Financial Expertise of the board of directors in companies with small market capitalization

Tilburg University School of Economics and Management Financial Expertise of the board of directors in companies with small market capitalization Name: Anna Vorobyeva ANR: 566793 Program: MSc Finance Supervisor: Prof. Marco Da Rin Defense date: 25 September 2014

Introduction. Corporate governance plays an important role in the overall firm s performance. Many studies try to investigate what are the optimal corporate governance policies and structures to make a company operate more efficiently. Board characteristics are one of the most important aspects of corporate governance. Choosing the optimal board of directors is a key driver of the firm s success. A lot of firms regard an appropriate board of directors as a competitive advantage. Among different views regarding board composition I want to focus on the financial expertise facet. It is widely believed that the financial expertise of the board can improve the firm performance. Directors with financial background can contribute their knowledge of the project valuation and financial instruments; hence the firm is able to undertake projects that will be more beneficial for it. In addition these directors can reduce the information asymmetry problem and improve the monitoring of the firm. Moreover as it is widely believed that the expert directors improve the firm efficiency and the firm policy their appointment can affect positively the firm s reputation. A lot of studies examine the expertise factor s influence on the firm. However these works look mostly on the board of large organizations whereas I believe that financial experts can be more advantageous for small companies. The problem of information asymmetry or reputational problems are more severe in these firms as there are no much available information and financial records on it. It is harder to attract investments for small firms. Therefore the effect of the appointment of the director with financial background should be more visible in firms with small market capitalization. Also in these organizations the board size tend to be smaller than in large organizations and the impact of the appointment of a director is more visible as well. For my analysis I choose those firm characteristics, which I believe are more important for small firms. These characteristics are investment opportunities measured as investment to cash flow sensitivity and investment to Tobin s Q sensitivity; the stock performance measured as cumulative stock returns, and changes in the firm value measured as changes in Tobin s Q. My tests show that financial expert directors improve investment opportunities for firms with small market capitalization as they reduce investment to cash flow sensitivity. Hence small organizations are better off from this type of directors appointments. However the results from the stock performance and changes in the firm value should be considered with cautious. Directors with financial background seem not to improve the stock performance of the firm, though these results are from the period of the financial crisis and can be due to the fact that before the crisis these directors were allowed to take more risk. This fact could cause worse performance of the firm during and after the financial crisis. These results stay when I investigate the influence of the financial expert in the firm over the time. 2

I cannot conclude about the obvious effect of the financial expertise on the firms, however some results suggest they partially improve the firm performance. Literature overview. Existing research shows that there is a correlation between board characteristics and firm s performance. A lot of studies examine which particular characteristics have a larger effect. Some of the researchers look at the structure of the board. Yermack s (1996) study of Fortune 500 industrial firms verifies a negative correlation between firm s value and the size of a firm s board of directors, consistent with small board size being associated with good governance. A lot of studies show that a higher proportion of outside directors is connected with strong corporate governance (e.g., Weisbach (1988), Rosenstein and Wyatt (1990), Brickley, Coles, and Terry (1994), Dechow, Sloan, and Sweeney (1996), Core, Holthausen, and Larcker (1999)). Adams, Almeida and Ferreira (2005) find that the interaction between executive characteristics and organizational variables has important consequences for the firm performance. They provide evidence that the firm performance is more variable as decision- making power becomes more centralized in the hands of the CEO. Pathan (2009) examines the relation between board attributes and bank risk- taking. He finds that bank risk decreases with board size, board independence, CEO power, and CEO equity ownership. Fich and Shivdasani (2006) present evidence confirming a common view among investors and policy advocates that serving on numerous boards can result in overstretched directors that may not be effective on any board. They test this hypothesis on a panel of large U.S. industrial firms from 1989 to 1995 and find that companies with a majority of busy outside directors displays significantly lower market- to- book ratios. Also Core et al (1999) find that busy outside directors provide CEOs with excessive compensation packages, which in turn leads to weaker firm performance. Other researchers look at individual characteristics of directors. Malmendier and Tate (2005, 2008), for example, study the link of CEO characteristics to corporate decision- making. They show that overconfident CEOs do more and worse acquisitions and their investment is more sensitive to cash flow. Bertrand and Schoar (2003) examine how individual top management s skills affect the whole firm s performance and corporate policy. Their results display that older managers tend to invest less and choose lower levels of leverage. Another set of researchers focus particularly on the financial expertise of the board and its effect on the firm s characteristics. Minton, Taillard and Williamson (2010) examine the performance and risk- taking behavior of a broad sample of US financial institutions both during and prior to the financial crisis and relate them to the financial expertise levels of their independent board directors. They find that prior to the crisis outside financial experts on the board were associated with higher risk taking and slightly above- average performance. Their results are consistent with the idea that banks with more financial expertise among 3

