اتحادیه های کارگری و پرخاشگری مالیاتی

We examine the impact of unionization on firms' tax aggressiveness. We find a negative association between firms' tax aggressiveness and union power and a decrease in tax aggressiveness after labor union election wins. This relation is consistent with labor unions influencing managers' in one, or both, of two ways: (1) constraining managers' ability to invest in tax aggressiveness through increased monitoring; or (2) decreasing returns to tax aggressiveness that arise from unions' rent seeking behavior. We also find preliminary evidence that the market expects these reductions around union elections and discounts firms that likely add shareholder value via aggressive tax strategies.

Do all of a firm's stakeholders benefit when corporate taxes are reduced? While the answer to this question is potentially yes, a more interesting issue that we focus on is whether firms' tax planning outcomes provide more or less benefit to some stakeholders or cause some stakeholders to bear more or less risk than others. Specifically, we examine an important stakeholder—labor unions—and investigate a previously unexplored link between labor unions and corporate tax aggressiveness. Because taxes are a significant cost faced by all of a firm's stakeholders, a narrow view of corporate tax policy could lead researchers to expect firms to invest in aggressive tax strategies with abandon. However, as Scholes, Wolfson, Erickson, Maydew, and Shevlin (2005) point out, and recent research into the agency view of tax avoidance shows, this view ignores the effect of nontax agency costs that can negatively affect firm value (Desai and Dharmapala, 2008 and Chen et al., 2010). More important, this view inappropriately treats diverse stakeholders as homogeneous in their utilities for tax aggressiveness without appealing to potential variation in risk preferences. Prior literature suggests that labor unions have distinct incentives, political motivations, and risk preferences that manifest in conservative investment and financial reporting (Faleye et al., 2006 and Leung et al., 2012). It follows that the observed level of corporate tax aggressiveness could at least partially reflect labor unions' presence and influence. More specifically, based on their incentives, political motivations, and risk preferences, labor unions likely prefer less tax aggressiveness and corporate outcomes reflect this preference. Prior literature also suggests that unions exhibit rent seeking behavior that draws managerial responses (Klasa et al., 2009, Matsa, 2010 and Connolly et al., 1986). Our study allows for the possibility that as the influence of unions and their ability to capture firm profits increases, managers face diminishing returns on marginal investment in tax aggressiveness. To the extent that tax aggressiveness increases after-tax cash flows that are vulnerable to union capture, managers could be less willing to invest in aggressive tax strategies. We acknowledge that if labor unions are successful in bargaining away a share of firms' after-tax profits, one might expect unions to prefer more tax aggressiveness, calling into question our predictions. However, this argument applies only if there is no risk or cost associated with tax aggressiveness and if all stakeholders exhibit similar risk preferences. While unions try to bargain for a share of firms' after-tax profits, they are unlikely to be indifferent to the various means of profit generation. In fact, prior research supports unions' assessments of managers' investment decisions that are consistent with a bondholders' perspective, generally suggesting that unions prefer less risk than shareholders or managers (Faleye et al., 2006 and Chen et al., 2012). Because tax aggressiveness can simultaneously impact a firms' tax risk (Hanlon and Heitzman, 2010 and Crocker and Slemrod, 2005) and hamper unions' ability to assess the extent of firms' tax risk due to agency costs (Desai and Dharmapala, 2006 and Desai and Dharmapala, 2008), we expect unions to prefer less tax aggressiveness. In addition, once union contracts are set, increased after-tax cash flows derived from tax aggressiveness accrue to shareholders and managers, but not necessarily immediately to unions. In the interim period before contract renegotiations, union members bear the risk of aggressive tax position reversals that could result in additional taxes and Internal Revenue Service (IRS) penalties without a concurrent benefit.1 To the extent that all stakeholders suffer from adverse consequences associated with tax aggressiveness, unions likely suffer more because their time horizon in firms is likely much longer than that of managers or shareholders who have the ability to leave firms or quickly sell their interests and invest elsewhere. Finally, our predictions do not necessarily require that unions have a direct impact on corporate tax policy. Given unions' potential capture of after-tax cash flows, managers are less likely to invest in aggressive tax strategies at the margin if they expect to share the benefits with unions.2 To test our assertions and expectations regarding the link between labor unions and corporate tax aggressiveness, we examine the association between tax aggressiveness and unionization rate, a proxy for union strength and bargaining power (Connolly et al., 1986, Bronars and Deere, 1991, Matsa, 2010 and Klasa et al., 2009), at the establishment and industry levels. We define industries following prior research (Connolly et al., 