Why do countries change the taxation of foreign-source income of multinational firms?

Abstract

Why do some countries continue to tax income that multinational firms create overseas, even as other countries increasingly adopt a system that only taxes income generated within the country? I argue that this phenomenon reflects an interaction between trade openness and the number of veto players. Increasing trade openness incentivizes governments to move to a territorial tax system, because firms that operate across borders want to avoid various tax liabilities in multiple countries. Yet countries with fewer veto players are more likely to move to a territorial tax system than those with many veto players. To test my hypothesis, I employ survival and logistic regression analyses of 15 advanced industrialized countries between 1981 and 2013. Overall the findings conform to the expectation: Economically open countries with fewer veto players are more likely to shift to a territorial tax system than those with many veto players.

Keywords

Territorial taxation worldwide taxation foreign-source income of multinationals trade openness veto players

Introduction

The European Union (EU) has recently investigated tax avoidance of many American multinational corporations (MNCs) such as Google, Apple, Amazon, Starbucks, and McDonald’s (Brundsden, 2016a). The United States (USA) has criticized the European Commission for these crackdowns, arguing that they target funds the MNCs owe to the US Treasury (Oliver and Brundsden, 2016). This tension has roots, in part, in the different international tax systems of the EU and the USA. The US tax system is worldwide, and it imposes taxes on foreign-source income of MNCs upon repatriation. Conversely, the EU has a territorial tax system that imposes taxes only on profits generated in its own jurisdictions and it exempts MNCs’ foreign profits from domestic taxation. Despite tax treaties to avoid double taxation, profits generated by US multinationals in Europe can, in principle, be taxed by both governments.

Over the last two decades, many countries in the Organization for Economic Cooperation and Development (OECD) have transitioned into a territorial tax system, but the USA and a few other countries continue to follow a worldwide tax system. This is puzzling, yet political science scholarship has ignored these differences. Previous research, mostly by business scholars, has focused on the effects of changes in international tax systems on firms’ behavior, but not what might drive these changes. This is partly because the previous studies have largely ignored the politics that translate economic factors into policy outcomes. It likewise reflects a separation between studies of corporate taxation from studies of taxation of foreign-source income of MNCs. Yet, as I describe in the theoretical section, the way governments tax foreign-source income of multinationals is not distinguishable from broader corporate taxation policy in the political science scholarship.

In this article, I first attempt to explain why some countries have switched to a territorial tax system and others have not by considering the impact of both economic and political features. Drawing upon the corporate taxation literature, I argue that increasing trade openness prompts countries to adopt a territorial tax system to attract investment by MNCs. Yet domestic institutions, veto players, counteract this effect across countries. I test the interactive effects of trade openness and veto players on the likelihood of changing a country’s international tax system in a sample of 15 advanced industrialized countries between 1981 and 2013, using survival and logistic regression analyses. The main findings are that trade openness pushes countries to move to a territorial tax system, but this effect is stronger for countries that have fewer veto players.

The next section provides background and an overview of the existing literature. The third section proposes theoretical perspectives to explain the change in the international tax system. The fourth section introduces data, variables, and methods. The empirical results are presented in the fifth section. The final section summarizes findings and discusses avenues for future research and implications.

Background and literature review

Foreign-source income refers to income that MNCs generate in a foreign country. Foreign Direct Investment (FDI) host countries tax this income based on their domestic law. Thus, to avoid imposing double taxation on their MNCs, FDI source countries either provide a tax credit for foreign income taxes paid or an exemption from home country taxation (see PriceWaterhouseCoopers, 2010).

Countries can generally be categorized as having either a worldwide or territorial system for taxing resident companies’ foreign income, although most have some mixed features.¹ Countries with worldwide tax systems impose taxes on the income resident firms produce through domestic or foreign activities. Countries with a territorial tax system impose taxes on resident firms based on income only created within the country, not on that generated in foreign subsidiaries. To understand substantive differences between worldwide taxation and territorial taxation, we can take a look at two examples of how different international tax systems impose taxes on foreign-source income residential firms in the USA, which has a worldwide tax system, and Japan, which has had a territorial system since 2009. Suppose that a US multinational firm operates a subsidiary in Japan. The Japanese subsidiary makes profits and sends part of its earnings to its parent company in the USA. Then, the Japanese subsidiary pays tax to the Japanese government on their total profit. The firm needs to pay taxes on the income sent to the USA, but has a foreign tax credit that is equivalent to the taxes paid to the Japanese government (Clausing, 2012: 704–705). By contrast, a Japanese multinational firm that operates a subsidiary in the USA pays taxes on the profits of the US subsidiary to the US government. It pays no taxes on the profits generated in the USA to the Japanese government.²

Over the past two decades, many OECD countries have transitioned to territorial tax systems, including almost all advanced industrial countries, Japan and the United Kingdom among them. Figure 1 displays the number of OECD countries with each type of system during the period 1981–2013, showing a clear pattern of increase in territorial systems since 1989.

Figure 1.

Trends in international tax systems in OECD countries between 1981 and 2013.