independent directors perform worse during the crisis, particularly for large commercial banks. Also they investigate that banks with more financial experts have more leverage. McCahery and Vermeulen (2013) examine the factors and board strategies that are associated with value creation and innovation by analyzing the composition of high- performance and high- growth companies. They look at the board composition from a different prospective by introducing in their analysis a new dimension associated with value creation. Therefore they focus on the expertise, skills and capabilities that can help a director to improve the performance of a firm. In their research they test the board composition of seventy venture capital backed companies that were involved in IPOs on US stock markets between 2011 and the first half of 2012 and show that venture capitalist (independent) directors continue to play a dominant role in the further development of the recently listed companies. They conclude that talented and experienced directors bring value to the firm. Another paper by Bertrand and Schoar (2003) finds out that manager style do matter for firm s policy and affects different firm s performance characteristics. Their results show that managers with MBA degree invest more, overall are more aggressive, choose lower capital expenditures and financial leverage. Rates of returns on assets are more than 1 percentage point higher for MBA graduates. Chevalier and Ellison (1999) study whether mutual fund performance is related to characteristics of fund managers. They use a sample of 492 managers and look cross- sectionally at how performance is related to observable characteristics of the fund manager. Authors conclude that managers who attended more selective undergraduate institutions have higher performance. A study by Guner, Malmendier and Tate (2006) focuses on the effect of the financial expertise of directors. They examine whether financial experts on the board influence corporate decisions. They analyze a sample of 282 publicly traded companies from 1988 to 2001 and make a conclusion that financial experts significantly affect corporate decisions, but only when their influence serves the interest of their own institutions. The study is conducted in three steps: firstly looking at the internal investment and loan financing, secondly looking at the external financing and financing with public securities and thirdly looking at the financial expertise and CEO compensations. The outcome from the obtained research questions are that a firm displays less investment- cash flow sensitivity and obtains larger loans when commercial bankers are on the board of the particular firm. Additionally firms with financial experts on their boards undertake worse acquisitions and are associated with larger bond issues. Finally from the third research question authors show that overall the financial expertise doesn t influence much the compensation policy. Also Kroszner and Strahan (2001) investigate what determines the presence of a commercial banker on the board of a non- financial firm. According to their paper bankers tend to be on the boards of large stable firms with high tangible capital ratios and low reliance of short- term debt financing. Thus many studies have shown that financial expertise plays an important role in the performance of the firm. However most of the corporate performance studies look at samples 4

of large organizations, whereas some of them point that looking at the effect in small organizations can be useful as well. Adams and Mehran (2003, 2008) find the positive correlation between board size and organization size, and Yermack (1996) suggests that smaller board is related to better governance; hence small firms may have better governance instruments. Bertrand and Schoar (2003) look at the sample of large US companies, as it is more likely that a successful CEO will be present on the board of a mature company, however they point that the effect of individual actions can be more visible in small firms. Guner, Malmendier and Tate (2008) highlight that the financial expertise of venture capitalists (outsiders) may benefit smaller, early- staged firms due to their higher innovativeness and professionalism. Taking into consideration these issues I want to contribute to the existent research and investigate the impact of financial expertise of directors on the firm s performance, but looking at the sample of small companies. Particularly, I want to investigate how directors with financial background affect investment opportunities, stock performance, and changes in the firm value. I assume these three characteristics can show the influence of the expertise due to several factors. Carpenter and Petersen (2002) reveal that the growth of a small firm is dependent on the internal finance. Therefore I want to test whether an expert can improve investment opportunities for the small organization. A lot of research papers find evidence of correlation between cash flows and investments (Meyer and Kuh (1957), Fazzari, Hubbard, and Petersen (1988), Gilchrist and Himmelberg (1995)) and demonstrate that investment to Cash flow sensitivity can be a good measure for firm s investment opportunities. Another possibility to examine investment opportunities is looking at investment Tobin s Q sensitivity. It is so called q theory. Keynes (1936) define the idea behind it as follows: there is no sense in building up a new enterprise at a cost greater than that at which a similar existing enterprise can be purchased; whilst there is an inducement to spend on a new project what may seem an extravagant sum, if it can be floated off the stock exchange at an immediate profit (p. 151). Grunfeld (1960) argues similarly that a firm should invest when it expects investment to be profitable and that an efficient asset market s valuation of the firm captures this expectation. Tobin (1969) built on this work by using a straightforward arbitrage argument: the firm will invest if Tobin s q, the ratio of the market valuation of a firm s capital stock to its replacement value, exceeds one. It s also worth to look at the change in a firm s value, which can be a consequence of improved investment opportunities or of an appointment of a director with better knowledge of valuation techniques. The third characteristic, stock performance is widely used as an overall performance measure of the firm. Therefore in this study I want to focus on such firm performance characteristics as the investment to cash flow sensitivity, investment to Tobin s Q sensitivity, stock performance and the change in firm s value. I choose these factors because, first of all, studying investment cash 5

flow sensitivity can be applicable to my sample as there could be a problem of information asymmetries. Small firms may be not recognized as a good investment and therefore it can be harder for them to increase external financing. Thus the role of internal financing is important for small firms. Some authors also point that high investment cash flow sensitivity is a sign of financial constraints faced by the firm (Fazzari et al. (1988)). And as higher investment- cash flow sensitivity is also observed in firms that are new or small (Shin and Kim, 2002; Carpenter and Petersen, 2002a; Carpenter and Guariglia, 2008; Hovakimian and Hovakimian, 2009), it may be the case that my firms are financially constrained and are due to lack of external financing. I want to investigate whether the board expertise can reduce this sensitivity and increase investment opportunities for small companies. There are several studies focusing on investment cash flow sensitivity and board characteristics. Guner, Malmendier and Tate (2008) find that the presence of commercial banker affects investment- cash flow sensitivity. Bertrand, M. and A. Schoar, (2003) also study the influence of a manager s fixed effect on the investment- cash flow sensitivity. They build panel data set with respect to individual managers and test firm s characteristic including CEO or CFO fixed- effects. They demonstrate that manager fixed- effects add explanatory power to the models with investment- cash flow sensitivity, investment- Tobin s Q sensitivity, returns on assets, even after running placebo tests of an active role of a manager. Several authors determine Q ratio as a predictor of investment opportunities of the company that is why I focus on investment to Tobin s Q sensitivity as well. For instance, Blundell et al (1990) estimate a Q model of investments over a panel of UK companies and find that Q is a significant factor in the explanation of the company s investments, even if the effect is small. Moreover Bertrand, M. and A. Schoar (2003) in their findings show that managers who are more investment- Q sensitive are also less investment- cash flow sensitive. They find positive relation between holding MBA degree and investment- Q sensitivity and also that MBA graduates on average respond more to Tobin s Q and less to cash flow availability when deciding about capital expenditures. Stock performance is widely used as a measure of firm s performance. Moreover directors with financial background may have more knowledge about financial instruments than directors without such a background. I use Minton, Taillard and Williamson (2010) measure of stock performance, cumulative returns: an aggregate gain or loss in the stock price over the time. They find that the stock performance is negatively correlated with the fraction of financial experts in the sample of commercial banks during the crisis period. Erkens, Hung and Matos (2012) also find that independent directors in financial organizations are associated with worse stock performance during the crisis. Another paper looking at the relation between board expertise and stock performance is DeFond, Hann and Hu (2005). They examine tree- day cumulative abnormal returns around of 702 newly appointed outside directors, which have a 6

