|
Article Excerpt
I. INTRODUCTION
The reason why Europeans work less than Americans is the subject of intense discussion in recent literature. Prescott (2004), Cardia, Kozhaya, and Ruge-Murcia (2003), and Roeger and De Fiore (1999) emphasize the negative effects of high taxes in Europe. Other authors, like Blanchard (2004) and Alesina, Glaeser, and Sacerdote (2005), point to differences in the desire for leisure. Cultural differences or differences in unionization and labor market regulations may be at the basis of a higher European taste for leisure. Freeman and Schettkat (2005) explain higher employment in the United States as the result of a much more extensive shift of traditional household production to the market. Lower taxes on wages and higher degrees of labor market flexibility in the United States may be among the factors that have contributed to this shift.
There is no doubt that these hypotheses contain relevant elements to account for the employment differences between Europe and the United States. However, they are also incomplete. First of all, each of these hypotheses will have difficulty explaining employment in the Nordic countries. Countries like Sweden, Norway, and Denmark also have high taxes, which increased during the last decades. According to Nicoletti and Scarpetta (2005), they also combine many of the European labor and product market rigidities. Yet, as we show in Table 1 for 19952004, the average employment rate in persons in the Nordic countries is about the same as in the United States. The employment rate in hours is lower than in the United States, but it is still much higher than in the core countries of the euro area or in "convergence" countries like Spain and Ireland. Second, these explanations ignore the other side of the performance gap between Europe and the United States. They neglect per capita growth differences. A coherent theory should also be able to explain why in the last decade most of the core countries of the euro area have had lower potential per capita growth than the United States, while again several of the Nordic countries have performed at least as well as the United States or even better (Table 1). (1)
Our contribution in this paper is double. First, we explain not only employment differences between the United States and Europe, as most of the above-mentioned literature does, but also employment differences within Europe. (2) Second, our explanation also accounts for per capita growth differences. A critical explanatory variable in our analysis is the composition of fiscal policy. Taxes are important, but not only taxes. The allocation of government spending to productive or nonproductive purposes is at least as important. By contrast, our results would suggest that any differences in taste for leisure or labor market rigidity are not critical.
The structure of the paper is as follows. In Section II, we develop a theoretical model that explains both employment and long-run growth within a coherent framework. We basically build on seminal work by Barro (1990). We show the impact of income taxes, transfers in the case of nonemployment and productive government expenditures. In Section Ill, we calibrate the model on actual data and compare its predictions with the facts reported in Table 1. It should be clear from the outset that we take fiscal policy differences across countries as given. We investigate their effects on employment and growth, but we do not try to explain why policy differs. This approach is not unusual in related literature like Roeger and De Fiore (1999), Cardia, Kozhaya, and Ruge-Murcia (2003), and Prescott (2004). Section IV concludes the paper. Anticipating, we find that our model explains well. Employment in the United States is higher than in the core euro area thanks to lower income tax rates and much lower transfers in the case of nonemployment. The Nordic countries have higher employment than the core euro area mainly thanks to a higher share of productive government expenditures and lower nonemployment transfers. The combination of high taxes, high nonemployment transfers, and relatively low productive government expenditures also clarifies why many core euro area countries have the lowest per capita growth rates. High productive expenditures and slightly lower taxes, financed by lower transfers, explain the relatively good growth performance of the Nordic countries. Growth in the United States is less outstanding mainly due to relatively low productive government expenditures.
II. THE MODEL
We investigate the relationship between fiscal policy, employment, and growth in a simple endogenous growth framework. We extend Barro (1990) by endogenizing the decision to work and by allowing two kinds of government expenditures: productive government expenditures and transfers related to nonemployment, where it should be recognized that nonemployment is a structural or equilibrium phenomenon in our model. Given ample empirical evidence that transfers related to structural nonemployment are a major determinant of (un)employment and growth, we consider their introduction to be a major strength of our model. (3)
Set-up of the Model
Consider a closed economy consisting of N identical individuals with infinite lives and perfect foresight and N identical perfectly competitive firms. Population is constant. Assuming an equal number of individuals and firms, each firm's output equals per capita output.
Firms. Firms produce output by using physical capital and effective labor. The production function exhibits constant returns to scale, which allows us to focus on a representative firm.
(1) [y.sub.t] = [K.sup.1.-[beta].sub.t][([h.sub.t][l.sub.t].sup.[beta],
where [y.sub.t], [k.sub.t],...
|
