The influence of corporate governance on REITs performance:
The role of the financial crisis
Master thesis
Name: M.J. Hendrix (Margot)
Student no: 5776457
Field: Finance & Real Estate Finance
Supervisor: Dhr. dr. M.I. Dröes
2.
Statement of Originality
This document is written by Student Margot Hendrix who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3. Abstract
Corporate governance and firm performance are extensively studied topics in the finance literature. The real estate market played a major role in causing the financial crisis and the crisis has ensured that new attention is paid to corporate governance. This thesis combines these two important elements by investigating the relationship between corporate governance and the performance of US equity REITs before versus after the crisis. In this thesis, 49 REITs are observed for the time period 2000-2013. To investigate whether or not the financial crisis influenced the relationship between corporate governance and REIT performance, three different research methods are used. The first research method investigates the average effect of corporate governance on REITS operating performance, measured by FFO. To measure corporate governance an own constructed corporate governance index is used. When controlling for several fixed effects, there are no significant results found for the existence of the relationship and the crisis effect. The second research method is a differences-in-differences research method which allows an ultimate test to analyze how high corporate governance groups and low corporate governance groups performed during the crisis. The results in this thesis points towards no statistically significant effect of the financial crisis in the relationship between corporate governance and REIT operating performance. The last research method concentrates on the REITs equity prices. No significant results are found that an investment strategy based on buying a portfolio containing high governance REITS and selling the low governance REITs portfolio results in significant outperformance.
4. Table of contents 1. Introduction ... 5 2. Literature review ... 7 2.1. Corporate governance ... 7 2.2. REITs ... 8
2.3. Turndowns and corporate governance ... 10
3. Data ... 11
3.1. Sample and sample period ... 11
3.2. Corporate governance measure ... 12
3.3. REITs performance data and control variables ... 13
4. Research method ... 16
4.1. Corporate governance and operating performance ... 16
4.2. Differences-in-differences ... 17
4.3. Corporate governance and equity prices ... 18
5. Results and robustness check ... 19
5.1. Results ... 19
5.2. Robustness check ... 22
6. Conclusion ... 24
References ... 27
5. 1. Introduction
The financial crisis started with the collapse of housing market in 2007 and the sub-prime mortgage crisis in the US. Kirckpatrick (2009) states in his research that the bad quality of corporate governance arrangements have had a significant contribution to the financial crisis. Financial institutions were able to develop complex instruments where the risk was incorrectly modeled and the credit ranking agencies have failed to exercise their role. This lack of transparency in the securitized debt products for real estate investments played an important role in the failure of corporate governance. Due to the failure of corporate governance mechanisms, the recent financial crisis highlighted the importance of high corporate governance. Given the addressed attention to the failure of corporate governance and the role of real estate during the financial crisis, it would be interesting to combine these topics.
Corporate governance and firm performance are extensively researched topics in the finance literature but there is little research done on the influence of corporate governance on firm performance during the period of the recent financial crisis. There are studies conducted on the role of corporate governance on firm performance during the Asian crisis in 1997. The papers of Lemmon and Lins (2003) and Mitton (2002) show that corporate governance indeed influenced firms financial performance during economic turndowns. The study of Erkens, Hung and Matos(2012) concentrates on the recent financial crisis but only investigates banking firms. They find that firms with more independent boards and higher institutional ownership were hit harder during the financial crisis. These banking firms are different from other companies: they can rely on the government support in recession times, which in turn could lead to even more excessive risk taking behavior. An important reference point for this thesis is the study of Bauer, Einzholtz and Kok (2010). They investigate the role of corporate governance in the real estate market and find that corporate governance did not contribute to REIT performance in the US. The study is conducted on a time sample that did not include the crisis period. It is interesting to see whether these results are maintained or have changed during the financial crisis. In particular, we would expect that low corporate governance results in lower financial performance during the financial crisis. Since the studies of Lemmon et al. (2003) and Mitton (2002) concentrate on the Asian crisis and Erkens et al. (2012) only investigates banking firms, there is a gap to fill in to investigate how REITs corporate governance structures contribute to performance in the financial crisis.
The aim of this thesis is to investigate whether or not the influence of corporate governance on firm performance in Real Estate Investment Trusts (REITs) has changed during the financial crisis, which is
6. summarized in the following research question: To what extent does the global financial crisis influence the relationship between corporate governance on the REITs performance?
To answer this research question a sample of 49 US equity REITs are observed for the time period 2000-2013. A corporate governance index is constructed and subsequently used to measure the corporate governance levels of the 49 US REITs. This corporate governance index contains 16 provisions, based on literature about governance indices and available data from the ISS database. Three research methods are used to investigate the relationship between the corporate governance index and firms performance. The performance is separated in the operating performance (FFO) and the performance of equity prices (the stock returns). The first and second research method concentrates on the operating performance, investigating if high governance results in more efficient operations. First, using an OLS regression model, the average relationship between corporate governance, firm performance and the effect during the crisis is studied. The second research method uses an differences-in-differences method. This method allows to really test if corporate governance does matter in the crisis period. There are two groups conducted, the high and low governance groups, based on their corporate governance index score in the period before the crisis. The performance of these two groups are analyzed in the period during the financial crisis. The third research method investigates if an investment strategy that buys a portfolios containing high corporate governance REITs and selling a low corporate governance shows abnormal returns. The high governance portfolio contains of REITs that have a high corporate governance score before the crisis and their observations during the crisis. The low governance portfolio contains of the REITs with low corporate governance scores before the crisis and their observations during the crisis.
In this thesis there are no significant results found for the existence of the average relationship between corporate governance, firm operating performance and the crisis effect. For the equity prices, there are no significant results found that an investment strategy based on buying a portfolio containing high governance REITS and selling the low governance REITs portfolio results in significant outperformance. These findings are in line with the paper of Bauer et al. (2010). With the differences-in-differences research method unexpected significant results are found. The low corporate governance group outperformed the control group with 29.5% during the crisis. The high corporate governance group underperformed the control group with 13.5% in this period. However, these results turned out to be not significant after a robustness check.
The remaining of this thesis is structured as follows: the next chapter shows a literature review about the corporate governance mechanisms and indices, their impact on firm performance and gives more insight about the corporate governance structure in REITs. Secondly, there is a literature
7. section about the corporate governance in economic turndowns. The third section describes the used data, the construction of the corporate governance index and shows some descriptive statistics. Section 4 explains the three different research methods used in this thesis and section 5 shows the results of these research methods and includes a robustness check. Last, this thesis ends with the conclusion and discussion of this research.
