Principal investigator: Bruno Benevit
Authors: Xavier Giroud and Joshua Rauh
Original title: State Taxation and the Relocation of Business Activity: Evidence from Establishment-Level Data
Location of the Intervention: United States
Sample Size: 27.6 million firms per year
Sector: Business
Primary Variable of Interest: Opening a business
Type of Intervention: Tax rate
Methodology: OLS, Poisson, DID
Summary
Tax policies can significantly shape the business landscape, affecting not only job creation but also capital formation and broader economic growth. Aiming to analyze the impact of corporate taxation, this study examined the impact of state taxes on the business activity of multi-state firms in the United States. Corporate taxation exerted a substantial impact on the labor market, with notable variations in employment in response to changes in fiscal policies. The influence of taxation on company capital proves to be a determining factor in economic dynamics, highlighting the sensitivity of investments to changes in tax policies.
- Policy Problem
Corporate taxation plays a crucial role in shaping employment, with fiscal policies directly impacting labor market dynamics. The relationship between taxation and capital proves to be a key element in corporate financial decision-making, influencing investments and resource allocation.
Thus, the impact of corporate taxation on these factors is commonly addressed in the public finance literature. These impacts can occur both in marginal incentives through effective marginal tax rates and the cost of capital, and in the choice of firm location through the average tax rate. However, the mobility of firms to more favorable business environments can mitigate the incidence of negative effects of this nature (GIROUD; RAUH, 2019). In this sense, multi-state companies may exhibit distinct behaviors in the face of the establishment of state taxes.
- Policy Implementation Context
In the United States, the structure of state corporate taxation is similar to federal tax legislation, where companies that decide to incorporate have the option of taxation at the entity level according to the corporate tax code at the federal level (GIROUD; RAUH, 2019). Among the federative entities, however, there is great heterogeneity with respect to state tax regimes.
This context further complicates the analysis of the effects of tax policy on corporate activity, given the varying state laws regarding how taxable income should be distributed to multinational companies. Unlike the federal tax treatment for multinational companies, states use distribution formulas that eliminate the need to track domestic prices, involving three different measures of economic activity: sales, payroll, and property. A company must first determine which states have the power to tax the business, considering its physical presence, and then apply the distribution formula in each state with a "nexus".
Companies incorporated under Subchapter C of the federal tax code (C Corporations) are required to pay taxes at the corporate rate. Under this regime, the owners of these companies paid individual taxes when they received dividends from the C Corporations or when they realized capital gains. On the other hand, similarly to unincorporated companies such as partnerships and sole proprietorships, companies incorporated under Subchapter S of the federal tax code (S Corporations) are considered pass-through entities. Pass-through entities do not pay taxes at the company level, passing all profits to their owners, who would otherwise have to pay taxes immediately on their profits. Finally, American companies can also organize themselves as Limited Liability Corporations (LLCs), providing some of the benefits of corporate organization, such as full liability protection, avoiding taxation at the firm level under the federal corporate tax code.
- Evaluation Details
This study used data at the establishment level through Census Bureau's Longitudinal Business Database (LBD) to examine the impact of state taxation on employment and capital. Establishments were defined as “unique physical locations where business is conducted,” for which the LBD provides data on employment, payroll, industry sector, location, and firm identifier. In addition, the sample considered was supplemented with data from Census of Manufactures (CMF) and of Annual Survey of Manufacturers (ASM). Both samples provide detailed data on establishment-level information, such as capital expenditures, total assets, and the value of remittances from a more restricted set of establishments. These two databases provide a richer set of establishment-level variables, although they are less comprehensive.
The primary sample for the study consisted of all establishments in the United States from 1977 to 2011 belonging to companies with operations in at least two states and with at least 100 employees present in the LBD. This sample consisted of 27,6 million establishment-year observations, equivalent to 647.000 firm-year observations. Additionally, a secondary sample was established considering establishments from CMF and ASM data, resulting in a new sample of 854.700 establishment-years (104.400 firm-years).
The identification of the legal form is based on the company's tax filing status with the data from Standard Statistical Establishment List (SSEL). SSEL provides a legal form of organization based on tax for all businesses through the LBD firm identifier. Businesses can be listed under one of seven possible legal forms. This study considered only the categories of (i) sole proprietorships, (ii) partnerships, and (iii) corporations. While categories (i) and (ii) are always considered pass-through entities for tax purposes, firms organized as corporations have the possibility of being C or S Corporations. Naturally, the tax regime and rates of many states have changed over this period, and other data sources were used to compile tax characteristics: a University of Michigan Tax Database, Tax Foundation, o Book of States, o Commerce Clearing House's State Tax Handbooks, o Census of Government State & Local Financesin addition to other data sources.
