University of Amsterdam
Amsterdam Business School
Master in International Finance
Master Thesis
The Impact of Board Busyness on Corporate
Performance
-Evidence from the UK
Gulimila Kurexijiang
September 2013
Acknowledgements
I would like to thank professor Chris Florackis for helpful guide, advice and suggestions during the whole process of completing my thesis. I would like to thank my parents for everything.
ABSTRACT
We examine the relationship between board busyness and corporate performance using 262 UK listed companies from the period 2002 to 2009. Board Busyness refers to the percentage of outside directors that hold multiple directorships in the board. We consider it busy if an outside director holds two or more board seats in different companies. Our finding supports a positive relationship between board busyness and Tobin’s Q, a market-based proxy of firm performance. This is in line with the expertise hypothesis, according to which board busyness is a good contributor to improving firm value as the company can get access to management expertise, external capital as well as network brought by highly skilled directors with multiple directorships. However, our result does not support any statistically significant relationship between board busyness and operating performance, as measured by ROA.
Table of Contents
1.Introduction... 5
2.Literature Review ... 9
2.1 Reputational Hypothesis ... 10
2.2 Board Busyness Hypothesis ... 11
2.3 Expertise Hypothesis ... 13
3.Data and Methodology ... 16
3.1 Sample and Data ... 16
3.2 Variables ... 17
3.3 Methodology ... 20
4.Empirical Analysis ... 23
4.1 Results of time fixed effect regression model ... 23
4.2 Discussion of regression result ... 24
5.Conclusion ... 30
References ... 32
1.Introduction
The effectiveness of corporate governance and its impact on corporate performance has been a topic of considerable controversy in the academic and business communities. Kirkpatrick (2009) pointed that the insufficient risk management and failure in corporate governance plays an important role in resulting in financial distress in the period of economic downturn. Besides, the prevalence of the bubble of high technology companies in the late 1990s also informed people with several corporate governance lessons, which leads to certain reform actions by international regulatory institutions. Jiraporn (2009) stated that the in the year of 2002, the establishment of the Sarbanes-Oxley act has raised awareness in corporate governance and warned directors to take more responsibility in monitoring the board. Therefore, the primary motivation of this paper is to investigate the impact of corporate governance on corporate performance. However, corporate governance is a wide range and is difficult to measure the relationship between the corporate governance as a whole and company performance. Therefore, instead of researching corporate governance in general, this paper focused on the impact of board busyness on corporate performance. Board busyness means the extent of the board controlled and monitored by outside directors that hold two or more board seats in different companies at the same time. As most research paper studying the same topic are concentrated on US companies, such as Fich and Shivdasani(2006) and Field, Lowry, Mkrtchyan(2002), this paper intends to fill the gap by investigating the relationship between board busyness and corporate performance in UK market.
The research question of this thesis is specifically related to three testable hypotheses on the basis of prior research field. The first one is board busyness
hypothesis, which concludes that outside directors working on multiple boards at the
same time may be too busy to monitor management effectively. According to busyness hypothesis, outside directors with multiple board seats may be lack of energy and time to serve the board well. The second one is expertise hypothesis, which is contrary to busyness hypothesis by suggesting that busy directors can contribute more on the company value by bringing and sharing expertise and other resources accumulated through their experience based on the board seats of different companies. The last one is reputational hypothesis, Fama and Jensen (1983) contends that multiple board appointments can be seen as a mirror of superior director quality. According to reputational hypothesis, busy outside directors deserve more board seats than others because they once helped their former employer to achieve a superior performance and hence got a better reputation as an outside director. As long as the market for directors is related to prior corporate performance, firm value appreciation can generate additional offers of board employment. Following Kaplanand Reishus (1990), Vafeas (1999) states busy outside directors can be viewed as high quality monitors as they have reputation, which can be seen as a kind of capital for the firms they served. Moreover, services on multiple directorships can provide a director with different aspects of monitoring experience. Overall, this paper mainly focused on the test of busyness hypothesis and expertise hypothesis as we investigated the impact of multiple directors on corporate value rather than the influence of prior firm value on
board busyness.
The objective of this paper is to find out whether the advantage of multiple directorships outweighs its cost. In doing so, we test the relationship between Tobin’s Q and proxy for board busyness, controlling other variables. Our paper relates to the study by Ferris (2003), Ferris (2003) finds that for investors and shareholders, having busy outside directors in the board is a good signal, implying that the enhanced experience or reputations of such directors is beneficial to firm performance. However, Fich and Shivdasani (2006) suggests that to the operation of certain company, hiring outside directors with various board seats come with certain costs. For instance, if most part of a board is occupied by outside directors with additional board seats, the monitor quality of a firm might reduce and is inclined to become more inefficient. Accordingly, if the multiple directorships can positively leading to a boost of corporate value, it means the premium brought by multiple directors to a company outweigh the cost company paid to an overcommitted board. Another objective of this paper is to get a new set of result concerning the data based only on UK firms and get a general clue of how the multiple board busyness impacts corporate performance in most large, mature and well-known companies in UK. Compared with prior research which this paper presents in the literature review part, this paper mainly follows the research methodology used by Fitch and Shivdasani (2006) to analyze panel data using fixed effect regression. However, compared with the model used by Fitch and Shivdasani (2006), this paper did not present firm’s age, CEO’s ownership in the board, board meetings and board interlocking models as control variables because of
the limitation on data collection. But dividends and supervisory directors are added as one of other control variables in the mode. Another distinction of this paper with Ferris, Jagannathan and A.C.Pritchard (2003) is that this paper does not focus on the test of reputational hypothesis, in which we have to use multiple directorships as a dependent variable, we did not include an event study either. Using multi variant framework plus a related event study, Ferris, Jagannathan, and A.C.Pritchard (2003) test the busyness hypothesis but find no evidence of a negative relationship between board busyness and corporate performance. This paper, despite gets a conclusion inconsistent with busyness hypothesis as well, finds essentially a positive relationship between board busyness and market-to-book ratio no matter on low level or high level of busyness, which can be called expertise hypothesis. Last but not least, the main contribution of this paper is the research objective only targets the large, mature and well-known non utility and non financial UK firms that listed on FTSE index, rather than companies under Forbes 500 or US firms.
