Interest on Equity versus Dividends: The Role of Shareholder Identity in Corporate Tax Avoidance.
| Autor | Colombo, Jefferson Augusto |
1 Introduction
Brazilian capital markets are characterized by (i) mandatory minimum dividend rules, (ii) high corporate tax rates, (iii) non-voting shares, (iv) high ownership concentration, (v) tax-deductible dividends (called "interest on equity"--henceforth "IOE"), and (vi) taxation that depends on the legal nature of the investor. This environment makes Brazil a unique laboratory in which to study a range of corporate decisions. In this paper, we study the distribution of corporate profits through either dividends or IOE, according to the identity of the controlling shareholder of the firm. In our sample, about 40% of the firms that could benefit from legal tax deductibility by paying out IOE choose to pay only regular, non-deductible dividends. That decision may be optimal from the standpoint of the controlling shareholders, but it may destroy value for minority shareholders. Aside from this potential agency problem, understanding the corporate choice between regular dividends and IOE is vital since there is currently a controversial national debate on imposing a dividend distribution tax and eliminating IOE (Law Project n. 2337, 2021). Therefore, our research highlights a topic of academic, managerial, and policymaking importance.
In particular, we investigate whether ownership structure is a determinant of the distribution of earnings (IOE vis-a-vis regular dividends) in Brazil. Through the numerical exemplifications shown in Appendix A (Table A1), we show that the cost of receiving cash payouts through IOE is significantly heterogeneous across shareholders. For corporate shareholders, IOE payments can increase or decrease the overall tax burden--it depends on whether the firm pays additional income tax or not, among other factors (see Appendix A for details); however, for individual shareholders and associated investment entities, (i) the optimal distribution should be 100% of their cash dividends in the form of IOE, so long as this distribution does not surpass the limits imposed by law. Since the optimal distribution, from a tax standpoint, depends on the beneficiary's legal nature, we develop testable hypotheses based on taxation and ownership structure. That is, we analyze decisions regarding the distribution of dividends vis-a-vis IOE by considering the specific nature of the controlling shareholder. Therefore, this study represents a step toward better understanding why many firms in Brazil that recurrently distribute cash dividends to shareholders do not choose IOE.
This study adds to the literature in some critical respects. First, unlike previous studies, we discuss the distribution of IOE from the perspective of both the beneficiaries (shareholders) and payers (investees) rather than the former or the latter alone. Second, as a reflection of potential measurement problems, we create a variable that measures the ratio between IOE distributed and the maximum allowed by law (IOE_IOE*), which permits more precise measurement of the size of the tax benefit obtained by each firm. Finally, by exploring the heterogeneity in the tax rate on IOE income according to the shareholder's identity, we contribute to the debate about the interplay between payout policy and agency problems (Chang, Kang, & Li, 2016; Mulyani, Singh, & Mishra, 2016), the role of ownership identity in shaping corporate outcomes (Alhababsah, 2019), and clientele effects induced by heterogeneity in tax preferences among shareholders (Portal & Laureano, 2017). Importantly, our results indicate that overall taxation (firm-level plus shareholder-level) is a key factor driving payout policy.
Overall, our results suggest that shareholder identity influences payout policy through the taxation channel. Consistent with ex-ante tax-driven hypotheses, the presence of institutional ownership (mutual funds, private equity funds, private pension entities) with 5% or more of voting rights increases the likelihood of a firm distributing earnings via IOE. These shareholders are precisely the ones that benefit most from IOE payments --receiving earnings in the form of IOE is tax-free for this type of investor (Law 9532/1997; Law 11053/2004). Moreover, our findings suggest that larger, more profitable firms with better corporate governance practices and more growth opportunities tend to distribute more cash earnings via IOE, a practice that increases the wealth of the average minority shareholder by alleviating total tax payments. Importantly, corporate governance results are restricted to firms listed in the Novo Mercado and Level II special segments of corporate governance of the Brazilian Stock Exchange (B3), whose rules are more demanding. (ii) When we include firms listed in the less stringent Level I special segment, the likelihood of using IOE diminishes.