financial expertise, assigned to audit committees. They find positive and significant CARs around the appointment of accounting financial experts to the audit committee but not around the appointment of nonaccounting financial experts or directors without financial expertise. In addition, positive CARs are only found when the appointing firms have strong corporate governance. Finally I look at the changes in firm s value. Similar to Minton, Taillard and Williamson (2010) I measure the change in firm s value as the change in firm s Tobin s Q ratio. Their results suggest negative correlation between the fraction of financial experts and the change in firm s value. Looking at the effect on the firm s value can help to understand overall influence of financial expert directors. Hypothesis. In my study I want to examine if financial experts on boards affect the performance of the company and whether this effect is positive or not. Financial experts can contribute to the firm their knowledge of financial instruments and project valuation hence the firm can benefit from their presence and increase its value; also knowledge of the financial sector can help to avoid excessive risk- taking. Moreover the appointment of the financial expert can have a positive impact on the firm s reputation. Therefore experts can influence several firm performance characteristics, such as: investment opportunities by increasing the firm s reputation and by choosing more carefully projects to undertake; stock performance by increasing firm s reputation as well, and by providing better financial instruments knowledge; firm s value can be affected by all the factors above too. There are a lot of studies looking at the presence of experts in large organizations, but for my analysis I look at the sample of micro- cap United Kingdom based firms. Market capitalization of these firms does not exceed 300 mill. pounds. Following Bertrand and Schoar (2003) argument I believe that in small organizations the effect of a financial expertise can be eliminated easier than in large companies. Small companies are young, growing companies or small businesses with corporate tradition. Therefore I assume that the influence of experts can be tracked more precisely in these firms. According to the list of firm characteristics that can be affected by the board expertise mentioned above I form three hypotheses. The first hypothesis tests the effect on the investment opportunities. In the previous section I figured two main factors that can determine investment opportunities: investment to Cash Flow sensitivity and investment to Tobin s Q sensitivity. Firstly I test the influence of the expertise on the investment to Cash flow sensitivity. I assume that in small firms the sensitivity will be high (Shin and Kim, 2002; Carpenter and Petersen, 2002a; Carpenter and Guariglia, 2008; Hovakimian and Hovakimian, 2009) and the relation between directors with financial background and investment cash flow sensitivity will be negative (Guner, Malmendier and Tate (2008), Bertrand, M. and A. Schoar, (2003)). Hence my hypothesis is that 7

the financial expertise reduces investment to cash flow sensitivity. As a next test for the first hypothesis I look at the investment Tobin s Q sensitivity. Following Bertrand, M. and A. Schoar, (2003) I suppose the increase in the explanatory power of the model when include financial experts. Therefore experts on the board improve investment opportunities measured by investment to Tobin s Q sensitivity too. The second hypothesis tests the stock performance effect. According to findings in Minton, Taillard and Williamson (2010) stock performance is negatively related to financial expertise. This is also consistent with findings in Erkens, Hung and Matos (2012). However both papers investigate the stock performance during the financial crisis. Hence I assume worse stock performance during the financial crisis and positive relation with the stock performance in the period excluding the financial crisis. As a last hypothesis I look at the change in the firm s value and follow Minton, Taillard and Williamson (2010) also. They find that the presence of experts predicts changes in the firm s value negatively, but looking at the years of financial crisis. Therefore I suggest that prior to the crisis the board expertise had a positive effect and after it had a negative effect. Sample and Data. In order to examine the impact of the financial experts on boards I construct the panel data set for United Kingdom based firms across 10 years. Descriptive statistics shows that average tenure of a director is 10 years therefore I assume that 10 years of observations should be enough to eliminate possible endogeneity problems concerning an appointment of the particular director to correct the corporate policy. Furthermore to correct this possible concern I exclude first years of an appointment of a financial expert by indicating an expert s dummy as zero in this year. I analyze companies with micro- cap, that means companies with market capitalization less than 300 mill. pounds. These companies are not traded in large amounts and considered to be risky investments. A period of 10 years also will reduce the effect of the high volatility of these firms. I examine a sample of publicly traded UK companies from 2002 to 2012. I am interested in companies with market capitalization equal or less than 300 mill. pounds. To be included in my sample a firm should appear in the Financial Times Ranking of biggest 500 UK companies at least twice with market capitalization equal or smaller than 300 mill. pounds. Overall the sample consists of 118 companies. I hand- collect biographical information on all board members of these companies using annual proxy statements. I code each outside director according to the following categories used in Guner, Malmendier and Tate (2008): (1) commercial banker (a director who is currently or previously employed by the commercial bank); (2) investment banker (a director who is currently or previously employed by the investment bank); (3) executive of a non- bank financial institution; (4) CFO (current or retired finance executive); (5) Accountant; (6) Finance professor; 8