In summary, our key finding suggests that board busyness can be seen as one of drivers to improve corporate performance. It reflects that busy outside directors, instead of undermining the monitor efficiency, is a good contributor to boost the market value of a firm. In the UK market, a high level of board busyness can be seen as a signal of upcoming good firm value. Field, Lowry, Mkrtchyan (2012) finds the expertise hypothesis applies to the IPO and ventured backed firms in the U.S, however, this paper finds the expertise hypothesis also applies to mature firms in UK. The reminder of this thesis is organized as follow: Section 2 reviews the most
recent literature relevant to the relationship between board busyness and firm performance. Section 3 describes the data, variables and methodology used to develop the research. Section 4 constructs the result of market-to-book ratio test and explains the findings using reliable source. The last part gets a final conclusion for this thesis.
2.Literature Review
The topic of multiple board directorships has already been a great deal of interests among researchers in the field of corporate governance. Adams, Hermalin, and Weisbach (2010) makes research on practical work on busy directors as a part of their study by making research on the literature related to boards of directors. Basically, the effect of board busyness on corporate performance can be divided into two major situations. The first one is reputational effect indicating that busy directors are inclined to be relatively high quality directors, their high skills and monitoring ability achieved through their prior experience serves to offset the effect of their lack of time. Therefore, major firms with large skills and good rankings on total assets, market value would like to employ those busy directors because of their good reputations and former experience. Another one is busyness hypothesis indicating that employing busy directors on a board is detrimental to the market value for a firm. However, The Geoff C and Gavin J (2006) has mentioned the benefits of multiple directorships. They mentioned the advantage may come from an individual level such as environment
scanning, information sharing or access to communication channel. The expertise brought by multiple directorships may boost the corporate value on both individual and company level, this can be called expertise hypothesis, which is contrary to the busyness hypothesis on the perspective of research result. The following part reviews some of the literature on those hypotheses.
2.1 Reputational Hypothesis
Fich and Shivdasani (2007) are more supportive of the reputational hypothesis by researching the reputation effect on the changes in the number of other directorships. In spite of no evidence of abnormal turnover from the board of firms accused of fraud, Fich and Shivdasani finds that once an outside director has involved in the board company with fraud record, there will be an obvious reduction in the number of other board appointments got by the director in future, which is a kind of reputational penalty.
Masulis and Mobbs (2013) found an interesting phenomenon that outside directors with multiple directorships would like to unequally distribute their efforts based on the directorship’s relative prestige by researching on reputation incentives in the director labor market. It demonstrates that experience independent directors are more likely to make more time and effort into those companies where they can get more prestigious directorships. From this point of view, we can know that the firm value will be obviously affected by the value of board busyness if the directors with multiple directorships value different board seats unequally. Companies will gain
more value appreciation and better operating efficiency from the busy directors who think the effort they make for the company will reward them with more reputation in future. Masulis and Mobbs found boards with greater composition by busy outside directors who value their board seats more highly are closely related to a lower possibility of forced CEO departure, greater forced CEO turnover sensitivity to performance, by studying these incentive effects of individual director at the company level.
Based on their research, there is a hint that reputational effect may affect the board composition, as more famous and well-known company may attract more independent director and get a high level of board busyness. The independent directors will then put more efforts on those companies in order to gain more reputation. In this way, Board busyness may positively influence the firm value.
Upon the study of the relationship between the element relating to the quality of monitoring of corporate management by multiple directors, represented by the number of directorships held by independent directors, Ferris and Jagannathan (2003) implies that reputation is crucial in the market for directors. However, they fail to find the negative relation predicted by the busyness hypothesis between the number of board memberships held by a director and subsequent corporate performance.
2.2 Board Busyness Hypothesis
Fich and Shivdasani(2006) suggests that there is a tendency of poor corporate governance and bad corporate management among the firms with boards relying
heavily on outside directors that sits on several boards of different companies at the same time. Using a panel of large U.S. industrial firms from 1989 to 1995, Fich and Shivdasani proved that companies with boards composed of numerous busy outside directors got significantly lower market-to-book ratios, lower operating profit margin, lower asset turnover ratio as well as lower operating return on sales. Talking about the theory behind the article, Fich and Shivdasani suggest that even though busy outside directors might be beneficial to the performance of the company because they are more experienced and reputational, there is a cost faced by firms that appoint busy outside directors. As the number of outside directors sitting on multiple board increases, boards are more likely to become distracted and monitoring quality is likely to decrease heavily. In addition, time is the one of most strict limits for busy directors, as busy directors may become ineffective monitors under pressing time constraints. Therefore, their results imply that it may not be optimal for the management of firms to choose the potential directors based on the number of other boards they sit on, since this may lead to an reduction in board management quality.
