KB), http://www.bafin.de/cln_179/nn_720486/EN/BaFin/Organisation/organisation__node.html?__nnn=true, accessed 15 August 2010.

Basic issues departament banking

supervision

Regulatory services / Human resources

Prevention of money laundering

Information technology

Risk and financial markets analysis

Consumer and investor protection; certification of retirement savings

contracts and particular legal issues

Insurance and Pension Fund Supervision

Occupational retirement provision;

supervision of Pensionskassen,

pension funds and health insurers

Supervision of property/casualty insurers; national insurance groups;

quantitative supervision

Supervision of life insurers and death

benefit funds

President

Presiden’s Office

Press and public relations / Information

management

Internal audit

International Policy / Affairs

Securities Supervision / Asset

Management

Basic issues relating to securities supervision; company takeovers;

major holdings of voting rights; reporting

Supervision of FSIs In accordance with the Banking Act (KWG) and the Securities

Trading Act (WpHG); supervision of credit

institutions in accordance with

WpHG; basic issues relating to the

interpretation and verification of rules of conduct (section 31 et

seq. WpHG)

Insider surveillance; ad hoc disclosure;

directors’dealings; stock exchange

competence centre; market surveillance

and analysis; prospectuses

Banking

Supervision

Supervision of major banks and selected commercial banks

Supervision of Landesbanks, savings

banks and building societies

Supervision of commercial banks,

regional and specialist banks, Pfandbrief

banks, stockbrokers, securities trading banks; Pfandbrief

competence centre

Supervision of credit institutions in the legal

form of registered cooperative societies

and housing enterprises with

savings schemes; issues relating to

currency conversion and accounting in DM

Investment funds

Supervision of international insurance

groups; financial conglomerates and

reinsurers; qualitative supervision; internal

models

Basic issues; VA policy; risk orientation

Integrity of the financial system

Cross-sectoral risk modelling

Central services

To fulfill all abovementioned tasks BaFin employs, as of April 2008, about 1 700[22]21 persons of which approximately two-thirds are civil servants at three levels – basic/middle, upper and higher – and one-third are regular employees. Most of them i.e. approximately 68% work in the main BaFin’s location in Bonn, 20% work in Frankfurt am Main and the remaining 10-12% work in national or international authorities and institutions or are on temporary leaves. While majority of the employees work in three main branches of Banking Supervision, Insurance and Pension Fund Supervision and Securities Supervision and Asset Management, around 300 people work in two remaining branches charged with administrative and coordinative duties[22].

The supervisory authority also utilizes trainees in order to fill in temporary posts as well as to train junior level workers to become part of higher and upper tier of workforce in the future. The training program is set up to provide a continuous education for all members of BaFin. This occurs by the means of clear career paths and targeted training syllabuses both inside (in-house trainings) and outside (external trainings) the organization. As of April of 2008, BaFin had 62 candidates for an entry from junior to upper Civil Service being trained in any of the following areas: administration, information technology, office communication, and media and information services[22]. Together with Bundesbank, BaFin also offers training practices for university students.

The demographic characteristics of its employees show a high degree of tolerance and appreciation of variety. Almost half of its workforce consists of women who recognize family-friendly working conditions as well as good professional opportunities. More than 75% of BaFin’s team is in the middle age category that is between 30 and 44 years of age and they come from various educational backgrounds including law, economics, public administration, computer science and mathematics. All these are necessary to adequately tackle problems faced by the supervisory office[22]. Supervisory Aims, Tasks and Functions

As it has been mentioned previously, BaFin is responsible for integrated financial supervision of German financial market which means an authority over supervising financial service providers, credit institutions, investment and insurance companies as well as some issues related to securities trading. BaFin’s primary objective of financial supervision is “to ensure the proper functioning, stability and integrity of the German financial market”[22]. As of May 2010, German supervisory office regulated 2 000 banks, 710 financial service providers, about 620 insurance companies, 28 pension funds, approximately 6 000 domestic investment funds and 73 asset management companies[22]. As part of BaFin’s solvency supervision, the office checks whether financial institutions are able to meet their obligations at all times, yet through its market supervision, it enforces fairness and transparency in the market[22].