Various economists and policymakers have debated the effects of the USA’s continued use of a worldwide system (Curtis, 2013; Gleckman, 2015). These discussions have focused on how the transition to a territorial tax system influences government revenue in the home country, competitiveness of the home country in the global market, and the flows of FDI.³ Supporters propose that a territorial tax system would alter the volume of outbound FDI and allocation among host countries. The transition would also help collect tax revenue and promote business competitiveness, reduce income shifting, increase receipt of accumulated foreign earnings, and increase domestic investment and jobs creation. Conversely, opponents claim that the transition would lead US firms to move or invest abroad rather than in the USA to gain the highest after-tax rate of return. Thus it would weaken the domestic economy with increased capital outflows and reduction in tax revenues and slow the growth of domestic employment (Matheson et al., 2013).

Empirical studies of transitions in other countries show that switching from a worldwide tax system to a territorial tax affects FDI flows (Clausing, 2009; Hong and Smart, 2010) and the location of MNC headquarters (Voget, 2011) and increases income shifting (Markle, 2016), repatriation to a home country (Hasegawa and Kiyota, 2017), and MNCs’ tax burden (Maffini, 2012). For example, Markle (2016) finds that firms under territorial tax systems are more likely to shift profits than under worldwide tax systems, but this effect disappears when worldwide tax systems defer the taxation of foreign earnings until repatriation. Maffini (2012) shows that corporations with affiliates in low-tax rate host countries have lower tax burdens if headquartered in a territorial country than those headquartered in worldwide countries. Voget’s (2011) findings suggest that MNCs are more likely to relocate their headquarters if they have to pay additional tax due in the home country upon repatriation of foreign profits.

Despite extensive studies on the firms’ behavior under the two different tax systems, little is known about what leads to the adoption of either system. Indeed, political science scholarship has paid little attention to why countries have different tax systems, although a few scholars have paid attention to double tax avoidance (Rixen, 2011; Genschel and Rixen, 2012). These studies primarily focus on how international tax governance has evolved historically and how the institutional trajectory was formed. They argue that the strategic structures and the sequences by which double taxation generates problems explain the development of international tax governance. But they do not address why some countries move to the territorial tax system and others do not. To explain the factors underlying this transition, I first propose a theoretical explanation that may account for the observed phenomena by taking into account the role of political factors in such a setting.

Argument

The literature on corporate tax policy has developed substantially over recent decades. However, no study to date has systematically explored countries’ change from worldwide taxation to territorial taxation. Research examining the determinants of corporate tax policy offers insight, however. I draw on two strands of literature here: tax competition arising from economic globalization; and the political factor, veto players.⁴

A large body of research demonstrates that economic globalization prompts national governments to compete for mobile capital, leading to reductions in corporate tax rates (Devereux et al., 2008; Leibrecht and Hochgatterer, 2012; Swank, 2016). The model of tax competition states that two countries share internationally mobile capital, suggesting that one country’s capital outflow represents capital inflows for the other country. Because capital responds to differences and changes in tax rates, each country tries to attract mobile capital from the other. Thus, the tax competition model predicts that tax competition leads to inefficiently low tax rates and public expenditure levels to attract or retain mobile capital.

As the tax competition model suggests, increased capital mobility arising from economic globalization leads to a race to the bottom because mobile asset holders can move across borders and thus force incumbent governments to compete for capital by reducing tax rates. Empirical research finds that there is a significant negative relationship between tax rates on capital income and a country’s degree of openness (Devereux et al., 2008). As a consequence of tax competition for mobile capital, taxes on mobile capital decrease while the tax burden shifts from mobile capital to relatively immobile factors such as labor or consumption.

The tax competition literature suggests that governments need to provide generous policy measures to firms to support economic growth (Devereux et al., 2008; Leibrecht and Hochgatterer, 2012; Swank, 2016). Many studies show that MNCs are the main driver of economic activities in the world economy because of their large revenue and productivity, their vertically integrated production, and their advantage in knowledge and capital-intensive activities over domestic companies, and that, as such, governments wisely treat them generously. Indeed, governments attempt to attract investment by MNCs by providing various policy tools such as lower corporate tax rates, generous tax incentives, and relaxed regulations, because it is essential to promote economic growth through job creation, higher wages through greater productivity, capital accumulation, export promotion, and greater productivity in domestic firms through new technology transfer or other positive spillovers (Markusen, 2002; Dunning and Lundan, 2008). Here I contend that governments adopt territorial tax systems to lure MNCs, much as they offer lower corporate tax rates and generous tax incentives with the same goal.

MNCs push governments to move to a territorial tax system to reduce their tax payments or to streamline tax procedures across borders. In particular, firms involved in intra-firm trade have an incentive to streamline the tax procedures responsible for multiple countries and to avoid multiple tax liabilities, and they therefore demand their own governments change how they tax profits that MNCs generate in multiple countries. Intra-firm trade through MNC affiliates across countries accounts for the vast majority of international trade across borders in the world economy (Ramondo et al., 2016; Yeaple, 2006). According to Grossman et al. (2006), intra-firm trade occurs when firms maximize the economies of scale in production, meaning that a few large affiliates dominate the production of inputs for the entire corporation and that smaller affiliates serve a foreign market primarily by minimizing transportation costs. Governments often bow to pressure to stimulate economic activity. Indeed, many large US MNCs have called for the adoption of a territorial system (Yang and Khimm, 2012), delaying repatriation and accumulating capital abroad; estimates say foreign affiliates of US companies held about $2.1 trillion in 2015 (Dharmapala, 2018). Governments tend to change their international tax system to respond to MNCs’ calls, as a way of attracting their investments to increase economic growth.