P. Jiraporn (2009), on the other hand, demonstrates the busyness hypothesis by examining the impact of directors’ busyness on board meeting attendance. The author concludes that companies with absent board seats are obviously smaller both in terms of sales as well as total assets. Besides, busy directors holding multiple directorships are more likely to be absent in the board meetings. Through a regression analysis, Jiraporn got the results that the estimated coefficient of the number of outside board seats is positive and statistically significant at the 0.01 level, from which we can know
that multiple directorships is following with the higher possibility of meeting attendance. Last but not least, Jiraporn argues that higher board busyness may exacerbate agency problem, from the perspective that a failure to attend board meeting is a kind of agency cost itself. While experienced outside directors may lead to a value increase for the firms, firms behavior also had a great effect on reputation of outside directors.
Fich and Shivdasani (2007) got the result that directors once served on company with fraud record are more inclined to lose job opportunities at firms with strict corporate governance regulation policy and most notably, their departure is following with value appreciation for these firms. The reputation hypothesis shows that outside directors will be given fewer board seats once their firms are involved in financial misconduct. More specifically, this hypothesis indicates that outside directors already involved in a fraud related company would like to spend more time monitoring this firm and also willing to give up on their other board seats so that they can safeguard their reputation in monitoring capability.
2.3 Expertise Hypothesis
While Fich and Shivdasani( 2006) find board busyness will lead to a bad corporate performance, Field, Lowry , Mkrtchya (2012) made a contrary conclusion and finds no evidence of a negative relation between board busyness and firm performance within any sample by considering the advantages offered by busy directors to the company, especially the IPO company in US. It is obvious that the newly public firms
usually have higher and more professional demands on their boards than large, mature firms, that’s why they need more busy outside directors with extra network, sills as well as access to various resources. Since newly public firms barely have experience in the aspect of navigating public markets, they are more inclined to rely heavily on their directors for guide and for advice on various aspects regarding finance, management and regulations. In this case, only directors who are more experienced and better connected are capable of navigate those young firms during the growing process, no matter how many other directorships they are holding at the same time. Another new perspective of their research is that Field, Lowry, Mkrtchyan (2012) examined the issue of busyness among the sample of venture-backed capital firms, as venture-backed IPOs are largely involved in the corporate governance of the firm. However, this thesis will only refer to the second part of Field, Lowry and Mkrtchyan (2012) in which whey study the effect of board busyness on a board sample of companies. Particularly, they regressed market-to-book ratio on board busyness interacted with a Forbes 500 dummy, and also control variables. To further investigate into the issues about whether the effects of busyness on corporate performance decline as firms become more mature and established, they also examined the relationship between board busyness and market-to-book ratio for both younger and older S&P 1500 firms.
The expertise hypothesis can be explained by the benefits brought by the multiple directorships, as multiple directorship has long been a specific issue in corporate governance, Nicholson and Kiel (2006 ) has presented the main benefits came along
with multiple directorships in their research about the relationship between corporate governance and firm performance among listed Australian firms. Busy directors are thought to be able to use their prior experience and soft skills to add value to the firms in three perspectives. First of all, they can act as a link between external shareholders and corporate management by extracting resources and obtaining support from external stakeholders critical to the firm performance. For instance, with assistance of highly networking busy directors, a firm can get access to external capital in a more attractive rate. Secondly, busy board members with different experience will be more familiar with the regulations and updated policy regarding the corporate governance, as they have connections to information channels and other environment. It is obvious that the company will advance in value appreciation by conducting business in a regulatory way. Thirdly, busy outside directors who are prestigious in certain industry or field are thought to play an important role in enhancing organizational reputations and fame. For example, there is a common fact that floating new companies are more likely to seek some experienced outside directors with established business reputations.
Furthermore, Nicholson and Kiel (2006) contends that investors should not see multiple directorships as the threat foreseen by some regulatory agencies, concerning the expertise and external resources, directors holding multiple board seats should be seen as an asset to the company and society as a whole.
3.Data and Methodology
3.1 Sample and Data
The primary sample consists of all UK –listed firms that appear in the FTSE index during the period 2002-2009. Then we impose screening criteria to exclude the companies with an industry code of 7000 and 8000, which refers to utility companies and financial companies respectively. The criteria is the same with Fitch and Shivdasani (2006) did in their paper, considering the board’s influence on the utility companies and financial companies may be limited by regulatory effects. Barclay’s announcement on 30th July, 2013 to launch 5 billion rights issue to reach the level of British regulatory requirements on leverage shows that corporate actions of financial institutions will be partially influenced by the regulatory effects rather than corporate governance. After excluding those two kinds of companies, we get a sample of 2112 observations with 262 companies. For each firm, we collect data on corporate governance characteristics and firm characteristics. Gathering from Database Board Ex, the four corporate governance characteristics are board busyness, average age of outside directors, board size and supervisory director. As for the characteristics for corporate performance, there are Tobin’s Q, ROA, cash holdings, investment, dividends and firm size. To calculate these variables, we collect data on market value, total assets, common equity, preferred equity, cash holdings and short-term investments, capital expenditure, dividends for common stockholders and preferred stock holders on DATASTREAM. Please see Table I for variable definition and item used in the appendix.
