Small firm governance and analyst following.

Fortin, Rich ; Roth, Greg

INTRODUCTION

A number of studies have investigated the factors that influence analysts' decisions to provide coverage of a firm's stock. These studies generally find that smaller firms and firms with lower trading volume are followed by fewer analysts. (1) The evidence suggests analysts have greater incentives to cover firms that are more likely to produce higher brokerage or investment banking income for the analyst's employer. Given the important role analysts play in transmitting firm information to investors, the reluctance of analysts to follow small firms potentially exacerbates the asymmetric information problem (between managers and outside shareholders) for these firms. (2)

In this paper we extend the literature on analyst behavior by examining the relationship between corporate governance and the degree of analyst following for small U.S. firms. We posit that, when considering which small U.S. firms to follow, analysts will prefer firms with superior shareholder rights. This preference is derived from analyst incentives and the likely consequences of poor corporate governance. Analysts have a well-documented incentive to provide coverage for firms that they view favorably. (3) Firms with weak governance may be viewed unfavorably because of their perceived high agency costs and their perceived inferior information disclosure. Although prior research has examined the relationship between ownership structure and analyst following, to our knowledge this is the first study to test whether a comprehensive measure of shareholder rights is related to the number of analysts covering small U.S. firms.

We sample 365 small firms (using the same size definition as that used in creating the S&P600 Small Cap Index) that have necessary I/B/E/S, Research Insight, and corporate governance data for the year 2002. We focus on small firms for two reasons. First, small firms have particular difficulty in attracting analyst coverage. Second, analyst coverage is probably more important in reducing information asymmetry for small firms, given the limited attention these firms receive in the financial press. We focus on the year 2002 because securities firms were cutting analyst positions at that time, forcing the shrinking population of professional analysts to make difficult choices regarding which firms to follow. (4) As our measure of corporate governance, we use the corporate governance index and corporate governance data provided by Professor Andrew Metrick at the Wharton School. These data have been used in many earlier studies. (5) We regress the number of analysts following a firm on the firm's corporate governance index and we control for several firm-specific factors potentially related to analyst following.

Our main finding is that firms with superior shareholder rights are followed by a greater number of security analysts. This relationship is robust whether OLS regression or negative binomial regression is used. Although we hesitate to make strong claims regarding causality, our evidence is consistent with analysts having a preference for following firms with better corporate governance. We find additional evidence that insider ownership is negatively related to analyst coverage. Bhushan (1989), Hope (2003a), and others have also observed a negative relationship between insider ownership and analyst following. Bhushan (1989) argues that the demand for analyst coverage decreases as the ownership of insiders increases. Hope (2003 a) argues that high ownership concentration is associated with managerial incentives to lower financial disclosure quality so that analysts won't trust a firm's reported earnings. Finally, we find that firm size, trading volume, and revenue growth are positively related to analyst coverage, whereas share price momentum is negatively related to analyst coverage.

LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT

Researchers have long been interested in corporate governance and the distribution of firm decision rights between managers and shareholders. (6) In this study we test the hypothesis that small firms with better corporate governance attract greater analyst coverage. Analysts have two strong incentives to follow firms with superior shareholder rights. First, analysts usually follow firms that they can favorably recommend (with a "buy" recommendation), so they may avoid firms with poor governance that are likely to suffer high costs of shareholder-management conflict. McNichols and O'Brien (1997) find that analysts initiate coverage of firms that they favorably recommend and discontinue coverage of firms that they have recently downgraded. Evidence provided by Gompers, Ishii, and Metrick (2003) and others suggest firms with weak governance underperform. (7) A second reason analysts have for avoiding firms with poor governance is these firms are more likely to produce incomplete or misleading financial disclosure. Analysts' careers depend, at least in part, on the quality of their earnings forecasts and investment recommendations. This dependency should lead analysts to cover firms whose managers make more credible information disclosures. Using a sample of firms sued in federal securities class actions, Griffin (2003) finds evidence that analysts decrease coverage of firms following corrective disclosures. His results suggest that analysts avoid covering firms when information quality is problematic or management credibility is in doubt. Lang, Lins, and Miller (2004) argue that firms with poor corporate governance are more likely to suffer high agency costs and are more likely to manipulate or withhold important firm information. They sample non-U.S. firms and find that concentrated family or managerial control of a firm (an indicator of poor governance) is associated with reduced analyst following.

