CEO Bonus: Alternative Performance Versus Gamesmanship

Crocker Liu and Desmond Tsang*

March 2012 (Preliminary Draft)

Abstract CEO bonus plans typically use net income as the performance measure in determining bonus. However, directors often get pressure from CEOs to adopt some alternative non-GAAP performance measures in bonus calculation. In this study, we examine the relationship between CEO bonus and Funds from Operations (FFO), an industry-specific non-GAAP alternative performance measure that resembles net income but excludes depreciation and certain nonrecurring, unusual items. Under the unique setting of Real Estate Investment Trusts (REITs), CEO bonus is explicitly determined based on FFO instead of net income. We examine whether REIT CEOs manipulate components of FFO to increase compensation and our results show that CEO bonus is significantly associated with manipulation of the specific non-GAAP FFO exclusions but not the GAAP component of FFO. We further show that industry and regulatory standards as pertains to FFO and bonus disclosures are both effective in reducing the impact of FFO manipulation on CEO bonus, while internal governance mechanisms do not seem to constrain the manipulation of FFO. Overall, our findings imply incentive compensation based on non-GAAP performance measures induces managerial opportunistic behavior where managers classify more discretionary items as non-GAAP exclusions. The imposition of specific industry and regulatory standards serves as important instruments to ensure the proper use of non-GAAP measures in compensation contract design.

Keywords: Performance Measurement, CEO Bonus Compensation, Non-GAAP Reporting, Earnings Management, Industry Guidance, Disclosure Regulation

*Crocker Liu, Cornell, School of Hotel Administration, 440 Statler, Ithaca, NY 14853, Tel : (607) 255-3739, chl62@cornell.edu and Desmond Tsang, McGill University, Bronfman Building, Rm 331, 1001 Sherbrooke Street West, Montreal, Quebec, H3A 1G5, Tel: (514) 398-5417, desmond.tsang@mcgill.ca. We acknowledge financial support from Cornell University and McGill University. We thank Sam Bakhshalian, Carrie Cheung, Nick Karali and Jing Zhang for their able research assistance.

2

Introduction

In recent years, the reporting of alternative performance measures has become

very common practices (Bradshaw and Sloan, 2002). These measures are often labeled as

“pro forma” or “street” earnings, and they are usually non-GAAP (non-Generally

Accepted Accounting Principles), unaudited performance measures provided by

management. Managers typically calculate these alternative performance measures by

excluding from GAAP net income some unusual or unexpected nonrecurring items (e.g.,

restructuring charges, extraordinary items) and they argue these measures contain useful

information that better reflects the persistent and core component of earnings. Though

compensation contracts are not usually tied to these alternative performance measures,

directors often get pressure from CEOs to consider these measures in bonus calculation.

As a result, some companies have allowed CEOs to exclude several unusual one-time

losses in setting their bonus targets in compensation contracts (Dechow, Huson and

Sloan, 1994; Gaver and Gaver, 1998). Lately, the practice of tying bonuses with these pro

forma, adjusted financial metrics has become the subject of public scrutiny due to the

alleged fraud trials on several high-profile companies such as Nortel Network Corp.1

While agency cost theory dictates performance-based compensation contracts can

effectively align interests of managers and shareholders (e.g., Smith and Watts, 1982;

Jensen and Murphy, 1990), prior finance and accounting literature also shows that

managers have more incentives to engage in earnings management activities to

manipulate financial information when CEO compensation is linked to firm performance

(e.g., Healy, 1985; Holthausen, Larcker and Sloan, 1995; Gaver, Gaver and Austin, 1995;

1 Nortel Network Corp. distributed huge bonuses to its top executives while the company was reporting net losses under GAAP. The bonuses were triggered when the company achieved its pro forma income targets (Sturgeon, 2012).

3

Cheng and Warfield, 2005; Bergstresser and Philippon, 2006). Given that non-GAAP

alternative performance measures are unaudited and are reported voluntarily by

management, studies have shown that the reporting of these alternative performance

measures can be driven by opportunistic reasons (e.g., Doyle, Lundholm and Soliman,

2003). Two recent studies (Isidro and Marques, 2010; Black, Black, Christensen and

Waegelein, 2011) have shown the design of compensation contracts has significant

influences on the reporting decisions of alternative performance measures, even when

compensation contracts are not explicitly linked to these measures, as managers want to

affect investors’ perception of firm performance through these measures. Hence, it is

highly plausible the adoption of non-GAAP alternative performance measures for bonus

purposes in companies such as Nortel Network Corp. could further increase the risks of

firms manipulating these measures.

In this study, we investigate earnings management behavior of firms under a

unique setting where CEO bonus is explicitly and predominantly determined based on a

non-GAAP alternative performance measure. The REIT industry is characterized by the

common use of an industry-specific alternative performance measure known as Funds

from Operations (FFO). Based on the recommended definition by the National

Association of Real Estate Investment Trusts (NAREIT), the calculation of FFO excludes

(from net income) depreciation and amortization expenses related to real estate

properties, their associated disposal gains and losses, and items that are unusual and/or

nonrecurring in nature, namely extraordinary items, impairment write-downs of

depreciable real estate properties, discontinued operations, and cumulative effects of

4

accounting changes.2 Although FFO is unaudited and is not prepared according to

GAAP, which in turn makes it more susceptible to manipulation, FFO is recognized as

the industry-wide standard performance measure for REITs (Sloan, 1998), and CEO

bonus is predominantly determined by FFO according to the annual compensation survey

issued by NAREIT. Prior academic evidence (e.g., Pennathur and Shelor, 2002;

Pennathur, Gilley and Shelor, 2005) also shows CEO compensation is positively related

to FFO.

The first objective of our study is to explore whether using FFO, a non-GAAP

alternative performance measure, in bonus contract design gives managers incentives to

alter this performance metric to maximize their CEO payoffs, and more importantly, how

managers would manipulate the measure should they decide to overstate it for

compensation purposes. When CEO compensation is tied to net income, managers who

want to manipulate earnings for bonus purposes have no other option but to manage the

GAAP-governed net income. However, as FFO comprises of net income and the

exclusion items, managers who want to manipulate FFO can exercise discretion in either

net income (i.e., the GAAP component) or the exclusion items (i.e., the non-GAAP

component), or both. On one hand, as the exclusion items of FFO is unaudited and non-

GAAP, managers may be more tempted to exercise their discretion in this component of

FFO, and they can do so by making ad-hoc, discretionary exclusions from net income to

calculate FFO. On the other hand, when firms make adjustments that are not in the

recommended NAREIT FFO definition, these ad-hoc exclusions are quite transparent. It

is also doubtful how much discretionary expenses a firm can exclude for managers to

achieve the intended earnings management effect. Hence, managers may still need to 2 See “White Paper on Funds from Operations” published by NAREIT in 2002.

5

manipulate net income (i.e., the GAAP component of FFO) for bonus compensation

purposes.

Utilizing a sample of 188 firm-year observations over the period of 2006-2009,

we test the relationship of CEO bonus compensation and FFO manipulation by regressing

CEO bonus on proxies of earnings management in the GAAP and non-GAAP

components of FFO. We measure manipulation in the GAAP component using the level

of discretionary accruals in the common component of FFO and net income, and

manipulation in the non-GAAP component using the ad-hoc exclusions from FFO,

captured by the deviations of actual FFO from the recommended FFO definition of

NAREIT. We show that when FFO is adopted as the performance benchmark of bonus

contracts, CEO bonus is significantly associated with the non-GAAP exclusions of

expense items from FFO but not with discretionary accruals in FFO. Our results imply

that when CEO bonus is tied to a non-GAAP measure, managers are more likely to

manipulate the non-GAAP component rather than the GAAP component for bonus

purposes.

If CEO bonus that is contingent on FFO performance prompts managers to

manipulate FFO, we wonder if there exists any effective regulatory force to govern such

self-serving managerial actions. We next analyze how industry and regulatory standards

as pertains to FFO and bonus disclosures might affect the relationship of CEO bonus and

FFO manipulation. In particular, we examine three standards: one voluntary and one

regulatory that govern the reporting of FFO, and one regulatory that applies to

compensation disclosures. First, NAREIT has made several attempts over the years to

improve the uniformity and transparency of FFO by issuing guidance on defining FFO.

6

For instance, NAREIT has presented a “recommended” definition of FFO in 2000 and

has repeatedly advocated its member firms to voluntarily follow its guidelines when

presenting FFO in their financial statements (NAREIT, 1999; 2002). We posit managers

that choose to disclose the adoption of the NAREIT definition of FFO are less likely to

manipulate FFO to boost CEO bonuses. Second, to curtail misreporting of non-GAAP

information, the Security and Exchange Commission (SEC) adopted Regulation G in

2003. The new regulation requires firms reporting non-GAAP information provide

reconciliation of their non-GAAP measure with the most directly comparable GAAP

financial measure. Yet, our sample shows not all firms have compiled with the SEC

requirements. We posit that REIT firms that adhere strictly to Regulation G are

committed to greater transparency in FFO reporting and managers of these firms are less

likely to manipulate FFO to increase CEO bonus. Finally, the SEC mandated new rules

on compensation disclosures in 2006 that require a new Compensation Discussion and

Analysis section in the proxy statements filed with the SEC. One particular aspect with

the expanded disclosure requirements is the explicit discussion of specific performance

targets used to determine compensation payout. We conjecture the noncompliance of the

new compensation disclosures via the nondisclosure of performance targets is associated

with firms whose managers are more likely to manipulate FFO for bonus purposes.

We hand-collect data from the proxy statements and 10-K filings of our sample

firms to construct our proxies of regulatory disclosure, and we empirically test the

interrelationships of CEO bonus compensation, FFO manipulation and regulatory forces.

Our empirical findings show the impact of FFO manipulation on CEO bonus is smaller

for firms with (i) explicit disclosures of the use of NAREIT definition of FFO, (ii) better

7

compliance to Regulation G by providing a reconciliation schedule of FFO and net

income, and (iii) better adherence to the SEC mandated compensation disclosures by

disclosing specific performance targets that determine CEO bonuses.