However, domestic responses to economic globalization can vary depending on domestic political institutions across countries. The economic globalization and domestic politics linkage literature shows that domestic politics plays a distinct role in mediating external pressures. For example, numerous political scientists point to the importance of domestic institutions in moderating the effects of electoral competition on taxes (see Basinger and Hallerberg, 2004; Swank, 2016). In line with this reasoning, it is plausible that even if globalization incentivizes governments to move to a territorial tax system, domestic politics may counter these effects. Domestic actors or institutions can constrain a government that might otherwise change its international tax regime.

Veto player theory explains how domestic political institutions influence policy change (Basinger and Hallerberg, 2004; Tsebelis, 1995). Veto players can make it hard to change policy proposals because an agreement among actors is essential to pass any law (Tsebelis, 1995). The number of veto players splits decision-making authority and limits the degree to which a change in the status quo affects policy decisions. Veto players make it hard to pass laws. Drawing upon veto player theory, Hallerberg and Basinger (1998) argue that multiparty legislatures and cabinets, and horizontally and vertically fragmented political authority, create opportunities for opponents of policy change to slow or block reform. The implication of veto player theory is that countries with few veto players will make changes more easily than those with many veto players.

The veto player theory discussed so far can provide insights as to why some countries are more likely to move to the territorial tax system than others. In line with the reasoning drawn from both the tax competition literature and veto player literature, I argue that veto players condition the effect of economic globalization, especially growing trade, on international tax system change. The underlying mechanism is straightforward: Governments confront pressures to change their systems to promote multinational firms’ investments for economic growth in response to economic globalization. Multinational firms tend to prefer a territorial tax system to streamline tax procedures and liabilities incurred in multiple countries, and governments respond to this demand to attract them to their jurisdictions. Yet the number of veto players constrains this effect because it hinders policy change. Because opponents of change can slow down or block the policy change, the effect of economic globalization has little impact on the tax system of countries with many veto players. Those with few are more likely to move to a territorial tax system.

In short, I propose that veto players constrain the effect of economic globalization on international tax system change. Because economic globalization prompts governments to reform in a way that favors multinational firms supporting a territorial tax system, governments may face pressure to move to a territorial tax system. Yet the number of veto players can hinder or facilitate change in policy. Countries with fewer veto players are more likely to adopt a territorial tax system than those with many veto players. This discussion leads to the following hypothesis:

Although growing trade openness tends to push countries to move to a territorial tax system, the effect is much greater in countries with fewer veto players than in those with many veto players.

Research design

Data and variables

To test my hypothesis, I compiled an original dataset of 23 OECD countries’ tax system between 1981 and 2013 based on a report released by PriceWaterhouseCoopers in 2013. The report includes information on when each country adopted a territorial tax system and which countries continue to employ a worldwide tax system. Because my interest is in transition, I eliminated the seven that adopted the territorial system before 1981. The final sample thus consists of 15 advanced industrialized countries.⁵ The unit of analysis is country–year.

The dependent variable is a dummy indicating the transition to a territorial system. I employ two measures capturing the transition to a territorial tax system. First, it refers to an event, coded ‘1’ if the event is observed and ‘0’ if not observed. The change to the territorial tax system is a nonrecurring event that occurs sometime between 1981 and 2013 or not at all. Once the event is observed, I treat subsequent observations on that subject as missing values. The second measure I use, transition, is a dichotomous variable equal to ‘1’ for the territorial tax system and ‘0’ for the worldwide tax system.⁶

The main independent variables are the number of veto players and trade openness. First, I use a common measure, CHECKS, which is available from the Database of Political Institutions (Beck et al., 2001), to capture the number of veto players. This measure is constructed from a count of distinct parties in government coalitions (for parliamentary systems) or in control of the executive and legislative chambers (for presidential systems). CHECKS accounts for institutional as well as partisan veto players (Dahl, 2014). The average number of veto players is higher in worldwide tax systems than in territorial tax systems. The other key explanatory variable is trade openness, which is widely used to capture the degree to which a country is integrated in the world economy in the international and comparative political economic literature. Trade openness is measured as imports plus exports over gross domestic product (GDP). Figure 2 indicates that countries with a territorial tax system are more open to trade than those with a worldwide tax system.

Figure 2.

Distribution of trade openness by international tax system.

I include several control variables that may influence both independent and dependent variables. Following up on the corporate taxation literature, domestic economic conditions such as unemployment, corporate tax rates, and debt are included. Studies show that a higher level of unemployment leads policymakers to provide more tax incentives to promote economic growth (Basinger and Hallerberg, 2004; Swank, 2016). As a way of retaining and attracting mobile capital to promote economic growth, countries may adopt a territorial tax system because firms move across borders more easily when they do not have to pay taxes on revenue generated in other countries. Accordingly, high unemployment rates should be positively associated with the transition to a territorial tax system. Similarly, the corporate taxation literature demonstrates that governments tend to cut corporate tax rates to stimulate economic activities, suggesting that lowering corporate tax rates may coincide with the transition to a territorial tax system. Furthermore, a high ratio of debt may indicate that governments need to collect more taxes for spending and thus have an incentive to keep up with a territorial tax system or a worldwide tax system. Existing studies show that the effect of the adaptation of a territorial tax system on revenue is mixed: some find that the adoption of a territorial tax system loses money (Yang and Khimm, 2012), but others argue there is no evidence of this (Dittmer, 2012). Supporting the claim that adapting a territorial tax system decreases tax revenue, Markle (2016) finds that multinationals will engage in tax-motivated income shifting under either system, but that it is more common under the territorial tax system than under the worldwide tax system. Nonetheless, the effect of corporate tax rates and debt on the transition to territorial tax reform remains an unresolved question. I measure unemployment rates as the percentage of the labor force that is unemployed. Public sector debt is measured as the ratio of debt to GDP. Corporate tax rates come from the OECD tax database, which combines the national and subnational corporate tax rates.