[Insert Table I about here]
After matching the variables data of corporate performance with corporate governance, and excluding the observations with non-applicable values, we get a final unbalanced panel that includes 1814 observations. In Table II there is data description for each variable. As can be seen from the table, the mean and media of board size for the 262 mature UK public listed firms is around 8, lower than the average board size (11) of 500 largest US companies represented by Fitch and Anil, Shivdasani (2006). This fact reflects that board size of mature UK firms is smaller than the board size of mature US firms, therefore, the multiple directors compose higher percentage among the board in UK. The market-to-book ratio is approximately 1.824 on average among the sample firms, indicating that the firms in our sample worth more investments in future as the market value of those companies outweigh their recorded underlying assets.
[Insert Table II about here]
3.2 Variables
This section outlines dependent and explanatory variables used in our empirical models. Detailed description for variables definitions are provided in Table I.
Firm Performance: We use market-to-book ratio, Tobin’s Q, as the proxy for firm
value to regress against board busyness and other controllable variables. Market-to- book ratio can measure both the value added by management as well as the value of
intangible assets such as future investment opportunities. Tobin's Q is the ratio between the market value and replacement value of the same physical asset. It is very common and practical to use Tobin’s Q as a representative of firm value, with considerable macroeconomic significance and usefulness, for example McConnell and Servaes (1990). In our study, we measure Tobin’s Q as the ratio of the book value of assets minus the book value of equity plus the market value of equity to the book value of assets. If Tobin's q is greater than 1.0, then the market value is greater than the value of the company's recorded assets. This suggests that the market value reflects some unmeasured or unrecorded assets of the company. High Tobin's q values encourage companies to invest more in capital because they are "worth" more than the price they paid for them. The average Tobin’s Q value of 264 companies in our sample is around 1.824. This measurement is consistent with Ferris, Jagannathan and Pritchard (2003), Fich and Shivdasani (2006) as well as most of the prior research.
Control Variables: A set of Controls are used in our empirical models to ward off
the effects of other variables on Tobin’s Q. Following prior research paper, our controllable variables are divided into Firm Characteristics (Investment, Cash Holdings, Firm Size, dividend payout) and Corporate Governance Characteristics (Board Size, SD and average age of board directors).
Board Size: We control for board size because Gilson (1990) finds that during periods of financial distress, firms are inclined to reduce board size, and Yermack (1996) documents a negative and significant association between company valuation and board size.
Board Age: we control this characteristic as directors who are older in ages might pose a retirement risk. For instances, board directors who are close to retirement age might look for opportunities to boost their current incomes to the largest extent by accepting excessive board seats. On the other hand, board directors close to retirement age might view multiple directorships as an interesting and challenging part-time job for their retirement years. The two situations above will result in different relationship between board age and firm performance, that’s why we put board age as one of our control variables.
Firm size can be a proxy for the intensity of monitoring required of the director. So we use the nature log of total asset as one of the firm variables, as bigger firm might requires a higher level of corporate management and may need the assistance of more experienced and skillful outside directors. Also, busy outside directors may need to devote much time and effort on a larger organization. Larger firms tend to have much more extensive network of outside directors than smaller companies. This view is consistent with Ferris, Jagannathan and Pritchard (2008) that multiple directorships are basically a large-firm Phenomenon.
Investment Opportunities, calculated by depreciation expense/sales, is another controllable variable for firm performance, which is the same with Fitch and Shivdasani (2006) did in their research. The rational is that firms with bigger growth opportunities are inclined to involve in new investment strategies to extend their business, with increasing firm value and better performance.
correlated with the probability of performance for firms, which is concluded by Couderc in his research about investigating the determinants and consequences of corporate cash holdings. In most cases, firms with better cash liquidity are more likely to outperform than the market, as firms with excess cash holdings are more likely than other firms to get access to corporate actions such as mergers and acquisitions, leading to value reallocation. Beside, Managers of companies that rich in cash may be less motivated to manage the company and hence undermine the management quality of the firm. Besides, Couderc implemented a bivariate probit model between cash holdings and firm’s profitability, and got a negative correlation between the two variables.
Besides, we also control the dividends in our model, as changes in dividend payout policy contain some signals regarding the earning and future capability of the firm. Miller. M and K. Rock (1985) pointed out that dividends are used by management intentionally as explicit signals about future earnings, thus an implication of the dividend signaling hypothesis is that dividends changes are positively related to future changes of firm profitability and earning, which means a firm’s stock price tends to grow when dividends for shareholder grows.
3.3 Methodology
This paper uses panel data regression in the research because of three reasons: First of all, the change of firm value and company performance involves a broader range of issues and tackles more complex problems, in this case it’s better and easier to use
panel data regression than pure time-series or cross-sectional data. Secondly, by combing cross-sectional and time-series data, one can increase the number of degrees of freedom, and thus the power of test. Thirdly, by structuring the model in an appropriate way, we can remove the impact of certain forms of omitted variable bias in regression result. In the paper I use both random model and time-fixed effect model to investigate the relationship between Tobin’s Q and board busyness.
At first, I use random effect model first because I believe differences across 192 companies may have some influence on dependent variable. Random effects assume that the entity’s error term is not correlated with the predictors which allows for time-invariant variables to play a role as explanatory variables. The result of random effect model can be found in the first regression model Q1 of Table III. As P>|z| is equals to 0.001, which is smaller than 0.05, we can see the board busyness has a significant positive effect on the firm value, and when the board busyness increased for one unit, the firm value increased by approximately 65.90%. Regarding the other control variables, the board size does not have significant impact and firm size does not have very significant impact on the firm value. However, the cash holdings, investments and dividends have positive effect on Tobin’s Q and statistically significant at 5% level.