Lastly, we also evaluate the impact of good internal corporate governance on the

relationship of CEO bonus and FFO manipulation, but we find that general corporate

governance mechanisms do not seem to constrain the manipulation of FFO for bonus

purpose. Overall, our findings imply that incentive compensation based on alternative

performance measures induces managerial opportunistic behavior where managers

classify more discretionary items as non-GAAP exclusions. The imposition of specific

industry and regulatory standards serves as important instruments to ensure the proper

use of non-GAAP measures in compensation contract design.

Our study contributes to the broad literature on non-GAAP reporting by providing

novel evidence on the association of CEO compensation and non-GAAP alternative

performance measures. Prior research generally supports the view that non-GAAP

disclosures contain useful information and are valued by investors (e.g., Bradshaw and

Sloan, 2002; Brown and Sivakumar, 2003; Bhattacharya, Black, Christensen and Larson,

2003; Gu and Chen, 2004; Bowen, Davis and Matsumoto, 2005; Bhattacharya, Black,

Christensen and Mergenthaler, 2007). However, other studies also provide ample

evidence showing the opportunistic reporting of non-GAAP performance measures (e.g.,

Doyle et al., 2003; Johnson and Schwartz, 2005; Abarbanell and Lehavy, 2007; Cohen,

Hann and Ogneva, 2007; Black and Christensen, 2009). Our paper is one of the few

studies to examine the opportunistic reporting of non-GAAP performance measures

driven by compensation concerns. Our findings have important implications to

8

compensation contract design, especially given that board of directors and compensation

committee members are increasingly open to the idea of using an “adjusted” net income

measure in setting performance targets for bonuses (e.g., Dechow et al., 1994; Gaver and

Gaver, 1998). Yet, there has been little research so far examining the relationship of CEO

bonuses and non-GAAP reporting. Two recent studies look into how compensation

concerns may affect the reporting decision of pro forma earnings (Isidro and Marques,

2010) and motivate aggressive calculation of pro forma earnings (Black et al., 2011).

Unfortunately, a severe data limitation on the pro forma studies is that it is difficult to

ascertain if these firms are in fact using the alternative performance measure to set the

bonus benchmark in compensation contracts. Also, there is little consensus on how firms

would define pro forma earnings. Our study utilizes the unique environment in the REIT

industry where firms uniformly adopt FFO in performance reporting and in bonus

contract design. Thus, it offers us with an experimental setting to provide insights on how

compensation can affect non-GAAP reporting when CEO bonuses are directly linked to

the alternative non-GAAP performance measure.

Second, we extend the literature that examines the association of CEO

compensation and earnings management. Prior research has found evidence suggesting

the likelihood of managers manipulating earnings can be related to different components

of CEO compensation including bonuses (e.g., Healy, 1985; Gaver et al., 1995;

Holthausen et al., 1995; Balsam, 1998) or equity incentives (e.g., Cheng and Warfield,

2005; Bergstresser and Philippon, 2006; Burns and Kedia, 2006). Our findings provide

evidence that CEO bonuses can encourage managers to manipulate earnings via an

alternative performance measure. More importantly, we show that compensation tied to

9

the non-GAAP measure can provoke managerial manipulative behavior through an

unconventional form of classification shifting. While accrual and real earnings

management has been the subject of considerable academic research, evidence on

earnings management with classification shifting is relatively scarce. Prior research (e.g.,

Beattie, Brown, Ewers, John, Manson, Thomas and Turner, 1994; Godfrey and Jones,

1999; Athanasakou, Strong and Walker, 2007) shows firms exercise discretions in

defining extraordinary items to achieve classification shifting for income smoothing.

McVay (2006) shows firms may engage in earnings management using classification

shifting of special items such that their “core” (pro forma) earnings can meet analyst

forecast. Fan, Barua, Cready and Thomas (2010) show classification shifting of special

items happens more often in the fourth quarter when managers are less able to manipulate

accruals. We show that, instead of manipulating discretionary accruals, managers who

have their CEO bonuses tied to the non-GAAP measure are more likely to engage in

earnings management by classifying expenses as unusual and nonrecurring that warrant

exclusions from the non-GAAP measure.

Third, we contribute to the literature on disclosure regulation. Healy and Palepu

(2001) note literature on disclosure regulation is relatively limited. Our study adds to the

limited literature on disclosure regulation related to both non-GAAP financial

information (e.g., Marques, 2006; Heflin and Hsu, 2008; Kolev, Marquardt and McVay,

2008; Fortin, Liu and Tsang, 2011) and compensation disclosure (e.g., Vefeas and

Afexentiou, 1998; Ke, Petroni and Safieddine, 1999; Lo, 2003; Craighead, Magnan and

Thorne, 2004; Robinson, Xue and Yu, 2009). Consequently, our research also responds

to Leuz and Wysocki (2008)’s call for further studies that evaluate the complementary

10

nature among different disclosure regulations. In particular, our study utilizes an ideal

setting that allows us to evaluate the importance of three alternative disclosure rules that

are symptomatically different from each other. While both the NAREIT guidance and

Regulation G govern the reporting of FFO, the NAREIT guidance represents voluntary

self-regulation and Regulation G is enforceable by the SEC. On the other hand, while

both Regulation G and the new compensation disclosure requirements are imposed by the

SEC, Regulation G oversees performance measurement and the compensation disclosure

requirements govern executive compensation. Hence, it is possible these regulatory

forces exert differential impact on REIT managers.

Finally, we add to the literature that examines the association of corporate

governance and opportunistic financial reporting. While there is a long literature that

provides evidence that better corporate governance constrains earnings management (e.g.,

Klein, 2002; Xie, Davidson and DaDalt, 2003; Bowen, Rajgopal and Venketachalam,

2008), research on corporate governance and non-GAAP reporting is relatively limited.

Two recent studies (Jennings and Murphy, 2011; Frankel, McVay and Soliman, 2011)

show that the association of corporate governance and opportunistic non-GAAP reporting

has been reduced following the imposition of Regulation G. Our study contributes to the

literature by showing that regulatory actions have also undermined the role of corporate

governance in ensuring effective compensation contracting.

The remainder of the paper is organized as follows: The next section presents the

institutional background of our research setting. The third section develops the hypothesis

and outlines the research design. The fourth section outlines the sample selection process

11

and presents descriptive statistics. We present our empirical results in the fifth section.

Finally, we offer some concluding remarks in the last section.

Institutional Background

The REIT industry has long argued that GAAP net income does not accurately

reflect firm performance due to the mandatory inclusion of accounting depreciation3 and

several nonrecurring items that provide little information in evaluating REIT

performance. Hence, NAREIT introduced the concept of FFO in 1991 as an alternative

non-GAAP performance measure to supplement net income in measuring firm

profitability. Since then, FFO has become the industry-wide standard measure reported

by almost all equity REITs. Prior research (e.g., Fields, Rangan and Thiagarajan, 1998;

Vincent, 1999; Graham and Knight, 2000; Stunda and Typpo, 2004; Hayunga and

Stephens, 2009) generally shows that both net income and FFO provide useful

information and are valued by market participants. However, due to the voluntary and

non-GAAP nature of FFO, there have been continued concerns that REIT managers use

FFO to mislead investors. Responding to the controversies, NAREIT has published

various guidelines on defining FFO in 1995 and 1999. Finally, in 2002, NAREIT issued a

“White Paper on Funds from Operations” reinforcing a recommended FFO definition for

its member firms. Baik, Billing and Morton (2008) show that these increased industry

efforts at self-regulation have reduced managerial discretion, and have increased the

uniformity and improved the transparency of FFO reporting. Yet, anecdotal evidence

3 For instance, Ben-Shahar, Sulganik and Tsang (2011) show accounting depreciation reported by REITs suffer from huge measurement errors, thereby distorting the value relevance of REIT net income. Kang and Zhao (2010) show REIT accounting depreciation deviates from economic depreciation to a greater extent than other industries.

12

shows that REIT managers’ compliance to the NAREIT-recommended FFO was far from

perfect (Romanek, 2003).

Concerned that companies provide non-GAAP performance measures to mislead

investors, the SEC adopted Regulation G in 2003 to govern non-GAAP reporting. The

regulation requires firms that report non-GAAP information to disclose the most directly

comparable GAAP financial measure with a reconciliation schedule of the non-GAAP

measure to this GAAP measure. The regulation also requires firms to explain why

management believes the non-GAAP information is beneficial to investors. Research that

examines the effect of Regulation G on non-GAAP reporting largely focuses on pro

forma earnings. These findings generally indicate a decreased likelihood of firms

reporting pro forma earnings (e.g., Marques, 2006) but an overall improvement in the

reporting quality of pro forma earnings (e.g., Heflin and Hsu, 2008; Kolev et al., 2008).

In the context of the REIT industry, Fortin et al. (2011) examine the impact of Regulation

G on FFO and show a uniform improvement in the quality of FFO disclosures subsequent

to the enactment of the regulation.

In 2006, the SEC had a substantial overhaul of executive compensation disclosure

regulation to improve the quality and quantity of executive compensation information

presented by management. The amended regulation that was set forth in Item 402 of

Regulation S-K requires disclosures in five categories: (a) option disclosures, (b)

Compensation Discussion and Analysis (CD&A), (c) a summary compensation table, (d)

exercises and holdings of previously awarded equity interests, and (e) post-employment

compensation. Companies are required to disclose specific quantitative or qualitative

performance-related targets in the CD&A unless such disclosure involves confidential

13

information and disclosing the information will have an adverse effect on the company.

However, if the company uses targets that it does not disclose, the company must provide

detailed discussions on how difficult or likely it will be for the company to achieve the

undisclosed targets. If the company decides to use performance targets based on a non-

GAAP financial measure, the company must also disclose how the measures are derived

the audited financial statements. A recent study by Robinson et al. (2009) shows non-

compliance with the new compensation disclosure requirement is associated with

excessive CEO compensation and higher media criticisms of CEO compensation.

Hypothesis Development & Research Design

Hypothesis Development

Healy (1985) shows that management bonus tied to an accounting number can

create incentives for manipulation of the accounting figure. Research has since shown net

income is subject to more severe manipulation when it is used as the target or benchmark

to determine bonus (e.g., Holthausen et al., 1995). In the REIT industry, bonus is directly

and explicitly tied to FFO. Given that FFO is a non-GAAP alternative performance

measure not subject to GAAP and audit, it can be susceptible to more management

discretion than the GAAP-governed net income measure. Hence, it is possible that FFO

can be used opportunistically to enhance firm performance such that firms can increase

performance-based compensation.