2. Literature review 2.1. Corporate governance
Larcker and Tayan (2011, p. 9) define corporate governance as “the collection of control mechanisms that an organization adopts to prevent or dissuade potentially self-interested managers from engaging in activities detrimental to the welfare of shareholders and stakeholders.” In particular, for listed firms the ownership and the management of the firm are separated. Fama and Jensen (1983) conclude that this separation could lead to differences in interest between the owners and the management (agency problem). There are several corporate governance mechanisms that try to align the management’s interest with the interest of the shareholders and these mechanism should reduce the moral hazard problem. In the literature a broad distinction is made between internal and external corporate governance mechanisms. The internal corporate governance includes board of directors, executive compensation, ownership structure, several monitoring committees and that the CEO is not the chairman of the board. The external corporate governance consists of the mechanisms for the market for corporate control (take over) and legal infrastructure (Chong-en Bai, Liu, Lu, Song & Zhang, 2004; Bianco, Ghosh & Sirmans, 2007).
2.1.1. Corporate governance and firm performance
The existing literature investigates different practices of corporate governance and shows different results on the firm performance and corporate governance relationship. Renders, Gaeremynck & Secru (2010) state that well-functioning and implemented corporate governance practices can reduce agency problems and therefore they expect a positive relationship between corporate governance ratings and firm performance. Earlier studies focus on specific governance mechanisms, such as stock ownership, board composition and CEO compensation have yielded inconclusive results. La Porta, Lopez-de-Silanes, Sheifler & Vishny (2002) investigate how shareholder protection, measured by the origin of countries laws and an index of legal rules, differs between countries and the effect of protection on firm valuation. They find that firms in countries with high minority shareholder protection face higher firm valuations. The controlling shareholders of the firms are able to monitor and control managers and the expropriate the minority shareholders. When these
8. minority shareholders are protected by law, the expropriation is limited which results in higher firm valuation. The investors recognize this better protection which increases the willingness to pay more and provide financing.
Klapper and Love (2004) investigate the influence of corporate governance and its role in the legal environment in different countries. They conclude that the firms country’s legal environment does matter in such a way that this legal environment has a positive relation to the degree of firm-level governance: the overall average of firm-level governance is higher in countries with better legal protection. The other relationship that they find, is that higher corporate governance positively influences operating performance and firm valuation (measured by Tobin’s Q). Besides, Klapper and Love suggest that higher firm-level governance, compared to the countries average, has greater influence in countries with weak legal protection.
Another very often cited paper in the finance literature that investigates the relationship between corporate governance and firm performance is from Gomper, Ishii & Metrick (2003). The authors measure the quality of corporate governance by calculating an index (the G-index) that reflects the level of shareholder rights for the time period 1990-1999. They find an abnormal return of 8.5% per year for investing in a portfolio that holds more firm’s shares with strong shareholder rights and less firm’s shares with weak shareholder rights. Also the firm valuation (measured by Tobin’s q) decreases with 11.4 percentage-point with a one-point difference in the G-index. Where Gomper et al. (2003) investigate the relationship based on 24 provisions of shareholder rights, Bebuck, Cohen and Ferrel (2009) focus their research on a set of six provisions and built their E-index to study this relationship. They find the same relationship as Gomper et al. (2003): investing in lower E-index (better shareholder protection) shares results in abnormal returns of 7%. However, they don’t find evidence that the other 18 provisions used in the Gomper et al. (2003) contribute to this relationship. Contrary, Core, Guay & Rusticus (2006) conclude that there is no evidence for the existence of the relationship, indicating that this relationship is sample specific. In their study they use more recent data.
2.2. REITs
Real Estate Investment Trusts (REITs) are companies that own and/or finance commercial real estate. There is a distinction between non-listed privately traded REITs and the listed publicity traded REITs. A REIT owns and manages a portfolio of real estate properties. The individual REITs are able to distinguish themselves by specialization in property type and geographically. Most REITs specialize in a single property type but some hold portfolios consisting of different property types. The same holds for the geographically specialization, concentrating on a single region, state or country while
9. others diversify by investing in different regions and even different countries. Besides these specialization differences three different kinds of REITs are distinguished: equity REITs, mortgage REITs and hybrid REITs. The equity REITs own and manage real estate properties, lease out their properties and receive rents as income. The mortgage REITs concentrate on the financing part of real estate. They originate or purchase residential and commercial mortgages and residential and commercial mortgage backed securities (RMBS and CMBS) and earn their income from the interest on these financing investments. The hybrid REITs are a combination of equity and mortgage REITs1.
2.2.1. US REITs and their structure
REITs differ in their structure from other listed companies because they have to operate under certain requirements to be qualified as a REIT. They have to pay out 90% of their taxable income and in turn the shareholders pay income taxes on these dividends. Their investments must be for 75% in real estate. Another restriction is the shareholder composition: there is a minimum of 100 shareholders and the five largest shareholders are not allowed to have together a stake of 50% in a particular REIT. Next to this, REITs can differ in their management form. There are two management forms: the internally and externally advised structures (Ambrose & Linneman, 2001). In externally advised REITs there is an external third party involved that is concerned with the asset management services. This third party makes externally advised REITs more prone to conflicts of interest. Here the external REIT managers might have other incentives and interests than that of the REITs shareholders. There are studies that have investigated the relationship between REITs performance and internal versus external advised REITs and the differences in incentives. The study of Cannon and Vogt (1995) find significant results that advisor REITs (external managed) underperform the internal managed REITs (self-administered). Cappozza and Seguin (2000) find the same results: externally managed REITs underperform the internal managed with 7% per year. They explain this underperformance due to the fact that the external managed REITs have higher corporate-level expenses, in particular higher interest expenses. The external managers have an incentive to increase the assets and issuing under their management because their compensation pages are based on this. The increased use of debt could not be in the best interest for the REIT.
2.2.2. REITs, corporate governance and firm performance
Where corporate governance is excessively studied in the finance literature, there are less papers that concentrate on the corporate governance in the real estate industry. The literature section about the relationship between corporate governance and firm performance shows the mixed and inconclusive results. The same holds for the literature that addresses attention to the existence of
1
10. this relationship in REITs. The section above about the REITs structure introduces the agency problem within the different management forms of the REITs. With the introduction of the Tax Reform Act in 1986, where REITs are now allowed to be internally advised, most of the REITs changed their management form into internal. One of the main reason was to reduce the agency problem (Benefield and Pyles, 2009). However, the literature does not stop here, more recent papers have investigated the relationship between different corporate governance mechanism and firm performance in the real estate sector. In the real estate literature there is the same trend where most papers concentrate only on a single mechanism of corporate governance. Gosh and Sirmans (2003) concentrate on internal corporate governance mechanisms: board independence and ownership structure. They find that independent boards, block ownership and institutional ownership weakly increase performance. Feng, Gosh & Sirman (2005) design an index that measures governance based on board size, the number of outside directors and if the board is not chaired by the CEO. They indicate that high index scores, reflecting good governance, resulted in better performance. These results only holds for the best and worst board index scores.
