رقابت برای سرمایه هنگامی که نیروی کار ناهمگون است

This paper investigates the impacts of capital mobility and tax competition in a setting with imperfect matching between firms and workers. The small country attracts less firms than the large one but accommodates a share of the industry that exceeds its capital share—a reverse home market effect. This allows the small country to be more aggressive and to set a higher tax rate than the large one, thus implying that tax competition reduces international inequalities. However, the large country always attains a higher utility than does the small country. Our model thus encapsulates both the “importance of being small” and the “importance of being large”. Last, tax harmonization benefits to the small country but is detrimental to the large one.

During the last two decades, OECD countries have experienced very high increases in foreign direct investments (OECD, 2003). As economic integration gets deeper, these investments are likely to become more responsive to differentials in corporate tax rates. The empirical evidence collected by Mooij and Ederveen (2003) confirms the idea that governments vastly use such an instrument to influence firms’ locational choices. Building on that observation, the literature on fiscal competition studies how governments choose their tax rates in order to attract firms (Wilson, 1999). Because the outcome of fiscal competition crucially hinges on the international mobility of capital, it seems promising and reasonable to build on the microeconomic underpinnings of firms’ locational choices. New economic geography (NEG) aims precisely at explaining how firms do interact to form clusters within a few regions (Fujita et al., 1999 and Baldwin et al., 2003). It is, therefore, natural to tackle the process of fiscal competition by using the main ingredients of NEG, namely increasing returns, market size, and imperfect competition. This is the road taken recently in various papers (Baldwin and Krugman, 2004, Ottaviano and van Ypersele, 2005 and Borck and Pflüger, 2006). However, very much like in NEG, they all have chosen to focus on the product market. Yet, recent empirical contributions suggest that labor-market pooling is one of the main explanations for the existence of firms’ clusters (Dumais et al., 2002 and Rosenthal and Strange, 2004). The specificity of human capital being itself the main reason for imperfect matching between firms and workers, we find it natural to study how skill mismatch affects the spatial distribution of firms through both firms’ locational choices and the working of local labor markets. When the labor force is heterogeneous in the skill space, firms are able to set wages below the marginal productivity of labor by differentiating technologies. Hence, they operate on imperfectly competitive labor markets. As firms also exhibit scale economies at the plant level, it appears that our setting blends the main ingredients of NEG. However, we obtain results that are very different from existing ones. First, we show that the large country's residents enjoy a higher utility level than do those of the small country. Yet, though the large country has more firms than does the small one, competition on local labor markets hinders the large country to have a more than proportionate share of firms. We thus get a reverse home market effect: the share of firms in the larger region is less than the share of consumers in that region. Put together, the last two results mean that our model encapsulates both the “importance of being small” and the “importance of being large”, two aspects that have been emphasized in distinct papers. Second, the few existing studies focussing on tax competition between asymmetric countries predict that the large country sets a higher corporate tax rate than does the small one (Bucovetsky, 1991, Wilson, 1991, Haufler and Wooton, 1999 and Ottaviano and van Ypersele, 2005). This prediction does not necessarily fit well the real world, however. In their analysis of the effective corporate tax rates set in several OECD countries, Devereux et al. (2002) report much more mixed results. For example, their Fig. 7 reveals that, if the effective average tax rates in Germany, Japan, and the United States are higher than those set in Austria, Finland and Sweden, those prevailing in Belgium and Greece are higher than those in France and the United Kingdom. There is, therefore, a need for a different approach. This is what we accomplish in this paper where we show that the small country levies higher corporate tax rate than does the large country, the reason being that the small country enjoys a reverse home market effect that allows it to follow a more aggressive tax policy. Hence, by focussing on the microeconomic underpinnings of local labor markets, we are able to establish that the main implication of the above-mentioned studies is reversed. Such a prediction seems to fit well what we observe in the European Union. To see it, consider the six founding EU-members (Belgium, France, Germany, Italy, Luxembourg, and the Netherlands), the economies of which are fairly well integrated. Fig. 1 and Fig. 2 show the relationship between the share of corporate taxes in total tax receipts and, respectively, these countries’ GDP and population size. Clearly, both relationships are downward sloping and correlations between variables are very high.Third, our analysis has three major redistributional implications. The first one is that tax competition leads to redistribution from the large to the small country. This is due to the fact that, because of the reverse home market effect, the small country is able to tax both domestic and foreign capital. However, the large country always reaches a higher utility level than does the small country. This sharply differs from the result obtained in the existing studies where the small country typically attains higher utility under tax competition (Bucovetsky, 1991 and Wilson, 1991). The next implication is that tax competition reduces international inequalities in the sense that the large country's residents prefer the no-tax outcome, whereas those of the small country are better off under tax competition. The final implication is that tax harmonization leads to redistribution from the large to the small country. In other words, countries are bound to disagree on the need to allow for tax competition as well as on the choice of a common tax rate. Our last contribution is methodological in nature. Even in the case of simple games such as those by Wilson (1986) and Zodrow and Mieszkowski (1986), showing the existence of a (pure-strategy) Nash equilibrium in tax games is known to be a very problematic issue.2 Here, we propose a new approach to show that our tax game has a unique Nash equilibrium. It is worth stressing that this is done without imposing ad hoc restrictions. This in turn allows us to compare the fiscal competition outcome to both the autarky and no-tax cases on solid grounds, unlike many existing contributions. The remaining of the paper is organized as follows. The model is presented in Section 2. In Section 3, we study the international distribution of capital in the no-tax case, whereas the process of fiscal competition is discussed in Section 4. Section 5 concludes by discussing extensions and policy implications.

