Coveting thy neighbor’s manufacturing: the dilemma of state income apportionment

Abstract

This paper investigates the economic impact of the apportionment formulae used to divide corporate income taxes among the states. Most apportionment formulae, by including payroll, turn the state corporate income tax at least partially into a payroll tax. Using panel data from 1978–1994, the results show that this distortion has an important effect on state-level employment. For the average state, reducing the payroll weight from one-third to one-quarter increases manufacturing employment around 1.1%, concentrated in durable goods manufacturing and with larger effects in the long-run. The results also suggest that apportionment changes have important negative externalities on other states. On average, the aggregate effects of apportionment formula changes are close to zero.

Introduction

Faced with the continuing devolution of resources and responsibilities from the federal government back to the states and the decline of manufacturing employment throughout the country, state governments have repeatedly changed their tax systems in the last 20 years to encourage employment and investment. Enterprise zones, tax concessions and corporate rate cuts have been common and have also been the subject of extensive academic research.¹ Most work has matched aggregate data on employment or investment to general tax measures like the average corporate tax burden in a state or the statutory corporate tax rate but found negligible effects of tax policy. More recent studies looking at micro-level investment decisions or at more specific tax incentives, such as Bartik, 1985, Papke, 1991, or Hines (1996), have found larger effects.

An important feature of state taxation that has been relatively neglected in empirical work is the state income apportionment formula by which companies allocate their national income across state tax jurisdictions.² This formula usually attributes income based on a firm’s geographic distribution of payroll, property, and sales. McLure (1980) first demonstrated that formula apportionment largely transforms the state corporate income tax into three separate taxes on the factors in the apportionment formula. Since payroll is usually one of the factors, the formula could have important effects on employment.

State legislatures and their constituents seem to understand this and have actively attempted to modify their states’ apportionment formulae to stimulate employment and investment. The theoretical work of Gordon and Wilson (1986) and later Anand and Sansing (1997), however, indicates that the tax choices of individual states can have negative externalities on their neighbors and this may put states into a prisoner’s type dilemma that can lead to a series of beggar-thy-neighbor policies.

The impact of state tax policy on economic performance and the externalities of tax policy on other states are central to public finance and the apportionment formula provides an excellent place to examine such issues. In addition, the rising interest in inter-jurisdictional tax issues as well as recent proposals to convert the taxation of multinational companies to a cross-country formula apportionment system similar to the system used within the United States has generated considerable interest among policy makers in the economic effects of formula apportionment.³

In this paper, we use panel data for the U.S. states from 1978 to 1994 and control for a variety of non-tax factors to examine the relationship between employment and state apportionment formulae. In doing so, our results establish two important facts about these state tax policies.

One, we provide the first robust evidence that the apportionment formula has a significant real effect on a state’s economy. The results suggest that the payroll weight is a significant determinant of state employment. Although there may be other unobserved policy changes contributing to the result, we find that for the average state, reducing the payroll weight from one-third to one-quarter increases manufacturing employment by approximately 1.1%. Further, we show that these significant employment effects imply that although increasing the sales weight in a state may lead to corporate income tax revenue losses (see Pomp, 1987), the increased employment generates an indirect source of additional personal income tax revenue. The results suggest that this additional revenue reduces and may even exceed the corporate revenue loss for some recently proposed formula changes.

Two, we are able to explore the externalities that one state’s apportionment formula has on other states. We show that while the within state employment effect of changing the apportionment formula is large, the aggregate effect is close to zero. Employment does increase in states that reduce their payroll weight but the same change reduces employment in other states. In this way our results show why it is difficult to maintain harmony in the formulae across states, a fact fully consistent with theoretical predictions and with the ongoing trends in apportionment decisions. The negative externality suggests that the U.S. as a whole might be better off moving away from a decentralized system to a nationally uniform apportionment formula along the lines of Canada.

The paper proceeds in six parts. Section 2 presents a simple theoretical model of how formula apportionment affects the employment decisions of firms and describes the existing literature in the area. Section 3 discusses the panel data and the empirical specification. Section 4 presents the results. Section 5 discusses the revenue implications and measures the external effects. Section 6 concludes.