From the random effect model we can a get general outline of the relationship between Tobin’s Q and board busyness, however, fixed effect model is more plausible when the entities in the sample effectively constitute the entire population. As our sample is made up of 162 non-financial and non- utility UK companies under the
FTSE 600 index, a fixed effect model are conducted, which is also similar to the research conducted by Fitch and Shivdasani (2006). There are two assumptions behind the fixed effect model: 1. something within the individual may impact or bias the predictor or outcome variables and we need to control for this. 2. Those time-invariant characteristics are unique to the individual and should not be correlated with other individual characteristics.
In the analysis, the fixed-effects model controls for all time-invariant differences between different companies, so the estimated coefficients of the fixed-effects models cannot be biased because of omitted time-invariant characteristics like company culture. To take the time-invariant factors such as macroeconomic condition and government policy into account, I use time-fixed effect model analysis. After adding the year dummies in the model, the model that I use in the paper would be as follows:
"𝑇𝑜𝑏𝑖𝑛′𝑠 𝑄𝑖𝑡 = 𝛽1𝐵𝑜𝑎𝑟𝑑 𝑏𝑢𝑠𝑦𝑛𝑒𝑠𝑠𝑖𝑡 + 𝛽2𝐵𝑜𝑎𝑟𝑑 𝑠𝑖𝑧𝑒𝑖𝑡 + 𝛽3𝑆𝐷𝑖𝑡 + 𝛽4𝐴𝑔𝑒𝑖𝑡 + 𝛽5𝐶𝑎𝑠ℎℎ𝑜𝑙𝑑𝑖𝑛𝑔𝑠𝑖𝑡 + 𝛽6𝐷𝑖𝑣𝑖𝑑𝑒𝑛𝑑𝑠𝑖𝑡 + 𝛽7𝐼𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡𝑠𝑖𝑡 + 𝛽8𝐹𝑖𝑟𝑚𝑠𝑖𝑧𝑒𝑖𝑡 + 𝛿2𝑇2 + ⋯ + 𝛿𝑡𝑇𝑡 + 𝑢𝑖𝑡"
The year dummy from 2002 to 2008 is used to absorb the effect specific related to each year, so that we can estimate the pure effect of board busyness. The result of the fixed effect model will be analyzed in next section. To further confirm the utilization of fixed effect model rather than random effect model, the Housman Test has been conducted to compare random effect analysis and fixed effect analysis. The null hypothesis of Hausman test is that the preferred model is random effect and the alternative hypothesis is fixed effect. The result of Housman Test can be found in
TABLE IV. It can be seen that the Prob>chi2 =0.0000<0.05, which rejects the null hypothesis and indicates the unique error (ui) are correlated with regressors. Therefore, we constructed the time-fixed effect model to control for time effects because unexpected variation or special events will affect the Tobin’s Q.
[Insert Table III about here]
4.Empirical Analysis
4.1 Results of time fixed effect regression model
To further develop our hypothesis that board busyness can positively affect corporate performance, this paper represents the result of two time fixed effect models and robust test regarding the relationship between Return on asset and the same explanatory variables. The first time fixed effect model test the relationship between low level of board busyness and market-to-book ratio, the result of which is represented as Q2 in Table IV. More specifically, the firm’s total assets capture the firm size, which simultaneously influenced a firm’s risk and its observed pattern. This model assumes that a high Tobin’s Q is indicative of good management and governance. As we have shown in the variable definition, we set the market-to-book ratios as the dependent variable. We calculate the market-to-book ratio as the market value of the firm’s equity at the end of the year plus the difference between the book value of the firm’s assets and the book value of the firm’s equity at the end of the year, divided by the book value of the firm’s assets at the end of the year. This calculation closely follows that of Smith and Watts (1992). The regressions control for corporate
governance and financial characteristics likely to affect firm performance. By controlling the board size, firm size, investment, dividends, cash holdings, average age of board size, the fixed effect specification is employed to control for unobservable attributes, such as company’s history, culture, and product mix, which potentially affect firm performance. By adding year dummy variables in our model, we capture time variation as well.
Furthermore, to test the existence of nonlinearity in the data and to check whether the impact turns from positive to negative from low level of busyness to high level of busyness, this paper add additional set of time fixed effect regression model with an additional explanatory variable, board busyness * board busyness in the model, the result of which is represented as Q3 in Table IV. Last but not least, similar to the previous research paper, this thesis also conducted a robust test of the relationship between ROA and board busyness as the last regression model, result of which is represented as Q4 in Table IV.
[Insert Table IV about here]
4.2 Discussion of regression result
As the model of Q1 is explained in Section 3.3, this part mainly uses the model Q2 and Q3 to explain our result. The R-square of both Q2 and Q3 is approximately 30%, it means our explanatory variables as a whole influence 30% of changes happened on Tobin’s Q and our models are valid in panel data analysis, despite a little bit lower
than the R-square in the research of Fitch and Shivdasani (2003) ,which is around 38%.
From the result of time fixed effect Q2 and Q3, we can see that at low level of board busyness, the relationship between board busyness and Tobin’s Q is coefficient equals to 0.59 with a p value equals to 0.005, which means at a 5% significant level, when the board busyness increased by 1%, the Tobin’s Q increased by 59%. From this result, we can see that as low level of board busyness, board busyness has positive and statistically significant impact on corporate performance. However, at high level of busyness, the coefficient is 1.017 with p value equals to 0.029, which means the positive impact of board busyness on corporate performance does not turn to negative. Those two results indicate that among the public UK firms under the FTSE index, the board busyness always have a positive impact on corporate performance and does not turn to negative at high level of board busyness. This result is consistent with expertise hypothesis. Besides, this result can not sufficiently prove reputational hypothesis cause we are studying the impact of board busyness on corporate performance, rather than studying whether a good firm performance can attract busy board directors.