If REIT managers do manipulate FFO for compensation purposes, a more

interesting question is how they would achieve their goals. Unlike the net income

measure, managers who intend to manipulate a non-GAAP alternative performance

measure have the option of exercising discretion in the calculation of the non-GAAP

14

component, the GAAP component, or both. As the recommended definition of FFO by

NAREIT remains voluntary in nature, firms can choose not to follow the standard

definition and report an ad-hoc, firm-defined FFO by excluding some additional items

that managers deem appropriate. In fact, NAREIT has specifically stated due to the

diversity of practices and nature of nonrecurring items, some items (e.g., debt

restructuring expenses, straight-line rent expense) are not considered as recommended

exclusions but firms would have the discretion to exclude these items from FFO provided

that managers have good justification. However, such flexibility also increases the

potential of managers manipulating FFO through the selective inclusion or exclusion of

items in their firm-defined FFO measures. As a result, we expect REIT firms who have

their CEO bonuses linked to FFO would manipulate the non-GAAP component of FFO

to increase bonus compensation. Alternatively, the required provision of reconciliation

under Regulation G between the non-GAAP and the GAAP measures has made these

non-GAAP adjustments more visible to financial statement users. Consequently,

managers may not be as motivated to include any ad-hoc adjustments in the non-GAAP

component. We define our first hypothesis as below:

H1: CEO bonus compensation tied to an alternative non-GAAP performance

measure is positively associated with the level of manipulation in the non-GAAP

component of this measure.

While REIT managers can manipulate the non-GAAP adjustments in FFO to

increase performance-based compensation, they can also engage in earnings management

of the GAAP component in FFO to affect the non-GAAP performance measure. Note

non-GAAP alternative performance measures (e.g., pro forma earnings, EBITDA, FFO)

15

are typically obtained by making adjustments to a GAAP measure (i.e., usually net

income). Hence, earnings management on the GAAP measure can affect firm

performance measured by both the GAAP and non-GAAP metrics. In the context of the

REIT industry, managers can affect FFO via the manipulation of net income. On one

hand, one can argue that managers may not want to manipulate the GAAP-governed net

income measure for compensation purposes when they have the alternative choice of

manipulating the non-GAAP component. On the other hand, prior research (e.g., Doyle et

al., 2003) shows large GAAP-non-GAAP differences constitute a detrimental signal to

the future value of the firm, so there may be limit to managers on how much they would

want to manage the non-GAAP component. This is especially true for the REIT industry,

when there exists a recommended standard definition of FFO by NAREIT that allows

investors to obtain a rough estimate of the FFO-net income differences. Moreover, it is

also questionable how much discretionary expenses a firm can exclude from FFO for

REIT managers to achieve the intended earnings management effect. Therefore, earnings

management via the GAAP component may remain necessary for managers to manage

FFO to increase CEO bonuses. We define our second hypothesis as follows:

H2: CEO bonus compensation tied to an alternative non-GAAP performance

measure is positively associated with the level of manipulation in the GAAP component

of this measure.

We next consider the effectiveness of existing regulatory efforts in constraining

potential FFO manipulation. While there is little standard that defines pro forma earnings,

REITs are provided with detailed guidance by the industry organization NAREIT on the

compilation of FFO. These self-regulatory efforts have proved to be useful in improving

16

the transparency of the FFO measure. For instance, Baik et al. (2008) show the frequency

of managers using FFO to meet or beat analysts forecast has been greatly reduced

subsequent to the increased industry efforts to promote a uniform definition of FFO in

2000. However, since the industry guidance is voluntary in nature, managers who want to

manipulate FFO for bonus purposes would likely not follow the NAREIT-recommended

guidelines in defining FFO. We therefore expect the association between CEO bonus and

FFO manipulation to be lower when firms voluntarily adhere to the industry guidance

provided by NAREIT, and we define the first part of our third hypothesis as follows:

H3A: The adherence to industry self-regulatory efforts reduces the association

of CEO bonus compensation tied to an alternative non-GAAP performance measure and

the level of manipulation in the measure.

Prior research shows Regulation G is effective in constraining opportunistic

reporting behavior of pro forma earnings (e.g., Kolev et al., 2008) as well as FFO (e.g.,

Fortin et al., 2011). Yet, the SEC has not brought forward any enforcement action with

regard to the noncompliance of Regulation G until recently.4 Our sample shows that

some firms did not comply with the key requirement of Regulation G by failing to

provide a reconciliation of the non-GAAP measure and the GAAP measure. Both Baik et

al. (2008) and Fortin et al. (2011) have shown reconciliation is an effective device to

improve the transparency of FFO reporting. We conjecture that managers who have little

intention to manipulate FFO to affect CEO compensation would have no problem

following Regulation G disclosing the reconciliation schedule, while managers who want

to manipulate FFO may have a stronger incentive to omit disclosure of such important

4 The SEC filed its first enforcement action under Regulation G on November 12, 2009 to SafeNet, Inc., claiming that the company made improper adjustments to the company’s recurring expenses without factual support (Katz, 2009).

17

information. Hence, we expect the association between CEO bonus and FFO

manipulation would be lower when firms comply with the reconciliation disclosure

requirement of Regulation G, and we define the second part of our third hypothesis as

follows:

H3B: The compliance to Regulation G reduces the association of CEO bonus

compensation tied to an alternative non-GAAP performance measure and the level of

manipulation in the measure.

Anecdotal evidence shows that even when compensation disclosure is mandatory

and the potential repercussions for violating such requirement could be severe, some

firms would still provide incomplete and even fraudulent compensation disclosure,

eventually leading to enforcement actions brought by the SEC (Wood and Missal, 2006).5

We conjecture that firm managers who want to manipulate FFO to increase CEO bonuses

have stronger incentives to obscure compensation disclosure. We focus on the disclosure

of targets and benchmarks as they are the most relevant and quantifiable factors for bonus

determination (Indjejikian and Nanda, 2002). We expect the association between CEO

bonus and FFO manipulation would be lower for firms with compliance to the SEC

compensation disclosure requirement revealing their targets and benchmarks used in

determining bonus, and we define the last part of our third hypothesis as follows:

H3C: The compliance to the SEC compensation disclosure requirements reduces

the association of CEO bonus compensation tied to an alternative non-GAAP

performance measure and the level of manipulation in the measure.

5 For instance, the SEC has initiated enforcement proceedings against General Electric Co. in 2004 and against Tyson Foods Inc. in 2005.

18

Finally, we examine the impact of internal corporate governance on the

association of FFO manipulation and CEO bonus. Prior evidence (e.g., Klein, 2002; Xie

et al., 2003; Bowen et al., 2008) shows that opportunistic earnings management is

associated with weak governance quality. Extant research (e.g., Boyd, 1994; Core,

Houlthausen and Larcker, 1999; Cyert, Kang and Kumar, 2002) also shows weaker

corporate governance structure is associated with greater agency problem and higher

executive pay. In the context of non-GAAP reporting, Mbagwu (2007) shows that board

independence is positively associated with quality of non-GAAP measures. However,

Jennings and Marques (2011) and Frankel et al. (2011) show that subsequent to

Regulation G, the association between corporate governance and the level of opportunism

on non-GAAP reporting declines as the regulation acted as an effective substitute to

internal corporate governance. Nonetheless, we define our fourth hypothesis in

alternative form, as follows:

H4: Better internal governance reduces the association of CEO bonus

compensation tied to an alternative non-GAAP performance measure and the level of

manipulation in the measure.

Research Design

Our first two hypotheses (H1 and H2) examine whether REIT managers

manipulate FFO to increase CEO bonuses. We use the following regression models:

BONUS = α + β1 DACC + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6

CEO_OWN + β7CEO_DIR+ β8 CEO_TENURE + ε (1)

BONUS = α + β1 FFOMANI + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH +

β6 CEO_OWN + β7CEO_DIR+ β8 CEO_TENURE + ε (2)

19

Where the dependent variable is total annual bonus awarded to CEO. Our key variables

of interest are DACC and FFOMANI. DACC represents manipulation in the GAAP

component of FFO. Since the GAAP component of FFO is essentially net income, we use

the commonly used proxy of earnings management in net income, measured by the level

of discretionary accruals, to estimate the level of manipulation in the GAAP component

of FFO. We follow the accounting and finance literature (e.g., Bergstresser and

Philippon, 2006; Cohen, Dey and Lys, 2008) and measure DACC using the modified

Jones model, as proposed in Dechow, Sloan and Sweeney (1995).6 We use signed instead

of absolute discretionary accruals as firms would have positive discretionary accruals if

managers want to increase CEO bonus compensation. FFOMANI measures the level of

manipulation in the non-GAAP component of FFO. We follow the real estate literature

(e.g., Zhu, Ong and Yeo, 2010; Anglin, Edelstein, Gao and Tsang, 2012) and measure

FFOMANI as the percentage deviation of actual FFO from normal FFO, where normal

FFO is defined as net income adjusted for real estate depreciation, real estate disposal

gains and losses, extraordinary items and results from discontinued operations (as in the

NAREIT definition). If managers manipulate both components of FFO to increase CEO

bonus compensation, we would observe positive and significant coefficients for both

DACC and FFOMANI.

We include other variables that affect the level of CEO bonus compensation in the

REIT industry. As REITs’ CEO bonuses are directly tied to FFO, prior research has

shown a positive relation between FFO performance and bonus (e.g., Pennathur and

Shelor, 2002; Pennathur et al., 2005; Griffin, Najand and Weeks, 2012). We measure

6 Alternatively, in unreported analysis, we also conduct our study using alternative accrual measures calculated based on the methodology proposed in Dechow and Dichve (2002) and obtain similar results.

20

FFO performance with the variable NAREIT_FFO, measured as normal FFO per share.

