«Kathleen A. Farrell* University of Nebraska - Lincoln and Drew B. Winters** Texas Tech University Using a broad-based sample of small businesses, we ...»
An Analysis of Executive Compensation
in Small Businesses'*'
Kathleen A. Farrell*
University of Nebraska - Lincoln
Drew B. Winters**
Texas Tech University
Using a broad-based sample of small businesses, we analyze the relation between accounting-
based firm performance measures and executive compensation for S-corporations, and C-
corporations. After controlling for the potential endogeneity associated with the choice of
organizational form, we find a positive relation between executive pay and ROA in S- corporations and C-corporations. We also find a positive relation between executive pay and total asset turnover but the relation is stronger for S-corporations. We document a positive * We are grateful for helpful comments and suggestions from Chris Anderson, Linda Miles and seminar participants at the University o f Nebraska-Lincoln, University of Kansas, and the 2001 Financial Management Association meeting, Kathleen A. Farrell is an Associate Professor of Finance at the University of Nebraska-Lincohi. Dr. Farrell’s primary research interests are in executive compensation, executive turnover and other corporate governance issues. Dr. Farrell’s publications have appeared in such journals as iht Journal o f Business, Journal o f Accounting and Economics, Journal o f Corporate Finance, National Tax Journal, Journal o f Banking and Finance, Journal o f Financial Research, and Advances in Financial Economics. Dr. Farrell received her Ph.D. in finance from the University of Georgia, and a BBA in fmance and accounting from Kent State University.
Drew B. Winters is a Full Professor and the Jerry S. Rawls Professor of Finance at Texas Tech University. Dr.
Winters' research focuses on the intraday and daily behavior of short-term interest rates, and factors that influence the interest rate banks charge on lines o f credit. His publications have appeared in several academic journals including the Journal o f Business, Journal o f Accounting Research, the Journal o f Banking and Finance, the Journal o f Financial Research, Quarterly Journal o f Business and Economics, and the Economic Review of the Federal Reserve Bank of St. Louis. Dr. Winters has a Ph.D. in fmance from University of Georgia, a M.B.A. in fmance and MIS from University o f Georgia, and a B.S. in computer science and accounting from Duke University.
2 An Analysis of Executive Compensation in Small Businesses (Farrell & Winters) relation between executive salaries and more diffuse ownership, owner managers, and when the founder is the current owner. We find a negative relation between executive salaries and firms with greater than fifty percent family ownership.
I. Introduction and motivation Executive compensation issues attract attention in both the academic and political arenas. Much of the public interest focuses on escalating compensation for CEOs of publicly traded companies with an increased divergence in pay between CEOs and lower level employees. As described by Murphy (1999), the academic literature has evolved primarily because of the emergence and general acceptance of agency theory and the increased accessibility to executive compensation data. In general, studies find that compensation increases with firm size and executive compensation is positively related to firm performance (e.g., Lewellen and Huntsman, 1970; Ciscel and Carroll, 1980; Murphy, 1985), although the economic magnitude of the change in pay relative to performance has been questioned (Jensen and Murphy, 1990). In addition, studies have focused on firm characteristics that impact both compensation levels and compensation structure (e.g., Smith and Watts, 1992; Gaver and Gaver, 1993).
The current compensation literature focuses on large, publicly traded firms. For the most part, a void exists in the literature regarding small firm compensation practices, partially due to the difficulty in obtaining data on compensation in small, privately held firms. Also, small firms tend to have less pronounced agency problems since many small firms have owner/managers. Fama and Jensen (1983a, 1983b) argue that the residual claims of privately held firms are largely restricted to owner/managers or to agents, such as family members or business associates, with a special relationship with the owner. Therefore, they argue that agency problems between owners and managers may be mitigated in privately held firms.
However, not all privately held firms are run by owner/managers in which case the agency problems may still exist with a need for owners to monitor management (e.g., Ang, Cole and Lin, 2000).
Only recently has research begun to focus on closely held firms and the unique incentive problems they may face. For example. Bates, Jandik and Lehn (2000) analyze the promotion incentives (tournament theory) and executive compensation of executives in familycontrolled public firms and find that non-family executives in family owned firms are faced with diminished promotion incentives. In their study of publicly-traded and privately-held property-liability insurance companies, Ke et al. (1999) find a positive, significant relation between CEO pay and ROA in publicly-traded insurers but the relation is insignificant for privately-held insurers. They conclude that privately-held insurers do not rely on compensation contracts linking pay to accounting based performance to mitigate agency problems but instead may rely on direct monitoring combined with subjective performance measurement.
Focusing on privately-held small C corporations, Cavalluzzo and Sankaraguruswamy (2005) find evidence that executive compensation is more closely related to sales to assets in privately-held small corporations with more disperse ownership than those with more concentrated ownership. They also find that the sensitivity o f compensation to sales to assets is weaker if the firm is a family owned business and the manager is a shareholder.
The goal of this paper is to analyze the unique issues faced by small firms that are organized as C-corporations and S-corporations. These organizational forms differ in their ownership structures, tax treatment, and in the type of agency problems they face. Specifically, The Journal of Entrepreneurial Finance & Business Ventures. Vol. 12, Iss.3 3 we analyze the factors that influence the compensation structure of small firms while considering the intra-dependence between the choice of organizational form and the optimal structure of compensation contracts.
Our paper differs from Cavalluzzo and Sankaraguruswamy (2005) in the following ways. First, we analyze both S and C Corporations and restrict our sample to firms with less diffuse ownership. Specifically, we exclude C corporations (approximately five percent of the C-corporate sample observations) with more than thirty shareholders. Our goal is to compare small firm compensation practices while controlling for diffuse ownership of C corporations.