The studies above concentrate on single mechanisms of corporate governance. In line with the finance literature, for the real estate industry there are papers that include an overall measure of corporate governance. The paper of Bianco et al. (2007) investigates the relationship using the G-index. This G-index is a proxy for the external corporate governance, including provisions that prevent shareholders for hostile takeovers. They conclude that hostile takeovers are very rare for REITs because of their structure and therefore there is less need in REITs to create such barriers. Bauer et al. (2010) use another measure of corporate governance, the Corporate Governance Quotient index (which incorporate both the internal and external corporate governance mechanisms). In their study they find that the corporate governance is positively influenced firm value (Tobin’s Q). However, very weak evidence is found for the existence of the relationship between corporate governance and operating performance but not for the equity prices. They indicate that the restrictions for REITs reduce the agency problems. The required 90% dividend payout reduces the free cash flows left over for managers to spend. They need to find other funding sources to finance their operations and forcing managers to disclose information. The 75% real estate related investment set some boundaries to invest in other industries (Bianco et al. 2007). These restrictions seem to act like substitutes for corporate governance mechanisms that are used for the other listed firms. They denote this as the ‘REIT-effect’ (Bauer et al., 2010).
2.3. Turndowns and corporate governance
The question that arises is whether this ‘REIT effect’ still exists in recent financial crisis where the real estate industry played a major role. Bauer et al. (2010) base their research on the time period before
11. the financial crisis. This makes it interesting to investigate if the role of corporate governance changed during the recent financial crisis in US equity REITs. There are papers that investigate the link between corporate governance and performance in economics turndowns. Mitton (2002) uses proxies for disclosure quality to measure corporate governance. He concludes that firms with higher disclosure quality show significant better stock price returns during the Asian financial crisis. In addition, Lemmon et al. (2003) examine the impact of management ownership and control rights.
In Asian firms the pyramid structure is very common. Due to these pyramid structures, the actual and control rights are separated and by this allow investors to exercise actual control with little actual ownership rights (Bebchuck, Reiner & Triantia, 2000). This separation leads to agency problems where the controlling shareholders have incentives to engage into projects in their own interest and bear less of the cost of these investments. Lemmon et al. (2003) find that during the Asian crisis, the stock returns of pyramid structure firms perform 12 percentage point worse compared to other firms. However, these pyramid structures are typical for Asian ownership structure which makes these results harder to apply to US companies and in this particular case the US REITs.
There is little literature that investigate corporate governance mechanism during the recent global financial crisis. Erkens et al. (2012) investigate the role of corporate governance and the performance of banking firms during this crisis. They base their research on financial firms in a variety of countries and find that the firms in the presence of more independent boards and higher institutional ownership were hit harder during the financial crisis. These two components of corporate governance induce more risk taking behavior and show that corporate governance has an important impact on firm performance during the financial crisis. It seems that the controlling and monitoring role of independent boards don’t work in the banking sector before the crisis. An explanation for this outcome is that independent directors favor managers to increase shareholders returns through large risk taking decisions. These shareholders find it beneficial to increase risks because they do not bear the social cost of bankruptcy. Since the social net only applies for the financial institutions, it is hard to assert these findings to the real estate sector.
3. Data
3.1. Sample and sample period
The data used in this research are obtained from three different databases. The ISS (former Risk Metrics) database is used for the corporate governance data. The REITs performance data are obtained for the SNL database and the stock return data from the CRSP Ziman REIT database. After combining these three databases and excluding outliers, this results in an unbalanced panel data
12. sample for 49 US equity REITs for the period 2000-2013. This data sample enables to investigate the crisis effect, to compare the influence of corporate governance before and during (and after) the recent financial crisis. In this research the start of the financial crisis is 2007, in line with Erkens et al. (2012).
3.2. Corporate governance measure
As mentioned in the literature review, there are different corporate governance measures. Bianco et al. (2007) highlight the importance to use a corporate governance measure that both include the internal and external mechanism. The Corporate Governance Quotient (CGQ) used in the Bauer et al. (2010) paper is an example that includes both mechanisms. In the CGQ 55 provisions are included. These 55 provisions are rated by questionnaires. Unfortunately, the CGQ index is no longer available and with the available data it is not possible to reconstruct the CGQ. Therefore, an own corporate governance index is constructed, based on the approach of Aggarwal, Erel, Stulz & Williamson (2007). This approach consists of 44 corporate governance attributes. A substantial part of Aggerwal et al.’s attributes that are used can be found in the ISS database. In Appendix 1 there is an overview of the 44 provisions used by Aggerwal et al. (2007) with an indication of which provisions are available in the ISS database. As is shown, there are some provisions not available in the database, especially the audit and compensation and ownership subcategories lack data. Table 1 gives an overview of the total provision available per subcategory. In total, 16 provisions of Aggerwal et al.’s attributes can be included from the WRDS ISS database.
Table 1
Overview number of provision per subcategory.
Subcategory
ISS
dataset Aggarwal et al. (2007)
Board 10 25
Audit 0 3
Anti-take over 6 6
Compensation & ownership 0 10
Total 16 44
To compute the corporate governance index, dummy variables are used. If the answer to the question of the attributes is ‘yes’ then the provision gets a 1-score. When the answer is ‘no’, the dummy variable gets a 0-score. A total score is calculated by summing up all the provisions scores. Notice that in this research a higher corporate governance index is positive in contrast with the construction of the G-index where higher rates reflect lower corporate governance. Figure 1 shows
13. the results of the corporate governance data per year for the period 2000-2013. Before 2007, the corporate governance data are only available for the even years.
Figure 1 shows an increase in the corporate governance mean values for the period during the crisis. Especially in the year 2007-2009 there are higher mean values for the corporate governance measure where the crisis was at its highest. Looking at the period during and after the crisis, it seems that the corporate governance is relatively stable over time, except of the year 2000. For 2002, there are no observations available for the corporate governance-measure.
Figure 1: Mean values corporate governance of US equity REITs; 2000-2013.
Note: This figure shows the corporate governance mean values per year, based on data from US equity REITs
during the period 2000-2013.