We include firm size and leverage, measured respectively as the natural log of market

capitalization and the ratio of debt to total assets, and expect positive coefficients (e.g.,

Ghosh and Sirmans, 2005; Feng, Ghosh and Sirmans, 2007; Eichholtz, Kok and Otten,

2008; Feng, Ghosh, He and Sirmans, 2010). Prior research (e.g., Davis and Shelor, 1995)

shows firm growth (measured by change in total assets) has a positive impact on CEO

compensation. Hence we include GROWTH, measured as the percentage one-year growth

in total assets. We also include variables that capture CEO characteristics. We follow

Feng et al. (2007), Feng et al. (2010) and Griffin et al. (2012) and include CEO_OWN

(measured as the percentage of shares owned by the CEO), CEO_DIR (dummy variable

equal to one, zero otherwise, if the CEO serves a dual role as director of the board), and

CEO_TENURE (measured as the number of years the CEO has served the firm).7 We

expect positive effects from these variables. Since property types of REITs are important

determinants of a REIT’s operating structure and they affect CEO compensation (e.g.,

Hardin, 1998), we classify our sample firms by their REIT types (i.e., retail, office,

industrial residential, diversified, and specialized) and include property type fixed effects

in the regressions.

To test our third hypothesis on the effect of regulatory forces on the association of

CEO bonus and FFO manipulation, we augment (1) and (2) with the following models:

7 Alternatively, we include additional CEO characteristics, such the age of the CEO, in our regression specifications. However, we do not include age in our main analysis because age and tenure are highly correlated. Nonetheless, in results unreported, we find the inclusion of these additional control variables does not change our main findings.

21

BONUS = α + β1A DACC (FFOMANI) + β1B D_NAREIT + β1C D_NAREIT* DACC

(FFOMANI) + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN +

β7 CEO_DIR+ β8CEO_TENURE + ε (3)

BONUS = α + β1A DACC (FFOMANI) + β1B D_RECON + β1C D_RECON*DACC

(FFOMANI) + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN +

β7 CEO_DIR+ β8 CEO_TENURE + ε (4)

BONUS = α + β1A DACC (FFOMANI) + β1B D_FFO + β1C D_FFO* DACC

(FFOMANI) + β1D D_FFOGROWTH + β1E D_FFOGROWTH* DACC (FFOMANI) +

β1F D_WFFO + β1G D_WFFO* DACC (FFOMANI) + β2 NAREIT_FFO + β3 SIZE + β4

LEV + β5 GROWTH + β6 CEO_OWN + β7 CEO_DIR+ β8 CEO_TENURE + ε (5)

We introduce several new variables to capture firms’ compliance to the NAREIT

definition of FFO and to SEC’s Regulation G and compensation disclosure. In

specification (3), D_NAREIT is a dummy variable equal to one (zero otherwise) when

REIT managers disclose the adoption of the NAREIT definition when defining FFO. Our

variable of interest is the interaction term of D_NAREIT and FFO manipulation, and we

expect a negative coefficient for β1C because voluntary disclosure of the use of NAREIT

definition implies managers are more committed to the transparency of FFO reporting,

and these managers should be less likely to manipulate FFO to increase CEO bonus. In

specification (4), D_RECON is a dummy variable equal to one (zero otherwise) when

REIT managers comply with Regulation G and provide a reconciliation between FFO and

net income. We also expect the interactive effect of D_RECON and FFO manipulation to

be negative as the provision of reconciliation schedule reduces the opportunity that REIT

managers can manipulate FFO to affect CEO bonus. Finally, we examine the impact of

22

the new compensation disclosure under Item 402 of Regulation S-K in specification (5).

Specifically, we focus on the disclosure of two benchmarks with regard to bonus

determination. D_FFO is a dummy variable equal to one (zero otherwise) if firms have a

target FFO level and managers decide to disclose this target. Alternatively, some firms

may have their bonus targets tied to growth/change of FFO instead of level.

D_FFOGROWTH is a dummy variable equal to one (zero otherwise) if firms have a

target growth/change of FFO and managers decide to disclose this target. Finally, we find

that a substantial portion of our sample firms also disclose the weights on different

factors they use when determining CEO bonus.8 We include D_WFFO, a dummy

variable equal to one (zero otherwise) when firms discuss the weighting scheme of how

bonus is determined.

Finally, we examine the effect of general internal corporate governance

mechanisms on the association of CEO bonus and FFO manipulation with the following

model:

BONUS = α + β1A DACC (FFOMANI) + β1B CG_INDEX + β1C CG_INDEX*DACC

(FFOMANI) + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN +

β7 CEO_DIR+ β8 CEO_TENURE + ε (6)

Where CG_INDEX is equal to one (zero otherwise) if the Corporate Governance Index

Score reported by Institutional Shareholder Services (ISS) for the company in that given

year is greater than the sample median.9 If general corporate governance mechanisms

8 In fact, the compensation disclosure requirement states that firms should address in their CD&A how each element of compensation is determined in terms of the amount and formula, if applicable. Hence, the weighting of factors in determining bonus is also a required, albeit less explicit, disclosure requirement. 9 We elect to construct a dummy variable to measure overall corporate governance as we want to compare the significance of internal governance with the external regulation variables. However, we also conduct a robustness analysis using the raw corporate governance score as our CG_INDEX measure and find our main inferences are unaffected.

23

constrain earnings management activities motivated by bonus concerns, we expect the

interaction effect of CG_INDEX and FFO manipulation to be significant and negative.

Data & Sample Selection

Our sample firms include 257 REITs over the period of 2006-2009 in the Capital

IQ database. Our sample period starts in 2006 as we intend to examine firms’ compliance

to and the complementary nature of different regulatory actions imposed by NAREIT and

the SEC and the latest regulation on compensation disclosure came into effect in 2006.

We eliminate 55 mortgage REITs from the total sample as performance measurement of

this sub-industry sector is drastically different and FFO is typically not reported as a

supplementary measure by this sector. We collect CEO compensation as well as CEO

characteristics and financial data for these firms from Capital IQ. Corporate Governance

Index Score is obtained from ISS. We require non-missing data on all regression

variables, and our final sample contains 54 distinct firms and 188 firm-year observations.

We then hand-collect information on their compliance to NAREIT and the SEC

disclosure requirements from companies’ SEC filings. Firms’ disclosure of the use of

NAREIT definition of FFO and FFO reconciliation are found in the Management

Discussion and Analysis (MD&A) section in the 10-K filings, and we obtain disclosure

information on executive compensation in the CD&A section from the annual proxy

statements.

Table 1 (Panel A) provides descriptive statistics for the total sample of 188 firm-

year observations. On average, CEO bonus amounts to $432.61 (in thousands) and

represents 44% of total cash compensation.10 DACC has a mean of -0.00065 and

10 Capital IQ defines cash compensation as the sum of salary and bonus, and non-cash compensation as the sum of stock awards, option awards, non-equity incentives, pension change and other compensation. Our

24

FFOMANI has a mean of 28%.11 As for the disclosure variables, we find that an average

of 86% of firms explicitly states they follow the NAREIT definition in defining FFO.12

An average of 89% of firms adheres to Regulation G with the provision of a

reconciliation schedule between FFO and a GAAP performance measure. D_FFO and

D_FFOGROWTH have means of 0.51 and 0.31 respectively. Note it is possible that firms

determine CEO bonuses using target level of FFO, growth of FFO, or both.13

Approximately half (49%) of firms discloses the weighting of FFO in setting bonus. The

average value of CG_INDEX is 0.53. Of the control variables, the sample firms report an

average NAREIT_FFO of 2.18 per share, SIZE of 7.61 and leverage ratio of 60.98%. The

CEO owns an average of 0.90% of the company’s shares and 93% of CEO also serve as

the director of the board. Finally, an average CEO has tenure of 8.67 years.

Table 1 (Panel B) reports Pearson correlations for variables used in the regression

analysis. Univariate correlations show that FFOMANI is positively correlated to CEO

bonus, and bonus is also correlated to CG_INDEX, NAREIT_FFO, SIZE and GROWTH.

Interestingly, we observe pretty low and mostly insignificant correlations among the

disclosure variables. These correlations suggest that firms’ compliance to one regulation

does not imply compliance to another regulation. Hence, we believe our disclosure

sample firms report an average CEO cash compensation of $992.14 (in thousands) and non-cash compensation of $2292.49 (in thousands). Overall, stock awards, salary and bonus represent the three largest components of compensation for REIT CEOs. 11 By construction, signed discretionary accruals have a mean close to zero. 12 We acknowledge there is a possibility that firms are actually following the NAREIT definition of FFO without explicitly disclosing the use of the NAREIT definition. On the other hand, firms that disclose the use of NAREIT FFO definition nonetheless report actual FFOs that are substantially different from NAREIT_FFO. However, our variable D_NAREIT is not intended to capture actual conformance to the NAREIT definition. Instead, D_NAREIT measures the disclosure quality with regard to the industry self-regulation assuming all REITs follow the NAREIT definition of FFO to a certain extent. 13 In un-tabulated results, we find that the average disclosure of either FFO target or FFO growth target is 0.70. We discuss the empirical findings of aggregating our measures of compensation disclosure in next section.

25

variables capture the impact of different regulatory efforts instead of the overall

disclosure quality of the company.

Empirical Results

Main Findings

Table 2 present the empirical results from regression equations (1) and (2). All

regressions are estimated using ordinary least square with robust standard errors

clustering at the company level. Columns I and II show results of specification (1) with

DACC as our key variable of interest. In column I, we find that signed discretionary

accruals are unrelated to bonus compensation. We obtain the same findings after we

include the control variables in column II. The findings imply REIT managers do not

manage discretionary accruals to affect CEO bonus when bonus contract is tied to a non-

GAAP alternative performance measure. Our results provide interesting contrast to extant

literature that documents a positive association of CEO compensation and discretionary

accruals when compensation is directly linked to a GAAP measure (e.g., Holthausen et

al., 1995; Balsam, 1998). Of the control variables, we find significant association of

bonus to firm size (at 1% level). We also observe higher bonuses for CEOs who also

serve as a director on the board (at 10% level).

Columns III and IV in Table 2 present results of specification (2). We find that

FFOMANI is significantly positively related to CEO bonus. The findings suggest that

when bonus is explicitly tied to a non-GAAP measure, managers would be tempted to

manipulate the non-GAAP component of the measure instead of the GAAP component.

