اثربخشی و ارزیابی به حداقل رساندن مالیات سیاسی

We find evidence suggesting that corporate lobbying for tax purposes over the period 1999–2009 is one method by which firms managed corporate taxes. Furthermore, tax management strategies employed by these politically active firms were valued by shareholders. Firms lobbying on tax issues have lower book effective taxes and greater discretionary permanent differences in GAAP and IRS taxable income. Investors place a premium on lobbying activities for tax purposes unless the firm already has a low effective tax rate or very high book-tax differences. We conclude that lobbying political officials is one method by which firms manage risks attendant an aggressive tax strategy.

Firms and their insiders spend substantial resources to influence US political processes, and this political influence is sought with aim to increase firm revenues and investment opportunities or to decrease costs (Stigler, 1971 and Grossman and Helpman, 1994). An important avenue through which firms attempt to gain influence is via contracts with lobbyists, who market to firms their connections with politicians, regulators, and other political entities.1 Over $30 billion has been spent on lobbying activities over 1998–2010, and a large majority of these resources were dispersed by firms and their constituencies.2 Clearly, political influence and lobbying activities are important to firms, and a reduction in taxes owed can either improve the feasibility of firm investment opportunities or decrease firm operating costs (or both). Tax policies under consideration by Congress and regulators attract much of the political activity by firms and their lobbyists.3 According to publicly available data collected by the Senate Office of Public Records, tax issues are the third most popular reason for lobbying over 1998–2010.4Fig. 1 shows the number of lobbying clients (i.e., legal entities: firms, trade groups, individuals, etc.) by the top twelve issue categories over 1998–2010. Tax issues are secondary only to budget appropriations and health issues, and are lobbied by more entities than are defense, transportation, and energy issues. In this paper, we ask whether empirical proxies for tax avoidance are directly related to tax lobbying activity, and whether shareholders value this tax minimization strategy. In our first set of results, we test the hypothesis that one outcome of a successful lobbying effort is a lower tax burden by analyzing firm lobbying activity for tax purposes and several empirical proxies for tax avoidance, including book effective tax rates and discretionary permanent book-tax differences. We find that lobbying Congress for tax purposes has a 150–230 basis point negative effect on the firm’s book effective tax rate, and a 30–60 basis point positive effect on discretionary permanent book tax differences scaled by book assets. These results hold for a number of robustness checks, including controls for selection and propensity score matching, and are consistent with the conclusions drawn in Richter et al. (2009), who find that general lobbying activity (i.e., lobbying for any purpose) is related to lower book effective tax rates. Richter et al. (2009) find that lobbying firms have on average between 0.5% and 1.6% lower effective tax rates relative to non-lobbying firms. We improve upon and extend Richter et al.’s important first step by analyzing several proxies for a firm’s tax minimization strategies, and show that lobbying affects not just effective tax rates but also book tax differences. While effective tax planning, and even sheltering, will manifest through lower effective tax rates, this measure is affected by financial reporting conventions (through total tax expense) and may not fully represent discretionary tax avoidance activity. Accordingly, we analyze a measure of discretionary tax planning as described in Frank et al. (2009). Our empirical strategy also addresses some of the important econometric issues not addressed in Richter et al., namely short panel bias and the use of potentially weak instruments in selection models. Specifically, Richter et al. estimate a dynamic panel regression of firms’ effective tax rates and model firm fixed effects, making their results potentially sensitive to short-panel bias (i.e., Nickell (1981) bias).5 Richter et al. also attempt to control for selection effects due to a firm’s decision to lobby, but use the natural log of the number of firm employees as an identification instrument. This instrument is highly correlated with firm size, a common determinant in regressions explaining tax outcomes (e.g., Chen et al., 2010 and McGuire et al., 2012). We use a different set of instruments that are more likely to be related to the lobbying decision, but less likely related to tax outcomes, and we provide further evidence of robustness by employing propensity score matching. We also have an improved measure of tax-motivated lobbying for our sample firms. Richter et al. use a dummy variable and the level of lobbying expenditures for any purpose, and relates that activity to book effective tax rates. Our measure of lobbying is a dummy variable indicating that the firm lobbied specifically for tax purposes, and we relate this activity to our expanded set of tax outcomes. In our second set of results, we investigate whether shareholders value these political tax strategies. This is an important extension of research on tax avoidance because the literature has thus far found mixed results on the value of corporate tax strategies (e.g., Desai and Dharmapala, 2009 and Hanlon and Slemrod, 2009). A shortcoming of the tax avoidance literature is typically identification – tax returns are not observable, so tax avoiders must be estimated or researchers must rely on ex post revelations of tax aggressiveness. We find that shareholders value tax-motivated lobbying, unless the firm ex ante has a low effective tax rate or very large book-tax differences.6 These results are also important not only due to the substantial corporate resources allocated to both taxes and lobbying, but also because tax avoidance and corporate lobbying research suggests that these activities are separately associated with firm agency risks.7 Tax strategies are motivated by management as a means to minimize costs, but there can be risks attendant these strategies, and lobbying policymakers is one method by which firms can manage this risk. The rest of the paper proceeds as follows. In the next section we review related literature and develop our hypothesis. Section 3 describes the research design. Section 4 presents our results and discusses additional analyses. Section 5 concludes.