As the relationship between board size and market-to-book ratio, we find a non-significant negative result (coefficient=-0.00693, p-value=0.733 in Q2, coefficient=-0.0096, p-value= 0.693 in Q3). This result, while not statistically significant, is consistent with Yermack (1995), which finds an inverse relationship between board size and firm value based on evidence from 452 US firms during the
period 1985-1992. Yermack (1995) stated that firms with smaller board can provide better executive compensation and achieve better operating efficiency. As our result is insignificant between board size and market-to- book-ratio, we failed to get any strong evidence regarding the effect of board size on market-to-book ratio. However, we do find a negative and statistically significant impact of board busyness on ROA (coefficient= -0.60196, p value=0.002), which is consistent with Yermack (1995). After controlling the board size, we still get a positive relationship between board busyness and firm value, it probably means compared with a large board with overcommitted directors, most UK firms prefer busy outside directors with external benefits and contributes.
Regarding the relationship between other control variable and Tobin’s Q, we failed to find a statistically significant relationship between Tobin’s Q and board age (coefficient= -0.00953, p-value= 0.225). It reveals that it is not age of directors, but expertise of director matters in the board. The interesting part is that we did not find any relationship between SD (supervisory director) and Tobin’s Q at the low level of board busyness (coefficient=-0.77489, P-value=0.01), which turns to positive at high level of board busyness (coefficient=-0,7895, p-value= 0.009). This finding implies that supervisory director plays a more important role in monitoring the board activity and further improving firm valuation at higher level of board busyness, which is consistent with Florackis(2006), which also finds a positive relationship between non-executive director and corporate performance. The company characteristics (cash holdings, dividends, investment and firm size), on the other hand, show positive and
statistically significant relationship with Tobin’s Q. Consistent with the firms in prior research, mature UK firms with high cash volume are more likely to outperform, as firms with excess cash holdings are more likely than other firms to engage in acquisition, leading to a value destruction. Also, the result of dividends and corporate performance (coefficient=13.9, p-value=0) meets our expectation that higher dividends for the shareholders can better predict upcoming stock price increasing and value appreciation. Based on our evidence, the increasing in shareholders’ dividends is positively related to corporate performance and earnings among the largest UK firms in our sample. Last but not least, this result shows a positive impact of investment opportunities (depreciation expense/sales) on corporate performance (coefficient= 3.8, p-value=0), which is consistent with Fitch and Shivdasani (2006) and meets our prediction that investment opportunities plays an important role in boosting firm value.
Obviously, this time fixed effect model in this paper failed to indicate the busyness hypothesis and got opposite result with Fitch and Shivdasani(2007), whose test result shows a significant negative relationship between busy board and company performance. Our result, based on a totally different sample, indicates that firms are not harmed by executives with substantial external commitments, most notably, it shows firms with multiple directors are more highly valued as measured by the market-to-book ratio. This result is consistent with Miwa and Ramseyer (2000), which contends that presence of directors holding multiple board seats was strongly related to firm performance. Our test result confirms the expertise hypothesis that
outside directors with multiple board seats are good contributors to the corporate performance. In the model of ROA, the coefficient for the busy board indicator variable is positive but not statistically significant(0.8723711, p-value= 0.668), therefore, based our sample representing the UK mature firms from the period 2002-2009, we failed to find any relationship between board busyness and Return on Asset.
Several plausible explanations could account for the positive association between board busyness and firm value. We think the reason for expertise hypothesis is that advantages of multiple directors to most UK firms outweigh the cost for those firms to hire busy directors. First of all, the advantages of a board with high board busyness may come on a personal level. For instance, after sitting on the board seats of different companies, outside directors can easily get access to external resources and hence will make it much easier for the company they serve to get a reduction in the costs on coordinating and planning resources .The efficient sharing of information can boost the corporate value. It is common that those outside directors are more inclined to serve as directors on well-known and reputational companies, therefore, the expertise and resources they get from their previous experience also can apply to the next position. Similarly, directors connecting to strategically related firms are more likely to provide better advice and guide, which is positively related to firm performance. Although our panel data analysis cannot test reputational effect, some theories behind the reputational effect can be used to explain expertise hypothesis. For example, as the corporate performance record achieved by an outside director will
become a step stone for his/her next position, he will work harder to pursue a better performance, no matter how many seat he has to work on. Also according to the reputational effect, multiple directorships benefit firms by increasing a firm’s legitimacy, since a company’s reputation is linked to that of its board.
Secondly, according to Nicholson and Kiel (2006), busy outside directors are thought to be able to use their multiple links to contribute to value appreciation for the company. The busy directors in well-known companies tend to take board seats only on companies in big names because of reputational effect, therefore, they tend to have more experience and expertise on monitoring the management of big firms and can easily get access to major capital sources or network that might add value to the growth of a company. Thirdly, the well-experience busy directors will get the board to access to capital, such as a new fund or external contacts. It also has a positive effect on entire corporate system because of the dissemination of innovation through network.
At last, the positive impact of board busyness on corporate performance reflects the features of corporate governance in UK Markets. Based on our result, the UK market values the board busyness, as the board busyness is beneficial to improving the efficiency of corporate governance, rather than bring cost to the company. In accordance with conclusion got by the Abdullah and Page(2009) about corporate governance and corporate performance of UK FTSE 350 companies, the strategic decisions making of companies appreciate multiple directorships in a mature system of governance regulation like the one in UK.