(7) executive of a non- financial firm; (8) lawyer; (9) consultant; (10) non- corporate worker (politics, non- executive, academia career); (11) other (a director who doesn t fall in any of these categories). Some directors appear to have not only one qualification, thus one director can fall into different categories. I compile financial data on these companies using COMPUSTAT Global Annual, as a proxy I take items used in Guner, Malmendier and Tate (2008). They measure capital as property, plants, and equipment (item 8), investment as capital expenditures (item 128), cash flow as earnings before extraordinary items (item 18) plus depreciation (item 14), and normalize investment and cash flow by lagged capital. Tobin s Q is the market value of assets normalized by total assets (item 6), where market value is total assets plus market equity (number of shares outstanding multiplied by closing price of the share) minus book equity (item 60). Return on assets (ROA) is income divided by the average of current and lagged total assets, where income is earnings before extraordinary items (item 18) plus interest expense (item 15) plus income statement deferred taxes (item 50). Return on equity (ROE) is net income (item 172) scaled by the average of current and lagged book equity. I exclude companies with negative Net Income to eliminate non- profitable firms. Book leverage is defined as long term debt (item 9) plus debt in current liabilities (item 34), divided by long- term debt plus debt in current liabilities plus stockholders equity (item 216). Market leverage is long term debt plus debt in current liabilities, divided by the market value of assets. Board size is natural logarithm of the total number of directors and Independence is a fraction of independent directors: number of independent directors divided by board size. Also I use Altman s z- score as an index of bankruptcy likelihood. It s calculated as follows: (3.3 (Operating Income before depreciation Depreciation and Amortization) + Sales + 1.4 Retained Earnings + 1.2 Working Capital )/Total Assets where operating income before depreciation (item 13), depreciation and amortization (item 14), sales (item 12), retained earnings (item 36), working capital (item 121). Also I identify large acquisitions that can affect corporate governance of a company by making a ratio equal to sum of the deal divided by total assets, plotting a graph allows to distinguish outliers. I record financial expert dummies as 0 in the first years of the financial experts appointment. After excluding big acquisitions, financial firms, first directors years, missing data and companies with negative net income resulting sample contains 108 companies. 9

Table 1. Summary Statistics. The sample period is 2002-2012. The Table provides summary statistics for firm- years. Firm characteristics are from Compustat Annual: Total assets are total assets (6), Capital is property, plant, and equipment (8), Investment is capital expenditures (128), scaled by lagged Capital, Cash flow is earnings before extraordinary items (18) plus depreciation (14), scaled by lagged Capital. Tobin s Q is the market value of assets over total assets (6), where market value is total assets plus market equity (25*closing price of the stock, downloaded from Orbis database) minus book equity (60). Return on Assets (ROA) is income ((18+15+50) divided by the average of current and lagged total assets. Return on Equity (ROE) is net income (172) scaled by the average of current and lagged book equity. Altman s z- score is defined as 3.3 times the difference in operating income before depreciation (13) and depreciation and amortization (14) plus sales (12) plus 1.4 times retained earnings (36) plus 1.2 times working capital (121), divided by total assets (6). Book leverage is interest- bearing debt (9+34) divided by operating assets (9+34+216). Market leverage is interest- bearing debt divided by the market value of assets. Board size is the number of directors. Fraction of independent directors is the ratio of outside directors over board size. Female is the dummy variable indicating one if the female director is present on the board and zero otherwise. Insider: CEO, CFO, Accountant are dummy variables indicating if at least one director hold one of these positions. Outsider: Commercial banker, Investment banker, Executive of non- bank financial company, CFO, Accountant, Finance professor, Executive of a non- financial company, Lawyer, Consultant, Non- corporate are dummy variables indicating if at least one director has the corresponding career. Moreover one director can have more than one qualification. Variable Obs Mean Median Std.Dev. Insider CEO 728 0,977 1 0,151 CFO 728 0,894 1 0,308 Accountant 728 0,027 0 0,164 Outsider Commercial?Banker 728 0,095 0 0,293 Investment?Banker 728 0,078 0 0,269 Executive?of?nonEbank?financial?company 728 0,221 0 0,415 CFO 728 0,323 0 0,468 Accountant 728 0,411 0 0,492 Finance?Professor 728 0,040 0 0,196 Executive?of?a?nonEfinancial?company 728 0,883 1 0,321 Lawyer 728 0,048 0 0,214 Consultant 728 0,077 0 0,267 NonEcorporate 728 0,254 0 0,436 Age 728 54,321 54,250 4,444 Tenure 728 10,205 9,845 3,355 Female 728 0,309 0 0,465 Number?of?other?Directorships 728 40,023 36,938 26,053 Board?size 728 6,945 7,000 1,913 Fraction?of?independent?Directors 727 0,465 0,429 0,234 Firm?characteristics Total?assets?( M) 728 253,330 164,700 314,721 Capital?( M) 728 63,860 23,752 144,256 Investment?( M) 619 0,505 0,200 1,950 Cash?Flow?( M) 619 1,930 0,610 4,203 Q?lagged 619 1,252 1,085 0,789 ROA 619 0,076 0,066 0,052 ROE 619 0,250 0,132 2,353 Book?leverage?( M) 728 0,272 0,214 0,295 Market?leverage?( M) 728 0,173 0,126 0,183 Altman's?zEscore 728 2,029 1,935 1,229 10