In the case of REITs, managers would deviate from the recommended NAREIT

definition of FFO by incorporating discretionary positive exclusions. Of the control

26

variables, we find that SIZE, CEO_DIR as well as GROWTH are significantly associated

with bonus.

Finally, we include both DACC and FFOMANI in the same regressions and we

report the results in columns V and VI. We obtain the same findings that DACC remains

insignificant and FFOMANI remains significant.

We next examine the impact of regulatory forces on the association of bonus and

FFO manipulation. As we find that REIT managers only manipulate the non-GAAP

component of FFO to increase bonus, we simply present the results with FFOMANI as

the explanatory variable in subsequent analysis.14 Table 3 reports results of regression

specification (3). Column I (II) presents results of the regression of BONUS on the

variables of interest without (with) the control variables. We again find significant and

positive association between bonus and FFOMANI. It is interesting that firms that

disclose the use of NAREIT FFO definition offer higher level of bonus to their CEOs, as

the coefficient of D_NAREIT is marginally significant at 10% level. Our key variable of

interest in equation (3), the interaction term of D_NAREIT and FFOMANI, is significant

at 5% level. It has a coefficient of -120.94 while FFOMANI has a coefficient of 215.40,

meaning that the impact of FFOMANI on CEO bonus is reduced by more than 50% when

firms voluntarily disclose the adherence to the industry self-guidance in defining their

non-GAAP measure.

Table 4 reports results of regression specification (4). While we do not find the

provision of a reconciliation schedule between FFO and net income affects bonus, we

find that the association between CEO bonus and FFO manipulation is significantly lower

14 We find that the inclusion of the disclosure variables does not change our conclusions with regard to the insignificance of DACC. Hence, the results of specification (3) to (6) with DACC as the explanatory variable are not tabulated but are available from the authors upon request.

27

for firms that disclose the reconciliation. The results are consistent with prior research

(e.g., Baik et al., 2008) and suggest that reconciliation improves the transparency of non-

GAAP reporting. Increased transparency in turn discourages managers to adjust the non-

GAAP measure opportunistically upward to increase CEO bonus.

Table 5 reports results of regression specification (5) and we find that all three of

our proxies for the quality of disclosure with regard to bonus determination have

statistically significant interaction effects with FFOMANI on CEO bonus. The findings

indicate that firms are less likely to manipulate FFO to boost CEO bonus when managers

are more willing to comply with the SEC compensation disclosure and reveal their FFO

targets and bonus weighting methods in the companies’ CD&A. Our results extend the

findings in a recent study by Robinson et al. (2009), which show noncompliance with the

new compensation disclosure is associated with excess CEO compensation. D_FFO,

D_FFOGROWTH and D_WFFO have coefficients of -110.81, -204.44 and -232.31

respectively, which suggest that disclosure of the weights of FFO in bonus calculation

actually plays an important role in improving the transparency of the alternative

performance measure. Since firms may have their bonus thresholds determined by the

level of FFO, growth of FFO, or both, we construct an alternative variable D_TARGET

and assign a value of one (zero otherwise) if a firm has disclosed any bonus target in

terms of level or growth. We also construct an alternative disclosure measure for bonus

compensation (D_COMP), measured as the sum of the variables D_FFO,

D_FFOGROWTH and D_WFFO. In results un-tabulated, we find both D_TARGET and

D_COMP have significant negative interactive effect with FFOMANI on CEO bonus.

28

Finally, in Table 6, we report results of regression specification (6). Consistent

with prior literature (e.g., Boyd, 1994; Core et al., 1999; Cyert et al., 2002), we find the

level of corporate governance is negatively related to CEO compensation. However, we

do not find any evidence that corporate governance constrains the opportunistic reporting

of FFO for bonus purposes. Our findings confirm with the views in recent studies in the

REIT industry that show corporate governance has little effect on accruals earnings

management (Anglin et al., 2012) and is only weakly related to firm performance (e.g.,

Hartzell, Sun and Titman, 2006; Bauer, Eichholtz and Kok, 2010). And the results could

be the consequence of regulation being the substitute for good corporate governance

(e.g., Jennings and Marques, 2011; Frankel et al., 2011). To ensure the robustness of our

findings, in unreported analysis we also adopt an alternative self-constructed corporate

governance measure and find similar results.15

In summary, our empirical analysis shows that when bonus compensation is

linked to a non-GAAP alternative performance measure, managers would be motivated

by bonus concern and would engage in earnings management of this particular measure.

However, they would be more likely to achieve their goal via the manipulation of the

non-GAAP adjustments instead of the GAAP component. We further find that regulatory

forces exerted by both the industry and the SEC, and on both non-GAAP reporting and

compensation disclosure, are effective in constraining opportunistic financial reporting

for bonus purposes.

15 Since 2010, ISS decomposes its governance score into four broad categories: Audit, board structure, shareholders’ concern and compensation concern. Each subcategory is ranked as either ‘low,’ ‘medium’ or ‘high’. We assign numerical score of -1, 0 and 1 for the three rankings respectively and sum the score of the four subcategories as our alternative governance measure. Hence, the value of our alternative corporate governance variable can range from -4 to 4. As we only have data for 2010, we assign the same score for a given firm over the period of 2006-2009.

29

Additional Analysis

An interesting issue is how different regulatory forces compete and complement

with each other in constraining opportunistic financial reporting for bonus purposes.

Hence, we also run an augmented regression model including D_NAREIT, D_RECON,

D_FFO, D_FFOGROWTH, D_WFFO and their interaction effects with FFOMANI in

one equation. In results un-tabulated, we find that the interaction effects of D_RECON

and D_WFFO with FFOMANI remain highly significant, while the interaction term of

D_NAREIT and FFOMANI has become insignificant.16 The findings indicate that the

imposition of external regulation represents the most important safeguard in warranting

the proper use of non-GAAP measures in bonus determination.

Next, we consider other variables that may affect CEO bonus. While economic

theory dictates the presence of institutional investors mitigate the agency problem

between shareholders and managers, evidence on the effect of institutional investors on

CEO compensation is mixed in both the general market setting (e.g., David, Kochhar and

Levitas, 1998; Hartzell and Starks, 2003) and the REIT setting (e.g., Eichholtz et al.,

2008; Feng et al., 2010). Nonetheless, we include IO, measured as the percentage of

shares owned by institutional investors, as an additional control variable. As REITs in the

U.S. are required by law to distribute 90% of their taxable income as dividends to

investors, dividend yield may be an important determinant of CEO compensation

(Eichholtz et al., 2008). We therefore include dividend yield as an additional variable in

our regressions. Our key findings are unaffected by the inclusion of these additional

controls.

16 Though the interaction terms of FFOMANI with D_FFO and D_FFOGROWTH are insignificant in this augmented regression, the interaction term of FFOMANI and D_TARGET becomes significant at 10% level when we combine D_FFO and D_FFOGROWTH into the aggregate measure D_TARGET.

30

Over the sample period of 2006-2009, REITs were going through perhaps the

largest financial crisis in the history of the industry. To control for this macroeconomic

time series effect, we first include additional year dummies to our regression

specifications and we find our results remain intact. Given the seriousness of the crisis to

the real estate industry, we further examine the impact of the crisis on CEO bonus

compensation and FFO manipulation in the following model:

BONUS = α + β1A FFOMANI + β1B CRISIS + β1C CRISIS*FFOMANI + β2

NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN + β7 CEO_DIR+ β8

CEO_TENURE + ε (7)

Prior research (e.g., Devos, Ong, Spieler and Tsang, 2012; Devos, Spieler and

Tsang, 2012) show the REIT industry was hit hard over 2007 and 2008 before bouncing

back in 2009. Hence, we define CRISIS as a dummy variable equal to one (zero

otherwise) for firm-year observations in the years of 2007 and 2008. Table 7 reports

results of regression specification (7). We find that during the financial crisis period, the

association between CEO bonus and FFOMANI is lower than before and after the crisis.

The results imply increased scrutiny on firms during the market downturn has limited the

opportunity that managers can manipulate the performance measure to affect CEO

compensation.17

Lastly, our study has so far focused on the examination of CEO bonus as it is

directly linked to the non-GAAP performance measure in the REIT industry. Prior

studies (Isidro and Marques, 2010; Black et al., 2011) suggest that CEO compensation is

related to opportunistic non-GAAP reporting even when the non-GAAP measures are not

17 An alternative explanation is that firms care less about manipulating FFO to affect CEO bonus during the financial crisis.

31

used in compensation contracts, as managers may try to overstate the non-GAAP

measures to affect market perceptions. If this conjecture is true, we should observe

positive relations of FFO manipulation and other components of CEO compensation. We

test this conjecture with the following model:

CEO_COMP = α + β1 FFOMANI + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5

GROWTH + β6 CEO_OWN + β7CEO_DIR+ β8 CEO_TENURE + ε (8)

Where CEO_COMP represents one of other six components (i.e., salary, stock awards,

option awards, non-equity incentives, pension change, and other) of CEO compensation

reported in the Capital IQ database. Regression results of specification (8) presented in

Table 8 show that none of the other components of CEO compensation is significantly

related to FFOMANI.18 The findings reinforce that it is specifically the inclusion of the

non-GAAP measure in bonus contract design that motivate opportunistic reporting of

FFO in our sample firms.

Concluding Remarks

In recent years, the reporting of non-GAAP alternative performance measures

along with audited GAAP performance measures has slowly become a norm for many

firms and in many industries. Our study utilizes the unique REIT setting where the

industry has moved one step further incorporating its industry-specific non-GAAP

performance measure known as FFO in bonus compensation contract design. Despite that

FFO is commonly considered as more reliable than other non-GAAP pro forma measures

18 We attribute differences of results in our study and the above-mentioned pro forma studies to sample differences. Firms that report the pro forma measure may have used the non-GAAP measure, or some adjusted net income measures that are highly correlated to pro forma, in determining CEO compensation. Unfortunately, it is rather difficult for authors of these pro forma studies to identify which firms in their samples tie CEO compensation directly to pro forma performance. In fact, Isidro and Marques (2010) have acknowledged this data limitation in their study.