As a control variable, I also include government partisanship that might affect the veto players and change in international tax system. The ideological distance between veto players is important in shaping policy change because it is much easier to change policy when there is a small ideological distance among veto players than when there is a vast distance (Basinger and Hallerberg, 2004). Left-leaning governments are less likely to open the market and shift to the worldwide tax system because they want to protect their market. Thus, I anticipate that left-leaning governments may be less likely to transition than right-leaning governments. I measure government partisanship as the percentage of cabinet posts controlled by left-wing parties in each country. This measure is commonly used in the political science literature (Shin, 2017; Swank, 2016), and data for the study period is available from the Comparative Parties Dataset (Swank, 2016). The index ranges from 0 to 100, where ‘0’ represents full government controlled by right-wing parties and ‘100’ indicates full government controlled by left-wing parties.

Despite these control variables that might affect both independent and dependent variables, there is a question of post-treatment bias, in that there could be a problematic association between treatment and control variables (Gelman and Hill, 2006). If one of the control variables affects the treatment effect, it could generate a misleading treatment effect. For instance, corporate tax rates may influence the degree of trade openness while at the same time affecting the change in international tax systems. Or government partisanship could influence trade openness. Excluding these control variables in the model estimation is one way to address this concern. For this reason, I present the estimates with and without these control variables in the ‘Results’ section. As I will show, the results do not differ. Summary statistics for all variables appear in Table A1 for data covering the dependent variable, event, and in Table A2 for data covering the dependent variable, transition, in the Online Appendix.

Model specification

Because the dependent variable is a duration of time until a country moves to a territorial tax system, I employ a survival analysis to explore when a country shifts to a territorial tax system. For the survival analysis, Cox Proportional Hazard Models are commonly used because there are no strict assumptions about survival distribution. In the survival analysis, the hazard rate describes the probability of any individual experiencing the non-repeating event at time t, given that this individual has not yet experienced it before time t. The hazard rate consists of two components. The first is that the baseline hazard rate functions as the Intercept. The second is that the influence of predictor variables is parameterized in the form of a regression, which is the Cox Proportional Hazards Regression Model.

The Cox model, however, unduly assumes that all individuals in a sample are homogenous. The notion of frailty is introduced to capture population heterogeneity. Frailty captures an unobserved random proportionality factor that modifies the hazard function of an individual. As an extension of the Cox Proportional Hazard Model, I employ frailty models to consider unobserved heterogeneity for survival data (Box-Steffensmeier and Jones, 2004). A frailty model is a random effect model for time-to-event data, and the random effect (the frailty) has a multiplicative effect on the baseline hazard function. The regression model can be further written as follows:

\begin{matrix} T r a n s i t i o n {(y)}_{i t} = α + β_{1} t r a d e o p e n n e s s_{i t} + β_{2} v e t o p l a y e r s_{i t} \\ + β_{3} t r a d e o p e n n e s s \times v e t o p l a y e r s_{i t} + β_{i} c o n t r o l s_{i t} + γ_{i} \end{matrix}

where Transition(y)_it = 1 if the event occurs in country i at time t, and 0 if the event has not occurred; γ_i indicates unobserved heterogeneity across countries.⁷ In addition to the survival analysis, I also use a logit model to examine why some countries move to a territorial tax system and others do not, because the dependent variable is dichotomous, indicating whether or not a country adopts a territorial system (transition). The results reported in Table 2 remain unchanged if I run a logistic regression model.

Results

Table 1 presents the empirical results. All estimates are obtained from the Cox regression model, including a shared frailty for countries to model unobserved country heterogeneity. The coefficients of covariates estimated by the Cox regression model display their effects on the baseline hazard. A positive coefficient indicates that the covariate increases the hazard rate, accelerating a time of transition to a territorial tax system. Conversely, a negative coefficient means that covariates decrease the hazard rate, delaying the move to a territorial tax system. The theoretical predictions are that trade expansion helps move to a territorial tax system and that countries with many veto players are less likely to shift to a territorial tax system. What is more interesting is, domestic institution—the number of veto players—mediates the effect of trade openness on the transition, supporting my hypothesis.

Table 1.

Estimated effects of veto players and trade openness on the transition to a territorial tax system.

	Model 1	Model 2	Model 3	Model 4
Trade openness	0.0018	0.0928^*	0.0138	0.1279^*
Trade openness	(0.0108)	(0.0419)	(0.0161)	(0.0607)
Veto players	−0.2317	0.7226	−0.3732	0.7419
Veto players	(0.4082)	(0.3699)	(0.5222)	(0.4829)
Trade openness × veto players		−0.0193^*		−0.0237^*
Trade openness × veto players		(0.0088)		(0.0120)
Corporate tax rate			9.7864	10.7708^*
Corporate tax rate			(5.2461)	(5.3973)
Unemployment			0.1078	0.1328
Unemployment			(0.1045)	(0.1087)
Debt			−0.0147	−0.0106
Debt			(0.0136)	(0.0142)
Left-leaning government			−0.0011	−0.0049
Left-leaning government			(0.0089)	(0.0087)
Akaike information criterion	93.6149	92.5849	78.3520	76.8823
R ²	0.0492	0.0602	0.0979	0.1104
Maximum R²	0.2847	0.2847	0.2906	0.2906
Number of events	12	12	11	11
Number of observations	298	298	263	263
PH test	0.3656	0.6361	0.1994	0.3944

Note: Dependent variable is whether the transition to a territorial system occurred. Estimates are obtained from the Cox Proportional Hazard Models with fragility. SEs are in parentheses.

p < 0.05.

