journal of economic geography volume 6 issue 5 2006 [doi 10.1093_jeg_lbl003] wojcik, d. --...

8/10/2019 Journal of Economic Geography Volume 6 issue 5 2006 [doi 10.1093_jeg_lbl003] Wojcik, D. -- Convergence in corp

1/22

Journal of Economic Geography 6 (2006) pp. 639660Advance Access published on 9 May 2006 doi:10.1093/jeg/lbl003

Convergence in corporate governance: evidencefrom Europe and the challenge for economicgeographyDariusz Wojcik * , **

AbstractUsing a dataset on corporate governance ratings of the 300 largestpublicly traded European companies from 17 countries, I analyse the dyn-amics of corporate governance between 2000 and 2004 focusing onconvergence. Within the structure of corporate governance, shareholdersrights and duties and takeover defences have changed little while ratingsfor board structure and functioning, and particularly for disclosure, haverisen in every country and industry. Continental companies have narrowedthe gap in relation to the UK and Ireland, and there is evidenceofconvergencewithin individual countries and industries. Nevertheless, the Europeancorporate governance landscape is still diverse, with differences betweencountries overwhelming differences between industries.

Keywords: corporate governance, convergence, rating, EuropeJEL classifications: G15, G34, K00, P51Date submitted: 1 March 2005 Date accepted: 20 January 2006

1. Introduction

After Parmalat and Ahold followed the Enron and Worldcom corporate scandals in theUnited States, the term corporate governance has reinforced its presence in the head-lines of European media. There is no reason to be surprised by the medias hunger forcorporate bad news or to consider that corporate scandals are a new feature of capit-alist economies; however, there are reasons to believe that corporate governance hasbecome prominent in policy-making and research. As the number of people and insti-tutions owning shares grows in the US and in Europe, the treatment of shareholders by

companies has become more important than ever before (NYSE, 2000; FESE, 2002) onboth sides of the Atlantic. In the United States, the Sarbanes-Oxley Act of 2002 imposesstricter disclosure and audit requirements on corporations. In Europe, the policyobjective is to develop a model of corporate governance that spurs innovation andproductivity (EC, 2003). Disagreement on what model to choose in Europe has leadto a heated debate concerning such issues as the European Company Statute, the regu-lation of takeovers and the adaptation of the International Accounting Standards.

* Jesus College, Oxford, UK.

** Department of Geography, University College London, 26 Bedford Way, London WC1H 0AP, UK.email < [email protected] >

# The Author (2006). Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]

a t F or d h a m Uni v e r s i t y

on J a n u a r y1 6 ,2 0 1

3

h t t p : / / j o e g . oxf or d j o ur n a l s . or g /

D o wnl o a d e d f r om
http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/http://joeg.oxfordjournals.org/


2/22

According to Shleifer and Vishny (1997, p. 738) corporate governance deals with theways in which suppliers of finance to corporation assure themselves return on theirinvestment. Using a broader definition, the OECD describes corporate governanceas a set of relationships between a companys board, its shareholders and otherstakeholders (OECD, 1999, p. 1). According to another definition corporate govern-ance, is concerned with the institutions that influence how business corporations alloc-ate resources and returns. Specifically, a system of corporate governance shapes whomakes investment decisions in corporations, what types of investments they make, andhow returns from investments are distributed (OSullivan, 2001, p. 1). Theoreticalresearch in financial economics applies a narrow definition of corporate governan-cearguably because it can be modelled more easily. Most corporate governanceliterature, however, including textbooks, stresses the broad concept of corporate gov-ernance (compare Jensen and Meckling, 1976 with Monks and Minow, 2004 or Mallin,2004). Recognizing the complexity of the concept of the firm, corporate governancedebate thus comes close to the approach of economic geography, according to which,

the firm is indeed a messy constellation of multiple identities, contestation of power,and shifting representations (Yeung, 2003, p. 451).

What benefits do insights into corporate governance offer economic geography?Insight into disclosure tells us how firms present themselves to the public. Boardstructure and functioning, as well as the rights and duties of shareholders, reveal thedistribution of power, including even gender relations. Put differently, the buildingblocks of corporate governance are as much about the financial bottom line of corpor-ate performance, i.e. profit and return to shareholders, as they are about communica-tion, conversations, and discourses, all central to the discursive concept of the firm, nowwidely accepted within economic geography (Schoenberger, 1994; Thrift, 1996; ONeilland Gibson-Graham, 1999). The value of research on corporate governance is illus-trated by Dicken and Malmberg (2001, p. 347): the institutions and processesof governancethe sets of institutions, rules, and conventions that form the regulatorycontext of industrial systems, firms and territoriespervade all aspects of thefirm-territory nexus.

I do not suggest that economic geographers have ignored corporate governanceresearch in the past. First, they share some intellectual roots with corporate governanceresearchers. Firm theorists including Ronald Coase and Olivier Williamson areregarded as classic contributors to corporate governance research, and their workshave also been discussed thoroughly in economic geography (compare Jensen, 2000with Scott, 1983). Corporate governance is also addressed by research on power andgender relations within companies, the roots of trans-national companies and the gov-ernance of financial institutions (see Barnes and Sheppard, 2000; Clark et al., 2000).The point is that economic geographers often talk about corporate governance, withoutmentioning the term or referring to corporate governance research, despite potentialbenefit from doing so. In light of the current dynamics of research on corporategovernance, 1 the time is ripe for economic geography research to examine corporate

1 Consider the geographical spread of research on corporate governance, which started in the earnest in theUnited States in late 1970s following Jensen and Meckling (1976). In late 1980s it has moved to other

leading developed economies: Germany, Japan, and the UK. Only in late 1990s it has started to becomeworldwide including the emerging economies.

640 Wojcik


on J a n u a r y1 6 ,2 0 1

3




3/22

governance concepts and literature more explicitly (Wojcik, 2003; Clark andWojcik, 2005a).

This article directly addresses the developments in corporate governance, with a focuson convergence, which has been a major preoccupation of economic geographers in theEuropean context, mainly from the perspective of regional economic growth or insti-tutional development (Martin and Sunley, 1998; Rodr guez-Pose, 1998; Martin, 2001;Clark, 2003). The objective of the article is to make-up for the deficit of empiricalevidence on convergence and simultaneously to disentangle the complex concept of corporate governance. Using a unique and comprehensive proprietary dataset oncorporate governance for the largest 300 publicly traded European companies for2000 and 2004, the following research questions are addressed:

1. What was the state and structure of corporate governance in 2004 in Europe as awhole and how did it compare across countries and industries?

2. What has changed in the state and structure of corporate governance between2000 and 2004 and how did the change compare across countries and industries?

3. Has there been any convergence in corporate governance in Europe as a whole,or within individual countries and industries?