Data is transformed into firm- years. 9% and 7% of firms have commercial or investment banker on the board respectively. Other outsiders with financial background make up: 21% of companies have executives of non- bank financial companies, 31% current or former finance executives, 38% accountants, 4% finance professors. Average age is 54 years and average tenure is 10 years. Female directors are present in 30% of firms. Average board size is 7. Following Guner, Malmendier and Tate (2008) I split the sample into firms with and without financial experts, and separately for each type of the expert. Subsample with financial experts has more assets, capital and tend to invest more. Cash flows are higher as well, however Tobin s Q is slightly smaller. The ratio of returns on assets is the same for both subsamples, but the ratio of returns on equity is larger for the one without financial experts. Firms with expertise on the board have slightly more leverage. Altman s z- score is relatively equal for both samples. When I divide my sample according to each type of the expert I have following results. First subsamples with and without commercial bankers on the board show that firms with bankers are bigger in terms of assets, generate higher Cash Flows, have significantly higher rate of return on equity. Firms without commercial bankers on the board tend to invest more, have higher levels of book and market leverage. However average Tobin s Q, Altman s z- score and ROA are almost the same. Companies with investment bankers on the board have more assets, though invest and generate Cash Flows less than organizations without investment bankers. For ROA, ROE, Q and z- score ratios the pattern stays the same. Subsamples with executives of non- bank financial firms and accountants show the same statistics as the one with commercial bankers, except the fact that firms with outside accountants invest slightly more and with financial firms executives have slightly more book leverage. If the current or retired finance executive is present on the board firm has less assets, Cash Flows, lower ROA and ROE, slightly more leverage and slightly higher Q ratio. Organizations with finance professors have less assets, invest and generate Cash Flows more, however Q, ROA and ROE ratios are lower. For the last subsample there are only 29 observations, therefore for the rest of the analysis I will not use finance professor variable. Generally results from commercial bankers subsamples are similar to those in Guner, Malmendier and Tate (2008). However they found that firms with commercial bankers generate less cash flow and have higher average Q ratio and z- score. Kroszner and Strahan (2001) also found that firms with bankers have more assets and higher Tobin s Q. Interestingly the rate of return on equity is significantly higher for companies with bankers, accountants and executives of non- bank financial firms. The difference between ROA and ROE ratios are big, which can indicate that firms from my sample are highly indebted. 11

Table 2. Summary statistics on divided sample. The sample period is 2002-2012. The Table provides summary statistics for firm- years. Firm characteristics are from Compustat Annual: Total assets are total assets (6), Capital is property, plant, and equipment (8), Investment is capital expenditures (128), scaled by lagged Capital, Cash flow is earnings before extraordinary items (18) plus depreciation (14), scaled by lagged Capital. Tobin s Q is the market value of assets over total assets (6), where market value is total assets plus market equity (25*closing price of the stock, downloaded from Orbis database) minus book equity (60). Return on Assets (ROA) is income ((18+15+50) divided by the average of current and lagged total assets. Return on Equity (ROE) is net income (172) scaled by the average of current and lagged book equity. Altman s z- score is defined as 3.3 times the difference in operating income before depreciation (13) and depreciation and amortization (14) plus sales (12) plus 1.4 times retained earnings (36) plus 1.2 times working capital (121), divided by total assets (6). Book leverage is interest- bearing debt (9+34) divided by operating assets (9+34+216). Market leverage is interest- bearing debt divided by the market value of assets. Board size is the number of directors. Fraction of independent directors is the ratio of outside directors over board size. Female is the dummy variable indicating one if the female director is present on the board and zero otherwise. Insider: CEO, CFO, Accountant are dummy variables indicating if at least one director hold one of these positions. Outsider: Commercial banker, Investment banker, Executive of non- bank financial company, CFO, Accountant, Finance professor, Executive of a non- financial company, Lawyer, Consultant, Non- corporate are dummy variables indicating if at least one director has the corresponding career. Moreover one director can have more than one qualification. Financial'expert'='0 Financial'expert=1 Variable Obs Mean Median Std.Dev. Variable Obs Mean Median Std.Dev. Insider Insider CEO 231.000 0.996 1.000 0.066 CEO 497.000 0.968 1.000 0.177 CFO 231.000 0.909 1.000 0.288 CFO 497.000 0.887 1.000 0.317 Accountant 231.000 0.000 0.000 0.000 Accountant 497.000 0.040 0.000 0.197 Outsider Outsider Commercial'Banker 231.000 0.000 0.000 0.000 Commercial'Banker 497.000 0.139 0.000 0.346 Investment'Banker 231.000 0.000 0.000 0.000 Investment'Banker 497.000 0.115 0.000 0.319 Executive'of'nonGbank'financial'company 231.000 0.000 0.000 0.000 Executive'of'nonGbank'financial'company 497.000 0.324 0.000 0.468 CFO 231.000 0.000 0.000 0.000 CFO 497.000 0.473 0.000 0.500 Accountant 231.000 0.000 0.000 0.000 Accountant 497.000 0.602 1.000 0.490 Finance'Professor 231.000 0.000 0.000 0.000 Finance'Professor 497.000 0.058 0.000 0.235 Executive'of'a'nonGfinancial'company 231.000 0.883 1.000 0.322 Executive'of'a'nonGfinancial'company 497.000 0.883 1.000 0.321 Lawyer 231.000 0.004 0.000 0.066 Lawyer 497.000 0.068 0.000 0.253 Consultant 231.000 0.078 0.000 0.269 Consultant 497.000 0.076 0.000 0.266 NonGcorporate 231.000 0.247 0.000 0.432 NonGcorporate 497.000 0.258 0.000 0.438 Age 231.000 54.522 54.429 4.811 Age 497.000 54.227 54.167 4.265 Tenure 231.000 11.099 10.857 3.420 Tenure 497.000 9.789 9.571 3.245 Female 231.000 0.260 0.000 0.439 Female 497.000 0.332 0.000 0.476 Number'of'other'Directorships 231.000 37.095 28.000 30.052 Number'of'other'Directorships 497.000 41.385 39.500 23.880 Board'size 231.000 6.580 6.000 2.033 Board'size 497.000 7.115 7.000 1.833 Fraction'of'independent'Directors 231.000 0.416 0.400 0.183 Fraction'of'independent'Directors 496.000 0.488 0.500 0.252 Firm'characteristics Firm'characteristics Total'assets'( M) 231.000 217.579 141.400 268.279 Total'assets'( M) 497.000 269.947 177.822 333.065 Capital'( M) 231.000 50.627 25.951 66.540 Capital'( M) 497.000 70.010 21.300 168.311 Investment'( M) 197.000 0.340 0.205 0.491 Investment'( M) 422.000 1.055 0.196 10.124 Cash'Flow'( M) 197.000 1.494 0.550 2.541 Cash'Flow'( M) 422.000 2.360 0.761 6.669 Q'lagged 197.000 1.355 1.150 0.909 Q'lagged 422.000 1.182 1.027 0.708 ROA 197.000 0.079 0.071 0.050 ROA 422.000 0.079 0.065 0.053 ROE 197.000 0.228 0.155 0.487 ROE 422.000 0.304 0.131 2.894 Book'leverage'( M) 231.000 0.234 0.173 0.238 Book'leverage'( M) 497.000 0.290 0.248 0.316 Market'leverage'( M) 231.000 0.158 0.102 0.180 Market'leverage'( M) 497.000 0.180 0.139 0.184 Altman's'zGscore 231.000 2.016 1.925 1.258 Altman's'zGscore 497.000 2.035 1.942 1.216 12