PH test = Proportional Hazard test; SEs = standard errors.

Models 1 and 2 present the results without control variables and Models 3 and 4 report estimates with control variables. As expected, trade openness has a positive effect on the change of the tax system, which is consistent with the theoretical expectation, suggesting that trade openness accelerates the transition. When it comes to the effect of veto players on the transition to a territorial tax system, my theory suggests that countries with many veto players will be less likely to move to a territorial tax system. The signs are consistent with the theoretical expectation, suggesting that countries with many veto players are less likely to move to a territorial tax system.

The main interest of this article, however, is an interaction term that includes trade openness and the number of veto players. As seen in Model 2 and Model 4, the coefficient on the interaction term is negative and statistically significant, suggesting that the effect of trade openness on international tax system change varies across countries. As countries increase their trade flows, the effect of veto players decreases. The result is that countries with a greater number of veto players are less likely to move to a territorial tax system, providing support for the main hypothesis.

To provide substantive effects of veto players, I calculate the marginal effect of veto players on international tax system change conditional on trade openness based on Model 4 in Table 1 and display it in Figure 3. As seen in Figure 3, veto players have a negative effect on international tax system change, meaning that countries with many veto players delay the transition to a territorial tax system. Such an effect is also more significant for countries that have high degrees of trade openness.

Figure 3.

Marginal effect of veto players on transition.

Figure 4, however, presents the marginal effect of trade openness on international tax system change conditional on veto players. As it shows, trade openness has little impact on tax system change when countries have fewer veto players. As veto players increase, the effect of trade openness on the tax system decreases, implying that countries with a greater number of veto players, coupled with a high degree of trade openness, are more likely to have a worldwide tax system.

Figure 4.

Marginal effect of trade openness on transition.

The main finding shows that trade openness has a significant impact on the transition to a territorial tax system. As explained in the Research design section, the transition relates to the growth of intra-firm trade driven by MNCs in OECD countries. The intra-firm trade is one type of intra-industry trade: Goods and services are traded within multinationals across borders. This intra-firm trade accounts for a significant portion of trade in the OECD countries, which is mainly driven by MNCs that transfer goods through their production networks across borders. For example, most trade in manufactured goods among OECD countries is subject to the intra-trade type. In particular, in Europe and North America, intra-firm trade accounts for approximately 60–70 percent of total trade in manufacturing (Bonturi and Fukasaku, 1993). Given the growing importance of intra-firm trade by MNCs, countries will face pressure to change the taxation of foreign profits of MNCs from both multinational firms and domestic citizens to attract investment and promote economic growth. Thus, trade openness may have a significant impact on the transition to the territorial tax system.

In addition to this statistical analysis, I provide illustrative cases based in the UK and the USA examining what drives countries’ decision to adopt a territorial tax system. Both countries confront firms’ demand that they adopt a territorial tax system but the US tax system continues to be worldwide whereas the UK changed to a territorial tax system in 2009, citing concerns about firms’ competitiveness and the desire to improve their tax system’s competitiveness (Tax Foundation, 2012). As this study shows, the USA has a slightly higher number of veto players than the UK. As well, there were UK firms that threatened to leave the country, and the 2009 tax reform was expected to retain some of them (Houlder, 2009; Tax Foundation, 2012). By contrast, the USA has a worldwide tax system with a mixed feature, although corporate tax reforms the Trump administration has considered would include international taxation (see Dharmapala, 2018). US MNCs have called for the adoption of a territorial tax system, and have delayed repatriations and accumulated a significant amount of foreign earnings overseas in anticipation of such a change. Although US policymakers and business leaders have made a series of proposals to move to a territorial tax system for a long time,⁸ the Trump administration has started to reform international taxation.

Beyond the effects of trade openness and veto players on international tax system change, the effects of control variables on international tax system change are worth discussing. First, unemployment, as expected, is positive across all models. This means that higher unemployment rates encourage a transition to a territorial tax system. The reason might be that countries need to create jobs and thus are more willing to attract investment by MNCs in their home country. Although theories of the effect of territorial tax reform on revenue are mixed, the finding shows that corporate tax rate positively correlates with the tendency to transition to a territorial tax system. Debt has a negative sign, suggesting that it decreases the likelihood of transition to the territorial tax system. As a political control variable, the variable partisanship contributes to a shift to the territorial tax system. The results mean that right-wing governments are more likely to move to the territorial tax system than left-wing governments. Except for the variable corporate tax rate, most of variables are not statistically significant, and thus have little impact on the transition to the territorial tax system.

In sum, the main finding suggests that, in general, countries with a more open economy are more likely to adopt a territorial tax system than a worldwide tax system. This effect, however, is not homogeneous across countries. Greater domestic constraints decrease the effect. This finding implies that more open economy makes it possible for firms to move across borders easily and make the international tax system in their favorable direction. Yet domestic institutions constrain the power of firms, because of veto players that hinder change in the policy-making process.