Pertinent to all supervisory functions, the aim consists of ensuring that “only authorized companies offer their services on the market and that” their management meets “professional qualifications” as well as “that institutions comply with the statutory and regulatory principles for the banking business”[22]. In particular for the insurance business, BaFin must assure that interests of policyholders are protected in a long-term horizon, taking into account the longevity of these contracts. Moreover, it ensures that capital from policies is invested safely and profitably. For the purposes of asset management supervision, it enforces financial reporting rules and guarantees that market conditions are fair and transparent[22].

The supervisor also performs an investor protection function as it accepts complaints from financial market consumers and tries to resolve them by verifying if organizations follow statutory rules and rulings. BaFin also cooperates with other EU financial market bodies to create a uniform regulatory environment and to originate and improve on regulatory laws. Finally, it performs a representative role of a Germany as a European financial centre[22].

While maintaining highest quality standards, BaFin’s legal responsibility is based upon six pillars[23]22:

1. Risk-based resource allocation: Due to scare resources, BaFin supervises undertakings that have a highest probability of occurrence as well as whose size of loss is high. In other words, BaFin does not supervise all financial market activities.


2. Made-to-measure supervision with a sense of proportion: BaFin is a believer in a common-sense, market-oriented and coordinated financial market and so it supervises institutions using these principles depending on the size and scope of financial undertakings.

3. Responsibility for Europe: BaFin cooperates with other EU member authorities based on standard of mutual trust and upholds harmonized European Union supervisory practices to ensure stability of EU market.

4. Closeness to practical realities: BaFin’s role is to maintain an open dialogue with supervised institutions in order to promote self-regulation and draw up new laws in cooperation with all stakeholders.

5. Objectivity and transparency: All institutions are treated equally by the supervisory office, while applicable laws are made to be readily understandable by all. Breaches of the rules are punishable firmly, objectively and with BaFin’s tasks in mind.

6. Cost-consciousness: Since BaFin is funded by organizations it supervises, it is charged with being cost effective and thus at maintaining tight cost control.

Interrelations Between Supervisory Office and the Financial System

BaFin as one of its primary activities sees a cooperation between itself and other supervisory institutions of the EU member countries to create a single common European market. At the European level, the organization participates in three sector-specific committees – a banking supervisory committee (CEBS), an insurance and occupational pension supervisory committee (CEIOPS) and a securities regulation committee (CESR) – the same ones in which the Polish Supervisory Office holds seats. At an international level, alike PFSA, BaFin is engaged with three regulatory bodies: an insurance supervision (IAIS), securities regulation (IOSCO) and the Basel Committee on Banking Supervision. Besides these, the German regulator is present in other cross-sector organizations such as Financial Stability Board and the Joint Forum.

BaFin is deeply devoted to cooperation with the German central bank – the Bundesbank. Both institutions have prepared a common document, a “Memorandum of Understanding”, to outline each other’s duties and relationship between them. Deutsche Bundesbank has also prepared and issued, yet based on joint discussions with BaFin, a document entitled “Guideline on the execution and quality assurance of the ongoing supervision of credit and financial services institutions by the Deutsche Bundesbank”. REALIZATION OF SECONDARY AIMS Sharing Knowledge

Sharing knowledge pertains to a communication and an exchange of information among supervisory offices and the stakeholders such as consumers, supervised institutions and the government. The German and Polish supervisors organize conferences and seminars to achieve various aims. BaFin is targeted at arranging events of general market interest, where specially formed task forces work on solving problems. An example of that is a Financial Action Taskforce on Money Laundering (FATF) which has worked in 2008 on identifying legal loopholes in prevention and resolution of money laundering[24]. Thus, BaFin carries every day work in groups dedicated to solving precisely outlined problems. Large conferences, as for instance, the “Islamic Finance Conference” held in English on 29 October 2009 in Frankfurt, Germany23 are aimed at attracting various branches of financial sector and presenting a topic to a wider audience. This year up to October 2010, BaFin also prepared two other conferences, one dedicated to “Enabling Microinsurance Markets” in May of 2010 and “IOSCO Annual Conference 2010” in June.