DATA AND METHODOLOGY

We gather an initial sample of small U.S. firms from Research Insight for the year 2002. Our definition of "small" is the same as that used by Standard & Poor's for the purpose of creating the S&P600 Small Cap Index. The firms in this index have a total market value of equity ranging from $300 million up to $1.5 billion. From this initial sample we eliminate all exchange traded funds, all closed end funds, and all non-U.S. firms.

We also eliminate all firms for which information is unavailable regarding the number of analysts following the firm and the firm's corporate governance index. Analyst coverage data were drawn from I/B/E/S. The number of analysts following the firm is defined as the number of analysts providing annual earnings forecasts for the firm. The firm's corporate governance index value is drawn from Professor Andrew Metrick's website at the Wharton School. The construction of this index is described in detail in Gompers, Ishii, and Metrick (2003) and, as noted, these data have been used in many earlier studies. The original data for the index comes from publications of the Investor Responsibility Research Center (IRRC). The IRRC provides information on 24 different corporate governance provisions which Gompers, et al. (2003) divide into five categories: tactics for delaying hostile bidders; voting rights; director/officer protection; other takeover defenses and state laws. Gompers, et al. (2003) construct the index by adding one point for every provision that decreases shareholder rights. Thus, as the value of the governance index increases, a firm's shareholder rights (or corporate governance quality) decreases. For the year 2002, Professor Metrick provides the index values for 1,894 firms on his website.

To control for factors (other than corporate governance) that might influence analysts' incentives to follow firms, we also gather data on the following variables: the percentage of all board members who are insiders (8); the percentage of outstanding shares owned by inside board members; firm size; trading volume; price-book ratio; share price momentum; and revenue growth. The motivation underlying inclusion of these control variables is discussed in the results section. Data concerning board composition and inside board member ownership are drawn from proxy statements. All remaining data are drawn from Research Insight. After eliminating firms that lack data on any of the above variables, our final sample consists of 365 small U.S. firms. In some regressions we include the number of industry segments in which the firm operates. Data for this variable are also drawn from Research Insight and inclusion of this variable restricts some model specifications to a 312 firm sample.

To test the relationship between analyst following and corporate governance, we regress the number of analysts following the firm on the governance index and the control variables mentioned above. Rock, et al. (2000) specifically advise that negative binomial regression be used when estimating cross-sectional, analyst-following regressions. However, Bhushan (1989) and Barth, et al. (2001) use standard OLS for their regressions of analyst following. We provide results using standard OLS regression and negative binomial regression. Our main conclusion regarding analyst following and corporate governance is unaffected by this choice.

RESULTS

Summary statistics for all variables appear in Table 1. As noted, we selected a sample of firms from Research Insight such that they meet the size restrictions of firms appearing in the S&P600 Small Cap Index. The 365 firms in our final sample range in size from $302 million to $1.498 billion and have a mean size of $766.5 million. The average number of analysts following sampled firms is 9, with a range from 0 to 35. The governance index variable ranges from 3 (indicating superior shareholder rights) to 18 (indicating inferior shareholder rights).

we control for several factors that might influence analysts' incentives to cover firms. "Firm Size" is the total market value of equity. As noted, many earlier studies find that larger firms have greater analyst following. "Trading volume" is the number of shares of the firm's stock traded in the year 2002. We include this control variable because Barth, et al. (2001) and Jegadeesh, et al. (2004) find evidence that highly traded stocks, which generate greater income for brokerage firms, attract greater analyst following. We also include "Momentum" (one-year raw stock returns), one-year revenue growth, and the price-book ratio as indicators of high growth or "glamour stocks." Prior researchers, such as Jegadeesh, et al. (2004), find that sell-side analysts have a bias in favor of glamour stocks. "Business Segments" is the number of major business segments in which the firm operates, as defined by Research Insight. We include this as a proxy variable for firm complexity, which Bhushan (1989) finds is negatively related to analyst following.