Second, when we investigate the relation between executive pay and different accounting based firm performance measures for privately held firms, we control for the potential endogeneity associated with the choice to organize as an S-corporation and a C-corporation and compensation practices using two stage least squares.
We conduct our tests on the data from the 1993 National Survey of Small Business Finances (NSSBF) sponsored by the Federal Reserve Board and the Small Business Administration. The survey is of a nationally representative sample of small businesses with less than 500 employees and contains 4,637 usable survey responses (according to PriceWaterhouse). Specifically, we analyze S-corporations and privately-held C-corporations.^ Consistent with previous studies of large publicly traded companies, we find a positive relation between firm size and executive salaries for our full sample of privately-held firms.
We also find a positive relation between executive salaries and return on assets independent of organizational form. Decomposing return on assets into profit margin and total asset turnover, we find a positive relation between executive salaries and total asset turnover. However, when we interact corporate form with total asset turnover, the result suggests that C corporations pay less for the efficient use of assets relative to S corporations. Therefore, our results are more consistent with Holmstrom’s (1979) second best solution to the contracting problem which suggests using imperfect estimators of an agent's actions when complete monitoring is impossible or prohibitively costly. In the case of S corporations and C corporations, total asset turnover appears to be the better proxy for cash flows available to owners than ROA, which is more typically used for large publicly traded firms, even after controlling for owner managers, family ownership and diffuse ownership. Our results differ firom Ke, et.al. (1999) who find that privately-held insurers do not rely on compensation contracts linking pay to accounting based performance to mitigate agency problems. It may be that agency costs and/or information asymmetry problems are less severe in the insurance industry and thus, Ke, et. al.’s findings can not be generalized to a broad based sample of privately held small firms.
The paper is organized as follows. In Section II, we discuss the characteristics of different organizational forms and the implications with respect to compensation practices. In Section III, we describe the data and our empirical model. Section IV includes a discussion of our results followed by our conclusions in Section V.
II. Different organizational forms Since our focus is on executive compensation across different organizational forms of small businesses, we begin with a discussion of the characteristics associated with SWe do not discuss the characteristics of sole proprietorships because, while this organizational form is present in the survey data, none of the sole proprietorships provide the accounting data from the survey questions necessary to be included in this study. Similarly, partnerships do not provide sufficient compensation data to be included in this study.
4 An Analysis o f Executive Compensation in Small Businesses (Farrell & Winters) corporations and C-corporations. Drawing on the differences in firm structure, we describe the theoretical arguments that may predict compensation contracts across these different organizational forms. Then we discuss firm performance measures available for small businesses.
2.1 Characteristics of different organizational forms A corporation is a legal entity created by a state that is distinct from its owners and managers, has unlimited life, ease o f transferability of ownership and limited liability. Scorporation, as defined by Subchapter S in the Internal Revenue Code, is a federal income tax election made by a corporation by filing a form with the IRS. Otherwise, the S-corporation is formed in the same manner as a C-corporation under state law. The election is limited to firms that currently have 100 or fewer stockholders for tax years beginning in 2004 (per CCH Business Owners Toolkit). In 1993, the date of our sample, the restriction was for 35 or fewer stockholders (Denis and Sarin, 2002). S-corporate status requires the corporation to be domestic and the firm may issue only one class of stock. Subchapter S shareholders may not be nonresident aliens or a nonhuman entity. After the ownership restrictions, an S-corporation can be considered a hybrid organizational form that combines characteristics o f a regular corporation and a partnership. Unhke a C-corporation, an S-corporation does not pay corporate income tax, but instead passes all of its income on to the stockholders for payment of taxes in a manner similar to a partnership. The income of a S corporation is subject to only one tax, at the individual level. Unlike the S corporation, the C corporation shareholders have the tax benefit of deferring taxes on income retained in the firm. If earnings are not paid out as dividends, they are not taxed at the personal level until the shareholder realizes capital gains income (e.g., Denis and Sarin, 2002). Therefore, the primary differences in organizational form between Scorporations and C-corporations relate to taxes and the number and type o f owners.
Given the differences and similarities between S-corporations and C-corporations, the question becomes what determines the choice to set up the firm as a S or C corporation and what is the appropriate compensation of the executives? Various factors such as number of shareholders, growth opportunities, business risk, size, and lines o f business may contribute to both the choice of organizational form and level or structure of compensation. Additional factors likely to influence compensation include owner-manager, family ownership, and founder status.
We expect S-corporations to have fewer shareholders given they are faced with restrictions on the maximum number o f shareholders. Therefore, S-corporations are less likely to have a highly diffiise ownership structure. As originally argued by Berle and Means (1932), a diffuse ownership structure may diminish incentives to shareholders to monitor management since the cost of monitoring likely outweighs the benefit. Since more diffuse ownership suggests greater agency problems, tying compensation to performance may mitigate some agency costs in more diffusely held firms. In addition, given the restriction on nimiber of shareholders, owners that choose subchapter S status may not expect to have as great a need to raise additional capital from new shareholders. Those choosing C-corporate status may expect to be in need of a larger shareholder base to support expected growth. Since ownership structure may also impact agency costs, we specify a variable defined as number of owners in the firm.
The shareholder restriction may also suggest differences in the perceived growth opportunities associated with the firm. If a firm has high growth opportunities, the owners may The Journal of Entrepreneurial Finance & Business Ventures, Vol. 12, Iss.3 5 choose C-corporate status for the potential to draw on a larger shareholder base. Many Ccorporations draw from their employee pool to raise additional capital or to provide ownership incentives through private stock grants? High growth opportunities have also been found to be associated with a stronger pay-performance relation (e.g., Smith and Watts, 1992; Gaver and Gaver, 1993) suggesting that C corporations may be more likely to have a stronger payperformance relation than S corporations.