Are family firms more tax aggressive than non-family firms?

Abstract

Taxes represent a significant cost to the firm and shareholders, and it is generally expected that shareholders prefer tax aggressiveness. However, this argument ignores potential non-tax costs that can accompany tax aggressiveness, especially those arising from agency problems. Firms owned/run by founding family members are characterized by a unique agency conflict between dominant and small shareholders. Using multiple measures to capture tax aggressiveness and founding family presence, we find that family firms are less tax aggressive than their non-family counterparts, ceteris paribus. This result suggests that family owners are willing to forgo tax benefits to avoid the non-tax cost of a potential price discount, which can arise from minority shareholders’ concern with family rent-seeking masked by tax avoidance activities [Desai and Dharmapala, 2006. Corporate tax avoidance and high-powered incentives. Journal of Financial Economics 79, 145–179]. Our result is also consistent with family owners being more concerned with the potential penalty and reputation damage from an IRS audit than non-family firms. We obtain similar inferences when using a small sample of tax shelter cases.

Introduction

The government (federal, state, and local) takes a greater than one-third share of a firm's pre-tax profits. Given the significance of this tax cost to the firm and shareholders, it might be expected that tax aggressiveness is desired by shareholders. However, this argument ignores potential non-tax costs that can accompany tax aggressive activities (Scholes, Wolfson, Erickson, Maydew, and Shevlin, 2005). In this paper, we study the implications of non-tax cost considerations arising from the unique agency conflict in family firms for their tax management activities. Specifically, we investigate whether family firms are more or less tax aggressive than non-family firms.¹

As is common in the literature, we define family firms as firms where members of the founding family continue to hold positions in top management, are on the board, or are blockholders of the company.² Founding family presence implies a greater agency conflict between large and minority shareholders and a smaller agency conflict between owners and managers as compared to non-family firms. The nature and extent of agency conflicts, such as the costs arising from hidden actions of managers, can affect the level of tax aggressiveness. Accordingly, prominent researchers (Scholes, Wolfson, Erickson, Maydew, and Shevlin, 2005; Desai and Dharmapala, 2004, Desai and Dharmapala, 2006) call for more research to examine tax aggressiveness within an agency context. The different agency conflicts in family firms versus non-family firms enable us to examine this issue.

In addition, in their review of the empirical tax literature Shackelford and Shevlin (2001) point out that insider control and other organizational factors, such as ownership structure, are important, but understudied, determinants of tax aggressiveness.³ We fill in the void by examining the impact of founding family ownership on tax aggressiveness. The presence of the founding family leads to a different ownership structure compared to non-family firms and thus provides a unique setting to examine the impact of insider control on tax aggressiveness activities. Lastly, family firms are an important component of the economy—32–46% of S&P 1500 firms are classified as family firms depending on the definition of family firms—and are thus of interest in their own right.

To determine the level of tax aggressiveness, firms trade off the marginal benefits against the marginal costs of managing taxes. The marginal benefits include greater tax savings, whereas the marginal costs include the potential penalty imposed by the IRS, implementation costs (time/effort and transaction costs of implementing tax transactions), and agency costs accompanying tax aggressive activities (including rent extraction). Desai and Dharmapala, 2004, Desai and Dharmapala, 2006 argue that the critical characteristics of tax aggressive activities are complexity and obfuscation. Such complexity could be used to mask rent extraction, such as earnings management, related-party transactions, and other perquisite consumption behavior.⁴ If this is the case, shareholders will price protect themselves in an efficient capital market. That is, the potential rent extraction of tax aggressiveness, while beneficial to the decision maker, comes with a non-tax cost: a price discount on the firm's stock imposed by external shareholders. This non-tax cost is particularly poignant in family firms due to their larger dominant-small shareholder conflict: family owners have greater opportunities for rent extraction, but at the same time non-family shareholders, anticipating self-dealing, can penalize family members’ self-dealing by discounting the share price.

Under the ceteris paribus condition, the difference between family and non-family firms in tax aggressiveness depends on the impact of the differential characteristics of family owners versus managers in non-family firms on the benefits and costs of tax aggressiveness. Because family owners have substantially higher holdings, they benefit more from tax savings or rent extraction that can be concealed by tax aggressive activities, but at the same time, the potential price discount is also more costly for them. In addition, due to their much larger equity ownership and their much longer investment horizons, family owners are more concerned with the potential penalty imposed by the IRS and the reputation damage from being involved in a tax related lawsuit. Thus, both the benefits and costs appear to be higher for family owners than for managers in non-family firms. Accordingly, ex ante, it is unclear whether family firms will be more or less tax aggressive than non-family firms, and we examine this issue empirically.