The remainder of this manuscript is structured as follows. In the next section, we present a brief literature review on payout policy. In section 3 we discuss the Brazilian context, with emphasis on ownership concentration, interest on equity, and tax treatment of IOE and dividend cash distributions. In section 4 we describe our sample and method. Section 5 shows the empirical findings, and in section 6 we draw our conclusions.
2 Literature Review
2.1 Agency costs and payout policy in the recent international literature (iii)
Easterbrook (1984) pioneered the study of agency costs on payout policy. In his seminal paper, the author proposes that dividends--a financial phenomenon that is hard to explain--could mitigate perverse incentives that managers have to expropriate shareholders. Shareholders ultimately bear the agency costs implied by their relationship with managers. Two particular costs are the cost of monitoring and the cost of managerial risk-aversion. Dividends may offer a less expensive (partial) solution to these problems than other instruments. In particular, the continued outflow of cash from the company to shareholders implied by a stable dividend policy limits the number of resources available at the discretion of managers, since financing new projects would force the firm to tap the capital markets. Whenever the firm needs to raise cash in the market--either by contracting new debt or by issuing new shares--it subjects itself to independent market monitoring.
Similarly, managers are usually more risk-averse than shareholders, since their wealth depends mostly on their human capital, which is closely tied to the firm's survival. Therefore, managers may choose safer--and less profitable--projects than shareholders would. Shareholders have few mechanisms to assure that managers pick riskier projects. But both managers and shareholders can also control corporate risk by adjusting the firm's debt. Ceteris paribus, managers would prefer to keep financial leverage low, which would benefit creditors at the expense of shareholders. On the other hand, shareholders would prefer to raise leverage to the limit allowed by the debt covenants. Thus, a constant outflow of cash to shareholders under the form of dividends allows shareholders to prevent managers' risk aversion by preventing the debt-equity ratio from declining. As Farre-Mensa, Michaely, and Schmalz (2014) nicely summarize it, "continual exposure to the discipline of external financial markets reduces agency costs" (Easterbrook, 1984, p. 107).
The original insights of Easterbrook (1984) have spawned a rich body of literature. Research on global corporate governance characterizes controlling shareholders' influence on minority shareholders as a cutting-edge topic (Barka & Hamza, 2020; Bebchuk & Weisbach, 2010), especially in emerging markets (Kearney, 2012). Although payout policy has received considerable attention in the literature, not many papers have focused on the role of investor protection and ownership patterns, especially in emerging markets. For instance, Benavides, Berggrun, and Perafan (2016) study dividend payout policies in six Latin American countries and find that both the target dividend payout ratio and dividend smoothing are higher in countries with better governance indicators. Bradford, Chen, and Zhu (2013) investigate the effect of state and pyramid ownership on the dividend policies of listed firms in China and find that the state-controlled firms pay higher dividends while pyramid-controlled firms pay lower dividends, as the firm's chain of control lengthens.
A few recent papers have focused on the interplay between ownership structure, agency problems, and payout policy. Renneboog and Szilagyi (2015) study the role of dividend payout as an agency control device in the Netherlands, a stakeholder-oriented governance country. Concerning emerging markets, Mulyani et al. (2016) examine the roles of dividends and leverage in mitigating agency problems within family firms in Indonesia and find that, compared to non-family firms, family firms tend to maintain lower dividend payouts and higher leverage. Chang et al. (2016) study institutional ownership use of dividend payouts as a monitoring tool to mitigate firms' agency problems and find support for the role of dividend payments as a monitoring device under high agency costs, consonant with explanations based on the potentially conflicting interests of controlling shareholders (Portal & Laureano, 2017).
Recent empirical evidence also suggests that taxation has no impact on dividend payments (Khan, Jehan, & Shah, 2017). Still, this evidence is not conditional on heterogeneous tax status according to ownership identity. By identifying different tax preferences according to the legal nature of the shareholder, we provide fresh evidence on the role of taxes in payout policy. The hypothesis driving the present study is that many firms may not enjoy the tax benefits of IOE precisely because the tax law discourages controlling shareholders from using it.
The paper closest to this one is that of Boulton, Braga-Alves, and Shastri (2012). However, there are...
Para continuar a ler
PEÇA SUA AVALIAÇÃOCOPYRIGHT GALE, Cengage Learning. All rights reserved.