Investment to Cash Flow sensitivity. First of all, I look how the presence of the financial experts on the board affects investment decisions. Internal investments are important in financing of small enterprises as small companies primarily have no widely available information concerning their history and financial track records. It s rather difficult for these companies to obtain external financing; therefore internal financing plays a key role. Investment to Cash flow sensitivity is one of the most important indicators of the external financial constraints of corporations. I start with OLS model similar to one in Guner, Malmendier and Tate (2008). Dependent variable is Investment, determined as capital expenditures scaled by capital. Independent variables are: Cash Flow, set of dummies for financial experts (commercial bankers, Investment bankers, executives of non- bank financial firms, CFOs, accountants), and controls. Control variables include the firm and board size, the fraction of independent directors, fixed- effects for year, firm or industry. Firm size and board size are important control variables because larger firms tend to have larger boards and hence more outside directors. I cluster standard errors at the firm level to correct for a possible correlation of errors. Columns 1 and 2 present regressions without financial experts. Cash flows predict Investments negatively. Tobin s Q predicts Investments positively, however the coefficient is not statistically significant. Firm size has positive and statistically significant coefficient in the first regression with firm fixed- effects. Board independence is positively related to Investments. The interaction between the board size and Cash Flows is positive and significant. Columns 2 and 4 show regressions with financial experts dummies. Commercial banker dummy has a positive statistically significant coefficient (in the regression with fixed firm effects) as well as finance executive dummy (in both regressions). The presence of an accountant has negative effect on investments, though statistically significant only at the 90% confidence level. Cash Flows still have a negative and statistically significant effect on Investments, however the effect in the model with firm fixed- effects is smaller, therefore the presence of the commercial banker or the finance executive on the board influence investment- cash flow sensitivity. Moreover models with financial expert s dummies have slightly higher R- squared, which means that adding these variables adds to an explanatory power of the model. Except for the Cash Flow variable all signs are consistent with findings in Guner, Malmendier and Tate (2008). Experts on the board reduce investment- cash flow sensitivity. However the interpretation can be reversed and firms with less sensitivity choose commercial bankers or finance executives; or financial experts can decline directorship in an investment- cash flow sensitive firm. Guner, Malmendier and Tate (2008) explain this in terms that organizations with higher investment- cash flow sensitivity are less healthy and try to strengthen their policy. 13