3.3. REITs performance data and control variables
As an indicator for REITs operating performance, Fund From Operations (FFO) are used in this research. Because this thesis only investigates REITs, FFO is the best performance measure that is applicable for REITs. According to NAREIT: “FFO is the most commonly accepted and reported measure of operating performance. Equal to a REIT’s net income, excluding gains or losses from sales
0 2 4 6 8 10 Me a n co rp o ra te g o ve rn a n ce 2000 2004 2006 2007 2008 2009 2010 2011 2012 2013
14. of property and adding back real estate depreciation.“2 The FFO performance measure is an indicator if the firm operates more efficient under better corporate governance ratings. Besides the firm operating performance, the influence of corporate governance on REITs equity return is investigated. All the performance and control variables data are obtained from the SNL database. In the literature section the corporate governance problem between internal and external advised REITs is explained. In this thesis, all the included REITs are internally managed. Therefore there is no dummy variable for this incorporated.
Finally, the data needed for testing the two portfolios using the four factor Carhart model are obtained from the Kenneth French Data Library. Table 2 gives an overview of the definition and calculation of the REITs performance data and the control variables.
Table 2
Overview of the used variables and their description.
Variable Description
FFO Funds from operations.
CG Corporate governance measure as discussed above.
Crisis Dummy variable crisis: value 1 the year 2007 and after; value 0 year 2006 and before. CG*Crisis Variable of interest: interaction variable corporate governance measure and crisis effect. Market-to-book Price as a percent of book value per share.
Dividend pay out ratio Dividends declared during the period as a percent of earnings per share.
Dividend yield The most recent dividend, annualized and expressed as a percent of the security's price. D/E ratio Total debt, senior and subordinated, as e multiple of equity.
Size Logarithm total assets.
ROA Return on average assets; net income as a percent of average assets.
Table 3 Panel A shows the descriptive statistics for the REITs performance data and the control variables for the whole sample for the period 2000 till 2013.
2
https://www.reit.com/investing/reit-basics/glossary-reit-terms Table 3 Panel A
Descriptive statistics US equity REITs, whole sample; 2000-2013.
Variable Mean SD Minimum Maximum Observations
FFO 324,268 301,961 32,468 1,812,227 265
ROA 3.93 2.71 0.77 29.94 265
Market-to-book 244.26 97.63 91.40 579.42 265
Dividend pay-out ratio 158.98 87.87 23.68 493.15 265
Dividend yield 4.83 1.83 0.60 11.28 265
D/E-ratio 1.33 0.76 0.04 4.06 265
15. Panel B and C in Table 3 show the descriptive statistics for the two different groups that are used in the differences-in-differences research method. In section 4 it is further explained how these two groups are conducted. When panel B and C are compared, there are higher mean values for the operating performance measures FFO and ROA for the high corporate governance group and lower values for the low governance group. The tables give some indication that the high corporate governance REITs outperform and the low corporate governance REITS show underperformance.
To get a preliminary insight about the relationship between corporate governance scores and the performance measure, a correlation matrix is conducted. This correlation matrix, Table 4, gives some indication about the expected signs in the regression outputs. This table shows that there is a positive but very weak correlation between corporate governance and FFO but this correlation is not significant. Next to this, the table also reflects that FFO and size are highly correlated. Besides, this correlation matrix gives information about possible multicollinearity problems. It turns out however that there is no indication for multicollinearity problems.
Table 3 Panel B
Descriptive statistics US equity REITs for the high corporate governance treatment group.
Variable Mean SD Minimum Maximum Observations
FFO 516,208 412,610 83,642 1,276.60 17
ROA 3.86 1.54 1.75 7.55 17
Market-to-book 258.18 75.40 152.05 404.24 17
Dividend pay-out ratio 135.62 99.87 56.88 493.15 17
Dividend yield 5.14 2.22 2.80 10.32 17
D/E-ratio 1.64 0.42 0.90 2.62 17
Size 15.88 0.87 14.34 16.93 17
Table 3 Panel C
Descriptive statistics US equity REITs for the low corporate governance treatment group.
Variable Mean SD Minimum Maximum Observations
FFO 307,336 181,518 137,132 1,030,852 27
ROA 2.42 1.00 1.07 4.76 27
Market-to-book 209.70 79.97 103.72 449.24 27
Dividend pay-out ratio 149.56 102.39 67.61 459.46 27
Dividend yield 5.66 2.22 0.67 11.05 27
D/E-ratio 1.72 0.83 0.89 3.76 27
Margot Hendrix Master thesis The influence of corporate governance on REITs performance: The role of the financial crisis
16. Table 4
Correlation matrix for the operating performance data and control variables.
CG FFO ROA Dividend pay-out Dividend yield D/E-ratio Size
CG 1.000 FFO 0.0472 1.000 ROA 0.0636 0.0728 1.000 Dividend pay-out 0.0032 -0.2067* -0.5411* 1.000 Dividend yield -0.1161 -0.2747* -0.0190 0.2435* 1.000 D/E-ratio -0.1460* -0.002 -0.3089* 0.1231* 0.0623 1.000 Size 0.0294 0.8060* -0.1954* -0.0869 -0.2066* 0.1612* 1.000 * 5% significance level 4. Research method
4.1. Corporate governance and operating performance
The research method used for this thesis follows the existing literature about corporate governance and firm performance. Bauer et al. (2010) focus on REITs performance and use a corporate governance index that include both internal and external mechanisms. Their research method is the first step in investigating the relationship in this thesis. This regression model measures the average effect of the corporate governance measure on the firm operating performance measure. The operating performance that is tested in this research is funds from operations (FFO). The reason for choosing this performance measure is that FFO is the best performance measure for REITs as explained in section 3, the data section.
The control variables included in the OLS regression are size (logarithm of total assets), market-to-book value, dividend pay-out ratio, dividend yield and debt-equity-ratio (D/E). For the variables FFO and corporate governance (CG), the logarithm values are used in the regression. The first model that is regressed in this thesis is:
(1) LnFFOit = β1CGit + β2Dcrisis+ β3CGit*Dcrisis + β4Sizeit + β5D/Eit + β6Divyieldit +
β7DivPayoutit + β8M/B-valueit + αi + λt + λt*Dindustryi + εit
To come to this final first regression, four separate steps are made. First, a model is estimated without controlling for any fixed effects. The second step controls for firm fixed effects. Because these firm fixed effects capture the industry fixed effect, the industry effects are not included in this regression. In the third step year fixed effects are added. Finally, to come to the final regression model, the interaction between industry times year effects is added to control for any industry specific time trends.
17. The included control variables are based on the literature that has investigated the relationship between mechanisms of corporate governance and firm performance. The logarithm of assets is the control variable for firm size. This control variable is used in many papers (e.g. Bauer et al. 2010; Bianco et al. 2007; Gomper et al. 2003). The other control variables are based on the research of Bianco et al. (2007). Excessive debt ratios can affect performance in a negative way. Therefore, the D/E-ratio is included. The market-to-book value is a proxy for growth and is expected to positively affect performance.