Of course, the largest 300 publicly traded companies are only a section of Europeaneconomies made up of millions of enterprises. The total employment in these companiescan be estimated at about 15 million, compared with the total employment in the EU of over 200 million people. 2 Although the data are restricted to the European corporateelite, their relevance for the understanding of corporate governance in European eco-nomies will be discussed in the concluding section.

The rest of the article is structured as follows. Section 2 discusses arguments in favourand against convergence. Section 3 introduces the data and methodology. Section 46present empirical results relating to each research question: Section 4 present the pictureof European corporate governance in 2004; Section 5 adds a dynamic dimension to thispicture, analysing changes between 2000 and 2004; Section 6 focuses on convergence.Section 7 concludes.

2. Convergence in European corporate governance

The article focuses on convergence to an idealized model of Anglo-American origin,not to any set of corporate governance characteristics. First, the former is the topic of almost the whole theoretical and empirical literature on convergence in corporate gov-ernance. Second, the tool I have used to measure corporate governance in firms, asdiscussed in Section 3, captures the conformity to an idealized Anglo-American model.The term idealized, instead of ideal, stresses a positive character of the article. It isnot an objective of the article to evaluate the Anglo-American corporate governance or judge whether convergence in corporate governance is good or bad. The current sectiondefines briefly the Anglo-American model and summarizes arguments on why we

2 The estimate of employment in the largest 300 publicly listed European companies is based on data from

Financial Times 500 (www.ft.com/FT500); the estimate for EU employment is based on the assumptionthat employment represents 50% of total population.

Convergence in corporate governance 641


on J a n u a r y1 6 ,2 0 1

3




4/22

should or should not anticipate convergence to this model. It also discusses thesignificance of convergence in the context of economic geography, and reviews theempirical literature on convergence in European corporate governance, leading to ahypothesis.

2.1. Concepts and theories

A distinction is commonly made between two generic regimes of corporate governancein developed economies (Shleifer and Vishny, 1997). In a regime of closed governancecorporations have concentrated ownership, with controlling owners (mostly wealthyfamilies, the state or banks) disciplining the management of the firm. In an openregime ownership in corporations is dispersed, with arms length relationships betweenshareholders and managers, whose interests are aligned with the interests of sharehold-ers through stock market-based compensation. In addition, in an open regime man-

agers are disciplined by the threat of takeover. Consequently, the public market forcorporate stock influences firms in an open regime to a much higher degree than in aclosed regime. The open regime is said to prevail in the United States and other Anglo-American economies; the closed regime in most developed economies including Europeexcept the UK and Ireland (La Porta et al., 1998). This is a crude but important dis-tinction, since the convergence debate is focused on the alleged shift away from theclosed to the open regime.

The driving forces and barriers to convergence are subject to a heated debate(OSullivan, 2003). The view rooted in neo-classical economics starts with the premisethat enhanced global product and labour market competition combined with financialintegration leads firms to converge on a set of best practices in corporate governance. 3

The best practice is defined as perfected Anglo-American corporate governance,with its primary objective of maximizing the shareholder value, improving access of savers to investment opportunities and firms access to external funds (Jensen, 2000;Hansmann and Kraakman, 2004). In contrast, the varieties of capitalism perspectiveclaims there is no best way to organize an economy. The forces of competition andfinancial integration do operate, but before they make an impact on corporate govern-ance in a specific place, they are filtered through existing, mostly nationally basedinstitutions. As specific configurations of institutions are likely to respond differently,even if exposed to similar pressures, corporate governance will continue to differdepending on the institutional context in which it is embedded (Hollingsworth and

Boyer, 1997; Hall and Soskice, 2001).Political theory of corporate governance provides another source of scepticism

about convergence, claiming that we cannot understand the evolution of corporategovernance without acknowledging the central role of the state (Roe, 2002). To give

3 One of the arguments for the superiority of the open regime is that dispersed shareholdings meanshareholders wealth depends on more diversified portfolios of investments (held directly or throughinstitutions such as pension funds and mutual funds) than in the case of closed regime withconcentrated ownership. Since the risk of a more diversified portfolio is lower, shareholders requirelower return relative to the risk. This in turn lowers the cost of capital faced by corporations, and

makes capital for risky ventures more available. The argument is to hold particularly in the conditionsof global capital market integration (see Errunza and Losq, 1985; Rajan and Zingales, 2001).

642 Wojcik


on J a n u a r y1 6 ,2 0 1

3




5/22

an example, according to Roe, the dispersed ownership structure in large US companiesresulted mainly from the popular will and political decision to prevent the concentra-tion of power in financial conglomerates. 4 The conclusion of political theory is that amajor change in corporate governance must be conditioned by a major political trans-formation. Combining political and economic approaches, the theory of path depend-ence suggests that as a necessary condition for convergence to an open regime, thepotential benefits of the shift have to outweigh the actual benefits of control accruingto controlling owners, whose interests are often backed by politicians (Bebchuk andRoe, 2000; Morck and Steier, 2005). To the current stock of theories and predictionsI would add the proposition that convergence to the Anglo-American system does nothave to rely on its purported superior economic efficiency. Jeffrey Gordon (2003) sug-gests that the corporate governance with no barriers to hostile takeovers contributes tothe European integration project, because it creates a level playing field for corpora-tions competing across national boundaries.

If we consider the foregoing definitions of corporate governance and its generic types

seriously, the significance of convergence in corporate governance is clear. A shift to amodel that maximizes shareholder value involves a transfer of power from alliances of corporate insiders to public shareholders, with the corollary of larger and deeper publicstock markets. The potential impact, however, does not stop there, as the corporategovernance system at any spatial scale is an important part of the broader socio-economic system. In this vein Christopherson (2002) indicates complementaritiesbetween corporate governance regimes and firm networks as well as labour markets.Comparing the United States and the German regimes, she claims, that in the former,strategic alliances among firms are more short-term, and corporate expansion is basedmore on hostile, in contrast to friendly, takeovers (Christopherson, 1999). Allen andGale (2002) suggest that the longer-term employment is more compatible with a closedthan with an open regime of corporate governance. Gertler (2001) lists mergers andacquisitions, the possibility of which depends on corporate governance, as one of the major channels of international convergence in industrial practices, including tech-nology. As an example he refers to research by Leyshon and Pollard (2000) indicatinghow the similarities in corporate governance environment facilitated the transpositionof organizational and technology innovations pioneered by US banks to UK banks.Accordingly, convergence in corporate governance also facilitates convergence in otherparts of the socio-economic system.