5.Conclusion
Following the research methodology of Fitch and Shivdasani(2007), this paper conducted a time fixed effect regression to investigate the impact of board busyness on corporate performance based on a sample of 1814 observations of 262 FTSE listed companies during the period 2002-2009. The proxy for corporate performance is Tobin’s Q, which is consistent with the measurement used by Ferris, Jagannathan and Pritchard (2003), Fich and Shivdasani (2006), etc. The independent variables are board busyness and control variables including board size, board age, supervisory directorships, investment, cash holdings, firm size, dividend as well as year dummy variables. The result of the regression shows that the board busyness has a positive and statistically significant impact on corporate performance. To further research whether the impact will turn to negative at high level of board busyness, this paper added another set of time fixed effect regression with additional independent variable board busyness*board busyness. The result indicates that the impact did not turn to negative at high level of busyness, suggesting that among the UK firms we study, the impact of board busyness is always a good contributor of firm valuation. This finding is inconsistent with busyness hypothesis proved by Fitch and Shivdasani(2007) , while consistent with expertise hypothesis, which states that the benefit of busy outside director brought to the firm outweigh its cost. At last, this paper also conducted a regression between ROA and board busyness and found no evidence of the relationship between them.
value by stating that busy outside directors can reduce the costs of coordination and resource planning by sharing information and links, as the efficient sharing of information can boost the corporate value. In addition, independent directors sitting on the boards of different companies at the same time tend to have more experience and expertise on monitoring the management of big firms and can easily get access to major capital sources or network that might add value to the growth of a company. At last, based on our result, the UK market values the board busyness, as the board busyness is beneficial to improving the efficiency of corporate governance, rather than bring cost to the company.
In summary, this thesis contributes to the topic of relationship of corporate governance and firm value by studying on the data of FTSE listed firms, as the corporate governance and board composition is increasingly important research area since the financial crisis. However, this paper also has its limitations since it only researches the time period 2002-2009 because of the difficulties of data matching between the firm variables and corporate governance variables. The difference between this paper and prior research is that we failed to find any sufficient relationship between ROA and board busyness, although we find a positive relationship between board busyness and Tobin’s at both low and high level of busyness.
References
Abdullah, A. and Page, M., 2009, Corporate Governance and Corporate Performance: UK FTSE 350 Companies. The Institute of Chartered Accountants of Scotland.
Abrahamsen, S. and Balchen, T.W., 2010, Do Dividends Predict Future Firm Performance?
BI Norwegian School of Management – Thesis.
Adams, R.B., Hermalin, B.E., and Weisbach, M.S., 2010, The role of boards of directors in corporate governance: a conceptual framework and survey. Journal of Economic Literature 2010, 48:1, 58-107.
Couderc, N., 2005, Corporate Cash Holdings: financial determinants and consequences. JEL
Classification: C33, G32.
Eliezer M. Fich and Anil Shivdasani, 2007, Financial fraud, director reputation, and shareholder wealth. Journal of Financial Economics 86,306–336.
Fich, E.M. and Shivdasani, A, 2006, Are Busy Boards Effective Monitors? The Journal of
Finance, VOL. LXI, NO. 2.
Field, L. Lowry, M. and Mkrtchyan, A., 2012, Are busy boards detrimental? Journal of
Financial Economics.
Florackis, C., 2005, Internal corporate governance mechanisms and corporate performance: evidence for UK firms. Applied Financial Economics Letters, 1:4, 211-216.
Harris, I.C. & Shimizu, K., 2004, Too busy to serve? An examination of the influence of overboarded directors. Journal of Management Studies, 41(5), 775-798.
Jiraporna, P. Wallace N. Davidson III, DaDalt, P. and Ning, Y.X., 2009, Too busy to show up? An analysis of directors’ absences. The Quarterly Review of Economics and Finance 49, 1159–1171.
Kale, J.R. Kini, O. and Payne, J, D., 2012, The Dividend Initiation Decision of Newly Public Firms: Some Evidence on Signaling with Dividends. Journal of Financial and Quantitative
Analysis, Vol.47, No.2.
Kiel, Geoff C. and Nicholson, Gavin J, 2006, Multiple directorships and corporate performance in Australian listed companies. Corporate governance: An International Review
(http://eprints.qut.edu.au)14(6) :pp. 530-546.
Kirkpatrick, G., 2009, The corporate governance lessons from the financial crisis. OECD 2009, ISSN 1995-2864.
Miller, M. and K. Rock, 1985, Dividend Policy under Asymmetric Information, Journal of
Finance 40, 1031-1051.
Miwa, Y. and Ramseyer, J.M., 2000, The value of prominent directors: Lessons in corporate governance from transitional Japan. Harvard law school, John M. Olin
program in Law, Economics & Business, Discussion paper 267.
Smith, C. W. Jr., Watts, R. L., 1992, The investment opportunity set and corporate financing, dividend and compensation policies. Journal of Financial Economics, 32, 263-292.
Yermack, D., 1996, Higher market valuation of companies with a small board of directors. Journal of Financial Economics 40, 185-211.