To estimate the relationship between the corporate governance quality and analyst following, Although the corporate governance quality measure created by Gompers, et al. (2003) is very broad, it does not include information on ownership structure or board composition. Accordingly, we include the control variables "insider ownership" and "board insiders." Insider ownership is the percentage of total shares outstanding held by inside board members. Board insiders is the percentage of board members who are full time employees of the firm. Bhushan (1989) and Hope (2003a) find a negative relationship between insider ownership and analyst following. This can be explained by a reduced demand for analyst coverage as the ownership of insiders increases (Bhushan, 1989) or by analysts' skepticism of the financial disclosure quality of firms with concentrated managerial ownership (Hope, 2003a). Regarding the variable board insiders, analysts may view firms with greater insider board representation as having weaker outside monitoring of management and/or weaker financial disclosure.

Ordinary Least Squares (OLS) regression results using White-corrected standard errors are shown in Table 2 and Negative Binomial (NB) regression results are shown in Table 3. In all model specifications, using OLS regression or NB regression, the governance index is negatively related to the number of analysts covering the firm. This relationship is consistently significant at the 0.05 level or better. Our evidence suggests that, among small U.S. firms, those with better corporate governance have greater analyst following. Although we cannot be certain about the direction of causality, the most reasonable interpretation of our evidence is that analysts exhibit a preference for covering firms with better governance. (9)

The results concerning the control variables mostly support findings by earlier researchers. Firm size and trading volume are both consistently, positively related to analyst following. This evidence supports earlier studies and the notion that analysts have greater incentives to follow firms that generate higher income for the analyst's employer-firm. Results from Model (4) in Tables 2 and 3 show that the number of business segments in which the firm operates is negatively related to analyst following. Bhushan (1989) reports similar results. His interpretation of the evidence is that analysts' incentives to cover firms decreases as the cost of analyzing firms increases. The cost of analyzing firms is assumed to rise with firm complexity and the number of business segments is a proxy variable for firm complexity.

We also find evidence using OLS and NB regression that insider ownership is negatively related to analyst following. These results are somewhat weaker, generally significant at the 0.10 level or better, but they are consistent with the findings of Bhushan (1989) and Hope (2003a). Bhushan (1989) argues that demand for analyst services decreases as insider ownership increases. Hope (2003a) argues that financial disclosure is likely to be less credible from firms with a high concentration of managerial ownership.

The evidence regarding "glamour" firms is somewhat mixed. Our indicators for glamour firms include those with high price-book ratios, high revenue growth, or positive share price momentum. Jegadeesh, et al. (2004) finds that analysts are more likely to recommend firms with these characteristics. Because existing evidence also suggests that analysts are more likely to follow firms that they can recommend favorably, we would normally expect a positive coefficient on all of these proxy variables for glamour firms. However, we find only modest evidence that revenue growth is positively related to analyst following, no evidence that price-book ratio is related to analyst following, and additional evidence that share price momentum is negatively related to analyst coverage. The findings from Model (3) in Tables 2 and 3 show that revenue growth has a positive coefficient, but the result is only statistically significant using NB regression. Using the same model, share price momentum is negatively related to analyst following at p = 0.057 using OLS regression and at p = 0.009 using NB regression. Price-book ratio is not significant in any model estimation. Our mixed evidence regarding glamour firms may reflect reduced analyst incentives to cover such firms in the latter period of the 2000-2002 bear market when many glamour firm stocks produced especially low returns. Finally, the percentage of board members who are insiders does not enter significantly in any of model estimations.

CONCLUSIONS

We use a sample of small U.S. firms to test whether the quality of corporate governance is related to the degree of analyst following. We focus on small firms because: (1) they have particular difficulty in attracting analysts' attention; and (2) given the limited coverage of small firms in the financial press, analyst coverage of them is arguably more important in mitigating the information asymmetry problem between managers and outside shareholders. Our measure of corporate governance quality is broad and has been used in many recent studies. However, this is the first study (to our knowledge) that relates a comprehensive measure of corporate governance quality to analyst coverage. After controlling for a variety of firm-specific factors that could influence analysts' incentives to provide coverage, we find that firms with stronger corporate governance are followed by a greater number of analysts. We argue that analysts prefer to cover firms with better governance because these firms have lower expected agency costs and higher expected financial disclosure quality.