The Polish supervisor – the PFSA – applies a different strategy to raise awareness about financial market and solve problems arising in it. It operates an educatory CEDUR – as it was mentioned in part II of this paper –, a centre whose aim is to (a) organize conferences and seminars, (b) propagate primary, secondary and higher level of education regarding financial markets and (c) maintain an educational platform. The conferences and seminars that CEDUR conducts pertain primarily to four financial sectors


that is capital, insurance, pension and banking. These are dedicated to various consumer groups, including Poland’s administration of city, municipal, vovoidship and country levels.[25] However, by looking at the topics24 most of them are rather dedicated to increasing knowledge about current rules adopted by the Polish banking sector or up-to-date problems facing the market, than on a long-term perspective on the market developments and ways to improve financial market conditions.

Polish supervisory office also in 2009 implemented a so-called Teaching Initiative for Financial Sector – TIFS. The aim of this organization is to prepare seminars on current financial market topics, exchange information with EU supervisory counterparts as well as to help implement best practices in Poland from the EU market. As it is quite a new initiative, so far it conducted only three seminars that is (1) “Financing the Real Estate Market - in the context of the financial crisis” in October of 2009, (2) “SREP-process and risk management. Investment market case” in April of 2010, and (3) “Insurers' investments activity - new challenges for supervisory institutions” in October of 201025.

Given that, the German supervisor uses a different approach than its polish counterpart. While German’s BaFin concentrates on resolving current problems in smaller groups i.e. task forces whose work results in special recommendations to BaFin’s governmental authorities, the PFSA prefers to prepare solutions individually within departments – there appears to be “a grey zone” image on how the decisions are made and then to increase the awareness about solutions to wide groups of market participants. Likely, these approaches are different as a result of longevity of the supervisors and of varying needs of developed and developing financial markets. Consulting with the Market

Polish supervisor conducts weekly meetings between managers of the PFSA and the managers of financial institutions as well as financial intermediaries. These meetings are divided into four groups – banking, capital markets, insurance and pension – and they are aiming at increasing market communication between the supervisory and the consumers and also at accessing their satisfaction from being a part of the market. As a result of each meeting an announcement is published including the minutes and recommendations. However, it should be once again emphasized that these groups focus on current issues, rather than longer-term solutions.

On the contrary, BaFin does not provide open market meetings or at least no information on it is available to the public. Some forms of consultations are likely, yet it seems like they are conducted at hoc, not on regular basis. Moreover, both supervisory offices utlise a computerised method for registering complaints from the market. The German supervisory authority has a widespread system of complaints regarding the supervised bodies. On its website, BaFin advises consumers how to process with complaints regarding the financial markets – it provides information on the content of complaint as well as appropriate addresses for filing them with BaFin26. It also has a database of companies which can be utilized to access data on German financial intermediaries.

The Polish counterpart also provides the information on supervised financial companies but rather in a table format. In addition, it accepts complaints from the public at its website through a special form27 as well as in the form of a written letter of complaint or in person at its Warsaw office. The PFSA’s website is somewhat deficient in comparison to BaFin because the latter provides answers to most frequent questions asked by consumers and gives an advice on how to proceed with a complaint starting with a company itself and treating the supervisory office as a last resort. Even though, it is important to point out that both institutions are visibly interested in communication with its stakeholders and, in case of any problems, available to help. Contributing into Law Formation

Preparing a new law for future improvement of regulation of financial market is an aim of both authorities. The PFSA over the 2008-2010 initiated about five legal acts, fifteen resolutions and opinionated numerous legal acts of which most are, unfortunately, short-term oriented and pertain to current business matters. Only few are longer-term solutions such as e.g. „A legal act project regarding a public offer and conditions on implementation of financial instruments to public exchange as well as


regarding public companies and legal act pertaining to financial instruments exchange” filed with Polish Ministry of Finance in December of 2009. Although this is a very important part of the PFSA activities, this could be a place for preparation of legal fundamental institutional changes. Instead, it seems like it is a place to fix shortcomings of Polish present legal framework.

In Germany on the converse, while the BaFin also takes care of closing legal loopholes such as, for example, as a result of the BVerwG ruling, the Act on the Continuing Development of Pfandbrief Law, whose aim was to strengthen customer protection in the Cover Pool schemes. But it has also began much greater changes, for example when on May 18, 2010 issued decrees “prohibiting (i) naked short-sales of eurozone government bonds, (ii) naked short-sales of 10 German financial stocks, and (iii) entering into uncovered credit default swaps linked to Euro zone government debt28.