2.2. Empirical evidence from Europe

First, we need to distinguish between de jure and de facto convergence. Regarding de jure convergence, Mallin (2004, p. 207) suggests, there does seem to be convergenceon certain common core principles based usually around the OECD Principles of Corporate Governance. Focusing on, Europe Wymeersch (2002) identifies such signs

4 According to Roe (2003) political factors have affected the concentration of US incorporations intoDelaware. Incorporations imply tax revenues and a booming corporate law business, both importantfor this tiny state with little industry or services. The average citizen in Delaware has a stake in keeping

the revenues and in the rules that produce those revenuesbut not much of a stake in what the rulesactually are (2003, p. 594).



on J a n u a r y1 6 ,2 0 1

3




6/22

of de jure international convergence as the spread of similar corporate governancecodes, similarities in forthcoming legislation, and progress in work on the European13th Company Directive. Considering de facto convergence he concludes (Wymeersch,2002, p. 244): whether these developments also mean that more fundamental changeshave occurred, and that the patterns of the business firms have come closer to eachother, remains doubtful.

This article emphasizes de facto convergence, for which we can distinguish severalgroups of studies. The first focuses on the evolution of ownership structures and showsa falling degree of ownership concentration in major European countries (Van der Elst,2000; Wojcik, 2003). The second group treats corporate governance more comprehens-ively using case studies of individual countries. For example, research on corporategovernance in Germany demonstrates change towards some parameters of theAnglo-American model (Hopner, 2001; Vitols, 2003). OSullivan (2003) documents aconsiderable change in corporate governance in Germany and France, with the growinginfluence of the stock market in both countries. The third group of studies uses cross-

country data to demonstrate a high level of diversity and country-specific characterist-ics of corporate governance. These studies are mostly static in character, however,falling short of testing the issue of convergence (Pedersen and Thomsen, 1997;Doremus et al., 1998; Khanna et al., 2002; Doidge et al., 2004).

Studies on the diversity of corporate governance across industries are also inconclus-ive on convergence. At a theoretical level, Becht and Mayer (2001) suggest that sectorswhere investment projects are longer are better served by management that is stable andnot constantly threatened by takeovers. Hansmann and Kraakman (2004) propose thatyoung companies, operating in industries in a rapid process of change, are more likelyto embrace the Anglo-American shareholder model. The latter proposition is based onthe assumption that there is now a consensus on the superiority of the Anglo-Americanmodel. Companies created after this consensus has been reached are likely to follow it,while older companies, established under different models, need to change. Empiricalevidence on the influence of sector on convergence hardly exists, and the few exceptionsfocus on ownership structures or case studies of individual industries in selectedcountries (Palepu and Khanna, 2001; Wojcik, 2003; Clark and Wojcik, 2005b).

2.3. Towards a hypothesis

From the empirical and theoretical literature, I would expect to find a limited degreeof convergence in European corporate governance. Although this follows the findingsin the majority of empirical studies, its conceptual motivation requires some elabora-tion. The political theory of corporate governance supports the likelihood of conver-gence in Europe. The project of the European Economic and Monetary Union, and thenearly completed privatization of state-owned industries, might be viewed as parts of apolitical transformation facilitating a shift to corporate governance focusing on themaximization of shareholder value. Both processes enable European companies tobroaden their capital base on an international basis. A growing section of Europeansocieties also become shareholders requiring more rights, and demanding more trans-parency of corporate governance (FESE, 2002). Whether shareholders own shares dir-ectly or through institutional investors such as mutual funds or pension funds, it is

plausible to anticipate that their role will rise and reinforce the model of shareholderprimacy.

644 Wojcik


on J a n u a r y1 6 ,2 0 1

3




7/22

Such a hypothesis of convergence, does not imply that a linear scale, from Anglo-American minority shareholder friendliness to Continental European minority share-holder unfriendliness, is an ideal basis for studying change in corporate governance. Itdoes, however, offer a practical basis for discerning the directions of change in corpor-ate governance. Such changeprobably also involves mixing, and novel recombina-tions of elements from different systems, which dichotomous corporate governanceratings cannot assess. The objective of the article thus goes beyond simply decidingwhether convergence in European corporate governance is taking place. It addressesthe deficit of empirical evidence on the dynamics of European corporate governance forpatterns of diversity, change and convergence across countries and industries.

3. Data and methodology

The article uses proprietary data on corporate governance ratings of the largestEuropean companies in 2000 and 2004 provided by Deminor Rating SA (hithertoDeminor), a corporate governance rating agency headquartered in Brussels with officesin major European cities. This section describes how the ratings are constructed; pre-sents the sample of companies covered by Deminor and this article; introduces theapproach used to measure convergence in corporate governance ratings; and finallydiscuss the positive, in contrast to a normative, character of the article.

3.1. Corporate governance ratings

The objective of Deminor ratings is to provide information to investors about acompanys corporate governance standards and practices. Selected aggregate resultsof Deminor rating activity are available in the public domain, through publishedreports and website, while details are available on subscription. The main users of Deminor ratings are institutional investors, both European and non-European, whouse them to inform their investment decisions. Deminors customers are mostly insti-tutions that invest money on behalf of millions of individual small shareholders.

Deminor distinguishes four building blocks of corporate governance, referred to ascategories. The first category, shareholders rights and duties, captures the extent towhich shareholders, including small ones, may impact corporate decisions. The secondcategory, takeover defences, assesses if the company implements barriers againstpotential hostile takeovers, thus sheltering the management from the threat of replace-ment. Disclosure measures the availability and quality of information on corporategovernance. The fourth category, board structure and functioning, evaluates suchcriteria as the independence of board members and their remuneration. Each categoryconsists of subcategories, described in the Appendix, based on over 300 criteria.Deminor analysts use exclusively publicly available information, with corporate web-sites, stock exchange announcements and press articles as the main sources. For eachcategory a company scores between 0 and 10 points, and their sum gives the totalcorporate governance score. On the basis of the scores, referred to as notations,Deminor assigns a rating from 1 to 5 for each category. In this article I use notationssince they provide finer detail. Because notations are an intermediate step in the ratingprocess, however, throughout the article I refer to them as ratings.

The ratings are prepared annually, the first in its current format in 2000. This is therating I use in conjunction with the latest rating available, that for 2004. The year of the



on J a n u a r y1 6 ,2 0 1

3




8/22

rating stands for the year of its preparation; thus the ratings used in the article refermostly to corporate information for the first half of 2000 and the first half of 2004.

3.2. The sample of companies

Deminor aims at rating all companies that are constituents of the FTSE Eurotop 300index. The index consists of the largest 300 European companies according to marketcapitalization (stock market price multiplied by the number of outstanding shares). 5 Insome cases, e.g. when a company undergoes a merger or acquisition, Deminor is unableto obtain sufficient information to rate a company. The ratings for 2000 cover 259, andthe ratings for 2004 cover 296 companies. I analyse companies according to the countryof incorporation and industry, using the Industry Classification Benchmark preparedby FTSE in collaboration with Dow Jones. This comprises 10 industries, 18 supersect-ors, 39 sectors and 104 subsectors. Given the size of the sample of companies, in thisarticle I restrict myself to the use of industries.

