Where do U.S. Multinationals Pay Taxes?
Tax payments (or non-payments) of U.S. foreign affiliates to host governments
Nathan Jensen
Assistant Professor
Department of Political Science
Washington University in St. Louis
Abstract:
The literature on tax competition across countries for mobile firms has largely ignored the firm. In this paper I utilize aspects of the bargaining literature on firm-government relations to examine the firm and country level determinants of tax policy. Using a confidential data set on firm tax payments from the U.S. Bureau of Economic Analysis I show that: 1) there are large variations within countries on the tax burdens faced by firms, 2) parent level factors strongly influence the ability of firms to be exempted from corporate taxation and 3) political institutions strongly affect both exemptions from corporate taxes and the level of corporate taxes paid.
Acknowledgements:
Funding for this project was provided by the Weidenbaum Center on the Economy, Government, and Public Policy at Washington University in St. Louis.
The statistical analysis of firm-level data on foreign-owned U.S. affiliates was conducted at the International Investment Division, Bureau of Economic Analysis, U.S. Department of Commerce under arrangements that maintain legal confidentiality requirements. The views expressed in this paper are those of the author and do not necessarily reflect official positions of the U.S. Department of Commerce
1. Introduction
On May 11th, 2007 German steelmaker ThyssenKrupp’s announced plans to build a $3.7 billion steel plant in Mobile, Alabama employing 1,700 workers. ThyssenKrupp was awarded an extremely generous investment incentive package of $810 million in up-front subsidies and exemptions from the state corporate taxes that could amount to as much as $3.7 billion.[1] The investment also came with a contract binding the Alabama government to shield the company from specific types of new taxes on fuel.
These investment subsidies aren’t new; location incentives packages to induce investment have been utilized to attract capital since at least 1160.[2] Yet there is a growing uneasiness that the mobility of capital is leading to an escalation of these incentive packages, straining government resources and leading to taxpayer revolts.[3]
This concern over incentive wars are part of a broader question in political science and economics on how the mobility of capital affects the capacity of governments to tax and spend. The inability to tax capital can lead to efficiency enhancing reductions in government intervention in the economy, economically inefficient provision of public goods, or distributional consequences where governments shift the burden of taxation from mobile factors (capital) to immobile factors (land and labor).
In this paper I do not directly address the depth of this competition or the normative implications of the (in)ability of governments to tax footloose capital. Rather I attempt to refocus this debate from the competition between countries for mobile capital, to an exploration of complex bargaining between individual firms and host governments over taxes. This focus on government-firms bargaining is by no means new; rather it is a throwback to an influential area of scholarship in management and economics. My goal is to bring this bargaining literature to the political science debates on capital mobility and tax competition.
I focus on firm level (foreign affiliate) analysis of the determinants of taxes paid by U.S. multinationals by utilizing a data resource measuring the amount of taxes paid to foreign governments. This confidential data from the Bureau of Economic Analysis includes the complete universe of the investments of U.S. multinational corporations abroad. I focus on the country level and firm level determinants of corporate taxation in non-OECD countries, and leave an analysis of tax payments to OECD governments to future research.
This firm level perspective offers a descriptive view on the variation in firms’ corporate tax payments. First, there is a tremendous amount of variation in the amounts, rates, and complete exemptions from corporate taxation within countries. Two firms in the same host country pay vastly different amounts in taxation. Second, there are important parent level effects that are captured in the data. There is a very strong correlation between the tax activity of foreign affiliates from the same parent firm. An excellent predictor of whether or not a firm gets exempted from taxation is that ability of the parent firm to get exemptions for foreign affiliates in other countries. Third, as shown in the empirical analysis, political institutions have a strong impact on taxes paid by firms, even after taking into account national level tax policy.
This firm-level data analysis of all U.S. investments has advantages over the existing literature. First, this is one of the few studies to directly address taxes at the firm level. Second, my project doesn’t focus on the firms that achieved government subsidies, or the allocation of subsidies from specific subsidy program. My analysis focuses on the actual tax burden faced by firms. What explains differences in tax payments across foreign affiliates?