4.2. Differences-in-differences
The differences-in-differences research method provides a suitable way to see how the recent financial crisis has influenced the relationship between corporate governance and firm performance. This research method allows to investigate how REITs with low or high corporate governance scores before the crisis perform in the period during the crisis. Do the REITs with high corporate governance scores in the period before the financial crisis benefit from it during the period of financial turmoil? Or does low corporate governance harm the operating performance during this period? It is basically an extension to equation (1), because it adds a control group. This is a better way to investigate the crisis effect, where is expected that low corporate governance performs worse during the crisis. For the control group there is no effect expected. The expected effect for the best and the worst governance firms is in line with the literature (e.g. Feng, Gosh & Sirman, 2005).
To investigate this crisis effect, there are two groups observed for the two time periods. There are two ‘treatment’ groups, one for the low corporate governance and one for the high. They receive the ‘treatment’ in the crisis period. The low corporate governance ‘treatment’ group consists of REIT observations that have a corporate governance score below 5 before the crisis and have observations in the period during the crisis. This low corporate governance group contains 10% of the observations in the dataset. The high corporate governance ‘treatment’ group consists of REIT observations that have a corporate governance score above 8 before the crisis and contains less than 10% of the total observations in the used dataset. These separation limits are based on the descriptive statistics (Figure 1). The control group contains the observations that do not have a corporate governance score below 5 or above 8 in the period before the crisis. The final model for the differences-in-differences method is as follows:
(2) LnFFOit = β1 LowCGtreatmenteffect + β2HighCGtreatmenteffect + β3 Sizeit + β4D/Eit + β5Divyieldit+ β6DivPayoutit + β7M/B-valuei t+ αi + λt + λt*Dindustryi + εit
Again, the final model is build up in smaller steps. The first step investigates if there is a relationship between corporate governance and the dependent variable FFO. This regression does not include the
18. crisis effect or any fixed effects. The second step uses the ‘treatment’ variables as explained above but still not controlled for any fixed effects. The third step controls for firm fixed effects and time effects. To come to the final model, the time trend variable is added.
4.3. Corporate governance and equity prices
The two research methods above concentrate on the influence of corporate governance on the operating performance. To investigate if the stock market values corporate governance, the relationship is investigated on REITs equity prices. The Carhart (1997) four factor model is used to analyze if a particular portfolio shows abnormal returns. This analysis investigates if a high governance portfolio outperforms the low governance portfolio to establish an investment strategy that buys the high governance portfolio and sells the low governance.
The same limits as in the differences-in-differences analysis are used to separate these two portfolios: the low with a corporate governance score below 5 and the high with a score above 8.
For the period before the crisis, there are not sufficient data available to run the Carhart four factor model. As a consequence, the ’treatment’ group variables, as explained in the differences-in-differences section, are used to create this two portfolios. The high governance portfolio contains of REITs that have a high corporate governance score before the crisis and their observations during the crisis. The low governance portfolio contains of the REITs with low corporate governance scores before the crisis and their observations during the crisis. So, the portfolio construction is based on the corporate governance score before the crisis. In this regression of the Carhart model it is investigated if the portfolios with respectively low or high governance before the crisis show abnormal returns during the financial crisis.
The model specification of the Carhart model is as follows:
(3) Rt = α + β0 (Rm − Rf )t + β1 (SMB)t + β2 (HML)t + β3 (MOM)t + εt
Since the corporate governance data are annual data, all variables used in this regression are annual averages. In this model, Rt is the average annual return from the portfolio minus the average annual US T-bill rate. The annual average return includes dividends. The variable Rm is the equal weighted average annual return of the market, taken from on the Ziman CRSP database.
Here the SMB factor measures the differences between the returns of small stocks and big stock. The HML factor measures the differences between the returns of value stocks (high book to market
19. ratio) and growth stocks (low book to market ratio). The factor MOM shows the difference between the winners and losers portfolio. These data are available from the Kenneth French Data Library.3
5. Results and robustness check
5.1. Results
This section shows the results and analyses of the results from the three different research methods.
5.1.1. Corporate governance and operating performance
Results from the regression outputs of equation (1) are displayed in Table 5. This table shows the regression results based on the researched method used in the Bauer et al. (2010) paper.
Table 5
Regression output investigating the average effect of corporate governance on FFO equation (1).
The main variable of interest from these regression models is CGt*crisis: the interaction term CG times the crisis dummy variable. This coefficient measures the differences in the effect of a 1%
3
http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html
Dependent variable: Ln FFO (1) (2) (3) (4)
LnCGt -0.459*** -0.112 0.0170 0.142 (0.139) (0.0772) (0.105) (0.215) Crisis -0.429 0.132 (0.349) (0.195) LnCGt*crisis 0.255 -0.0368 -0.152 -0.391 (0.178) (0.101) (0.122) (0.277) Sizet 0.946*** 0.808*** 0.781*** 0.592*** (0.0272) (0.0526) (0.0594) (0.101) D/E-ratiot -0.230*** -0.113*** -0.101*** -0.164*** (0.0480) (0.0308) (0.0334) (0.0516) Dividend yieldt 0.0493*** 0.0189*** 0.0189** 0.0278** (0.0107) (0.00671) (0.00789) (0.0133) Dividend pay-out ratiot -0.00071*** -0.00082*** -0.00077*** -0.00067***
(0.00018) (0.00014) (0.00017) (0.00018) Market-to-book valuet 0.00226*** 0.000967*** 0.00103*** 0.00117***
(0.000273) (0.000128) (0.000174) (0.000284)
Constant -1.568*** 0.181 0.433 4.193***
(0.435) (0.791) (0.904) (1.431)
Firm FE No Yes Yes Yes
Year FE No No Yes Yes
Industry*Year FE No No No Yes
Observations 265 265 265 265
R-squared 0.946 0.795 0.807 0.879
Number of REITs 49 49 49 49
20. increase in the CG measure for the crisis period compared to the period before the crisis. This interaction term reflects whether or not corporate governance has a different effect on FFO during the crisis. In the first column, the interaction term does show the expected positive sign but this result is not significant at none of the significance levels. The LnCGt-variable does show a significant and negative relationship between corporate governance and dependent variable Ln FFO. This is contrary to the expectations. A 1% increase in the CG-score results in a decrease of 0.45% of the dependent variable FFO. This contradicting result is an indication of reverse causality; where financial performance influence the level of corporate governance. It could be the case that better operating performance induces higher corporate governance. Further, the first regression output shows the expected negative sign for the crisis variable, but this is not significant. All the control variables do show significant results. The second column controls for firm fixed effects. The interaction variable now shows a negative sign and is still not significant. Again, the control variables are significant and the signs of the coefficients are in the same direction as in the first regression output. The crisis variable has not the expected sign when controlled for firm fixed effects. The third column shows a negative sign for the interaction variable, but this is not significant. The sign of the corporate governance variable, LnCGt, is positive but not significant. The final column shows the same directions of the signs as in the third regression output. The interaction variable has the unexpected negative sign and the LnCGt – variable is positive. These coefficients are not significant. Also in this
output, the control variables are significant.