- Method
The study analyzed the impact of taxes on business activity using the distinction between S corporations and C corporations. This empirical strategy was established to explore the fact that the corporate tax code directly affects only companies organized as C corporations, unlike companies organized as S corporations, partnerships, or sole proprietorships, which are only affected by the individual tax code and other business taxes. The analysis was conducted only on companies with establishments in multiple states according to their organizational forms at the federal level.
To this end, Ordinary Least Squares (OLS) and Poisson regression models were initially established to estimate the interaction between the two types of taxes (corporate and personal) and the two types of corporations (C and transfer entities). The dependent variables related to business activity analyzed included the number of establishments, at the firm-state-year level, and the number of employees and capital, at the establishment-year level. To measure the effects resulting from relocation to other states, new models were considered with the inclusion of terms that controlled for the average tax rate of all other states (excluding the observation state). Non-tax factors and fixed year and firm-state effects were controlled as covariates.
Additionally, several strategies were implemented to verify the robustness of the results. First, the authors performed estimates focusing on companies with a presence in multiple states in order to minimize potential effects of organizational change in response to state taxes. Second, the impact of 161 major state tax changes (at least 100 basis points) on the number of establishments was estimated using a difference-in-differences (DID) model. Third, the motivation for changes in federal personal income tax rates was verified as established by Romer and Romer (2010), classifying changes as "exogenous" or "endogenous," depending on the motivation. To this end, news coverage regarding tax changes was analyzed to verify the classification of the changes, as well as the impact of two federal reforms (Economic Recovery Tax Act from 1981 e Tax Reform Act 1986).
- Main results
Estimates from the main analysis revealed that a 1 percentage point change in the state corporate tax rate corresponds to an average change of 0,5% in the number of establishments per Corporation C. Similarly, a 1 percentage point change in the state personal tax rate affects the number of establishments per passing entity by 0,4%. No significant correlations were identified between the activity of passing entities and corporate tax rates, nor between corporate activity and personal tax rates.
In the analysis regarding the number of employees per establishment, the results were similar. The effective marginal rate has a more significant impact than the nominal rate, especially in margin-intensive companies. When focusing on manufacturing firms, similar patterns were observed for capital, although with elasticities 31%-35% lower for capital.
Regarding analyses in response to tax changes in other states where companies operate, the results indicated that approximately half of the effects of tax variations are offset by the relocation of activity between states. These results indicate that tax competition between states is economically relevant.
Regarding robustness analyses, the analysis to verify organizational changes in firms in response to tax changes indicates that the response of companies is more significant when the physical location of employees and properties has a greater weight in the allocation of the tax burden to a given state. The effects remain significant even in cases where greater weight is given to taxation linked to the location of sales. The analysis considering major tax changes demonstrated that these changes have impacts similar to those found in the complete sample, where approximately half of the effects are felt in the fiscal year of the change and the full force in the following year. When analyzing changes in response to the federal tax reforms of 1981 and 1986, effects of comparable magnitude to other major changes in corporate and personal tax rates were identified.
- Lessons in Public Policy
This article explored the impacts of changes in state tax regimes on multistate corporations in the United States, examining both the extent and intensity of their economic responses. By analyzing firms organized as C Corporations and pass-through entities, we observed that variations in corporate and personal tax rates across states result in significant adjustments in the number of establishments, employees, and capital. It was observed that even when companies have non-tax reasons for locating in different states, state tax rates proved to play a crucial role in location decisions. Notably, approximately half of the identified responses are attributable to the relocation of business activities to states with more favorable taxation. The study's evidence highlights the economic relevance of state tax policies and their significant implications for business behavior.
References
GIROUD, X.; RAUH, J. State Taxation and the Relocation of Business Activity: Evidence from Establishment-Level Data. Journal of Political Economy, v. 127, no. 3, p. 1262–1316, 2019.
ROMER, CD; ROMER, DH The Macroeconomic Effects of Tax Changes: Estimates Based on a New Measure of Fiscal Shocks. American Economic Review, v. 100, no. 3, p. 763–801, 2010.