2. Tax Competition and MNCs
There is an extensive literature in political science and economics that explores the drivers of tax competition across governments for mobile capital. States, in an attempt to attract higher levels of international investment and deter domestic investors from “voting with their feet” by relocating to a lower tax jurisdictions, strategically manipulate tax policy to attract investment. Although a full review of this growing literature is beyond the scope of this paper, I will concentrate on highlighting some of the major contributions.
In the public finance literature, the classic Zodrow-Mieszkowski model and Wilson model explores how governments weigh the benefits of higher tax rates (usually in terms of providing public goods) against the loss of potential mobile capital that is deterred by high tax rates.[4] Governments in these models are maximizing the welfare of the representative household, charging a common tax rate on international capital. This doesn’t require that all governments charge the same tax rate, yet government optimization decisions on tax rates are affected by the after tax rates of return to capital in other countries.[5]
This incentive to attract capital can thus lead to an inefficient level of public good provision or shifting of the burden of taxation. The standard models of Zodrow and Mieszkowski (1986) and Wilson (1986) show how the competition for capital can lead to the under-provision of public goods. In the political science literature this is often referred to the “race to the bottom” where the competitive beggar thy neighbor tax cuts can lead to inefficient levels of capital taxation.[6]
These pressures for reductions in tax rates can also lead to a positive impact on welfare. As pointed out by Wilson and Wildasin, the standard models “assume welfare-maximizing governments and models of an economy that would be fully efficient if capital were not interregionally mobile. This seems to stack the deck against tax competition.” (Wilson and Wildasin 2004, 1066). As argued by Brennan and Buchanan (1980) this competition can have a positive impact on efficiency if tax competition disciplines governments from excessive taxation. Governments are given incentives to help firms maximize the after tax returns to capital, making important decisions on how to raise revenue, weighing the benefits of public goods provision against the loss of capital in response to higher rates of corporate taxation. The Leviathan of government may want to expand the scope of the state, but capital mobility punishes governments for excessive taxation.
No matter what the label or one’s normative position on corporate taxes, the competition for capital will drive down corporate tax rates. This competition can be welfare enhancing (as in the Leviathan models), have a negative impact on country welfare by leading to an under-provision of public goods, or have distributional consequence as the burden of taxation is shifted from mobile capital to other groups in society.[7] More important for this study, these models have focused on how the competition for capital affects government decisions on national levels of corporate taxation.
Even with this competition for international capital, scholars have pointed out the lack of a clear race to the bottom in corporate taxation. One reason for the lack of national-level convergence is that the assumption that governments maximize the welfare of the representative household, while theoretically convenient, misses many of the complex political issues surrounding corporate taxes. Even if markets “demand” a reduction in corporate tax rates, this doesn’t natural lead to tax reduction. If politicians want to “survive” in office, tax policy will be driven by electoral, not efficiency considerations.
Numerous political scientists have theoretically and empirically explored international tax competition and tax rate convergence across countries, mostly focusing on OECD countries. Swank and Steinmo (2002) finds a convergence in statutory corporate taxes, yet the effective tax rates and corporate tax receipts have remained relatively stable over time. Hays (2003) finds some convergence in corporate tax rates, yet there is little evidence for the race to the bottom. Swank (2004, 2006) finds that tax competition is a factor that drives international tax policy, specifically in a shift towards market conforming tax rates triggered by corporate tax reforms by the Reagan administration. Yet, this form of tax competition and diffusion is more nuanced than the traditional race to the bottom literature.
This competition for capital is only one part of government’s decisions on corporate taxation. Recent work by Pinto and Pinto (2006) explores how the complementariness of foreign capital with domestic labor and domestic capital affect tax policy. Thus partisan incumbent governments (representing domestic capital or domestic labor) have differing incentives on the tax rates charged to multinationals. Tax competition affects government decisions on tax rates, but the partisanship of governments affects the incentives of politicians to induce capital inflows.