In summary, these secondary areas of supervisory activities are very important, especially in light of turbulent financial markets and changing economic conditions. It is essential that these institutions stay in close contact with their customers and supervised companies as well as all other stakeholders in order to not only successfully resolve short-term dilemmas but also to prepare for and later address long-term strategic and institutional fundamentals of financial markets. It seems like the Polish supervisory authority is still not fully ready to fulfill this objective, although as can be seen by their current work, it tries to develop its internal structures to account for this target as well. CONCLUSION

The aim of the paper provided above was to conduct a comparative analysis of two supervisory offices – the Polish Financial Supervisory Authority and the German Federal Financial Supervisory Office – which were relatively recently organized as a result of merger of cross-sectoral supervisory offices. While at first sight, these seem quite alike, and at second look, they are not completely so. Both of them possess organizations and tools necessary to conduct daily supervisory routines and to solve current problems in the financial market. Both emphasize the importance of highly qualified and dedicated workforces and spend a lot of time and effort to increase competences of their employees through trainings and seminars. In both offices, communication with key stakeholders is essential although here the most important difference is already visible. While Polish FSA is aimed at communicating about current issues, the BaFin is more focused on long-term strategic information exchange. This is, for example, visible in the kinds of workshops, seminars and conferences organized by these two contrasted institutions. The former being more oriented toward present with a numerous brief seminars to increase financial awareness or solve current problems in comparison to the latter focused on organizing few large conferences per year in order to attract highly specialized individuals to find answers to long-term strategic problems. As it was mentioned in the article, the reason for this dialectic may come from the nature of Polish versus German financial markets in which the first is still developing and the second is already well developed. However, this should be a warning sign to the Polish authorities since as the Polish market develops, the PFSA will need to re-sharpen its focus on more institutional aspects to help their market compete with other European financial markets rather than just to aid the market in its smooth functioning within the Polish market. ENDNOTES

1. Among these risks are whether the group as a whole has an adequate capital and whether it has adequate systems and controls for managing its risks.

2. For example in data collection and processing and personnel administration. 3. There are important synergies between the data required for banking supervision and monetary policy

purposes which may outweigh the synergies between the data necessary for banking supervision and for other financial intermediaries’ supervision.

4. The fundamental consideration should be clarity of supervisory objectives rather than the number of agencies involved in the regulation. A unified supervisory agency with ill-defined objectives might be more difficult to hold to account than sectoral supervisory agencies with clearly specified objectives.


5. For instance more inflexibility and bureaucracy. This issue, however, is more likely to depend on quality of the organization, management and staff.

2. For example, if depositors are protected from a loss in the event of bank failure, then the clients of all other financial institutions may expect to be treated in an equal manner.

3. Naturally, agency may have many other powers that are not reflected in this list. 4. IAS – International Accounting Standards, IFRS – International Financial Reporting Standards. 5. www.manymany.info 6. http://www.knf.gov.pl/en/About_us/International_Cooperation/index.html, accessed 9 September 2010. 7. The Committee of European Securities Regulators. 8. The Committee of European Insurance and Occupational Pensions Supervisors. 9. Committee of European Banking Supervisors. 10. To fulfil Lamfalussy procedure’s requirements CESR, CEIOPS and CEBS hold open, public hearings on

draft documents. The open hearings are published on committees’ webpage with deadline for commenting on the drafts.

11. The PFSA participates in the implementation of the assistance projects. 12. The International Organization of Securities Commissions. 13. The International Association of Insurance Supervisors. 14. The International Organisation of Pension Supervisors. 15. Banking Supervisors from Central and Eastern Europe. 16. The Training Initiative for Financial Supervision was established by the PFSA in 2009. It is a training

centre for supervisors of all sectors of the financial market (banking, capital market, pension and insurance) from European countries.

17. As of May 2010, the source BaFin website, BaFin – About us (pdf/174 KB), http://www.bafin.de/cln_171/nn_721302/EN/BaFin/Functions/functions__node.html?__nnn=true, accessed 15 August 2010, provides a number of employees as 1900, however, without any further details.

18. These are a part of BaFin’s mission statement. 19. For more information visit the BaFin’s website at:

http://www.bafin.de/cln_171/nn_720790/SharedDocs/Veranstaltungen/EN/2009/islamic__finance__confernce.html.