Many companies in the sample operate in more than one country, including countriesoutside Europe. Although this has a potential impact on corporate governance, thefeatures covered in the ratings focus on central institutions within a company. A mul-tinational company has one set of shareholders irrespective of the number of branches,one board of directors, one set of consolidated financial statements and usually a com-mon disclosure policy. Therefore, assigning a multinational company to its home coun-try in analysing corporate governance should not generally be an oversimplification.

3.3. Measures of convergence

The measures of convergence originate from the literature on economic growth (see a

review in Islam, 2003), though their application has spread beyond economics. Thereare many variations of convergence models, but the basic measures are beta- and sigma-convergence. Beta-convergence measures the relationship between the initial incomelevel and the subsequent growth rate across territories. If there is convergence therelationship (captured usually through regression analysis) should be negative, withterritories with lower income level growing faster. Researchers such as Quah (1993)and Friedman (1994), however, pointed out that a negative beta does not imply areduction in the dispersion of cross-sectional income distribution. This led to theconcept of sigma-convergence, understood as the reduction in the standard deviationof the cross-sectional distribution of income level or growth rate. In this article I useboth concepts of convergence: beta-convergence, by analysing the relationship betweena company rating in 2000 and its absolute change between 2000 and 2004; sigma-convergence, by comparing the absolute and relative standard deviations of ratingsbetween 2000 and 2004. I apply the measures of convergence to the whole samplebut also to individual countries, groups of countries as well as individual industries.

To complete the methodological issues, I want to stress the positive character of the article, which neither claims that corporate governance convergence is good nor

5 Strictly speaking, the index consists of the largest 300 eligible European companies. To be eligible acompany must either have a free float of at least 15% or have a free float above 5% and market

capitalization greater than US $5 billion (US $2.5 billion if it is incorporated in an emerging marketcountry). For details of this and other European FTSE indices see FTSE (2004).

646 Wojcik


on J a n u a r y1 6 ,2 0 1

3




9/22

suggests what good corporate governance is. Throughout the article I intentionally usethe term good corporate governance rating instead of good corporate governance.No corporate governance rating is neutral, and Deminor rating is no exception beingaddressed mainly to institutional investors. My strategy is to make the criteria of Deminors ratings clear and explicit (see Appendix) and use this robust benchmarkaccompanied with a unique dataset to address the issue of convergence. Finally,I should stress that financial institutions, including rating agencies, are a still underusedsource of information in economic geography research (Wrigley et al., 2003; Clark,2004).

4. Corporate governance in 2004

I start the analysis of European corporate governance with the Deminor corporategovernance ratings for 2004. The upper part of Table 1 presents the median values of

the ratings by country, revealing a high degree of diversity.6

Sample companies in theUK and Ireland (also referred to as the British Isles) lead with the median total score of 32.1 and 30.4, respectively, compared with only 19.9 in the rest of Europe (referred to asthe Continent). Apart from the leadership of the British Isles, sharp differences alsoexist within the Continent, with leaders in total ratings including companies from Swe-den, Finland, The Netherlands, France, and Switzerland and laggards including firmsfrom Portugal, Greece, Luxembourg, and Denmark.

We gain more insight into the diversity of corporate governance by examining thefour rating categories. The scores are lowest for takeover defences, highest for disclos-ure, with shareholders rights and duties, and board structure and functioning in themiddle. The British and Irish corporations are leaders in all four categories, but theirdominance over the Continent varies considerably. It is strong in terms of takeoverdefences, where both the British and Irish firms reach a high median rating of 9,while companies from other countries score typically only one point; it is weak in theremaining categories with Finnish and Dutch firms not far behind in terms of disclosureas well as board structure and functioning, and Finnish and Norwegian companies interms of shareholders rights and duties.

Given that the British and the Irish sample companies command systematicallyhigher ratings than firms from the rest of Europe, I conduct the analysis of corporategovernance by industry separately for the British Isles and the Continent. Without thisdivision the distributions of company ratings within each industry are likely to have

wide ranges and similar median values. Potential differences between industries wouldthus be blurred. This method also enables comparison of differences between contin-ental industries with differences in the British Isles. Due to the small size of the sample amore detailed analysis of industries within individual countries is impossible.

6 As the distribution of ratings is not normal, and not symmetric, I used a non-parametric median methodtesting the null hypothesis that a given country or industry has the same median rating as the rest of thesample, i.e. all other countries or industries. This test does not require any assumptions about thedistribution of the variable. Based on calculated Chi-square values, asymptotic significance iscalculated which tells us how often we can expect a Chi-square value at least as large as the calculated

in similar repeated samples if there is no relationship between the medians. Simplifying, asymptoticsignificance of, for example, 0.04 tells us that the chance that the medians are equal is 4 out of 100.



on J a n u a r y1 6 ,2 0 1

3




10/22

The lower part of Table 1 presents median corporate governance ratings by industryfor the British Isles and the Continent. It is difficult to distinguish any industries that

are leaders or laggards in corporate governance. Indeed if we set the significance levelat 1% no single industry median is significantly different from the median of other

Table 1. Corporate governance ratings in 2004 (median values)

Country/industry N TotalShareholders

rights and dutiesTakeoverdefences Disclosure

Board structureand functioning

Austria 2 19.1 6.7 2.6 6.0 3.8Belgium 9 18.7*** 6.1** 1.0** 5.8*** 5.0**Denmark 5 17.8** 6.7 1.0** 6.2 3.6Finland 5 22.6 7.5 1.0 7.0 6.7**France 42 21.3* 6.5*** 1.0 6.9** 6.0Germany 32 19.6* 6.9 1.0*** 6.7** 4.5*Greece 6 17.3** 6.7 1.0** 5.3** 3.7**Republic of Ireland 7 30.4*** 7.8 9.0*** 7.6 7.0*Italy 25 18.6*** 5.7*** 1.0*** 6.6 5.1***Luxembourg 2 17.8 6.2 1.3 5.6 4.8The Netherlands 21 22.6 5.5** 3.8 8.1 6.6**Norway 5 20.3 7.7** 1.0 5.9** 5.2Portugal 4 16.9** 4.9 1.0** 6.6 4.6**Spain 17 19.8** 6.8 1.0*** 6.5** 5.1*Sweden 16 23.8 6.2 5.1 6.8** 5.2**Switzerland 17 21.1 7.0 1.0 5.9*** 5.7United Kingdom 81 32.1*** 8.0*** 9.0*** 8.1*** 7.3***Total 296 22.4 7.0 2.7 7.2 5.8

IndustryThe British IslesOil and gas 3 32.7* 8.0 9.0 8.3* 7.3Basic materials 7 32.6 8.0 9.0 8.4 7.3Industrials 9 31.4* 8.0 9.0 8.0 7.0**Consumer goods 11 31.3 8.0 8.0 8.2 7.4Health care 4 32.4 8.0 9.0 8.0 7.2Consumer services 21 31.7 8.0 9.0 7.9* 7.1Telecommunication 4 33.1 8.0 9.0 8.4 7.5Utilities 6 32.8** 8.0 9.0 8.2 7.5Financials 23 31.8 8.0 9.0 8.0 7.1Total 88 32.0 8.0 9.0 8.1 7.3