In this paper, we ask whether empirical proxies for tax avoidance are directly related to tax lobbying activity, and whether shareholders value this tax minimization strategy. In our first set of results, we test the hypothesis that one outcome of a successful lobbying effort is a lower tax burden by analyzing firm lobbying activity for tax purposes and several empirical proxies for tax avoidance, including book effective tax rates and discretionary permanent book-tax differences. We find that lobbying Congress for tax purposes has a 150–230 basis point negative effect on the firm’s book effective tax rate, and a 30–60 basis point positive effect on discretionary permanent book tax differences scaled by book assets. These results hold for a number of robustness checks, including controls for selection and propensity score matching, and are consistent with the conclusions drawn in Richter et al. (2009), who find that general lobbying activity (i.e., lobbying for any purpose) is related to lower book effective tax rates. Richter et al. (2009) find that lobbying firms have on average between 0.5% and 1.6% lower effective tax rates relative to non-lobbying firms. We improve upon and extend Richter et al.’s important first step by analyzing several proxies for a firm’s tax minimization strategies, and show that lobbying affects not just effective tax rates but also book tax differences. While effective tax planning, and even sheltering, will manifest through lower effective tax rates, this measure is affected by financial reporting conventions (through total tax expense) and may not fully represent discretionary tax avoidance activity. Accordingly, we analyze a measure of discretionary tax planning as described in Frank et al. (2009). Our empirical strategy also addresses some of the important econometric issues not addressed in Richter et al., namely short panel bias and the use of potentially weak instruments in selection models. Specifically, Richter et al. estimate a dynamic panel regression of firms’ effective tax rates and model firm fixed effects, making their results potentially sensitive to short-panel bias (i.e., Nickell (1981) bias).5 Richter et al. also attempt to control for selection effects due to a firm’s decision to lobby, but use the natural log of the number of firm employees as an identification instrument. This instrument is highly correlated with firm size, a common determinant in regressions explaining tax outcomes (e.g., Chen et al., 2010 and McGuire et al., 2012). We use a different set of instruments that are more likely to be related to the lobbying decision, but less likely related to tax outcomes, and we provide further evidence of robustness by employing propensity score matching. We also have an improved measure of tax-motivated lobbying for our sample firms. Richter et al. use a dummy variable and the level of lobbying expenditures for any purpose, and relates that activity to book effective tax rates. Our measure of lobbying is a dummy variable indicating that the firm lobbied specifically for tax purposes, and we relate this activity to our expanded set of tax outcomes. In our second set of results, we investigate whether shareholders value these political tax strategies. This is an important extension of research on tax avoidance because the literature has thus far found mixed results on the value of corporate tax strategies (e.g., Desai and Dharmapala, 2009 and Hanlon and Slemrod, 2009). A shortcoming of the tax avoidance literature is typically identification – tax returns are not observable, so tax avoiders must be estimated or researchers must rely on ex post revelations of tax aggressiveness. We find that shareholders value tax-motivated lobbying, unless the firm ex ante has a low effective tax rate or very large book-tax differences.6 These results are also important not only due to the substantial corporate resources allocated to both taxes and lobbying, but also because tax avoidance and corporate lobbying research suggests that these activities are separately associated with firm agency risks.7 Tax strategies are motivated by management as a means to minimize costs, but there can be risks attendant these strategies, and lobbying policymakers is one method by which firms can manage this risk. The rest of the paper proceeds as follows. In the next section we review related literature and develop our hypothesis. Section 3 describes the research design. Section 4 presents our results and discusses additional analyses. Section 5 concludes.