5.1.2. Impact of the financial crisis
Table 6 shows the regression outputs for the differences-in-differences research method. The first column shows a negative and significant relationship between corporate governance and FFO. A 1% increase in the CG-score results in an decrease of 0.0284% for FFO. The control variables are all significant. This model does not take into account the crisis dummy variable and is not controlled for any fixed effects. The second column incorporates the ‘treatment’ groups and the treatment effect of the crisis. Surprisingly, the low corporate governance group does show a positive and significant result. This low corporate governance group outperforms the control group, REITs that do not have CG-scores below 5 and above 8, with 29.5%. This is not the case with the coefficient of this group in the crisis period, which is reflected in the Low CG*crisis variable. This coefficient shows the expected negative sign but is not significant. The high corporate governance group shows the unexpected negative sign, the same holds for the High CG*crisis variable. Both coefficient do not show significant results. The crisis dummy variable is unexpectedly positive but not significant. Here, all the control variables are significant.
21. Table 6
Regression outputs for the differences-in-differences research method, equation (2).
Dependent variable: LnFFO (1) (2) (3) (4)
CGt -0.0284** (0.0116) Crisis 0.0276 (0.0431) Low CGt 0.295*** (0.0964) Low CGt*crisis -0.0783 0.0266 0.263*** (0.0994) (0.0551) (0.0768) High CGt -0.0764 (0.0663) High CGt*crisis -0.0515 -0.0983 -0.173*** (0.0937) (0.0908) (0.0522) Sizet 0.944*** 0.948*** 0.785*** 0.957*** (0.0275) (0.0258) (0.0619) (0.0255) D/E-ratiot -0.229*** -0.239*** -0.101*** -0.220*** (0.0498) (0.0443) (0.0336) (0.0503) Dividend yieldt 0.0509*** 0.0536*** 0.0181** 0.0590*** (0.0107) (0.0109) (0.00789) (0.0177) Dividend pay out ratiot -0.00071*** -0.00084*** -0.00077*** -0.00075***
(0.000187) (0.000165) (0.000173) (0.000267) Market-to-bookt 0.00219*** 0.00231*** 0.00100*** 0.00255***
(0.000287) (0.000244) (0.000181) (0.000342)
Constant -2.154*** -2.491*** 0.423 -2.000***
(0.428) (0.382) (0.912) (0.403)
Firm FE No No Yes Yes
Year FE No No Yes Yes
Industry*Year FE No No No Yes
Observations 265 265 265 265
R-squared 0.944 0.949 0.805 0.8192
Number of REITs 49 49 49 49
Note: clustered standard errors in parentheses. Significance levels *, **, *** are respectively at 10%, 5% and 1%. The third column only includes the two ‘treatment effect’ variables and the model is controlled for firm and time fixed effects. The coefficients of these variables do not show the expected sign, the low corporate governance coefficient is positive, the high corporate governance negative. In the last regression output the model is also controlled for time trends. Now the output shows significant results. Surprisingly, the low corporate governance group coefficient indicates that the low corporate governance group outperforms the control group during the crisis with 26.3%. An possible explanation for this result is that the low governance group has invested to improve their corporate governance during the crisis and benefits from the increased governance ratings. Also the high corporate governance group shows a significant results. The coefficient shows a unexpected negative sign. The high corporate governance group performs 17.3% worse during the crisis compared to the
22. control group. A possible explanation for these results is that the used corporate governance measure in this thesis contains data about board independence. More independent boards are associated with higher corporate governance scores. Erkens et al. (2012) find that firms with more independent boards suffered more during the crisis, due to the risk taking behavior before the crisis. Since the real estate market was booming in the period before the crisis, the more independent boards could have put more pressure to invest in more risky real estate projects, which results in a negative effect on performance during the crisis.
5.1.3. Corporate governance and equity prices
The results above concentrate on the operating performance measure. Next are the results of corporate governance on equity prices. The output uses the two portfolios that are constructed on the bases of the differences in differences group. Table 7 shows the results of the two portfolios that have high or low governance before the crisis and their performance during the crisis. From Table 7 can be seen that the portfolios do not show abnormal returns, there is no significant alpha (α). These results are in line with the Bauer et al. (2010) paper.
Table 7
Comparing the two portfolios based on their ratings before crisis.Estimation of the Carhart four factor model, equation (3). α Rm-Rf SMB HML MOM R² High CG -0.00189 0.945*** 0.0266 0.0182* 0.0278 0.997 (0.00532) (0.0419) (0.103) (0.00970) (0.0234) Low CG 0.00274 0.916*** -0.0832 0.0157 -0.0755* 0.969 (0.00989) (0.0757) (0.192) (0.0208) (0.0385) Difference -0.00463 0.029 0.1098 0.0025 0.1033 (0.00768) (0.0623) (0.1502) (0.0137) (0.0589) Note: Significance levels *, **, *** are respectively at 10%, 5% and 1%.
5.2. Robustness check
This section investigates if the results found above are solid by changing some of the characteristics of the model. In the data section is explained in detail why to choose for the FFO variable as dependent variable, since this performance measure is specific for REITs. Another often used performance indicator is ROA. The regression models that investigate the relationship between corporate governance and REITs performance are now regressed using ROA as the dependent variable. First, Table 8 shows the regression results following the Bauer et. al (2010) model. From this output can be seen that the main variable of interest, LnCGt*crisis, shows the expected positive sign
in the last column. In the regression model where FFO is used as dependent variable, the sign is negative. Here again, the interaction variable is not significant when controlled for all fixed effects.
23. When looking at the control variables, dividend yield and size do not show significant results for the last regression and the significance level of D/E ratio decreases. Comparing the R² values, this model with ROA has a much lower R² compared to the model with FFO.
Table 8
Regression output investigating the average effect of corporate governance on ROA, equation (1).