IndustryContinentOil and gas 7 22.4 6.1 1.0 7.1 6.1Basic materials 15 22.4 6.4 3.6** 6.6 5.2Industrials 28 22.0** 6.7 3.0** 6.9 5.4Consumer goods 31 19.7 6.8* 1.0 6.5 5.1Health care 12 19.4 6.6 1.0 6.8 5.2Consumer services 19 21.3 6.5 1.0 6.5 6.6*Telecommunication 18 19.9 6.6 1.0 6.8 5.0Utilities 12 19.6 6.2 1.0 7.0 5.0Financials 60 19.2* 6.3* 1.0 6.6 5.1Technology 6 19.6 6.5 1.0 6.7 5.3Total 208 19.9 6.4 1.0 6.7 5.2

Note: Signicance at 10% (*), 5% (**), and 1% (***) levelfor details of the test see footnote 6.Source: Authors based on Deminor and FTSE.

648 Wojcik


on J a n u a r y1 6 ,2 0 1

3




11/22

industries. The medians are also similar in the British Isles and on the Continent. Thisobservation holds for all Deminor categories, with the exception of takeover defences,where utilities lag behind in the British Isles, and basic materials and industrials lead onthe Continent. Differences between countries are thus more pronounced than betweenindustries. Every continental industry has a median score for board structure and func-tioning between 5.0 and 6.6, while for continental countries it varies between 3.6 and6.7. Furthermore, every continental industry has a median rating of disclosure between6.5 and 7.1, while for continental countries it differs from 5.3 to 8.1. Similarly, themedian shareholders rights and duties rating of continental industries falls within thenarrow band of 6.1 to 6.8, while for countries it stretches from 4.9 to 7.7.

To examine the significance of industries further we rank them in the British Isles andthe Continent according to their median rating, comparing the rankings between theContinent and the British Isles. The results show no traces of similarity. In terms of total ratings, for example, industrial companies perform well on the Continent but lagbehind on the British Isles. I have calculated Spearmans rank correlation, also known

as rho correlation, between the industrial ranking in the British Isles and the Continent,and found no significant correlations for any corporate governance category. In con-clusion, there is no sign of an industry effect on corporate governance holding acrossthe North Sea and the English Channel.

The foregoing analysis paints a picture of diversity. If there had been any conver-gence, it has not yet made corporate governance in large European firms similar acrosscountries. The findings confirm the established research on the difference between theAnglo-American model of corporate governance, with examples of the UK and Ireland,and corporate governance prevailing on the Continent (La Porta et al., 1998; Barcaand Becht, 2001). In addition, they show a complex diversity of corporate governancewithin the Continent. While companies from southern European countries seem tocommand lower ratings than those from northern countries, there is no simple geo-graphical pattern. Within Scandinavia, for example, there is a full range of ratings, fromhigh in Sweden and Finland, through average in Norway, to low in Denmark (see alsoWojcik, 2002). Overall, the map of European corporate governance in 2004 reveals ahigh level of diversity driven by country-specific factors, the significance of which isunderscored by the lack of significant differences between industries. In the followingsection I analyse how this map has changed between 2000 and 2004.

5. Change in corporate governance between 2000 and 2004In order to account for change in the composition of the sample between 2000 and2004, it is divided into three groups of companies. The first group (the core) consists of firms included in the sample in 2000 and 2004. The second group (drop-outs) representscompanies included in the sample in 2000 but not in 2004. A company can disappearfrom the sample due to de-listing, a takeover by another company or bankruptcy, andmost likely following its exclusion from the FTSE Eurotop 300 index as a result of arelative decrease in market value. The third group is novices, i.e. companies thatappeared in the sample after 2000, most probably following an increase in relativemarket value. As Table 2 shows, within only 4 years companies in the core improved

their scores considerably, while the median total ratings of drop-outs and novices didnot differ significantly from the ratings of the core in 2000 and 2004, respectively.



on J a n u a r y1 6 ,2 0 1

3




12/22

Table 2. Change in corporate governance ratings between 2000 and 2004 (median values)

Country/industry N TotalShareholders

rights and dutiesTakeoverdefences Disclosure

Board structureand functioning

Core in 2000 190 17.7*** 6.3** 2.0 4.7*** 3.9***Core in 2004 190 23.2 6.7 3.7 7.4 6.2Drop-outs 69 18.9 6.9* 4.0 4.6 3.5Novices 106 20.9* 7.3** 1.0* 6.6*** 5.4**

CountryBelgium 8 2.6 0.2*** 0.5 1.7** 0.5Denmark 3 7.9 0.4 0.0 2.9* 1.8Finland 3 2.7* 0.2 3.2 1.6 2.9*France 30 4.3 0.1*** 1.0 2.3 1.4Germany 22 3.0*** 0.2*** 1.6** 3.1*** 1.7Greece 1 8.7 0.3 9.0 0.8 0.1Republic of Ireland 3 2.2* 0.7 1.0 1.5* 1.0*

Italy 15 3.9 0.6*** 1.0 1.9 1.5The Netherlands 17 10.7*** 1.8** 1.0*** 3.2*** 3.3***Norway 1 6.4 1.9 0.8 2.6 2.7Portugal 3 7.7* 0.7 1.0 4.3* 2.5*Spain 9 6.7 1.2 1.0 3.4** 1.1Sweden 12 3.7 0.8 1.4 3.3** 1.8Switzerland 11 6.0*** 1.1* 0.0 3.5*** 3.4**United Kingdom 52 4.5 1.1*** 1.0*** 1.5*** 1.1***Total 190 4.6 0.6 1.0 2.3 1.5

IndustryThe British IslesOil and gas 3 12.4 1.1* 8.4 1.7 1.2Basic materials 4 7.0 1.2 2.5* 1.2 0.9Industrials 6 4.6 1.2 1.0 1.4 0.8*Consumer goods 8 4.2 1.1 0.5 1.5 1.6Health care 3 5.3 1.9* 1.0 1.7 1.2Consumer services 10 4.0*** 0.6* 1.0 1.6 1.2Telecommunication 3 13.9 2.8 9.0 1.6 0.9Utilities 5 14.2 3.0 9.0 1.7 1.3Financials 13 2.3 0.7 0.4 1.3** 0.5**Total 55 4.4 1.0 1.0 1.5 1.1

IndustryContinentOil and gas 4 4.1 0.5 1.0 2.2 1.1Basic materials 12 4.1 0.5 0.4 2.6 1.6Industrials 13 2.5** 0.2 1.5* 3.2 1.3Consumer goods 21 5.0 0.2 1.0 2.5 1.6Health care 6 5.1 0.2 0.0 3.5 2.6Consumer services 14 6.1* 0.4 1.0* 2.7 2.0Telecommunication 12 6.4* 0.4 1.0*** 2.7 2.0Utilities 8 6.1 0.9 0.0 3.5 1.3Financials 41 3.6 0.0 0.0 3.0 1.8Technology 4 7.3** 0.3 1.0 2.9 2.3Total 135 4.8 0.2 0.0 2.9 1.8

Note: For the four groups of rms at the top of the table the equality of medians was tested for the following pairs: Core 2000and Core 2004; Drop-outs and Core 2000; Novices and Core 2004. The results for countries and industries are for corecompanies only. Signicance at 10% (*), 5% (**), and 1% (***) levelfor details of the test see footnote 6.Source: Authors based on Deminor and FTSE.

