The Effect of State and Local Taxes on Economic Growth: A Time Series--Cross Section Approach

Results based on pooled time series and cross section data are presented, which indicate that state and local tax increases significantly retard economic growth when the revenue is used to fund transfer payments. However, when the revenue is used instead to finance improved public services (such as education, highways, and public health and safety) the favorable impact on location and production decisions provided by the enhanced services may more than counterbalance the disincentive effects of the associated taxes. These findings underscore the importance of considering the incentives provided by a state's expenditures as well as by its taxes. I N the past several years a number of states have implemented limitations on taxes or expenditures, while others have raised tax rates in response to the revenue shortfalls of the early 1980s. Although much of the ongoing debate over state fiscal structures revolves around expressed preferences over tax-expenditure combinations per se, the question of how state and local taxes may affect economic growth is often central to the discussion. Nonetheless, the empirical evidence remains sparse. Previous studies can be divided into two broad categories: assessments of the magnitudes of tax costs, and statistical analyses of the relationship between tax rates and growth. Cost studies have tended to minimize the role of taxes in business location decisions, either because state and local taxes are a relatively small component of total costs for most businesses, or because interstate tax differences are small relative to other cost differences between states.' Although these results suggest that the role played by taxes is a minor one, describing how large the tax disparities may be is not the same as ascertaining their economic effect. Of the statistical studies many, including Bloom (1955), Thompson and Mattila (1959), and Carlton (1979,1983), find no relationship between taxes and growth. However, Kleine (1977), Grieson (1980) and, generally, Grieson et al. (1977) report negative correlations. Romans and Subrahmanyam (1979) find growth to be negatively related to the fraction of revenues devoted to transfer payments and a variable which they claim measures personal tax progressivity, but positively related to business taxes. With the exception of Carlton (who estimates models of the birth and size of firms within metropolitan areas) and the Grieson et al. (1977) and Grieson (1980) studies (which consider taxes in New York City and Philadelphia), these authors use data from a single cross section of states. More recently, Newman (1983) and Plaut and Pluta (1983) both use cross sectional data from two time periods, the former finding negative effects of corporate tax rates and the latter yielding mixed results. In this study we utilize a time series (from 1965 through 1979) of cross sections of 48 states. Exploiting the richness of this data set, we are able to allow for otherwise unexplained differences between states which could be expected to bias results based on a single cross section. Moreover, this much larger sample provides the statistical power necessary to disentangle the separate effects of different categories of taxes, public expenditures, and transfer payments; a distinguishing feature of this paper is that we explicitly recognize the government budget constraint linking these fiscal variables and the deficit or surplus. In addition, we have taken account of the effects of labor force characteristics by including variables for wage rates, unionization, and population density. Our results indicate that tax increases significantly retard economic growth when the revenue is used to fund transfer payments; as a result, programs of income redistribution are more effectively undertaken at the federal than at the state and local level. On the other hand, when the revenue is Received for publication April 30, 1984. Revision accepted for publication January 22, 1985. *University of California, Davis. I am indebted to the Institute of Governmental Affairs at the University of California, Davis for financial support, to Adrienne Kandel for invaluable research assistance, and to two referees for their helpful comments. 1 See Williams (1967), Morgan and Brownlee (1974), Vasquez and deSeve (1977), and the Advisory Commission on Intergovernmental Relations (1981). See also Due (1961) for a survey of the early literature, including a discussion of the findings and limitations of survey results on this topic. [ 574 1 Copyright ? 1985 This content downloaded from 157.55.39.35 on Mon, 29 Aug 2016 04:45:38 UTC All use subject to http://about.jstor.org/terms STATE AND LOCAL TAXES AND ECONOMIC GROWTH 575 used instead to finance enhanced public services (such as highways, education, and public health and safety), the favorable impact on location and production decisions provided by the improved services may more than counterbalance the disincentive effects of the concomitant taxes. These findings underscore the importance of considering the incentives provided by a state's expenditures as well as by its taxes. We begin by sketching a model which provides the basis for the analysis and discussing appropriate estimation techniques. Then in section II we describe the data used and our hypotheses. The econometric results are presented in section III, followed by brief concluding remarks in the final