32

given the guidance of an industry organization NAREIT, we find that REIT managers

nonetheless manipulate FFO upwards in order to increase CEO bonus compensation. In

particular, these managers would manipulate the non-GAAP component of FFO by

making ad-hoc adjustments to the NAREIT FFO definition. We show that both self-

regulation by the industry and regulations set forth by the SEC are effective in curbing

the extent of managers manipulating FFO to increase CEO bonus. Yet, internal corporate

governance does not seem to provide strong enough control mechanisms to deter

managerial manipulative actions.

Overall, while non-GAAP performance measures reflect firm performance and

are legitimate potential factors to be considered in compensation contract design, our

study highlights the enhanced concerns of aggressive non-GAAP reporting when these

measures are further used to determine CEO compensation. We show that, in this

situation, internal corporate governance mechanisms would be insufficient and external

regulatory oversight and guidance would be needed to ensure fair reporting of non-GAAP

information.

33

References

Abarbanell, J. and R. Lehavy, 2007, “Letting the ‘Tail Wag the Dog’: The Debate over GAAP versus Street Earnings Revisited,” Contemporary Accounting Research, 24, 3, 675-723

Anglin, P., R. Edelstein, Y. Gao and D. Tsang, 2012, “What is the Relationship between

REIT Governance and Earnings Management?” Journal of Real Estate Finance and Economics, forthcoming

Athanasakou, V.E., N.C. Strong and M. Walker, 2007, “Classificatory Income

Smoothing: The Impact of a Change in Regime of Reporting Financial Performance,” Journal of Accounting and Public Policy, 26, 387-435

Baik B., B.K. Billings and R.M. Morton, “The Effect of Increased Transparency on

Manipulation and Value Relevance of Non-GAAP Disclosures by Real Estate Investment Trusts (REITs),” The Accounting Review, 2008, 83, 2, 271-301

Balsam, S., “Discretionary Accounting Choices and CEO Compensation,” Contemporary

Accounting Review, 15, 3, 229-252 Bauer, R., P. Eichholtz and N. Kok, 2010, “Corporate Governance and Performance: The

REIT Effect,” Real Estate Economics, 38, 1, 1-29 Beattie, V., S. Brown, D. Ewers, B. John, S. Manson, D. Thomas and M. Turner, 1994,

“Extraordinary Items and Income Smoothing: A Positive Accounting Approach,” Journal of Business Finance and Accounting, 21, 791-811

Ben-Shahar, D., E. Sulganik and D. Tsang, 2011, “FFO versus Net Income: Examining

the Dividend Relevance of REIT Performance Measures,” Journal of Real Estate Research, 33, 3, 415-441

Bergstresser, D.B. and T. Philippon, 2006, “CEO Incentives and Earnings Management,”

Journal of Financial Economics, 80, 3, 511-529 Bhattacharya, N., E. Black, T. Christensen and C. Larson, 2003, “Assessing the Relative

Informativeness and Permanence of Pro-forma Earnings and GAAP Operating Earnings,” Journal of Accounting and Economics, 36, 285-319

Bhattacharya, N., E. Black, T. Christensen and R. Mergenthaler, 2007, “Who Trades on

Pro-forma Earnings Information?” The Accounting Review, 82, 3, 581-619 Black, D.E., E.L. Black, T.E. Christensen and J.F. Waegelein, 2011, “The Effect of

Executive Compensation Contracts and Auditor Effort on Pro Forma Reporting Decisions,” Working Paper, Duke University, Brigham Young University and Emporia State University

34

Black, D.E. and T. Christensen, 2009, “U.S. Managers’ Use of ‘Pro-forma’ Adjustments

to Meet Strategic Earnings Benchmarks,” Journal of Business Finance and Accounting, 36, 3-4, 297-326

Bowen, R., A. Davis and D. Matsumoto, 2005, “Emphasis on Pro-Forma versus GAAP

Earnings in Quarterly Press Release: Determinants, SEC Intervention, and Market Actions,” The Accounting Review, 80, 1011-1038

Bowen, R.M., S. Rajgopal and M. Venkatachalam, 2008, “Accounting Discretion,

Corporate Governance, and Firm Performance,” Contemporary Accounting Research, 25, 2, 351-405

Boyd, B.K., 1994, “Board Control and CEO Compensation,” Strategic Management

Journal, 15, 335-344 Bradshaw, M. and R. Sloan, 2002, “GAAP versus the Street: An Empirical Assessment

of Two Alternative Definitions of Earnings,” Journal of Accounting Research, 40, 41-66

Brown, L. and K. Sivakumar, 2003, “Comparing the Quality of Two Operating Income

Measures,” Review of Accounting Studies, 8, 561-572 Burns, N. and S. Kedia, 2006, “The Impact of Performance-Based Compensation on

Misreporting,” Journal of Financial Economics, 79, 35-67 Cheng, Q. and T. Warfield, 2005, “Equity Incentives and Earnings Management,” The

Accounting Review, 80, 2, 441-476 Cohen, D.A., A. Dey and T.Z. Lys, 2008, “Real and Accrual-Based Earnings

Management in the Pre- and Post-Sarbanes-Oxley Periods,” The Accounting Review, 83, 3, 757-787

Cohen, D.A., R.N. Hann and M. Ogneva, 2007, “Another Look at GAAP versus The

Street: An Empirical Assessment of Measurement Error Bias,” Review of Accounting Studies, 12, 2/3, 271-303

Core, J.E., R. Houlthausen and D.F. Larcker, 1999, “Corporate Governance, Chief

Executive Officer Compensation, and Firm Performance,” Journal of Financial Economics, 51, 371-406

Romanek, B., 2003, “Regulation G’s Mixed Blessing for REITs,”

www.corporatecounsel.net

35

Craighead, J., M. Magnan and L. Thorne, 2004, “The Impact of Mandated Disclosure on Performance-Based CEO Compensation,” Contemporary Accounting Review, 21, 2, 369-398

Cyery, R.M., S-H Kang and P. Kumar, 2002, “Corporate Governance, Takeovers, and

Top-Management Compensation: Theory and Evidence,” Management Science, 48, 4, 453-469

David, P., R. Kochhar and E. Levitas, 1998, “The Effect of Institutional Investors on the

Level and Mix of CEO Compensation,” Academy of Management Journal, 41, 2, 200-208

Davis, B.J. and R.M. Shelor, 1995, “Executive Compensation and Financial Performance

in the Real Estate Industry,” Journal of Real Estate Research, 10, 2, 141-151 Dechow P.M. and I.D. Dichev, 2002, “The Quality of Accruals and Earnings: The Role

of Accrual Estimation Errors,” The Accounting Review, 77 (supplement), 35-59 Dechow, P., M. Huson and R. Sloan, 1994, “The Effect of Restructuring Charges on

Executives’ Cash Compensation,” The Accounting Review, 69, 138-156 Dechow, P.M., R. Sloan, and A. Sweeney, 1995, “Detecting Earnings Management,” The

Accounting Review, 70, 193-225 Devos, E., S.E. Ong, A. Spieler and D. Tsang, 2012, “REIT Institutional Ownership

Dynamics and the Financial Crisis,” Journal of Real Estate Finance and Economics, forthcoming

Devos, E., A. Spieler and D. Tsang, 2012, “Elective Stock Dividends and REITs:

Evidence from the Financial Crisis,” Working paper, University of Texas at El Paso, Hofstra University and McGill University

Doyle, J.T., R.J. Lundholm and M.T. Soliman, 2003, “The Predictive Value of Expenses

Excluded from Pro-forma Earnings,” Review of Accounting Studies, 8, 145-174 Eichholtz, P.M., N. Kok and R. Otten, 2008, “Executive Compensation in UK Property

Companies,” Journal of Real Estate Finance and Economics, 36, 405-426 Fan, Y., A. Barua, W.M. Cready and W. Thomas, 2010, “Managing Earnings Using

Classification Shifting: Evidence from Quarterly Special Items,” The Accounting Review, 85, 4, 1303-1323

Feng, Z., C. Ghosh and C.F. Sirmans, 2007, “Director Compensation and CEO

Bargaining Power in REITs,” Journal of Real Estate Finance and Economics, 35, 225-251

36

Feng, Z., C. Ghosh, F. He and C.F. Sirmans, 2010, “Institutional Monitoring and REIT CEO Compensation,” Journal of Real Estate Finance and Economics, 40, 446-479

Fields, T., S. Rangan and R. Thiagarajan, 1998, “An Empirical Evaluation of the

Usefulness of Non-GAAP Accounting Measures in the Real Estate Investment Trust Industry,” Review of Accounting Studies, 3, 103-130

Fortin, S., P. Liu and D. Tsang, “The Effect of SEC Intervention on Non-GAAP

Disclosures in the Presence of Industry Guidance,” Working Paper, McGill University and Cornell University

Frankel, R., S McVay and M. Soliman, 2011, “Non-GAAP Earnings and Board

Independence,” Review of Accounting Studies, 16, 719-744 Gaver, J., K. Gaver and J. Austin, 1995, “Additional Evidence on Bonus Plans and

Income Management,” Journal of Accounting and Economics, 19, 29-74 Gaver, J. and K. Gaver, 1998, “The Relation between Nonrecurring Accounting Charges

and CEO Compensation,” The Accounting Review, 73, 2, 235-253 Ghosh, C. and C.F. Sirmans, 2005, “On REIT CEO Compensation: Does Board Structure

Matter?” Journal of Real Estate Finance and Economics, 30, 397-428 Godfrey, J.M. and K.L. Jones, 1999, “Political Cost Influences on Income Smoothing via

Extraordinary Item Classification,” Accounting and Finance, 39, 229-254 Graham, C.M. and J.R. Knight, 2000, “Cash Flows vs. Earnings in the Valuation of

Equity REITs,” Journal of Real Estate Portfolio Management, 6, 1, 17-25 Griffin, J.M., M. Najand and H.S. Weeks, 2012, “What Influences the Changes in REIT

CEO Compensation? Evidence from Panel Data,” Journal of Real Estate Research, forthcoming

Gu, Z. and T. Chen, 2004, “Analysts’ Treatment of Nonrecurring Items in Street

Earnings,” Journal of Accounting and Economics, 39, 129-170 Hardin III, W.G., 1998, “Executive Compensation in ERITs: EREIT Size is but One