650 Wojcik


on J a n u a r y1 6 ,2 0 1

3




13/22

Focusing on the structure of change, median scores for shareholders rights andduties and for takeover defences have not significantly changed, leaving change inthe total score driven by an increase in excess of 50% in the ratings for disclosure aswell as board structure and functioning. The second part of Table 2, presenting themedian absolute change in ratings by country, confirms that ratings for takeoverdefences hardly changed, which reflects probably the high level of ownership concen-tration on the European continent acting as a barrier to hostile takeovers, as well as astalemate of European takeover regulation (Ferrell, 2003). In contrast, shareholdersrights and duties did change in individual countries, but the changes cancel each otherout at the aggregate level. In every country except Greece ratings for board structure andfunctioning increased, and disclosure ratings increased by at least 1 point. The improve-ment in these categories was consistentwith the exception of Greek sample firms, themedian total corporate governance rating increased in every country, with the Dutchand Swiss sample firms in the lead. The level of increase in disclosure and boards ratingson the Continent was nearly double that in the British Isles. Regarding the differential

pace of change by country, these results support the qualitative findings by OSullivan(2003), who claimed that changes towards the parameters of the Anglo-Americancorporate governance were more advanced in France than in Germany.

The lower part of Table 2 reports the median absolute change in ratings by industry.Since the values are quite similar between industries both in the British Isles and on theContinent, it is more difficult to find significant differences between industries thanbetween countries. Nevertheless, if we rank the industries according to the medianabsolute change in ratings for shareholders rights and duties, the ranking we obtainis similar between the British Isles and the Continent, with a Spearmans rho correlationcoefficient of 0.61, significant at 5% level. Both on the Continent and the British Islesindustries that improved most include utilities; those that improved least includefinancials. The interpretation of this pattern requires more detailed research.

To summarize the foregoing findings may be related to the issue of convergence.Although the high levels of diversity suggest necessary scepticism about convergencein European corporate governance, the high degree of recent change suggest that theremay be convergence trends at work. Although in 2004, as Table 1 shows, there was stilla large difference between the British Isles and the Continent, the latter has narrowedthe gap, particularly in terms of disclosure as well as board structure and functioning.The following section explores the issue of convergence in more detail.

6. Convergence in corporate governance between 2000 and 2004This section assesses whether there has been any convergence within the Continent andwithin the British Isles. As a proxy of beta convergence I use Spearmans rank correla-tion that measures the relationship between the starting rating of company corporategovernance in 2000 and its absolute change between 2000 and 2004. 7 The coefficients

7 Considering that ratings are defined over a closed number system, I have performed alternativecalculations of beta convergence. First, I removed from the sample all companies with rating 10 in anysubcategory in 2000. Second, I used square ratings (and differences between squared ratings) instead of

straight ratings. In both cases the results led to conclusions similar to those based on calculations usingstraight ratings for all companies, including those with ratings of 10 in 2000.



on J a n u a r y1 6 ,2 0 1

3




14/22

presented in Table 3 are negative and highly significant for all categories for both theContinent and the British Isles, indicating that companies that performed relativelypoorly in 2000 were catching up. The correlation coefficients are higher for the BritishIsles than for the Continent suggesting stronger convergence within the UK and Irelandthan within the rest of Europe.

Turning to sigma convergence I use the standard deviation of corporate governanceratings as well as the coefficient of variation (the standard deviation expressed as apercentage of the mean), and compare them between 2000 and 2004. The results pre-sented in Table 3 show both similarities and differences between the Continent and theBritish Isles. On the Continent standard deviation in absolute terms decreased foreach category with the exception of board structure and functioning, the coefficientof variation fell for all categories. Ratings for disclosure exhibit the highest level of sigma convergence with standard deviation falling from 29 to 15% of the mean rating.In the British Isles, we see coefficients of variation being halved for each category. Boththe absolute and relative decreases in standard deviation of ratings in the BritishIsles were much higher than on the Continent, confirming a strong convergence inthe British Isles compared with a weak convergence on the Continent.

In the calculations of beta and sigma convergence in this section I use data for core

companies, but the results are similar if we make calculations based on the full sampleof companies for both 2000 and 2004. The last step of the analysis involves sigmaconvergence within countries and industries. Table 4 presents the absolute and relativechange in standard deviation of total corporate governance ratings by country andindustry. The absolute standard deviation of total ratings decreased for sample firmswithin the UK, Germany, France, and Italy; as well as in eight out of nine sectors in theBritish Isles and in 6 out of 10 on the Continent. This is consistent with the weaker levelof convergence found within the Continent in relation to the British Isles. When weexpress standard deviation in relation to mean, however, the proportion of industrieson the Continent with evidence of convergence rises to 8 out of 10. Comparison of the

strength of sigma convergence between industries on the Continent and in the BritishIsles yields mixed results. Telecommunication firms are leaders of convergence on both

Table 3. Corporate governance convergence

Continent The British Isles

Category

Spearmans

rho

SD

2000

% of

mean

SD

2004

% of

mean

Spearmans

rho

SD

2000

% of

mean

SD

2004

% of

mean

Total 0.63 4.70 28 4.26 20 0.80 4.87 19 3.15 10Shareholders rights

and duties0.65 1.41 24 1.11 18 0.84 1.15 18 0.65 8

Takeover defences 0.78 3.49 114 2.62 90 0.79 3.62 55 2.53 32Disclosure 0.58 1.20 29 1.00 15 0.78 0.44 7 0.32 4Board structure

and functioning0.47 1.12 31 1.17 21 0.73 0.60 10 0.45 6

Note: Spearmans rho correlations are calculated between a company rating in 2000 and the change in the rating between2000 and 2004; all correlation coefcients are signicant at 1% level.

Source: Authors based on Deminor.

652 Wojcik


on J a n u a r y1 6 ,2 0 1

3




15/22

the Continent and the British Isles; on the other hand, health care companies are leaderson the Continent but laggards on the British Isles.

To summarize, there is significant evidence of beta and sigma convergence withinboth the British Isles and the Continent. Convergence within the British Isles is strong,while within the Continent it is weak, although it occurred within all major countries.The pace of convergence differs between different aspects of corporate governance, andis highest with regard to disclosure. Industries show no clear pattern of differences

either in terms of the absolute level of corporate governance ratings or in terms of their change.