Table 3. Sensitivity of Investment to Cash Flow: Baseline Regression. OLS regressions with Investment as the dependent variable, defined as capital expenditures normalized by lagged capital. Cash Flow is earnings before extraordinary items plus depreciation, normalized by lagged capital. Tobin s Q is the (lagged) ratio of market value of assets to book value of assets. Firm size is the natural logarithm of total assets. Board size is the natural logarithm of the number of directors on the board. Board independence is the ratio of outside directors to board size. Commercial banker, Investment banker, Executive of non- bank financial company, CFO, Accountant are dummy variables indicating if at least one director has the corresponding career. Moreover one director can have more than one qualification. All regressions include year fixed effects. All standard errors are clustered by firm. 1 2 3 4 Cash)Flow.1,02***.0,801*.0,983***.0,811*** [0,467] [0,434] [0,378] [0,343] ComBanker*Cash)Flow.0,257***.0,287** [0,098] [0,138] InvBanker*Cash)Flow.0,144.0,093 [0,156] [0,177] Evecutive)of)non.bank)financial)comp*Cash)Flow.0,049 0,012 [0,145] [0,161] CFO*Cash)Flow.0,298***.0,273*** [0,092] [0,095] ComBanker 0,429* 0,240 [0,227] [0,145] InvBanker 0,043.0,145 [0,16] [0,259] Evecutive)of)non.bank)financial)comp 0,403 0,245 [0,261] [0,253] CFO 0,408** 0,3* [0,207] [0,169] Accountant.0,307*.0,368* [0,17] [0,196] Tobin's)Q 0,053 0,128 0,018 0,064 [0,097] [0,093] [0,091] [0,088] Tobin's)Q*Cash)Flow 0,023 0,021 0,040 0,043 [0,046] [0,049] [0,035] [0,036] Firm)size 0,421*** 0,114 0,412*** 0,073 [0,208] [0,139] [0,189] [0,097] Firm)size*Cash)Flow 0,044 0,022 0,087*** 0,058 [0,072] [0,071] [0,043] [0,039] Board)size.0,254.0,750*.0,465.0,588* [0,608] [0,397] [0,55] [0,334] Board)size*Cash)Flow 0,772*** 0,694* 0,716*** 0,657*** [0,323] [0,365] [0,188] [0,24] Board)independence 0,844*** 0,733* 0,948*** 0,932*** [0,306] [0,399] [0,269] [0,397] Board)independence*Cash)Flow.0,717***.0,647*.0,890***.0,809*** [0,327] [0,374] [0,332] [0,377] Year)fixed.effects yes yes yes yes Firm)fixed.effects yes no yes no Industry)fixed.effects no yes no yes Observations 618 618 618 618 Adj.)R.squared 0.4319 0.5032 0.5183 0.5826 Robust)standard)errors)are)in)brackets.)Robust)t.statistics:)*)significant)at)10%;)**)significant)at)5%;) ***)significant)at)1% 14

However the average z- score of my sample is relatively high (2,029), which suggests that on average firms are not subject to bankruptcy. Moreover my sample consists of small companies, which are considered to be more investment- cash flow sensitive (Hovakimian (2009)). Descriptive statistics showed below is in the favor of the experts effect as well: investments vary a lot within firms, however board size changes little from year to year and mean tenure is 10 years, thus the turnover of directors is low. According to the model firms in my sample have negative investment- Cash Flow sensitivity. Hovakimian (2009) in his study of the investment- Cash Flow interaction obtains firm- level estimates and classifies firms into groups according to high, low or negative investment- Cash Flow sensitivity. He found that firms with the negative investment- cash flow interaction are more likely to be constrained; these are smallest and youngest companies. These findings are consistent with the description of my sample. As can be seen from descriptive statistics below there are 488 constrained firms in my sample comparing with 239 unconstrained. I measure firms constraints using the Kaplan- Zingales (KZ) index (Kaplan and Zingales (1997)) as described below. Decrease in investment- cash flow sensitivity can signify, for instance, that the presence of financial experts reduces information asymmetry and improve monitoring. If this is the case, following Guner, Malmendier and Tate (2008), it should be more evident if firms are financially constrained. I construct the Kaplan- Zingales (KZ) index for my sample (Kaplan and Zingales (1997)). It s computed as follows: KZ!" = 1,001909 CF!" K!"!! 0,2826389 Q!" + 3,139193 Leverage!" 39,3678 Dividends K!"!! 1,314759 C!" K!"!! where CF is cash flow, K capital, Q Tobin s Q, and C cash and short- term investments. Higher KZ values indicate greater financial constraints. I use sample median of the lagged KZ index to split firm years into a constrained and an unconstrained subsample. 15