Determinant,” Journal of Real Estate Research, 16, 3, 401-409 Hartzell, J.C. and L. Starks, 2003, “Institutional Investors and Executive Compensation,”

Journal of Finance, 58, 2351-2374 Hartzell, J.C., L. Sun and S. Titman, 2006, “The Effect of Corporate Governance on

Investment: Evidence from Real Estate Investment Trusts (REITs),” Real Estate Economics, 34, 3, 343-376

37

Hayunga, D.K. and C.P. Stephens, “Dividend Behavior of US Equity REITs,” Journal of Property Research, 2009, 26, 2, 105-123

Healy, P., 1985, “The Effect of Bonus Schemes on Accounting Decisions,” Journal of

Accounting and Economics, 7, 85-107 Healy, P. and K. Palepu, 2001, “Information Asymmetry, Corporate Disclosure, and the

Capital Markets: A Review of the Empirical Disclosure Literature,” Journal of Accounting and Economics, 31, 1-3, 405-440

Heflin, F. and C. Hsu, 2008, “The Impact of the SEC’s Regulation of Non-GAAP

Disclosures,” Journal of Accounting and Economics, 46, 349-365 Holthausen, R., D. Larcker and R. Sloan, 1995, “Annual Bonus Schemes and the

Manipulation of Earnings,” Journal of Accounting and Economics, 19, 29-74 Indjejikian, R.J. and D.J. Nanda, 2002, “Executive target Bonuses and What They Imply

about Performance Standards,” The Accounting Review, 77, 4, 793-819 Isidro, H. and A. Marques, “The Effect of Managers’ Compensation and Board Quality

on Non-GAAP Reporting Decisions,” Working Paper, City University London and Universidad Nova de Lisboa

Jennings, R. and A. Marques, 2011, “The Joint Effect of Corporate Governance and

Regulation on the Disclosure of Manager-Adjusted Non-GAAP Earnings in the US,” Journal of Business Finance and Accounting, 38, 3-4, 364-394

Jensen, M. and K. Murphy, 1990, “Performance Pay and Top-Management Incentives,”

Journal of Political Economy, 98, 2, 225-264 Johnson, B. and W, Schwartz, 2005, “Are Investors Misled by “Pro-Forma” Earnings?”

Contemporary Accounting Research, 22, 915-963 Kang, S-H and Y. Zhao, 2010, “Information Content and Value Relevance of

Depreciation: A Cross-Industry Analysis,” The Accounting Review, 85, 1, 227-260 Katz, D.A., 2009, “SEC Brings First Regulation G Enforcement Action,” The Harvard

Law School Forum on Corporate Governance and Financial Regulation Ke, B., K. Petroni, and A. Safieddine, 1999, “Ownership Concentration and Sensitivity of

Executive Pay to Accounting Performance Measures: Evidence from Publicly and Privately-Held Insurance Companies,” Journal of Accounting and Economics, 28, 2, 185-210

Klein, A., 2002, “Audit Committee, Board of Director Characteristics, and Earnings

Management,” Journal of Accounting and Economics, 33, 3, 375-400

38

Kolev, K., C.A. Marquardt, and S. McVay, 2008, “SEC Scrutiny and the Evolution of

Non-GAAP Reporting,” The Accounting Review, 83, 1, 157-184 Leuz, C. and P. Wysocki, 2008, “Economic Consequences of Financial Reporting and

Disclosure Regulation: A Review and Suggestions for Future Research,” Working paper, University of Chicago and University of Miami

Lo, K., 2003, “Economic Consequences of Regulated Changes in Disclosure: The Case

of Executive Compensation,” Journal of Accounting and Economics, 35, 285-314 Marques, A., 2006, “SEC Intervention and the Frequency and Usefulness of Non-GAAP

Financial Measures,” Review of Accounting Studies, 11, 549-574 Mbagwu, C., 2007, “Essays on the Value Relevance of Earnings Measures,” PhD

Dissertation, University of Saskatchewan McVay, S., 2006, “Earnings Management Using Classification Shifting: An Examination

of Core Earnings and Special Items,” The Accounting Review, 81, 3, 501-531 National Association of Real Estate Investment Trusts, 2002, “White Paper on Funds

from Operations,” Washington, D.C., United States National Association of Real Estate Investment Trusts, 1999, “National Policy Bulletin:

November 1999,” Washington, D.C., United States Pennathur, A.K., O.W. Gilley and R.M. Shelor, 2005, “An Analysis of REIT CEO Stock-

Based Compensation,” Real Estate Economics, 33, 1, 189-202 Pennathur, A.K. and R.M. Shelor, 2002, “The Determinants of REIT CEO

Compensation,” Journal of Real Estate Finance and Economics, 25, 1, 99-113 Robinson, J.R., Y. Xue and Y. Yu, 2009, “Mandated Compensation Disclosures and CEO

Pay,” Working paper, The University of Texas at Austin and George Washington University

Sloan, R., 1998, “Discussion of ‘Evaluating Non-GAAP Performance Measures in the

REIT Industry’,” Review of Accounting Studies, 3, 131-135 Smith, C. and R. Watts, 1982, “Incentive and Tax Effects of Executive Compensation

Plans,” Australian Journal of Management, 7, 139-157 Stunda, R.A. and E. Typpo, “The Relevance of Earnings and Funds Flow from

Operations in the Presence of Transitory Earnings,” Journal of Real Estate Portfolio Management, 2004, 10, 1, 37-45

39

Sturgeon, J., 2012, “Dunn Discussed ‘Optics’ of Bonus Payments, Trial Told,” The Financial Post, January 30

Vafeas, N. and Z. Afexentiou, 1998, “The Association between the SEC’s 1992

Compensation Disclosure Rule and Executive Compensation Policy Changes,” Journal of Accounting Public Policy, 17, 1, 27-54

Vincent, L., 1999, “The Information Content of Funds from Operations for Real Estate

Investment Trusts,” Journal of Accounting and Economics, 26, 69-104 Wood, R.E. and M.J. Missal, 2006, “The SEC’s New Executive Compensation

Disclosure Rules: Liability Concerns for Officers and Directors,” Corporate Officers and Directors Liability, 22, 6

Xie, B., W.N. Davidson III and P.J. DaDalt, 2003, “Earnings Management and Corporate

Governance: The Roles of the Board and the Audit Committee,” Journal of Corporate Finance, 9, 295-316

Zhu, Y.W., S.E. Ong, and W.Y. Yeo, 2010, “Do REITs Manipulate their Financial

Results around Seasoned Equity Offerings? Evidence from US Equity REITs,” Journal of Real Estate Finance and Economics, 40, 4, 412-445

40

TABLE 1 (Panel A) Descriptive Statistics of Total Sample

Descriptive Statistics for Total SampleVariables N Mean Median Standard Dev. Min MaxBONUS 188 432.61 0.00 900.41 0.00 7500DACC 188 -0.00065 -0.004 0.035 -0.072 0.25FFOMANI 188 0.28 0.035 1.45 -0.97 14.76D_NAREIT 188 0.86 1.00 0.35 0.00 1.00D_RECON 188 0.89 1.00 0.31 0.00 1.00D_FFO 188 0.51 1.00 0.50 0.00 1.00D_FFOGROWTH 188 0.31 0.00 0.46 0.00 1.00D_WFFO 188 0.49 0.00 0.50 0.00 1.00CG_INDEX 188 0.53 1.00 0.50 0.00 1.00NAREIT_FFO 188 2.18 2.00 1.72 -5.40 11.87SIZE 188 7.61 7.48 0.90 5.62 9.87LEV 188 60.98 63.45 14.04 11.20 87.20GROWTH 188 8.08 4.62 16.29 -27.10 146.80CEO_OWN 188 0.90 0.50 1.68 0.00 16.51CEO_DIR 188 0.93 1.00 0.26 0.00 1.00CEO_TENURE 188 8.67 7.00 5.09 1 22

This table reports sample statistics for 188 firm-year observations for 2006-2009. BONUS is the total bonus awarded to CEO in a particular year. DACC is signed discretionary accruals calculated using the modified Jones model. FFOMANI is the percentage deviation of the actual FFO from the NAREIT definition of FFO. D_NAREIT is a dummy variable equal to one if the firm discloses the use of the NAREIT definition of FFO in reporting its alternative performance measure, zero otherwise. D_RECON is a dummy variable equal to one if the firm reports a reconciliation schedule between net income and FFO, zero otherwise. D_FFO is a dummy variable equal to one if the firm discloses its target FFO for compensation purpose, zero otherwise. D_FFOGROWTH is a dummy variable equal to one if the firm discloses its target FFO growth for compensation purpose, zero otherwise. D_WFFO is a dummy variable equal to one the firm discloses the weight assigned to FFO when setting bonus, zero otherwise. CG_INDEX is a dummy variable equal to one if the Corporate Governance Index Score reported by ISS of the company in that given year is greater than the sample median, zero otherwise. NAREIT_FFO is FFO, as defined according to the NAREIT definition, divided by average common share outstanding. SIZE is the natural log of market capitalization. LEV is the leverage ratio (percentage) of the firm. GROWTH is the one-year growth (percentage) of total assets. CEO_OWN is the percentage of shares owned by the CEO. CEO_DIR is a dummy variable equal to one if the CEO also serves as the director of the board. CEO_TENURE is the number of years that the CEO has served the firm.