1986, Bronars and Deere, 1991, Matsa, 2010 and Klasa et al., 2009). Establishments are defined consistent with the National Labor Relations Board (NLRB) as the separate location for a company's business activity that has employees. Because not all employees at an establishment are represented by a labor union, bargaining power is defined as the bargaining unit size (number of employees covered by the union contract) divided by the total number of establishment employees. In both establishment and industry settings we find that union power is associated with a lower level of corporate tax aggressiveness. Next, we examine changes in tax aggressiveness around labor union elections in an event study setting, conditioning our analyses on whether a union wins or loses the unionization election. Employees at a firm become unionized when the union wins the election. Subsequently, these employees are represented by the union in the formation of a labor agreement through a collective bargaining process with firms' management. Consistent with our expectation, we find that tax aggressiveness decreases in the period after labor union election wins. This result provides further evidence that labor unions are negatively associated with tax aggressiveness. In an additional test, we present some preliminary evidence on the valuation implications of unions' impact on tax aggressiveness within the Desai and Dharmapala, 2006 and Desai and Dharmapala, 2008 framework. They predict that tax aggressiveness is value accretive to firms in the presence of strong corporate governance. In a weak corporate governance environment, managers could use the obfuscation inherent in tax avoidance activities to divert resources from shareholders. Consistent with the theoretical underpinnings of this framework, we find that the market discounts tax aggressiveness by well-governed firms around union elections. This result suggests that the market expects managers' rational response to potential union rent extraction, or unions' incentives to constrain managerial behavior, to reduce value increasing aggressive tax activities in well-governed firms. We make several contributions to the literatures on labor unions and tax aggressiveness. First, we echo scholars (Scholes et al., 2005, Desai and Dharmapala, 2004 and Desai and Dharmapala, 2006) who call for more research into the examination of tax aggressiveness within an agency context. Because unions cannot perfectly contract with firms' management, they are forced to entrust managers with the competent oversight of their claims on firm resources and the adequate deployment of their contributions to firms. This creates an agency dynamic not yet considered in the tax planning setting. Our study allows us to examine this dynamic and, therefore, adds to the empirical literature that examines tax policies within an agency framework (Chen et al., 2010, Desai and Dharmapala, 2006, Desai and Dharmapala, 2008 and Desai et al., 2007).3 Second, our findings are consistent with labor unions constraining managers' ability to invest in aggressive tax strategies and, therefore, adds to the literature that shows labor constraining managers' investment decisions (Connolly, Hirsch, and Hirschey, 1986) and risk - taking behavior (Faleye, Mehrotra, and Morck, 2006).4 Third, our results are consistent with labor unions' motive to reduce perceived or actual risk inherent in tax aggressiveness, strengthening the assertions of Agrawal (2008) and Leung, Li and Rui, (2012), who argue for labor unions' role in corporate governance and financial reporting conservatism. Finally, with respect to traditional tax research, our study presents another important determinant of tax aggressiveness. This furthers understanding of the cross-sectional variation in tax aggressiveness (Gupta and Newberry, 1997 and Dyreng et al., 2008) and why some firms do not pursue tax benefits more aggressively (Graham and Tucker, 2006 and Shevlin, 2002). We structure the remainder of this paper as follows. In Section 2, we place our study in the context of the existing literature and develop hypotheses. In Section 3, we describe the data, the proxies for tax aggressiveness, and the empirical design. In Section 4, we discuss our main results. We summarize our findings and conclude in Section 5.

We examine whether labor unions impact tax aggressiveness. Our hypotheses are drawn from the recent literature that examines labor unions' ability to constrain managerial decisions and managers' responses to reductions in the return on investments in aggressive tax strategies. We argue that labor unions' risk preferences and desire to maintain the value of their fixed claim, amidst an imperfect information environment with respect to managerial actions, lead them to prefer less tax aggressiveness. Given labor unions' proven role in impacting firm activity, we expect them to have the means to affect tax aggressiveness. Furthermore, the negative association between tax aggressiveness and unionization is indicative of unions' claims on tax savings from marginal investments in tax aggressiveness reducing the returns of these strategies to shareholders and managers. In our association tests, we find that unionization rates are negatively related to tax aggressiveness at both the establishment and industry levels. In our event study setting, we find that firms become less tax aggressive in the period after union election wins. Our results suggest that labor unions play an important role in influencing firms' tax policies.