We have obtained new results in tax competition in an otherwise standard model, which goes back at least to Salop (1979). Whereas the large country has more firms per capita than does the small country both under tax competition and tax cooperation in Ottaviano and van Ypersele (2005), we have seen that the large country always exports capital toward the small one. This is because we have a RHME, whereas the home market effect holds in Ottaviano and van Ypersele. This shows that the way firms choose their location is crucial in assessing the merits of tax competition. Stated differently, uncovering the various mechanisms that drive the mobility of firms across countries is needed to understand the possible implications of fiscal competition. An interesting implication of our framework is to shed light on the fact that fiscal competition might well trigger unemployment in a country. To see it, consider an economic environment in which some workers might not take a job, the setting being otherwise similar to the one described above. Specifically, we assume that workers get the same level of unemployment benefit b>0b>0 when unemployed. This implies that a worker supplies labor provided that her wage net of training costs is greater than or equal to b. Thisse and Zenou (2000) then show that the labor-market equilibrium involves unemployment in country i when 1<b+β/ni1<b+β/ni holds, namely when the number of firms located in this country is sufficiently small. In this case, the most distant workers on the skill circle refrain from working, thus implying that each firm acts as a monopsony in the labor market. Thus, capital mobility could foster unemployment in the large country that now has a smaller number of firms, thus qualifying Proposition 3. On the other hand, as fiscal competition leads to a reduction in the number of firms installed in the small country when compared to the no-tax case, it is now the small country that could experience unemployment at the taxation outcome, thus qualifying Proposition 7. At least three possible extensions are worth mentioning. First, countries could use the tax proceeds to subsidize workers’ training. In such a context, training costs would become lower and wages higher. However, lower training costs would make the corresponding country less attractive by reducing firms’ market power on the labor market. The following question thus suggests itself: To which extent does one country subsidy its labor force more than the other, and get a better trained labor force, according to its size? Second, introducing capital accumulation with the aim of studying the relationship between economic growth and skill mismatch appears to be a fairly natural topic to investigate. Last, some empirical evidence suggests that several countries tax discriminate between local and foreign firms instead of applying the same tax rate as in this paper (Huizinga and Nicodème, 2006). It would be interesting to revisit our model when national governments may use such additional instruments. These topics are left for future investigation.