To assess whether such an effect is robust to alternative model specifications and measures, first, I replicate all models reported in Table 1 using alternative measures of economic globalization: an aggregate indicator capturing economic globalization developed by Dreher et al. (2009); the total flows of inward and outward FDI as the ratio to GDP; and the outbound FDI flow as the ratio to GDP as suggested by Kerner (2009). Table A3 in the Online Appendix presents the results. As it shows, the coefficient for the interactive effects, the interactive effects between trade openness (economic globalization) and the number of veto players, remains unchanged (negative), but statistical significance disappears. Second, I run models through logistic regression and multilevel logit regression with a random-effect by country and report their results in Table 2. Furthermore, I also run models through multilevel logit regression with a random-effect by country and time and the fixed-effects model and report results in Table A4 in the Online Appendix. The results obtained from the logistic regression models are mainly similar to those obtained from the Cox Proportional Hazard Model reported in Table 1. The effect of trade openness on the transition is conditional on the domestic constraint, the number of veto players. It demonstrates that countries with a more open economy are likely to adopt a territorial tax system but that the effect is much greater in countries with weaker domestic constraints. However, other results estimated via the multi-level logit models accounting for country and year effects and the fixed-effects models appear changed.

Table 2.

Estimated effects of veto players and trade openness on the transition to a territorial tax system.

	Model 1	Model 2	Model 3	Model 4
(Intercept)	2.0662^**	−2.6114	−7.6842	3.8476
(Intercept)	(0.7573)	(1.4399)	(6.1819)	(8.8918)
Trade openness	0.0086^*	0.0805^***	0.1656^**	−0.0190
Trade openness	(0.0041)	(0.0194)	(0.0503)	(0.1104)
Veto players	−0.1163	0.8492^**	0.1549	−2.5945
Veto players	(0.0982)	(0.2760)	(0.3632)	(1.6441)
Corporate tax rate	−7.2047^***	−7.5584^***	−34.3024^***	−37.1239^***
Corporate tax rate	(1.3375)	(1.3672)	(8.9306)	(9.4530)
Unemployment rate	−0.0633^*	−0.0636^*	−0.4497^***	−0.4429^**
Unemployment rate	(0.0294)	(0.0308)	(0.1329)	(0.1356)
Debt	0.0078^*	0.0106^**	0.1373^***	0.1461^***
Debt	(0.0037)	(0.0039)	(0.0294)	(0.0312)
Left-leaning government	0.0006	−0.0003	−0.0002	0.0025
Left-leaning government	(0.0028)	(0.0029)	(0.0072)	(0.0075)
Trade openness × veto players		−0.0142^***		0.0421
Trade openness × veto players		(0.0038)		(0.0244)
Akaike information criterion	497.9367	481.1570	199.1825	198.3501
Bayes information criterion	526.1177	513.3639	231.3894	234.5829
Log likelihood	−241.9683	−232.5785	−91.5912	−90.1751
Deviance	483.9367	465.1570
Number of observations	414	414	414	414
Number of groups: country			15	15
Variables: country (Intercept)			122.8685	154.0271

Note: Statistical models: Dependent variable is the transition indicating a dichotomous variable as to whether a country adopts a territorial tax system or not. Models 1 and 2 estimated via logistic regression estimator and Models 3 and 4 estimated via multi-level regression estimator accounting for country effects. SEs are in parentheses.

***

p < 0.001, ^**p < 0.01, ^*p < 0.05.

SEs = standard errors.

Conclusion and implications

This article provides an explanation of why some countries move to the territorial tax system and others do not. Because trade expansion prompts governments to provide favorable conditions to multinational firms, countries are more likely to move to a territorial tax system. Yet this effect can vary depending on the number of veto players that can block policy change. I have argued that countries with fewer veto players are more likely to move to a territorial tax system than those with many veto players in response to economic globalization and trade openness.

The empirical findings show that trade openness, an aspect of economic globalization, accelerates the time to change to a territorial tax system. Moreover, countries with fewer veto players are more likely to switch to a territorial tax system, which is consistent with the expectation. When it comes to the interactive effect between veto players and trade openness, findings show that countries with fewer veto players and high degrees of trade openness are more likely to transition to the territorial tax system.

This article makes several contributions to the current literature: First, scholars of the international political economy who explore tax competition have focused mostly on its effects on domestic taxation (Swank, 2016), tax harmonization to manage harmful tax competition, the development of tax havens (Palan et al., 2013), and double tax avoidance (Rixen, 2011). Although taxation of MNCs have an important impact on government tax revenues and economic activities, international political economy research has not given the rules for taxing foreign-source income much attention. By taking into account the conditions of international tax system change, this paper can help broaden our understanding of the taxing of multinational firms.

Second, this article focuses on the causes of a change in international tax systems. The existing literature has generally ignored the role of politics in the change to a territorial tax system. This article argues that the political causes play a significant role in setting up the way countries tax foreign source earnings of MNCs. The findings shed light on the important role of domestic politics in policy changes, which resonates with a broad range of political science literature. While economic globalization strongly constrains the autonomy of national governments, domestic institutions still play a central role in countering such economic forces. By incorporating political factors into dominant economic explanations, I broaden our understanding of when and why countries shift from a worldwide tax system to a territorial tax system.