Appendix
Table I Definition of variables about firm characteristics
Table I Variable Definitions
Variable Name Definition Data Item Used
Tobin's Q
Ratio of the book value of assets minus the book value of equity plus the market value of equity to the book value of assets
DataStream items: MV,
WC03501, WC03451, WC02999
Dividends The ratio of total dividends to total assets
DataStream items: WC18192, WC01701, WC02999
Investments The ratio of capital expenditures to total assets
DataStream items: WC04601,WC02999
Cash Holdings The ratio of cash holdings to total assets
DataStream items: WC02001,WC02999
Firm Size The natural logarithm of the total sales
DataStream items: WC01001
Table II Data Description
Table II Data Description
Panel A provides descriptive statistics for characteristics of our sample firms. The sample consists of 1814 annual observations for 262 companies between 2002 and 2009. Companies are included in the sample if they are listed on FTSE 600 index and are not in the financial or utility industries. The table presents the mean, median, min, max and standard deviation for each variable.
Panel A
Variable Median Mean Min Max St.Deviation Corporate Governance Variable Board busyness 0.2 0.24051134 0 0.857143 0.1781297 Board Size 8 8.391952 3 20 2.577034 Age 54.6667 54.58388 37 68.3333 4.040508 SD 0.6 0.592471 0 1 0.135884 Firm Performance Variable
Tobin's Q 1.474727 1.824404 0.341106 28.69467 1.536103 ROA 7.26 7.44231 -124.2 90.45 10.92991 Operating Margin 0.051576 -0.04847 -180.583 2.016722 4.252394 Cash Holdings 0.079092 0.118456 0 0.834359 0.119212 Dividends 0.024142 0.032861 0 0.475674 0.037063 Investments 0.03938 0.052602 0 0.600957 0.050772 Firm Size 13.17818 13.31904 5.783825 19.3323 1.726204
Table III Result of Hausman Test
TABLE III Hausman Test
This table presents the running results of Housman Test. The null hypothesis of Hausman test is that the preferred model is random effect and the alternative hypothesis is fixed effect. As Prob>chi2 =0.0000<0.05, we reject the null hypothesis and confirm the upcoming utilization of fixed effect model.
Hausman Test Coefficents
Fixed(b) Random(B) Difference(b-B) S.E. board busyness 0.59057 0.42272 0.16785 0.07935 board size -0.00693 0.00657 -0.01351 0.01011 sd -0.77489 -0.76381 -0.01108 0.13222 age -0.00953 -0.00794 -0.00159 0.00145 cash holdings 1.34270 2.00413 -0.66143 0.14299 investments 3.82989 3.13285 0.69704 0.35850 dividends 13.99856 16.71930 -2.72074 0.36410 firm size 0.15807 -0.01822 0.17629 0.05073 year 2002 0.24475 0.11920 0.12555 0.03035 year 2003 0.19017 0.07184 0.11834 0.02543 year 2004 0.38611 0.26857 0.11754 0.01956 year 2005 0.41960 0.31796 0.10164 0.01268 year 2006 0.63797 0.57253 0.06545 0.00474 year 2007 0.57737 0.52842 0.04895 tear 2008 0.49664 0.04065 0.09016 Prob>chi2= 0
Table IV Results of market-to-book ratio tests
Table IV Market-to-book Ratio Test
This table presents fixed effects regressions of firm performance and busy outside directors. The first regressions use market-to-book ratio as the dependent variable. We calculate the market-to-book ratio as the market value of the firm’s equity at the end of the year plus the difference between the book value of the firm’s assets and the book value of the firm’s equity at the end of the year, divided by the book value of the firm’s assets at the end of the year. This calculation closely follows that of Smith and Watts (1992). Regression Q1 refers to the result of random effect model. Regression Q2 and Q3 refers to time effect model with year dummy variables. The difference between Q2 and Q3 is that Q3 add one more variable board busyness*board busyness to test the existence of nonlinearity. Regression ROA refers to the robust test of market-to-book ratio test and the dependent variable is Return on asset on the model of ROA. The sample is described in Panel A of Table II. We report White (1980) heteroskedasticity-robust p-values in parentheses below each coefficient estimate.
Panel B
Variable Q1 Q2 Q3 ROA
Corporate Governance Variable
Board busyness 0.42272 0.59057 1.01767 0.8724 (0.031) (0.005) (0.029) (0.668) Board busyness*Board busyness -0.73313
(0.303) Board Size 0..00657 -0.00693 -0.0096 -0.60196 (0.709) (0.733) (0.639) (0.002) Age -0.00794 -0.00953 -0.0094 -0.0788 (0.225) (0.156) (0.16) (0.222) SD -0.76382 -0.77489 -0.78957 -1.223 (0.005) (0.01) (0.009) (0.674) Firm Performance Variable
Cash Holdings 2.00413 1.3427 1.3441 15.969 (0) (0) (0) (0) Dividends 16.7193 13.999 13.9496 96.829 (0) (0) (0) (0) Investments 3.13285 3.82989 3.8082 -9.695781 (0) (0 (0) (0.195) Firm Size -0.01822 0.15807 0.15414 1.7713 (0.612) (0.011) (0.013) (0.003) Dummy Variable year2002 0.244755 0.24947 0.1333 (0.011) (0.009) (0.885) year2003 0.1901723 0.19267 0.39877 (0.038) (0.035) (0.65) year2004 0.3861066 0.38954 1.993
(0) (0) (0.019) year2005 0.419599 0.4233 2.3996 (0) (0) (0.003) year2006 0.63797 0.64102 3.0141 (0) (0) (0) year2007 0.57737 0.57955 4.25325 (0) (0) (0) year2008 0.04966 0.052993 2.2134 (0.525) (0.498) (0.003) R-square 0.4186 0.2863 0.294 0.2492