Table 4. Convergence of total ratings by country and industry

1 2 3 4 Change in SD

Country/Industry SD 2000 % of mean SD 2004 % of mean Absolute (31) Relative (42)

Belgium 1.50 10 2.43 13 0.93 4Denmark 3.59 29 1.29 7 2.30 21Finland 4.44 19 3.69 14 0.76 5France 4.54 25 3.71 16 0.83 9Germany 4.40 24 3.02 15 1.38 10Republic of Ireland 0.83 3 0.77 2 0.06 0Italy 3.79 22 2.65 14 1.14 8The Netherlands 3.37 25 3.98 17 0.60 9Portugal 1.02 12 0.46 3 0.56 9Spain 1.55 11 3.60 17 2.05 6Sweden 5.13 28 4.41 19 0.72 8Switzerland 5.74 39 6.62 29 0.88 10United Kingdom 4.88 19 3.23 10 1.65 9Total 6.21 33 5.85 24 0.36 8

IndustryThe British IslesOil and gas 5.58 25 0.55 2 5.03 24Basic materials 2.74 10 1.92 6 0.81 4Industrials 5.57 23 4.98 17 0.59 6Consumer goods 5.41 22 1.14 4 4.27 18Health care 0.07 0 0.45 1 0.38 1Consumer services 4.75 19 4.73 17 0.02 2Telecommunication 7.66 34 1.12 3 6.54 31Utilities 5.89 27 0.47 1 5.42 25Financials 2.53 9 0.97 3 1.56 6

IndustryContinentOil and gas 2.61 16 4.08 18 1.47 2Basic materials 4.78 26 4.52 21 0.26 5Industrials 3.50 16 3.74 15 0.24 1Consumer goods 4.60 31 4.02 19 0.58 11Health care 4.07 27 1.06 5 3.01 21Consumer services 5.03 31 4.54 20 0.49 12Telecommunication 5.37 36 3.46 17 1.91 20Utilities 4.11 29 3.45 17 0.65 12Financials 4.19 25 4.86 23 0.67 2Technology 1.36 11 3.38 16 2.02 6

Source: Authors based on Deminor and FTSE.



on J a n u a r y1 6 ,2 0 1

3




16/22

7. Conclusions and implications

The objective of the article was to assess convergence in European corporate govern-ance, addressing the deficit of empirical evidence on the dynamics of corporate change.To meet this objective I used proprietary data on corporate governance ratings of nearly 300 of the largest European companies in 2000 and 2004. The data providedby Deminor Rating SA offer an insight into the structure of corporate governance, withits building blocks comprising shareholders rights and duties, takeover defences, dis-closure, as well as board structure and functioning. The analytical steps undertaken onthe way to measure convergence in ratings focused on the state of corporate governancein 2004 and its change between 2000 and 2004 across countries, groups of countries andindustries.

The first conclusion is that convergence in European corporate governance is hap-pening. Between 2000 and 2004 almost all sample companies improved their ratings,but the rate of change has been uneven both across countries and across the building

blocks of corporate governance. Most important, the continental European companiesimproved their scores more than their counterparts in the British Isles, with the Dutch,the French, and the Swiss corporations as leaders of change on the Continent. Thestructure of change was different between the British Isles and the Continent. Whilein the British Isles companies improved their ratings for shareholders rights and dutiessignificantly, on the Continent the latter did not improve consistently, and ratings fortakeover defences did not change at all. The categories that improved in a spectacularway, and in all countries, were board structure and functioning and particularly dis-closure. In addition to continental European companies making a step towards closingthe gap between them and the Anglo-American corporate governance, there is evidenceon convergence within the Continent and within the British Isles. In the UK and Irelandcompanies with lower ratings in 2000 were catching up quickly, particularly in terms of disclosure. On the Continent disclosure also exhibited the highest degree of conver-gence, although the overall pace of convergence was much slower than in the BritishIsles. Nevertheless, evidence on convergence is found within every major Europeancountry.

The second major conclusion is that countries do matter, as reflected in a highlydiverse map of European corporate governance in 2004. The British and the Irishcompanies lead with high ratings in almost every category of corporate governance.This finding is consistent with the existing research on the divide between the Anglo-American and the continental European corporate governance. Nevertheless, the article

demonstrates that the diversity among companies within continental Europe is pro-nounced, including systematic differences between countries. In contrast, the diversityof corporate governance ratings across industries is strikingly small. It is possible that aclassification of only 10 sectors is too shallow to capture differences, and any classi-fication is imprecise for large companies with activities covering various industries. It isalso possible that sectors differ in their influence on corporate governance from countryto country. Notwithstanding the methodological issues, the evidence suggests thatcountry-specific factors affecting corporate governance and its evolution overwhelmsector-specific factors.

The key finding of the article is the evidence for convergence. The analysis of factorsdriving this, from political, product market and capital market integration to share-holder activism, is beyond the scope of the article (see Stulz, 1999; Clark, 2003).

654 Wojcik


on J a n u a r y1 6 ,2 0 1

3




17/22

Nevertheless, it is worth considering the role of corporate governance ratings in relationto convergence. Their existence reflects demand for standards. Rating agencies do notput direct pressure on rated companies to change corporate governance practices, buttheir customers and particularly institutional investors do. Indeed, there is evidence of institutional investors exercising such power. Hebb and Wojcik (2005) show how eco-nomic governance in emerging markets changes in response to the requirements of institutional investors, demonstrating in particular that country governments are a cru-cial link in the chain of institutional investors influence on companies. In addition,Hebb (2004) shows how institutional investors make firms adapt higher transparencystandards. These findings are consistent with the evidence of this article showing thesignificance of countries, and highlighting disclosure as the fastest progressing categoryof corporate governance in Europe.

While this article focuses on the largest European corporations, there are three waysin which the results may be indicative of broader processes of change in Europeancorporate governance. The first affects the evolution of companies. As smaller compan-

ies grow and consider improving their access to public capital markets and interna-tional investors, they are likely to model their corporate governance on the largestcompanies. In other words, as they grow, firms can reasonably expect to find them-selves affected by forces that already affect larger companies. Second, large corpora-tions play important roles in networks of firms, including smaller firms as theirsuppliers, customers or subcontractors. If the stock market punishes closed and rewardsopen corporate governance practices, large companies may seek to avoid associationswith small unlisted companies whose corporate governance is far from open (Clark andWojcik, 2005a). And vice versa, small companies, with ambitions to become listed, mayavoid associations with large companies with closed corporate governance practices.Third, we can think of the role of law, accountancy and audit firms providing compan-ies with services related to corporate governance. It is unlikely that these service pro-viders apply entirely different standards to large as opposed to smaller firms. Auditrequirements apply to tens of thousands of European firms, not only the publicly tradedones, and the Big Four companies increasingly dominate the market for audit services.