Third and finally, the current article has implications for policy. The EU has scrutinized tax inversions of multinationals, criticizing them for double tax avoidance (Brundsden, 2016b). Differences between territorial and worldwide tax systems largely drive this concern among MNCs. Although the USA has bilateral tax treaties with most EU members to avoid double taxation, the differences in their tax systems inevitably raise conflicts. A unified global tax system to tax MNC profits is essential to improve transparency and efficiency to the benefit of all of these nations.

Supplemental Material

IPS824677_French_Spanish – Supplemental material for Why do countries change the taxation of foreign-source income of multinational firms?

Supplemental material, IPS824677_French_Spanish for Why do countries change the taxation of foreign-source income of multinational firms? by Mi Jeong Shin in International Political Science Review

Supplemental Material

IPS824677_supplementary_material – Supplemental material for Why do countries change the taxation of foreign-source income of multinational firms?

Supplemental material, IPS824677_supplementary_material for Why do countries change the taxation of foreign-source income of multinational firms? by Mi Jeong Shin in International Political Science Review

Footnotes

Acknowledgements

I would like to thank Ambreen Chaudhri, Nathan Jensen, Chia-yi Lee, Adrian Lucardi, participants and discussants at the 2016 annual meeting of International Studies Association in Atlanta, the International Political Science Review Editors, and anonymous reviewers for their helpful suggestions and comments. Any remaining errors are my own. Online supplementary materials are also available on the author’s website, .

Funding

The author received no financial support for the research, authorship, and/or publication of this article.

Notes

Supplemental material

Appendix tables and figures can be found at .

ORCID iD

Mi Jeong Shin

Author biography

Mi Jeong Shin is an assistant professor in the School of Public Economics and Administration at Shanghai University of Finance and Economics. Her research focuses on the politics of international taxation and political economy of international trade and investment and has been published in Comparative Political Studies and International Interactions.

References

Basinger

Scott J

Hallerberg

Mark

(2004) Remodeling the Competition for Capital: How domestic politics erases the race to the bottom. American Political Science Review 98(2): 261–276.

Beck

Thorsten

Clarke

George

Groff

Alberto

Keefer

Philip

Walsh

Patrick

(2001) New Tools in Comparative Political Economy: The database of political institutions. World Bank Economic Review 15(1): 165–176.

Bonturi

Marcos

Fukasaku

Kiichiro

(1993) Globalization and Intra-Firm Trade: An empirical note. OECD Economic Studies 20 (Spring): 145–145.

Box-Steffensmeier

Janet M

Jones

Bradford S

(2004) Event History Modeling: A Guide for Social Scientists. New York: Cambridge University Press.

Brundsden

Jim

(2016a) Multinational Companies Face Tax Avoidance EU Crackdown. Financial Times, 8 February 2016.

Brundsden

Jim

(2016b) Europe Sets Out Plans to Curb Corporate Tax Avoidance. Financial Times, 28 January 2016.

Clausing

Kimberly A

(2009) Multinational Firm Tax Avoidance and Tax Policy. National Tax Journal 62(4): 703–725.

Clausing

Kimberly A

(2012) The Effect of Profit Shifting on the Corporate Tax Base in the United States and Beyond. National Tax Journal 69(4): 905–934.

Curtis

Dubay S

(2013) A Territorial Tax System Would Create Jobs and Raise Wages for U.S. Workers. The Heritage Foundation. 12 September. Available at: http://www.heritage.org/research/reports/2013/09/a-territorial-tax-system-would-create-jobs-and-raise-wages-for-us-workers.

10.

Dahl

Casper Hunnerup

(2014) Parties and Institutions: Empirical evidence on veto players and the growth of government. Public Choice 159(3–4): 415–433.

11.

Devereux

P Michael

Lockwood

Ben

Redoano

Michela

(2008) Do Countries Compete Over Corporate Tax Rate? Journal of Public Economics 92(5–6): 1210–1235.

12.

Dharmapala

Dhammika

(2018) The Consequences of the TCJA’s International Provisions: Lessons from Existing Research. University of Chicago Coase-Sandor Institute for Law & Economics Research Paper No. 856; University of Chicago, Public Law Working Paper No. 681. Available at: SSRN: https://ssrn.com/abstract=3212072 or https://dx-doi-org.web.bisu.edu.cn/10.2139/ssrn.3212072

13.

Dittmer

Philip

(2012) A Global Perspective on Territorial Taxation. Tax Foundation. Special Report No. 202. Available at: https://taxfoundation.org/global-perspective-territorial-taxation/

14.

Doupnik

Timothy S

Perera

Hector

(2012) International Accounting. Boston: McGraw-Hill.

15.

Dreher

Axel

Gaston

Noel

Martens

Pim

Van Boxem

Lotte

(2009) Measuring Globalization—Opening the Black Box: A Critical Analysis of Globalization Indices. Available at: http://www.sociostudies.org/books/files/globalistics_and_globalization_studies/129-147.pdf

16.

Dunning

John H

Lundan

Sarianna M

(2008) Multinational Enterprises and the Global Economy. Cheltenham, UK: Edward Elgar Publishing.

17.

Fox

John

Weisberg

Sanford

(2011) Cox Proportional-Hazards Regression for Survival Data in R. An Appendix to an R Companion to Applied Regression (2nd edition). Available at: https://socialsciences.mcmaster.ca/jfox/Books/Companion/appendices/Appendix-Cox-Regression.pdf

18.

Gelman

Andrew

Hill

Jennifer

(2006) Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge: Cambridge University Press.

