The Political Economy of Corporate Taxation

Abstract

ion is not necessary. Rather, we make our case on more substantive grounds. 9One reviewer suggested that corporate tax burden rather than tax rate is the most appropriate dependent variable. We disagree, because we would predict that corporate tax burdens would decline over time even when tax rates remain stable. This happens because government expenditures, largely a function of entitlements, have increased over time, and This content downloaded from 157.55.39.144 on Wed, 21 Sep 2016 06:36:45 UTC All use subject to http://about.jstor.org/terms THE POLITICAL ECONOMY OF CORPORATE TAXATION 217 makers can avoid shifting all the burden to individual income tax payers. So far we assume that members of Congress want to provide some firms with particularistic benefits, while keeping revenues from corporate taxation as high as possible. Together, this means that some firms will win and some will lose. We assume that who loses and gains will probably be a function of more micro-level (firm or industry) indicators of power (see Grier, Munger, and Roberts 1994). At least in the context of tax policy, business as a unitary interest group is inappropriate even though all firms want lower taxes (Lindblom and Woodhouse 1993; Martin 1991; Vogel 1989). Therefore, from these assumptions we hypothesize that the ECTR will not respond to any indicator of business political power in the aggregate; for example, the ECTR will be exogenous to aggregate indicators of business political strength. Following Quinn and Shapiro (199la) we use the aggregate number corporate political action committees (PACs) relative to other PACs as an indicator of business political influence. In contrast, we expect that aggregate business influence (via PACs) will be endogenous to the ECTR because as the tax rate increases, there are additional incentives for firms to organize and protect their interests (Grier, Munger, and Roberts 1994). We will state this argument as: Proposition 1: The ECTR will be exogenous to the political strength of business interests in the aggregate. Rather, an increase in the ECTR will lead to an increase in business political organizational activity.10 However, if firms are not capable of solving their collective action problem and forcing Congress to lower the ECTR, why do the rates remain at relatively low levels?11 That is, why does Congress not tax corporations at extremely high rates, given that these revenues could be redistributed to a majority of individuals who would then reward members of Congress with reelection? If our equilibrium theory explains corporate tax policy, then the answer can be found in Lindblom's argument that private firms have a preferred position in a capitalist democracy because they can withdraw investment. This does not, however, imply that business must directly organize a capital strike; collective disinvestment is usually unnecessary. Thus, structural dependence theory is difficult to test because of limited our theory indicates that the government is constrained in being able to increase the ECTR but has not been constrained in producing higher budget deficits and increasing social security taxes. Thus, a model of tax burden would be different from a model of the ECTR. 0We will test this proposition, but note that our theory suggests the overall impact of business interests on corporate tax policy will primarily be evidenced at a micro level, not at the macro level. This conjecture will be addressed in further research, but proposition one will face empirical scrutiny in this research. Steinmo (1993) argues that the United States is hyper-pluralist, resulting in high nominal corporate tax rates coupled with low revenue yields. This content downloaded from 157.55.39.144 on Wed, 21 Sep 2016 06:36:45 UTC All use subject to http://about.jstor.org/terms 218 John T. Williams and Brian K. Collins variation, and mixed findings based on limited variance are not surprising (Jacobs 1988; Quinn and Shapiro 1991a; Swank 1992). Our second proposition below relies on a standard economic mechanism: expected profitability will decline in response to shocks in corporate taxation, thus resulting in less investment. The market limits politics as Lindblom suggests. We do not expect to find a positive causal relationship in which declining investment motivates a lower ECTR because the ECTR influences levels of investment through altering expectations of future aftertax profits. Firms act upon these expectations to take account of projected tax policies.12 Therefore, we propose the following: Proposition 2: Investment is sensitive to the ECTR; a shock in effective tax rates will reduce levels of investment. Empirical analyses of Lindblom's theory typically test the hypothesis that decreased investment is associated with decreased tax rates (Jacobs 1988; Quinn and Shapiro 1991a, 1991b; Swank 1992). In adopting structural models that specify investment as an explanatory variable for taxation, the reciprocal influence of corporate taxation over aggregate investment is omitted and thus eliminates the expectational implications of Lindblom's structural dependence argument. If researchers resort to explaining low ECTR as a function of inordinate business power through the mechanism of disinvestment, they violate important exogeneity assumptions. For example, Quinn and Shapiro's (199la) test of the structural dependency theory is built on the assumptions of an IS-LM economic model. In using the static IS-LM model, they explicitly reject the possibility that business can act according to rational, or any other, expectations. The implication of this omission is that tax policy has no causal influence over the political or economic behavior of firms.3 In short, Quinn and Shapiro's (199la) structural model does not allow corporate tax policy to feed back and cause 2Przeworski and Wallerstein (1988) develop a formal model of structural dependence in which both business and wage earners are trying to maximize profits and consumption, respectively. The state must manage the conflicting goals while providing services to both the wage earner and business. In their static model, a tax increase does not substantially affect investment because the noncapital sources of investment increase to offset any decline in business investment. In their dynamic model, however, when business anticipated tax increases, investment declined substantially. After the tax policy was initiated, investment returned to prior levels. Thus, the difference in dynamic versus static analysis is crucial for a political economy of corporate tax policy. 3There is a bit of evidence supporting a dynamic ISLM framework (Gali 1992), but most, if not all, macroeconomists have abandoned the static ISLM setup except for expository and teaching purposes. Further, Gali's analysis does not speak to tax or fiscal policy, and our results focus more on tax policy. Our framework has more in common with Gali than Quinn and Shapiro's (199lb) because our setup and Gali's consider expectations to be impor-

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 29 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.