Table 4. Split- Sample Summary Statistics. The sample period is 2002-2012. The Table provides summary statistics for firm- years. Firm characteristics are from Compustat Annual: Total assets are total assets (6), Capital is property, plant, and equipment (8), Investment is capital expenditures (128), scaled by lagged Capital, Cash flow is earnings before extraordinary items (18) plus depreciation (14), scaled by lagged Capital. Tobin s Q is the market value of assets over total assets (6), where market value is total assets plus market equity (25*closing price of the stock, downloaded from Orbis database) minus book equity (60). Return on Assets (ROA) is income ((18+15+50) divided by the average of current and lagged total assets. Return on Equity (ROE) is net income (172) scaled by the average of current and lagged book equity. Altman s z- score is defined as 3.3 times the difference in operating income before depreciation (13) and depreciation and amortization (14) plus sales (12) plus 1.4 times retained earnings (36) plus 1.2 times working capital (121), divided by total assets (6). Book leverage is interest- bearing debt (9+34) divided by operating assets (9+34+216). Market leverage is interest- bearing debt divided by the market value of assets. Board size is the number of directors. Fraction of independent directors is the ratio of outside directors over board size. Female is the dummy variable indicating one if the female director is present on the board and zero otherwise. Insider: CEO, CFO, Accountant are dummy variables indicating if at least one director hold one of these positions. Outsider: Commercial banker, Investment banker, Executive of non- bank financial company, CFO, Accountant, Finance professor, Executive of a non- financial company, Lawyer, Consultant, Non- corporate are dummy variables indicating if at least one director has the corresponding career. Moreover one director can have more than one qualification. Unconstrained Constrained Variable Obs Mean Median Std.Dev. Variable Obs Mean Median Std.Dev. Insider Insider CEO 239.000 0.987 1.000 0.112 CEO 488.000 0.971 1.000 0.167 CFO 239.000 0.891 1.000 0.312 CFO 488.000 0.898 1.000 0.304 Accountant 239.000 0.013 0.000 0.112 Accountant 488.000 0.035 0.000 0.184 Outsider Outsider Commercial.Banker 239.000 0.117 0.000 0.322 Commercial.Banker 488.000 0.086 0.000 0.281 Investment.Banker 239.000 0.084 0.000 0.277 Investment.Banker 488.000 0.076 0.000 0.265 Executive.of.nonEbank.financial.company 239.000 0.163 0.000 0.370 Executive.of.nonEbank.financial.company 488.000 0.252 0.000 0.435 CFO 239.000 0.326 0.000 0.470 CFO 488.000 0.322 0.000 0.468 Accountant 239.000 0.418 0.000 0.494 Accountant 488.000 0.406 0.000 0.492 Finance.Professor 239.000 0.033 0.000 0.180 Finance.Professor 488.000 0.043 0.000 0.203 Executive.of.a.nonEfinancial.company 239.000 0.887 1.000 0.317 Executive.of.a.nonEfinancial.company 488.000 0.883 1.000 0.322 Lawyer 239.000 0.063 0.000 0.243 Lawyer 488.000 0.041 0.000 0.198 Consultant 239.000 0.071 0.000 0.258 Consultant 488.000 0.080 0.000 0.271 NonEcorporate 239.000 0.247 0.000 0.432 NonEcorporate 488.000 0.256 0.000 0.437 Age 239.000 54.458 54.429 4.801 Age 488.000 54.232 54.167 4.231 Tenure 239.000 10.748 10.429 3.110 Tenure 488.000 9.910 9.600 3.420 Female 239.000 0.297 0.000 0.467 Female 488.000 0.318 0.000 0.466 Number.of.other.Directorships 239.000 39.197 39.500 19.774 Number.of.other.Directorships 488.000 40.454 35.455 28.686 Board.size 239.000 6.686 7.000 1.652 Board.size 488.000 7.078 7.000 2.016 Fraction.of.independent.Directors 239.000 0.457 0.429 0.179 Fraction.of.independent.Directors 487.000 0.469 0.444 0.258 Firm.characteristics Firm.characteristics Total.assets.( M) 239.000 179.308 125.300 174.861 Total.assets.( M) 488.000 289.934 187.292 358.925 Capital.( M) 239.000 19.138 10.600 27.898 Capital.( M) 488.000 85.792 44.554 170.940 Investment.( M) 239.000 0.629 0.297 1.314 Investment.( M) 379.000 0.425 0.154 2.260 Cash.Flow.( M) 239.000 3.340 1.563 5.253 Cash.Flow.( M) 379.000 1.042 0.338 3.075 Q.lagged 239.000 1.386 1.153 0.889 Q.lagged 379.000 1.167 1.024 0.708 ROA 239.000 0.097 0.085 0.065 ROA 379.000 0.063 0.060 0.037 ROE 239.000 0.440 0.171 3.756 ROE 379.000 0.131 0.113 0.357 Book.leverage.( M) 239.000 0.179 0.093 0.216 Book.leverage.( M) 488.000 0.318 0.289 0.317 Market.leverage.( M) 239.000 0.106 0.045 0.143 Market.leverage.( M) 488.000 0.206 0.169 0.192 Altman's.zEscore 239.000 2.499 2.474 1.179 Altman's.zEscore 488.000 1.799 1.737 1.189 16

In constrained subsample average assets and capital are larger than in unconstrained subsample. The same pattern can be found in Guner, Malmendier and Tate (2008). Cash Flow is significantly smaller in constrained firms, consistent with Guner, Malmendier and Tate (2008). The level of leverage is higher for constrained firms. Interestingly there is no significant difference in the presence of financial experts in constrained and unconstrained firms, except for executives of non- bank financial companies, who are present more in constrained firms than in unconstrained (24% and 16% respectively). However small difference is present and, for instance, as well as in Guner, Malmendier and Tate (2008) commercial and investment bankers prefer unconstrained firms (11,7% and 8,6% versus 8,4% and 7,6% respectively). I present OLS regression similar to Guner, Malmendier and Tate (2008) with investments measured as capital expenditures scaled by capital as a dependent variable. I use the same set of controls as in the previous regression, but test the model for constrained and unconstrained firms separately. Results of regressions for two subsamples show that commercial bankers affect constrained firms more (coefficient is positive, higher and statistically significant), unlike in Guner, Malmendier and Tate (2008) where commercial banker dummy and its interaction with cash flows have no significant effect in constrained sample. The rest of financial expert s variables do not have significant results. However interaction of a non- bank financial executive dummy with Cash Flows is positive and statistically significant as well as the interaction between Commercial banker variable and cash flow. These results confirm the hypothesis that financial experts (commercial bankers) influence corporate governance mechanism in a positive way. This also can be consistent with the idea that this effect is tracked more in small firms, as in Guner, Malmendier and Tate (2008) there is no significant result from financial experts on the constrained sample. However, using the Kaplan- Zingales (KZ) index for measuring firms constraints can be challenging. For instance, Farre Mensa and Ljungqvist (2013) argue that this index can be not the proper one to determine whether the firm is financially constrained. In their study they test five most widely used measures of financial constraints: paying dividends, having a credit rating, Whited Wu index, Hadlock- Pierce index, and the Kaplan- Zingales index. According to Farre Mensa and Ljungqvist (2013) first 4 indices identify constrained firms as younger, smaller, less profitable, and less leveraged, but they also grow faster. According to Kaplan and Zingales constrained firms are not smaller than unconstrained, have lower market- to- book ratios, they invest more in fixed effects and spend less on R&D. All indices indicate that financially constrained firms are cut out of the capital markets. However, authors find that these firms do not experience inelastic supply curves of debt or equity capital (they have regular access to the public equity and bank- loan markets). 17