Dependent variable: ROA (1) (2) (3) (4)
LnCGt 0.934 1.507* 0.911 2.592 (0.811) (0.873) (0.929) (2.249) Crisis 1.971 1.777 (3.107) (2.204) LnCGt*crisis -0.935 -1.065 -0.698 0.218 (1.498) (1.178) (1.127) (2.207) Sizet -0.297 0.0448 -0.128 -0.0748 (0.288) (0.422) (0.645) (1.323) D/E-ratiot -1.445*** -0.713* -0.893** -1.047* (0.280) (0.371) (0.378) (0.541) Dividend yieldt 0.478*** 0.345*** 0.326** 0.243 (0.105) (0.126) (0.151) (0.225) Dividend pay-out ratiot -0.0172*** -0.0162*** -0.0161*** -0.0171***
(0.00259) (0.00270) (0.00332) (0.00445) Market-to-book valuet 0.0113*** 0.00623* 0.00504 0.00223
(0.00246) (0.00311) (0.00356) (0.00443)
Constant 6.103 0.858 4.561 -2.205
(4.039) (6.257) (9.510) (19.74)
Firm FE No Yes Yes Yes
Year FE No No Yes Yes
Industry*Year FE No No No Yes
Observations 265 265 265 265
R-squared 0.493 0.350 0.375 0.530
Number of REITs 49 49 49 49
Note: clustered standard errors in parentheses. Significance levels *, **, *** are respectively at 10%, 5% and 1%. Table 9 shows the results for the differences-in-differences research method with dependent variable ROA. In the last column of the regression outputs can be found that the low corporate governance group performed worse compared to the control group in the period during the crisis. This negative sign is expected but contradicting with the positive and significant sign in the differences-in-differences output with FFO as dependent variable. This coefficient is not significant in this output. The same is applicable for the high corporate governance group. This coefficient has the expected positive sign compared to the negative and significant sign in the FFO regression output. Also this coefficient does not show significant results.
24. Table 9
Regression outputs for the differences-in-differences research method and dependent variable ROA.
Dependent variable: ROA (1) (2) (3) (4)
CGt 0.0355 (0.0859) Crisis 0.172 (0.344) Low CGt -0.620 (0.444) Low CGt*crisis 0.422 0.482 -0.426 (0.466) (0.464) (0.436) High CGt 0.0341 (0.611) High CGt*crisis -0.278 -0.275 0.396 (0.900) (0.920) (0.492) Sizet -0.290 -0.275 -0.0950 -0.310 (0.280) (0.293) (0.658) (0.298) D/E-ratiot -1.464*** -1.425*** -0.901** -1.812*** (0.269) (0.281) (0.395) (0.389) Dividend yieldt 0.470*** 0.482*** 0.341** 0.529*** (0.102) (0.109) (0.152) (0.195) Dividend pay out ratiot -0.017*** -0.017*** -0.016*** -0.018***
(0.00235) (0.00257) (0.00340) (0.00419) Market-to-bookt 0.0114*** 0.0113*** 0.00496 0.0130***
(0.00234) (0.00248) (0.00366) (0.00282)
Constant 7.653 7.531 5.506 9.578**
(4.869) (4.640) (9.760) (4.801)
Firm FE No No Yes Yes
Year FE No No Yes Yes
Industry*Year FE No No No Yes
Observations 265 265 265 265
R-squared 0.491 0.493 0.375 0.4841
Number of REITs 49 49 49 49
Note: clustered standard errors in parentheses. Significance levels *, **, *** are respectively at 10%, 5% and 1%.
6. Conclusion
This thesis has investigated the relationship between corporate governance and US equity REITs performance. The aim of this thesis has been to study if the relationship between corporate governance and REITs performance exists and how the recent financial crisis has influenced this relationship. Past research has concentrated on this relationship in a booming real estate market. In order to study this relationship, an own constructed corporate governance measure was used to measure corporate governance, including 17 provisions.
25. To investigate the effect of the recent financial crisis, there are three different research methods used. Two of these methods concentrate on the REITs operating performance, the third on the REITs equity returns. The first research method follows the article of Bauer et al. (2010). In this thesis there are no significant results found for the existence of the relationship and the effect of the crisis on this relationship when using a standard fixed effects model. The second research method used in this thesis is the differences-in-differences method. This method assesses the influence of the financial crisis by including a control group. The differences-in-differences method shows unexpected results. The results indicate that the low corporate governance group outperforms the control group during the crisis with 29.5%. It was expected that low corporate governance harms operating performance. The high corporate governance group performed 13.5% worse compared to the control group during the crisis. It was expected that high corporate governance benefited during the crisis. However, robustness check shows that the results turned out to be not significant when using ROA as the dependent variable. The third research method used the Carhart (1997) four factor model to investigate the influence of corporate governance on equity prices. The two portfolios containing high governance and low governance REITs do not show significant abnormal returns. An investment strategy that buys a portfolios containing high corporate governance REITs and selling a low corporate governance portfolio does not result in significant outperformance. This was not in line with the Gomper et al. (2003) paper but showed same results as Bauer et al. (2010).
Based on these findings, it can be concluded that in this thesis no impact has been found between of the average effect of corporate governance on the operating performance of REITS. Also looking at equity prices, there is no relationship. However, other variables did seem to affect REITS performance such as size, dividend yield, debt/equity-ratio, dividend payout-ratio and the market to book value.
Not finding significant results for the relationship between corporate governance and REITs performance is in line with the findings of Bauer et al. (2010). They argue that the regulated REITS environment captures the effects of corporate governance. This can be underlined with the findings of Klapper et al. (2004) and La Porta et al. (2012), finding that corporate governance plays a more important role in firm valuation, especially in weakly regulated environments. Where this study only concentrates on REITs, it is not possible to conclude that corporate governance in the crisis has no effect on REITs performance due to their regulated environment. The expected positive effect of the financial crisis was based on the paper of Lemmon et al. (2003) and Erkens et al. (2012). As earlier mentioned in the literature section, these two studies have limitations to apply to the real estate sector. Since there are no other firms included in this research, the findings can’t be compared with other listed firms.
26. The results of this thesis have to be taken with care. The conclusions in this thesis are based on a small dataset with limited available (corporate governance) data which makes it hard to generalize the conclusions. The own constructed corporate governance measure lacked to fully incorporate all corporate governance aspects. The used governance index, for example, did not include the sub categories Audit and Compensation & ownership.
This research does only contain equity REITs. For further research it is interesting to also include mortgage REITs. Here the financial crisis might have played an even bigger role. Besides, this thesis investigates the US equity REITs. The reason for choosing US REITs is that the corporate governance data were only available for the US. Since the recent financial crisis has a global impact and in particular for the real estate market, further research can focus on other than US equity REITs, for example the European REITs.
27. References
Aggarwal. R., Erel. I., Stulz. R, and R.Williamson., 2007. Do U.S. firms have the best corporate governance? A cross-country examination of the relation between corporate governance and shareholder wealth. NBER Working paper Series.