19.

Genschel

Philipp

Rixen

Thomas

(2012) The International Tax Regime: Historical Evolution and Political Change. Technical Report. Social Science Research Network. Available at: http://www.academia.edu/16349692/The_International_Tax_Regime_Historical_Evolution_and_Political_Change

20.

Gleckman

Howard

(2015) A Look at the Territorial Tax Systems in Four Countries Finds No Magic Bullets, Forbes, 22 January 2015. Available at: http://www.forbes.com/sites/beltway/2015/01/22/a-look-at-the-territorial-tax-systems-in-four-countries-finds-no-magic-bullets/#5a8801687de2.

21.

Grossman

Gene M

Helpman

Elhanan

Szeidl

Adam

(2006) Optimal Integration Strategies for the Multinational Firm. Journal of International Economics 70(1): 216–238.

22.

Hallerberg

Mark

Basinger

Scott

(1998) Internationalization and Changes in Tax Policy in OECD Countries: The importance of domestic veto players. Comparative Political Studies 31(3): 321–352.

23.

Hasegawa

Makoto

Kiyota

Kozo

(2017) The Effect of Moving to a Territorial Tax System on Profit Repatriation: Evidence from Japan. Journal of Public Economics 153: 92–110.

24.

Hong

Qing

Smart

Michael

(2010) In Praise of Tax Havens: International tax planning and foreign direct investment. European Economic Review 54(1): 82–95.

25.

Houlder

Vanessa

(2009) Taxation Risks Business Exodus. Financial Times, 7 December.

26.

Kerner

Andrew

(2009) Why Should I Believe You? The Costs and Consequences of Bilateral Investment Treaties. International Studies Quarterly 53(1): 73–102.

27.

Leibrecht

Markus

Hochgatterer

Claudia

(2012) Tax Competition as a Cause of Falling Corporate Income Tax Rates: A survey of empirical literature. Journal of Economic Surveys 26(4): 616–648.

28.

Maffini

Giorgia

(2012) Territoriality, Worldwide Principle, and Competitiveness of Multinationals: A Firm-Level Analysis of Tax Burdens. Working Paper (WP 12/10). Available at: http://eureka.sbs.ox.ac.uk/3203/

29.

Markle

Kevin S

(2016) A Comparison of the Tax-Motivated Income Shifting of Multinationals in Territorial and Worldwide Countries. Contemporary Accounting Research 33(1): 7–43.

30.

Markusen

James R

(2002) Multinational Firms and the Theory of International Trade. Cambridge, MA: MIT Press.

31.

Matheson

Thornton

Victoria

Perry

Veung

Chandara

(2013) Territorial vs. Worldwide Corporate Taxation: Implications for Developing Countries. International Monetary Fund. Available at: https://www.imf.org/external/pubs/cat/longres_gsause.aspx?sk=40981&gsa=true

32.

Oliver

Christian

Brundsden

Jim

(2016) US Blasts Brussels Over Tax Probe Bias. Financial Times, 29 January.

33.

Palan

Ronen

Murphy

Richard

Chavagneux

Christian

(2013) Tax Havens: How Globalization Really Works. Ithaca, NY: Cornell University Press.

34.

PriceWaterHouseCoopers (2010) A Comparison of Key Aspects of the International Tax Systems of Major OECD and Developing Countries. Available at: https://s3.amazonaws.com/brt.org/archive/BRT_14_country_international_tax_comparison_20100510.pdf

35.

PriceWaterHouseCoopers (2013) Evolution of Territorial Tax Systems in the OECD. Available at: http://www.techceocouncil.org/clientuploads/reports/Report%20on%20Territorial%20Tax%20Systems_20130402b.pdf

36.

Ramondo

Natalia

Rappoport

Veronica

Ruhl

Kim J

(2016) Intrafirm Trade and Vertical Fragmentation in U.S. Multinational Corporations. Journal of International Economics 98 (C): 51–59.

37.

Rixen

Thomas

(2011) From Double Tax Avoidance to Tax Competition: Explaining the institutional trajectory of international tax governance. Review of International Political Economy 18(2): 197–227.

38.

Shin

Mi Jeong

(2017) Partisanship, Tax Policy, and Corporate Profit-Shifting in a Globalized World Economy. Comparative Political Studies 50(14): 1998–2026.

39.

Swank

Duane

(2016) Taxing Choices: International competition, domestic institutions and the transformation of corporate tax policy. Journal of European Public Policy 23(4): 571–603.

40.

Tax Foundation (2012) The United Kingdom’s Move to Territorial Taxation. Available at: https://taxfoundation.org/united-kingdoms-move-territorial-taxation/

41.

Tsebelis

George

(1995) Decision Making in Political Systems: Veto players in presidentialism, parliamentarism, multicameralism and multipartyism. British Journal of Political Science 25(3): 289–325.

42.

Voget

Johannes

(2011) Relocation of Headquarters and International Taxation. Journal of Public Economics 95(10): 1067–1081.

43.

Yang

Jia Lynn

Khimm

Suzy

(2012) Companies Quietly Push for Tax Break on Foreign Profits in ‘Fiscal Cliff’ Debate. Washington Post, 29 November.

44.

Yeaple

Stephen Ross

(2006) Offshoring, Foreign Direct Investment, and the Structure of U.S. Trade. Journal of European Economic Association 4(2–3): 602–611.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.33 MB

0.39 MB