This importance of domestic politics is central to the work of Basinger and Hallerberg (2004). They model tax competition for mobile capital as a tournament model, where competition affects politicians’ incentives for tax policy change, yet domestic politics affects the political costs of tax policy reform. Leftist governments and governments with multiple veto players are less likely to pay the political costs of reform and reduce corporate tax policy. Swank (2006) also argues that shifts towards market conforming tax policy are largely tempered by domestic politics.
The above studies all focus on national level rates of taxation or the level of aggregate corporate tax revenues. Li (2006) takes a new approach, focusing on the menu of tax incentives offered by governments to individual investors. He finds that political institutions are important drivers of the number of tax incentives a government can potentially offer an investor.
These works all provide important insights into tax competition between governments, yet they fail to explore the bargaining between individual firms and potential host governments. As I will show in the data section, firms within a single country are subject to very different tax burdens. Some firms pay close to the top corporate tax rate, while a larger percentage of firms are completely exempted from corporate taxation. Thus we have insights into how governments compete with each other, reducing rates of corporate taxation, but what determines the actual tax burden a firm faces within a country?
3. Bargaining Between Firms and Host Governments
An extensive literature in management explores the bargaining relationship between potential investors and host governments.[8] These studies focus on the initial entry decision of the firm, where potential host governments negotiate with multinationals over the conditions of investment, the operations of the firm, and the distribution of benefits (such as revenue sharing of natural resource production). Although the location decisions and types of investment are partially determined by economic and firm specific factors, there is considerable room for negotiation between firms and governments.
For example, in the 1970s and 1980s many host governments required specific “performance requirements” from investors, requiring the hiring of local managers, exporting final goods, etc while firms were assured monopoly protections from competitors.[9] Another example is the story of ThyssenKrupp’s investment used to motivate this paper. The large incentive package from the state of Alabama is conditional on the firm achieving and maintaining employment levels of 2,000 plus workers.
This bargaining between governments and firms isn’t static. Vernon’s (1980) obsolescing bargaining model identifies the complexity of this negotiation process where multinationals in extractive industries strike agreements over the distribution of future revenues. Yet, once these massive, high-fixed costs investments have been made firms can become hostages of the host government. Terms agreed to in the initial bargain becomes “obsolete” as host governments renegotiate contracts. One very recent example is Bolivia’s 2005 Hydrocarbon’s Law that effectively required the renegotiation of existing contracts with foreign multinationals in the natural gas sector.
Kobrin (1989) argues that this divide the dollar game over the benefits of multinational production, traditionally applied to investment in extractive industries, can also be expanded to analyze manufacturing agreements. Kobrin finds some evidence of this shift in bargaining power over time, although multinational resources, such as access to advanced technology, are important drivers of bargaining power.[10]
As noted by Ramamurti (2001), this bargaining literature has converged to a number of factors that affect the bargaining relationship. For the potential investor, the source of the MNCs bargaining power is: technology, product differentiation, ability to bring in capital, exports, product diversity, worldwide size/scale, and potential to play countries against on another. Thus multinationals firms with scarce resource prized by governments have an ability to demand better entry conditions, including looser regulations and the ability to invest in a wholly owned firm.
Host governments also have resources that affect their ability to negotiate with multinationals. These include the granting access to home market, especially key when market is big and growing rapidly. Even if the government has a small market that isn’t prized by multinationals, states may have access to natural resources, local labor or other country specific factors that are prized by multinationals. Finally the ability to offer incentives and potential to play MNCs against one another all affects this bargaining relationship.
This bargaining between governments doesn’t always resemble a divide the dollar game, where there is plenty of room for collaboration between firms and governments. As Luo (2001) points out, there is considerable room for cooperative relations between firms and governments. For this paper it isn’t necessary to assume a zero sum negotiation between firms and governments, rather I focus on the factors that lead to large reductions in corporate taxes for firms.