Ambrose, B. W., and P. Linneman, 2001. REIT organizational structure and operating characteristics. Journal of Real Estate Research, 21(3), 141-162.
Bauer. R. , Eichholtz,P., and R. Kok., 2010. Corporate Governance and Performance: The REIT Effect. Real Estate Economics. 36 (1), 1 – 29.
Bebchuk. L.A., Cohen, A., and A. Farell, 2009. What matters in corporate governance?. The Review of Financial Studies. 22(2), 783-827.
Bebchuk, L.A., Reiner, K, and G. Triantia, 2000. Stock Pyramids , and Dual Class Equity: Mechanism of Agency Cost of Separating Control and Ownership, NBER Working paper Series.
Benefield, D.B., and M.K. Pyles, 2009. Internally versus externally advises non-brokerage real estate firms. The Journal of Alternative Investments.
Bianco. C., Ghosh, C., and C.F. Sirmans, 2007. The Impact of Corporate Governance on the Performance of REITs. The Journal of Portfolio Management. 33 (5), 175-191.
Carhart. M., 1997. On persistence of mutual fund performance. Journal of Finance. 52(1), 57-82.
Cannon, S.E., and S.C. Vogt, 1995. REITs and their management: an analysis of organizational
structure, performance and management compensation. Journal of Real Estate Research, 10 (3), 297-318.
Capozza, D.R. and P.J. Seguin, 2000. Debt, Agency and Management contract in REITs: The external advisor puzzle. Journal of Real Estate Finance and Economics, 20(2), 91-116.
Chong-En Bai. Q. Liu. J.Lu. F. Song and J.Zhang. 2004. Corporate governance and market valuation in China. Journal of Comparative Economics. 32, 599–616.
Core. J., Guay, W., and T. Rusticus, 2006. Does weak governance cause weak stock returns? An examination of firm operating performance and investors' expectations. Journal of Finance. 61, 655– 687.
Erkens. D.H., Hung, M., and P. Matos, 2012. Corporate governance in the 2007–2008 financial crisis: Evidence from financial institutions worldwide. Journal of Corporate Finance. 18(2), 389–411.
Fama. E.F. and M.C. Jensen, 1983 . Separation of Ownership and Control. .Journal of Law and Economics. 26(2), 301-325.
Feng. Z., Gosh, C., and C.F. Sirmans.2005. How important is the board of directors to REIT performance?. Journal of Real Estate Portfolio Management. 1(3), 281-293.
Gompers. P., Ishii, I., and A. Metrick, 2003. Corporate governance and equity prices. Quarterly Journal of Economics. 118, 107-155.
28. Ghosh, C., and C.F. Sirmans, 2003. Board Independence. Ownership Structure and Performance: Evidence from Real Estate Investment Trusts. Journal of Real Estate Finance and Economics. 26. 287-318.
Kenneth French Data Library:
http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html
Kirckpatrick.G., 2009. The corporate governance lessons from the financial crisis. OECD Journal: Financial Market Trends. 3
Klapper, L.F. and I. Love. 2004, Corporate governance, investor protection and performance in emerging markets, Journal of Corporate Finance, 10 (5), 703-728
La Porta, R., Lopez-de-Silanes, F., Shleifer,A., and R. Vishny, 2002. Investor Protection and Corporate Valuation, The Journal of Finance 57,(3), 1147-1170.
Larcker, D. and B. Tayan. 2011. Corporate Governance Matters. Chapter 1. 1-21.
Lemmon. M. L., and K. Lins, 2003. Ownership structure. corporate governance.
and firm value: evidence from the East Asian financial crisis. The Journal of Finance. 58(4), 1445-1468.
Mitton. T., 2002. A cross-firm analysis of the impact of corporate governance during the East Asian financial crisis. Journal of Financial Economics. 64(2). 215-241.
Reit information, retrieved from Reit.com
https://www.reit.com/investing/reit-basics/guide-equity-reits https://www.reit.com/investing/reit-basics/glossary-reit-terms
Renders. A., Garenmynck, A., and P. Secru, 2010. Corporate Governance Ratings and Company performance: A Cross- European Study. Corporate Governance: An International Review. 18(2), 87– 106.
29. Appendix 1: Overview available data in the ISS database (X = available)
Provisions are retrieved from the Aggarwal et al.. (2007) paper.
BOARD Available in ISS (WRDS) database
1. All directors attended 75% of board meetings or had a valid excuse
X
2. CEO serves on the boards of two or fewer public companies
X
3. Board is controlled by more than 50% independent outside directors
4. Board size is at greater than five but less than 16 X 5. CEO is not listed as having a related-party
transaction
6. No former CEO on the board
7. Compensation committee comprised solely of independent outsiders
X
8. Chairman and CEO are separated or there is a lead director
9. Nominating committee comprised solely of independent outsiders
X
10. Governance committee exists and met in the past year
X
11. Shareholders vote on directors selected to fill vacancies
12. Governance guidelines are publicly disclosed
13. Annually elected board (no staggered board) X 14. Policy exists on outside directorships (four or
fewer boards is the limit)
15. Shareholders have cumulative voting rights X 16. Shareholder approval is required to
increase/decrease board size
17. Majority vote requirement to amend charter/bylaws (not supermajority)
X
18. Board has the express authority to hire its own advisors
19. Performance of the board is reviewed regularly 20. Board approved succession plan in place for the CEO
21. Outside directors meet without CEO and disclose number of times met
22. Directors are required to submit resignation upon a change in job
X
23. Board cannot amend bylaws without shareholder approval or can only do so under limited
circumstances
24. Does not ignore shareholder proposal
25. Qualifies for proxy contest defenses combination points
30. 26. Consulting fees paid to auditors are less than
audit fees paid to auditors
27. Audit committee comprised solely of independent outsiders
28. Auditors ratified at most recent annual meeting* ANTI TAKE OVER
29. Single class. common X
30. Majority vote requirement to approve mergers (not supermajority)
X
31. Shareholders may call special meetings X 32. Shareholder may act by written consent X 33. Company either has no poison pill or a pill that was shareholder approved
X
34. Company is not authorized to issue blank check preferred
X
COMPENSATION & OWNERSHIP
35. Directors are subject to stock ownership requirements
36. Executives are subject to stock ownership guidelines
37. No interlocks among compensation committee members
38. Directors receive all or a portion of their fees in stock
39. All stock-incentive plans adopted with shareholder approval
40. Options grants align with company performance and reasonable burn rate
41. Company expenses stock options
42. All directors with more than one year of service own stock
43. Officers’ and directors’ stock ownership is at least 1% but not over 30% of total shares outstanding 44. Repricing is prohibited