20. One can view the topics of conferences and seminars for the period 2007-2009 at CEDER webpage at http://www.knf.gov.pl/dla_rynku/edukacja_cedur/cedur/cedur.html#1.

21. For more information visit the TIFS’ website at: http://www.knf.gov.pl/o_nas/wspolpraca_miedzynarodowa/tifs/tifs.html.

22. For more information visit BaFin’s website at http://www.bafin.de/cln_179/nn_721994/EN/Consumers/Complaintscontacts/ComplaintoBaFin/complaintobafin__node.html?__nnn=true.

23. To view the list of possible forms depending on the subject one should visit PFSA’s website at http://www.knf.gov.pl/lista_formularzy.jsf.

24. Client Briefing, (28 May 2010), BaFin bans naked short-selling and uncovered sovereign CDS trading / new draft law proposes further restrictions including currency derivates - Update 2, p. 1.

REFERENCES Mały Słownik Języka Polskiego (1969). Warszawa: Wydawnictwo Naukowe PWN. Goral, L., Karlikowska, M. and Koperkiewicz–Mordel, K. (2003) Polskie prawo bankowe. Second edition. Warszawa: LexisNexis. Kallas, M., Lipowicz, L. and Niewiadomski, Z. (2002) Prawo administracyjne - czesc ustrojowa. Warszawa: LexisNexis. Daniluk, D. (1996) Regulacje i nadzór bankowy w Polsce. Warszawa: Biblioteka Menedzera i Bankowca.


Holopainen H. (2007) Integration of financial supervision. Bank of Finland Research, Discussion Papers 12/2007. Masciandaro D. (2005) Financial Supervision Unification and Central Bank Fragmentation Effect. www.ssrn. Abrams, R.K. and Taylor, M.W. (2000) Issues in the Unification of Financial Sector Supervision. International Monetary Fund, IMF Working Paper No. WP/00/213. Masciandaro, D. and Quintyn, M. (2009) After the Big Bang and Before the Next One? Reforming the Financial Supervision Architecture and the Role of the Central Bank. A Review of Worldwide Trends, Causes and Effects (1998-2008). Universita Commerciale Luigi Bocconi. “Paolo Baffi” Centre Research Paper Series No. 2009-37. de Luna Martínez, J. and Rose, T.A. (2003) International Survey of Integrated Financial Sector Supervision, www.ssrn.

Dz. U. of 2006, No. 157, item 1119, as amended. http://www.knf.gov.pl/o_nas/komisja/stanislaw-kluza.html, accessed 4 September 2010. Report on the Activities of the Polish Financial Supervision Authority in 2009 (2010). Warsaw: Polish Financial Supervision Authority. http://www.knf.gov.pl/o_nas/komisja/index.html, accessed 4 September 2010. http://www.knf.gov.pl/en/Images/organisational_chart_pfsa_tcm81-8078.pdf, accessed 4 September 2010. http://www.knf.gov.pl/en/dane_wspolne/Contact_us.html, accessed 4 September 2010. Sprawozdanie z dzialalnosci Komisji Nadzoru Finansowego w 2009 r. (2010). Warszawa: Komisja Nadzoru Finansowego. Report on the condition of Polish banks (2010). Warsaw: Polish Financial Supervision Authority. http://www.knf.gov.pl/en/About_us/International_Cooperation/index.html, accessed 9 September 2010. Schüler, M. (2004) Integrated Financial Supervision in Germany. Centre for European Economic Research. Discussion Paper No. 04-35. Finanzdienstleistungsaufsichtgesetz. BaFin (2010) The Executive http://www.bafin.de/cln_152/nn_721302/EN/BaFin/Organisation/TheExecutive/theexecutive__node.html?__nnn=true, accessed on 10 September 2010. http://www.bafin.de/cln_171/nn_721302/EN/BaFin/Functions/functions__node.html?__nnn=true, accessed 15 August 2010.


http://www.bafin.de/cln_179/nn_721300/EN/BaFin/Legalbasis/Missionstatement/missionstatement__node.html?__nnn=true, accessed 10 September 2010. BaFin (2008) BaFin’s Annual Report 2008, section VII Cross-sectoral responsibilities. Bonn: BaFin. http://www.knf.gov.pl/dla_rynku/edukacja_cedur/index.html, accessed 1 September 2010.


journal of applied business and economics€¦ · ©journal of applied business and economics 2012...

Documents