ACKNOWLEDGMENT

The authors gratefully acknowledge the contribution of Thomson Financial for providing analyst data, available through the Institutional Brokers Estimate System.

REFERENCES

Barth, M., R. Kasznik, and M. McNichols, 2001, Analyst Coverage and Intangible Assets, Journal of Accounting Research 39, 1-34.

Berle, A. and G. Means, 1932, The Modern Corporation and Private Property (Macmillan, New York).

Bhushan, R., 1989, Firm Characteristics and Analyst Following, Journal of Accounting and Economics 11, 255-274.

Bradley, D., B. Jordan, and J. Ritter, 2003, The Quite Period Goes Out with a Bang, Journal of Finance 58, 1-36.

Chang, X., S. Dasgupta, and G. Hilary, 2006, Analyst Coverage and Financing Decisions, Journal of Finance 61, 30093048.

Cliff, M. and D. Denis, 2004, Do Initial Public Offering Firms Purchase Analyst Coverage with Underpricing?, Journal of Finance 59, 2871-2901.

Coase, R., 1937, The Nature of the Firm, Econometrica 4, 386-405.

Core, J., W. Guay, and T. Rusticus, 2006, Does Weak Governance Cause Weak Stock Returns? An Examination of Firm Operating Performance and Investors' Expectations, Journal of Finance 61, 655-687.

Core, J., R. Holthausen, and D. Larcker, 1999, Corporate Governance, Chief Executive Officer Compensation, and Firm Performance, Journal of Financial Economics 51, 371-406.

Craig, S., 2003, March 26, Analyzing the Analysts: Amid Shrinking Research Pool, Companies Buy Their Coverage, The Wall Street Journal Online.

Denis, D. and J. McConnel, 2003, International Corporate Governance, Journal of Financial and Quantitative Analysis 38, 1-36.

Dittmar, A. and J. Mahrt-Smith, 2007, Corporate Governance and the Value of Cash Holdings, Journal of Financial Economics 83, 599-634.

Fama, E., 1980, Agency Problems and the Theory of the Firm, Journal of Political Economy 88,288-307.

Fama, E. and M. Jensen, 1983a, Separation of Ownership and Control, Journal of Law and Economics 26, 301-325.

Fama, E., and M. Jensen, 1983b, Agency Problems and Residuals Claims, Journal of Law and Economics 26, 327-349.

Gompers, P., J. Ishii, and A. Metrick, 2003, Corporate Governance and Equity Prices, Quarterly Journal of Economics 118, 107-155.

Griffin, P., 2003, A League of Their Own? Financial Analysts' Responses to Restatements and Corrective Disclosures, Journal of Accounting, Auditing and Finance 18, 479-518.

Hope, O., 2003a, Analyst Following and the Influence of Disclosure Components, IPOs and Ownership Concentration, Asia-Pacific Journal of Accounting and Economics 10, 117-141.

Jegadeesh, N., J. Kim, S. Krische, and C. Lee, 2004, Analyzing the Analysts: When Do Recommendations Add Value?, Journal ofFinance 59, 1083-1124.

Jensen, M. and W. Meckling, 1976, Theory of the firm: Managerial Behavior, Agency Costs, and Ownership Structure, Journal of Financial Economics 3, 305-360.

Klapper, L. and I. Love, 2003, Corporate Governance Investor Protection, and Performance in Emerging Markets, Journal of Corporate Finance 10, 703-728.

Klock, M., S. Mansi, and W. Maxwell, 2005, Does Corporate Governance Matter to Bondholders?, Journal of Financial and Quantitative Analysis 40, 693-719.

McNichols, M. and P. O'Brien, 1997, Self-selection and Analyst Coverage, Journal of Accounting Research 35, 167-199.

Rajan, R. and H. Servaes, 1997, Analyst Following of Initial Public Offerings, Journal of Finance 52, 507-529. Research in Commotion, 2007, June 23, The Economist, 83.

Rock, S., S. Sedo, and M. Willenborg, 2000, Analyst Following and Count-Data Econometrics, Journal of Accounting and Economics 30, 351-373.