These links between the largest companies and wider trends in corporate governanceopen opportunities for further research. One possible direction would be to analyse theenvironment of selected large corporations that, according to ratings such as Demi-nors, radically transformed their corporate governance. Such analysis could showwhether and how corporate governance change in the large company trickles downto the governance of its suppliers, subcontractors and customer firms. In addition, if conducted on a comparative, cross-country basis, such research could show whethercorporate governance changes that seem similar on the surface (and in terms of ratings)have the same substance and meaning.

One implication of this article, and simultaneously a challenge facing economic geo-graphy, is the importance of financial processes in the study of corporate transforma-tion. Let me use, simply for the sake of illustration, the framework of the bargainingrelationship between trans-national companies and host countries, applied to concep-tualize the dynamic territorial structure of firms (Dicken, 1999, p. 274). This needs toaccount for the power relations within the firm, and in particular the bargaining powerof shareholders. This adds complexity, since governments can represent shareholders,

while shareholders invest mostly through institutional investors, themselves large andoften trans-national corporations bargaining with the governments. Nevertheless, this



on J a n u a r y1 6 ,2 0 1

3




18/22

complexity is increasingly necessary as the power of shareholders and institutionalinvestors rises (Clark et al., 2005).

The significance of corporate governance for the understanding of corporations, andthe rise of shareholder friendly corporate governance in Europe have importantimplications, not only for the theory, but also for the practice of economic geography.In researching firms, and actors associated with firms, economic geographers have tra-ditionally focused on top managers and employees. Corporate governance research,developed mostly at the intersection of law and economics, stresses the relationshipbetween owners and managers without privileging one or the other as the subject of research. This is not to say that economic geographers researching the strategy of aparticular company should interview shareholders instead of interviewing managers.We need nevertheless to acknowledge the growing impact of shareholders on the devel-opment of particular companies and the whole corporate world. After all we call oureconomy capitalist not managerialist.

Acknowledgements

Jean-Nicolas Caprasse and Kristof Ho Tiu, Deminor Rating SA, provided data and invaluableguidance on corporate governance ratings. Rob Bauer, ABP Investments and Maastricht Uni-versity, and Nadja Gunster, Erasmus University Rotterdam made me aware of the data andprovided valuable feedback on my ideas. Shervin Setareh, Deminor Rating SA, offered perceptiveremarks on the draft. I have also benefited from the intellectual support of Gordon L. Clark andcomments from Tessa Hebb, both from University of Oxford. I would like to acknowledge thefinancial support of Jesus College and funding from the Social Sciences and Humanities ResearchCouncil of Canada within the project on Social Investment of Pension Funds University-UnionResearch Alliance, managed by Gordon L. Clark and Tessa Hebb. I would also like to thank

Peter Wood at the University College London for his thorough comments, as well as theanonymous referees. None of the above is responsible for any mistakes or omissions.

References

Allen, F., Gale, D. (2002) A comparative theory of corporate governance. Working Paper 03-27.Financial Institutions Center, Wharton School, University of Pennsylvania.

Barca, F., Becht, M. (eds) (2001) The Control of Corporate Europe . Oxford: Oxford UniversityPress.

Barnes, T. J., Sheppard, E. S. (2000) A Companion to Economic Geography . Oxford: Blackwell.Bebchuk, L., Roe, M. (2000) A theory of path dependence in corporate governance and

ownership. Stanford Law Review , 52: 121170.Becht, M., Mayer, C. (2001) The control of corporate Europe. In F. Barca and M. Becht (eds)The Control of Corporate Europe . Oxford: Oxford University Press, 145.

Christopherson, S. (1999) Rules as resources: how market governance regimes influence firmnetworks. In T. J. Barnes and M. S. Gertler (eds) The New Industrial Geography: Regions,Regulations, and Institutions . London: Routledge, 155175.

Christopherson, S. (2002) Why do national labor market practices continue to diverge in theglobal economy? The missing link of investment rules. Economic Geography , 78: 121.

Clark, G. L. (2003) European Pensions and Global Finance . Oxford: Oxford University Press.Clark, G. L. (2004) Money flows like mercury: the geography of global finance. Working Paper

01-04. School of Geography and the Environment, University of Oxford.Clark, G. L., Wojcik, D. (2003) An economic geography of global finance: ownership con-

centration and stock-price volatility in German firms and regions. Annals of the Associationof American Geographers , 93: 909924.

656 Wojcik


on J a n u a r y1 6 ,2 0 1

3




19/22

Clark, G. L., Wojcik, D. (2005a) Financial valuation of the German model: the negativerelationship between ownership concentration and stock market returns 19972001. EconomicGeography , 81: 1130.

Clark, G. L., Wojcik, D. (2005b) Path dependence and the alchemy of finance: the economicgeography of the German model. Environment and Planning A , 37: 17691791.

Clark, G. L., Gertler, M. S., Feldman, M. (eds) (2000) The Oxford Handbook of EconomicGeography . Oxford: Oxford University Press.Clark, G. L., Hebb, T., Wojcik, D. (2005) Institutional investors and the language of finance: the

global metrics of market performance. WPG 05-01. Oxford: School of Geography and theEnvironment, University of Oxford.

Deminor Rating (2003) Corporate governance research and rating service. www.deminor-rating.com.

Dicken, P. (1999) Global Shift: Transforming the World Economy . London: Paul ChapmanPublishing.

Dicken, P., Malmberg, A. (2001) Firms in territories: a relational perspective. EconomicGeography , 77: 345363.

Doidge, C. G., Karolyi, G. A., Stulz, R. M. (2004) Why do countries matter so much for

corporate governance? www.ssrn.com.Doremus, P. N., Keller, W. M., Pauly L. W., Reich, S. (1998) The Myth of the Global Corporation . Princeton: Princeton University Press.

Errunza, V., Losq, E. (1985) International asset pricing under mild segmentation: theory and test.Journal of Finance , 40: 105124.

European Commission (EC) (2003) Progress on Financial Services: Ninth Report. Brussels25.11.2003. http://europa.eu.int/comm/internal_market/en/finances/actionplan/.

Ferrell, A. (2003) Why continental European takeover law matters. Discussion Paper 454.Harvard Law School, Cambridge, MA.

FESE (2002) Share ownership structure in Europe 2002. Brussels: Federation of EuropeanSecurities Exchanges. www.fese.be.

Friedman, M. (1994) Do old fallacies ever die? Journal of Economic Literature , 30: 21292132.

FTSE (2004) Ground rules for the management of the FTSE Eurotop 100, FTSE Euro 100, FTSEEurotop 300, FTSE EuroMid Indices. www.ftse.com.Gordon, J. N. (2003) An international relations perspective on the convergence of corpor

journal of economic geography volume 6 issue 5 2006 [doi 10.1093_jeg_lbl003] wojcik, d. --...

Documents