Corporate Taxes and Wages: Evidence from the 50 States
Working Paper No. 8
Who pays corporate taxes remains one of the unanswered questions in public finance. The rationales for corporate income taxes include adding progressivity to the tax system and providing a backstop to the income tax. The former is premised on the notion that corporate income taxes are borne by owners of capital. The later presumes that without the corporate income tax, taxpayers may escape personal income taxes on capital.
Recent empirical research, however, has questioned whether corporate income taxes are indeed borne by owners of capital. This research draws on the substantial reductions in corporate tax rates internationally over the past several decades to estimate whether wages rates have risen the most in countries with the largest reductions in corporate tax rates. This research has found a substantial negative relationship between corporate tax rates and real wages with one study finding that a 1 percentage point reduction in corporate tax rates leads to a 0.8 percent increase in real manufacturing wages (Hassett and Mathur, 2006).
The intuition behind the research focusing on international changes in corporate taxes rests with one of the tenets of tax incidence: a tax will generally be borne by the least mobile factor. In an increasingly global economy, where capital flows freely across borders, but labor does not, the corporate income tax can be expected to be borne primarily by labor. Countries that are able to attract investment experience greater capital formation. Providing workers with more capital to work with increases their labor productivity, and, ultimately, their real wages.
This paper looks to the experience within the United States, rather than internationally, to empirically investigate this issue. In the aggregate, state and local corporate taxes as a share of state and local revenues has remained relatively constant at roughly 5 percent over the past several decades. In contrast, reliance on the corporate tax at the federal level has declined with corporate taxes comprising about 10 percent of total federal revenues as compared to 20 percent several decades ago.
Some states have increased their reliance on corporate taxes, while others have reduced their reliance. Moreover, the average state corporate tax rate has risen from 2.6 percent several decades ago to 4.4 percent today. This provides the exogenous variation in taxes needed to determine whether those states with the largest reductions in corporate tax rates have also had the largest gains in real wages. Using state level data also provides a stronger test of whether higher corporate taxes are borne by labor because labor is likely to be more mobile across the 50 states than internationally. Thus, a finding from this paper that is similar to the international experience would lend considerable support to the notion that corporate taxes are borne by labor rather than owners of capital.
The empirical strategy is to pool cross-sectional state level data from 1970 through 2007 to directly examine whether states with lower corporate taxes have tended to exhibit higher real wages. The empirical model also includes other factors that might influence wages rates, such as the degree of unionization, whether a state has a right to work law, and demographic features of the population. The model controls for both state and time effects to control for otherwise unobserved variables.
To anticipate the results, the paper generally finds a statistically significant, negative relationship between corporate taxes and the real hourly average earnings for production workers. A 1 percent increase in the average state and local corporate tax rate can be expected to lower workers’ real wages by 0.014 percent. This implies that for every one dollar increase in state and local corporate tax revenues, wages can be expected to fall by roughly 2.5 dollars. This result is robust to a broad range of model specifications, considerably smaller than some of the research based on the international experience, but roughly twice as large as the theoretically-based results reported by Harberger (2006). The paper also considers whether wage rates are sensitive to the corporate tax rate, although this relationship is not as robust across different model specifications.
The next section of the paper summarizes the recent research examining the relationship between taxes and wages. Section III of the paper describes a simple conceptual model. Section IV outlines the empirical model and data. The results are presented in Section V and Section VI concludes the report.