Shleifer, A. and R. Vishny, 1997, A Survey of Corporate Governance, Journal of Finance 52, 737-783.

White, H., 1980, A Heteroskedasticity-Consistent Covariance Matrix Estimator and A Direct Test for Heteroskedasticity, Econometrica 48, 817-838.

Rich Fortin, New Mexico State University

Greg Roth, New Mexico State University

ENDNOTES

(1) See, for example, Bhushan (1989), Rajan and Servaes (1997), Barth, et al. (2001), Bradley, et al. (2003), and Jegadeesh et al. (2004).

(2) Some earlier researchers, such as Chang, Dasgupta, and Hilary (2006), use low analyst following as a proxy for greater information asymmetry.

(3) See McNichols and O'Brien (1997), Rajan and Servaes (1997), Bradley, et al. (2003), and Cliff and Denis (2004

(4) Craig (2003) reports a 20% drop in the number of firms receiving analyst coverage in the two years following the March 2000 U.S. stock market peak. Investment banks slashed research budgets by 35% from 2000 to 2005 (The Economist, 2007).

(5) See Gompers, Ishii, and Metrick (2003), Klock, Mansi, and Maxwell (2005), Core, Guay, and Rusticus (2006), and Dittmar and Smith (2007). Professor Metrick's data are available at http://finance.wharton.upenn.edu/~metrick/data.htm.

(6) The agency problem arising from the separation of shareholder ownership and managerial control was recognized at least as early as Berle and Means (1932) and Coase (1937). This literature was further developed by Jensen and Meckling (1976), Fama (1980), Fama and Jensen (1983a, b) and others.

(7) For further evidence that poor governance is associated with lower valuations or firm underperformance, see Core, Holthausen, and Larcker (1999) and Klapper and Love (2004). For detailed surveys of the evidence on corporate governance around the world see Shleifer and Vishny (1997) and Denis and McConnell (2003).

(8) We define an "inside" board member as a board member who is a full time employee of the firm.

(9) We conducted (but do not show in the tables) several robustness checks. Many additional model specifications were used and in all cases the governance index is significant at the 0.05 level or better in both OLS and NB regressions.

Table 1: Summary Statistics

Shown are summary statistics for a sample of small U.S. firms. Each
firm was selected from Research Insight and has a total market value
of equity between $300 million and $1.5 billion. All variables are
measured for the year 2002. Analyst Following is the number of
analysts following the firm. Governance is the corporate governance
index created by Gompers, et al. (2003). Firm Size is the total
market value of equity. Trading Volume is the annual number of
shares traded. Momentum is the preceding one/year percentage raw
return on the firm's stock. Revenue Growth is the percentage change
in sales for the current year. Price/book is the market value of
equity divided by the book value of equity. Insider Ownership is the
percentage of outstanding shares held by inside board members. Board
Insiders is the percentage of board members who are full time
employees of the firm. Business Segments is the number of major
business segments in which the firm operates (as classified by
Research Insight). Analyst Coverage data are drawn from I/B/E/S.
Governance data are drawn from Professor Metrick's web site at the
Wharton School. All other data are drawn from Research Insight.

Variable N Mean Median

Analyst Following 365 9.09 8
Governance 365 8.69 9
Firm Size ($millions) 365 766.52 725.10
Trading Volume (millions) 365 128.99 60.36
Momentum (%) 365 -8.21 -11.67
Revenue Growth (%) 365 14.28 7.36
Price-book 365 4.26 2.82
Insider Ownership (%) 365 8.31 3.4
Board Insiders (%) 365 24.99 25
Business Segments 312 2.86 3

Variable Standard Minimum Maximum
 Deviation

Analyst Following 6.64 0 35
Governance 2.78 3 18
Firm Size ($millions) 331.49 302.27 1497.83
Trading Volume (millions) 230.75 .098 2182.29
Momentum (%) 111.31 -95.32 1897.88
Revenue Growth (%) 29.56 -32.58 385.98
Price-book 9.32 .12 156.75
Insider Ownership (%) 11.68 0 76.2
Board Insiders (%) 11.06 8.3 73.3
Business Segments 1.67 1 8