To examine the tax aggressiveness of firms, we rely on multiple measures of tax aggressiveness drawn from the literature. Specifically, we use two effective tax rate measures and two book-tax difference measures: effective tax rate (defined as total tax expense divided by pre-tax book income), cash effective tax rate (cash tax paid divided by pre-tax book income), the book-tax difference measure advanced by Manzon and Plesko (2002), and a residual book-tax difference measure developed in Desai and Dharmapala (2006). Firms that are more tax aggressive have lower effective tax rates (ETRs) and higher book-tax differences than other firms. As an additional test, we also use a factor analysis to extract one common factor from these four measures.

Using 3,865 firm-year observations from S&P 1500 firms in the period 1996–2000, we show that family firms exhibit lower tax aggressiveness than their non-family counterparts, as demonstrated by their higher effective tax rates and lower book-tax differences.⁵ This result holds both before and after we control for firm characteristics that are cross-sectionally associated with our tax aggressiveness measures: firm performance, leverage, loss carry forward, foreign income, tangible and intangible assets, equity income, firm size, market-to-book ratio, and industry fixed effects. Including these controls ensures that the documented difference in tax aggressiveness between family and non-family firms is not driven by fundamentals. For example, firms with loss carry forward will have a lower tax rate than other firms. Since, on average, family firms perform better than non-family firms (Anderson and Reeb, 2003), family firms will appear to have a higher tax rate and be less tax aggressive. However, such a difference is not due to tax planning of non-family firms or lack of tax planning in family firms. Controlling for those firm characteristics mitigates the concern that correlated omitted variables explain our results. Note that including those characteristics admittedly controls for some of the tax aggressive mechanisms. However, doing this simply controls for the average tax effects associated with these variables (e.g., firms with more foreign income on average have lower ETRs), and leaves much room to capture cross-sectional variation in firms’ tax planning activities (e.g., for firms with foreign operations, more tax aggressive firms will be more aggressive in transfer pricing to shift income). Our family firm variables also capture the difference between family firms and non-family firms in tax planning through means such as state tax planning, tax shelters, use of flow-through entities, and use of off-balance-sheet financing, to name just a few. We provide a detailed discussion in Section 3.3.

We further find that family firms without long-term institutional investors (as outside monitors) and family firms expecting to raise capital exhibit even lower tax aggressiveness. These results are consistent with family owners in these firms having even stronger incentives to reduce the perception of family entrenchment: family firms are willing to forgo tax savings to avoid the associated price discounts. In addition, we conduct many sensitivity tests and obtain similar inferences. Our results are robust to the alternative explanation of differential firm sophistication and hold after controlling for outside blockholder ownership and CEO ownership. We obtain similar inferences when using a sample of tax shelter cases.

Our paper contributes to the literature in the following ways. First, our evidence provides an important step toward a better understanding of the impact of equity ownership and agency conflict on firms’ tax reporting practices. Prior research generally focuses on the differences in firms’ tax reporting between private and public companies in a few selected industries, such as banks and insurers (e.g., Cloyd, Pratt, and Stock, 1996; Beatty and Harris, 1999; Mikhail, 1999; Mills and Newberry, 2001; Hanlon, Mills, and Slemrod, 2005). The general conclusion of these studies is that private companies are more tax aggressive. In contrast, we examine public firms (S&P 1500 firms) with diverse industry membership and employ proxies that capture the overall tax aggressiveness. Thus, our evidence can be generalized to a greater section of the economy. Interestingly we find family firms to be less tax aggressive. While family firms are similar to private firms in the concentration of ownership of selected individuals, the public nature of family firms gives rise to unique agency conflicts that can lead to differential non-tax cost concerns and hence differential tax aggressiveness.

To reconcile our findings with prior studies, we examine the banking industry further. We compare the tax aggressiveness of public family banks, public non-family banks, and private banks. (We focus on banks because financial data for private firms in other industries are not readily available.) Using both univariate and multivariate analyses, we find that private banks are the most tax aggressive, followed by public non-family banks, and then by public family banks. This reconciles our findings with prior research and is consistent with our conjecture that different agency problems and reputation concerns faced by private firms (as opposed to public family firms) lead to different tax aggressiveness.

Second, we extend the family firm literature by examining the impact of founding family presence on tax aggressiveness. Our evidence also corroborates Desai and Dharmapala, 2004, Desai and Dharmapala, 2006 argument and highlights the importance of taking into consideration (1) the complementarity between tax avoidance activities and rent extraction, and (2) the non-tax costs arising from the greater agency conflicts between large and small shareholders in family firms in studying the tax aggressiveness of family firms. Our results show that the non-tax costs arising from agency conflicts can have a significant impact on these firms’ tax management activities.

The rest of the paper proceeds as follows. In the next section we review related literature and develop our hypothesis. Section 3 describes our sample and research design. 4 Primary empirical analyses, 5 Corroborating analyses present our main results and corroborating evidence, and Section 6 reports additional analyses. Section 7 concludes.