TABLE 1 (Panel B) Pearson Correlations

Pearson CorrelationsVariables 9 16

1. BONUS 1.00

2. DACC 0.08 1.00

3. FFOMANI 0.17 ** -0.09 1.00

4. D_NAREIT 0.08 0.13 * 0.01 1.00

5. D_RECON -0.03 0.05 0.03 0.06 1.00

6. D_FFO -0.07 -0.02 0.00 0.08 -0.03 1.00

7. D_FFOGROWTH -0.05 0.03 0.04 0.04 -0.03 -0.18 ** 1.00

8. D_WFFO -0.11 -0.08 0.10 -0.18 ** -0.15 ** 0.38 *** -0.01 1.00

9. CG_INDEX -0.17 ** -0.01 -0.02 0.07 -0.05 0.11 0.00 0.11 1.00

10. NAREIT_FFO 0.18 ** 0.02 -0.17 ** 0.05 -0.03 0.12 * -0.10 0.11 0.09 1.00

11. SIZE 0.27 *** -0.15 ** 0.15 ** 0.05 -0.05 0.04 -0.04 0.05 0.00 0.30 *** 1.00

12. LEV 0.08 -0.13 * 0.04 -0.05 0.01 0.07 -0.12 0.07 0.07 0.24 *** 0.19 *** 1.00

13. GROWTH 0.39 *** 0.03 -0.13 * -0.05 0.05 -0.01 -0.09 -0.05 -0.02 0.19 ** 0.12 -0.03 1.00

14. CEO_OWN -0.03 0.13 * 0.01 -0.07 0.09 -0.02 -0.03 0.01 -0.05 -0.08 -0.26 *** -0.23 *** -0.06 1.00

15. CEO_DIR 0.05 0.02 0.04 -0.06 0.10 0.01 0.06 0.03 0.02 0.00 0.02 -0.02 -0.02 0.06 1.00

16. CEO_TENURE 0.01 0.09 -0.02 -0.03 0.00 0.06 -0.07 0.08 0.10 0.27 *** -0.02 0.13 * 0.05 0.27 *** -0.03 1.00

1 2 1110876543 15141312

This table reports Pearson correlations for the variables used in the regression analysis for 188 observations over 2006-2009. See Table 1 for variable definitions. *** significance at 1%, ** significance at 5%, * significance at 10% (2-sided test).

TABLE 2 The Impact of Discretionary Accruals and FFO Manipulation on CEO Bonus

BONUS = α + β1 DACC + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN + β7 CEO_DIR+ β8 CEO_TENURE + ε (1) BONUS = α + β1 FFOMANI + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN + β7 CEO_DIR+ β8 CEO_TENURE + ε (2)

Constant 433.91 *** -1297.16 * 403.29 *** -1330.51 ** 403.47 *** -1428.36 **

DACC 2009.68 2402.53 2402.13 2513.38FFOMANI 106.61 *** 99.74 ** 111.60 *** 101.99 **

NAREIT_FFO 16.16 35.90 33.66SIZE 221.21 *** 184.54 *** 197.32 ***

LEV 4.04 3.02 3.63GROWTH 18.75 19.60 * 19.48 *

CEO_OWN 11.56 13.56 10.44CEO_DIR 210.00 * 216.42 * 209.74 *

CEO_TENURE -5.09 -4.89 -6.35Property Type DummiesNR2

I II III IV VIV

188No

0.3278188Yes

0.0062188No

0.0293

YesNoYes

0.34570.03810.3368188 188 188

This table reports results of OLS regression for model specification (1) and (2). See Table 1 for variable definitions. ***, **, * indicate significance at 1%, 5%, 10% respectively (two-sided test).

TABLE X The Joint Impact of Discretionary Accruals and FFO Manipulation on CEO Bonus

BONUS = α + β1A DACC + β1B FFOMANI + β1C DACC*FFOMANI + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN + β7 CEO_DIR+ β8 CEO_TENURE + ε (X)

Constant 397.30 *** -1394.55 **

DACC 2444.61 2375.47FFOMANI 192.87 *** 189.78 ***

DACC*FFOMANI 3583.82 *** 3346.56 ***

NAREIT_FFO 41.39SIZE 180.13 ***

LEV 3.10GROWTH 19.81 *

CEO_OWN 17.63CEO_DIR 203.85 *

CEO_TENURE -4.98Property Type DummiesNR2

III

0.36250.0589188 188

YesNo

This table reports results of OLS regression for model specification (X). See Table 1 for variable definitions. ***, **, * indicate significance at 1%, 5%, 10% respectively (two-sided test).

2

TABLE 3

The Impact of FFO Manipulation and the Disclosure of the Use of NAREIT Definition on CEO Bonus

BONUS = α + β1A FFOMANI + β1B D_NAREIT + β1C D_NAREIT*FFOMANI + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN + β7 CEO_DIR+ β8 CEO_TENURE + ε (3)

Constant 206.39 *** -1609.25 **

FFOMANI 221.78 *** 215.40 ***

D_NAREIT 228.45 ** 250.95 *

D_NAREIT*FFOMANI -126.74 *** -120.94 **

NAREIT_FFO 34.96SIZE 179.11 ***

LEV 3.29GROWTH 20.07 *

CEO_OWN 18.31CEO_DIR 236.84 *

CEO_TENURE -4.03Property Type DummiesNR2

III

0.34670.0386188 188

YesNo

This table reports results of OLS regression for model specification (3). See Table 1 for variable definitions. ***, **, * indicate significance at 1%, 5%, 10% respectively (two-sided test).

3

TABLE 4

The Impact of FFO Manipulation and the Provision of Reconciliation Schedule on CEO Bonus

BONUS = α + β1A FFOMANI + β1B D_RECON + β1C D_RECON*FFOMANI + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN + β7 CEO_DIR+ β8 CEO_TENURE + ε (4)

Constant 446.54 ** -1156.74 **

FFOMANI 361.83 ** 426.09 ***

D_RECON -52.37 -138.08D_RECON*FFOMANI -258.48 * -330.76 ***

NAREIT_FFO 39.11SIZE 179.03 ***

LEV 2.98GROWTH 19.95 *

CEO_OWN 18.11CEO_DIR 230.44 *

CEO_TENURE -6.57Property Type DummiesNR2

III

0.3449188Yes

0.0329188No

This table reports results of OLS regression for model specification (4). See Table 1 for variable definitions. ***, **, * indicate significance at 1%, 5%, 10% respectively (two-sided test).

4

TABLE 5 The Impact of FFO Manipulation and Compensation Disclosure on CEO Bonus

BONUS = α + β1A FFOMANI + β1B D_FFO + β1C D_FFO*FFOMANI + β1D D_FFOGROWTH + β1E D_FFOGROWTH*FFOMANI + β1F D_WFFO + β1G D_WFFO*FFOMANI + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN + β7 CEO_DIR+ β8 CEO_TENURE + ε (5)

Constant 548.28 *** -1428.91FFOMANI 434.13 *** 428.14 ***

D_FFO -114.07 -66.63D_FFO*FFOMANI -198.89 *** -110.80 **

D_FFOGRWOTH -44.32 39.36D_FFOGROWTH*FFOMANI -138.87 *** -204.45 ***

D_WFFO -185.32 -141.17D_WFFO*FFOMANI -214.57 *** -232.31 ***

NAREIT_FFO 34.20SIZE 187.50 ***

LEV 3.86GROWTH 19.54 *

CEO_OWN 20.00CEO_DIR 232.83 **

CEO_TENURE -0.32Property Type DummiesNR2

III

1880.0681 0.3669

188YesNo

This table reports results of OLS regression for model specification (5). See Table 1 for variable definitions. ***, **, * indicate significance at 1%, 5%, 10% respectively (two-sided test).

5

TABLE 6

The Impact of FFO Manipulation and Corporate Governance on CEO Bonus

BONUS = α + β1A FFOMANI + β1B CG_INDEX + β1C CG_INDEX*FFOMANI + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN + β7 CEO_DIR+ β8 CEO_TENURE + ε (6)

Constant 568.20 *** -1067.97 *

FFOMANI 97.18 ** 84.94 *

CG_INDEX -312.97 ** -362.57 ***

CG_INDEX*FFOMANI 30.85 54.48NAREIT_FFO 43.19SIZE 171.67 ***

LEV 2.98GROWTH 19.35 *

CEO_OWN 5.50CEO_DIR 217.88 *

CEO_TENURE -0.64Property Type DummiesNR2

III

0.3736188Yes

0.0585188No

This table reports results of OLS regression for model specification (6). See Table 1 for variable definitions. ***, **, * indicate significance at 1%, 5%, 10% respectively (two-sided test).

6

TABLE 7 The Impact of FFO Manipulation and Financial Crisis on CEO Bonus

BONUS = α + β1A FFOMANI + β1B CRISIS + β1C CRISIS*FFOMANI + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN + β7 CEO_DIR+ β8 CEO_TENURE + ε (7)

Constant 536.47 *** -1269.16 **

FFOMANI 206.10 *** 222.63 ***

CRISIS -262.91 ** -67.77CRISIS*FFOMANI -135.11 *** -161.31 ***

NAREIT_FFO 33.26SIZE 169.93 ***

LEV 2.96GROWTH 19.93 *

CEO_OWN 16.70CEO_DIR 224.29 *

CEO_TENURE -2.60Property Type DummiesNR2

I II

0.3532188Yes

0.0663188No

This table reports results of OLS regression for model specification (7). See Table 1 for variable definitions. ***, **, * indicate significance at 1%, 5%, 10% respectively (two-sided test).

TABLE 8

The Impact of FFO Manipulation on Other Components of CEO Compensation

CEO_COMP = α + β1 FFOMANI + β2 NAREIT_FFO + β3 SIZE + β4 LEV + β5 GROWTH + β6 CEO_OWN + β7CEO_DIR+ β8 CEO_TENURE + ε (8)

CEO_COMPConstant -431.27 *** -4715.16 *** -2377.83 *** -257.78 -390.99 *** 452.77FFOMANI 15.00 -4.34 1.47 -5.47 1.93 145.97NAREIT_FFO 14.77 50.05 -60.41 -18.23 -11.06 -16.20SIZE 96.71 *** 630.72 *** 296.76 *** 82.51 ** 16.12 16.92LEV 3.69 *** 10.08 * 4.03 1.08 1.69 3.80GROWTH -1.49 *** -1.91 -1.07 -1.74 0.01 9.12CEO_OWN -4.93 79.00 * -23.52 -33.16 6.66 -12.58CEO_DIR 11.85 73.24 -108.81 -204.32 91.46 ** 130.54 **

CEO_TENURE 0.91 20.12 33.14 * -7.65 8.67 -15.26Property Type DummiesNR2

Salary Stock Awards Option Awards Non-Equity OtherPension Change

188Yes

0.2616188Yes

0.4425188Yes

0.1858

YesYesYes

0.13260.09020.1095188 188 188

This table reports results of OLS regression for model specification (1) and (2). See Table 1 for variable definitions. ***, **, * indicate significance at 1%, 5%, 10% respectively (two-sided test).