Table 2: Ordinary Least Squares Regressions of Analyst Following

Shown are the results of regressing analyst coverage on several
variables. The sample includes 365 small U.S. firms, each with a
total market value of equity between $300 million and $1.5 billion
at the end of year 2002. The dependent variable is the number of
analysts following the firm's stock. Governance is the corporate
governance index created by Gompers, et al. (2003). Firm Size is the
total market value of equity. Trading Volume is the annual number of
shares traded. Momentum is the preceding one-year percentage raw
return on the firm's stock. Revenue Growth is the percentage change
in sales for the current year. Price-book is the market value of
equity divided by the book value of equity. Insider Ownership is the
percentage of outstanding shares held by inside board members. Board
Insiders is the percentage of board members who are full time
employees of the firm. Business Segments is the number of major
business segments in which the firm operates (as classified by
Research Insight). All variables are measured for the year 2002.
Coefficient estimates are shown on the top row for each variable.
P-values are shown in parentheses and are calculated using White's
(1980) corrected standard errors.

 (1) (2) (3) (4)

Intercept 7.469 7.493 6.494 8.412
 (0.004) (0.000) (0.000) (0.000)

Governance -0.276 -0.302 -0.281 -0.293
 (0.000) (0.000) (0.009) (0.012)

Firm Size 0.003 0.003 0.003 0.004
 (0.001) (0.001) (0.000) (0.000)

Trading Volume 0.013 0.013 0.012 0.011
 (0.000) (0.000) (0.000) (0.000)

Momentum -0.008 -0.008 -0.007 -0.007
 (0.053) (0.040) (0.057) (0.038)

Price-book -0.006 -0.005
 (0.745) (0.799)

Revenue Growth 0.049 0.034
 (0.109) (0.204)

Insider Ownership -0.044 -0.041 -0.049
 (0.092) (0.091) (0.060)

Board Insiders 2.276 1.685 1.714
 (0.473) (0.574) (0.622)

Business Segments -0.667
 (0.001)

[R.sub.2] 0.295 0.305 0.350 0.392

N 365 365 365 312

Table 3: Negative Binomial Regressions of Analyst Following

Shown are the results of regressing analyst coverage on several
variables. The sample includes 365 small U.S. firms, each with a
total market value of equity between $300 million and $1.5 billion
at the end of year 2002. The dependent variable is the number of
analysts following the firm's stock. Governance is the corporate
governance index created by Gompers, et al. (2003). Firm Size is the
total market value of equity. Trading Volume is the annual number of
shares traded. Momentum is the preceding one-year percentage raw
return on the firm's stock. Revenue Growth is the percentage change
in sales for the current year. Price-book is the market value of
equity divided by the book value of equity. Insider Ownership is the
percentage of outstanding shares held by inside board members. Board
Insiders is the percentage of board members who are full time
employees of the firm. Business Segments is the number of major
business segments in which the firm operates (as classified by
Research Insight). All variables are measured for the year 2002.
Coefficient estimates are shown on the top row for each variable.
P-values are shown in parentheses.

 (1) (2) (3) (4)

Intercept 1.919 1.847 1.697 1.905
 (0.000) (0.000) (0.000) (0.000)

Governance -0.030 -0.033 -0.030 -0.031
 (0.019) (0.015) (0.020) (0.023)

Firm Size 0.000 0.000 0.000 0.001
 (0.000) (0.000) (0.000) (0.000)

Trading Volume 0.001 0.001 0.001 0.001
 (0.000) (0.000) (0.000) (0.000)

Momentum -0.002 -0.002 -0.002 -0.002
 (0.007) (0.006) (0.009) (0.009)

Price-book 0.001 0.001
 (0.800) (0.763)

Revenue Growth 0.007 0.005
 (0.000) (0.001)

Insider Ownership -0.005 -0.006 -0.007
 (0.094) (0.074) (0.038)

Board Insiders 0.522 0.441 0.445
 (0.125) (0.177) (0.229)

Business Segments -0.076
 (0.000)

Prob. > Chi (2) 0.000 0.000 0.000 0.000

N 365 365 365 312