اثرات استخدام و رشد اصلاح مالیاتی

Since the mid-1990s almost all OECD countries have engaged in fundamental reforms of their tax systems. There is a trend towards higher social security contributions and lower tax rates on personal and corporate income. This paper explores whether these tax policy measures are effective means for reducing unemployment and accelerating economic growth. Using a Pissarides type search model with endogenous growth, we analyze how savings and the incentive to create new jobs are affected by revenue-neutral tax swaps between wage income taxes, payroll taxes, capital income taxes and taxes levied on capital costs. In our framework, cutting the capital income tax (reducing the double taxation of dividend income) financed by a higher payroll tax turns out to be superior, such a policy mix fosters both employment and growth. Most other tax reforms imply a trade-off between employment and growth.

In the quest for effective means of reducing unemployment and/or accelerating growth, tax policy reforms have always been a key item on the policy agenda of many OECD countries. Almost all tax reforms of the last two decades can be characterized as rate reducing and base broadening reforms. Concerning the tax structure one clear trend is the growth in social security contributions. In most recent years this trend has come to a hold, but due to population aging the demand for an increase in these payroll taxes will probably grow significantly again. Another clear trend is the reduction in the rates of personal and corporate income taxes. The marginal tax rates for individuals with high wage income were eased between 2000 and 2003 by about 2.2 percentage points in the OECD countries and by 2.9 percentage points in the EU15. Similarly, between 2000 and 2003 the corporate income tax rates dropped on average by 2.8 percentage points in the OECD countries and by 3.4 percentage points in the EU15 (see OECD, 2004). Are these tax policy measures really helpful in alleviating the unemployment and growth problems? Our formal study sheds some light on this issue by analyzing the employment and growth effects of revenue-neutral tax reforms in a Pissarides (2000) type matching model with endogenous growth. In order to cover a broad range of actual reform plans currently discussed in many OECD countries, we integrate four taxes into the model, namely a wage income tax, a capital income tax, a payroll tax, and a tax on capital input. Our analytical framework merges three strands of literature. First is the literature on the employment effects of tax reforms. Most theoretical and empirical research finds a negative, but limited, linkage between employment and the tax wedge, defined as the difference between the gross labor costs that a firm has to pay and the after-tax wage income of workers (see Layard and Nickell, 1999 and Blanchard and Wolfers, 2000). Daveri and Tabellini (2000) emphasize the level at which wage bargaining takes place. Taxes on labor seem to matter less in countries where bargaining is either highly decentralized (as in the United States and the United Kingdom) or highly centralized (as in the Scandinavian countries and Austria). In these countries higher labor taxes are (partly) absorbed by a decline in gross wages limiting the increase in labor costs and thus the negative effect on employment. In the “continental European” countries, however, where the bargain is done at the industry level (France and Germany, for instance), the tax wedge has a large influence on labor costs and employment. Somewhat controversial is the importance of the structure of the tax wedge. According to Nickell (1997), only the overall size matters. On the other hand, the studies of Symons and Robertson (1990), Koskela and Vilmunen (1996) and Böhringer et al. (2005) indicate that, for example, a revenue-neutral shift from payroll taxes to a more progressive wage tax is good for employment. The second strand of research is on the growth effects of tax policies. In endogenous growth models the steady-state growth rate depends on the net rate of return on investment in human and physical capital, which, in turn, depends on various tax variables. The theoretical literature is reasonably clear. Almost all taxes, being it the components of the tax wedge, the capital income tax or the consumption tax, are detrimental to growth, since these taxes distort the investment decisions of agents with respect to physical and/or human capital. This has been shown, among others, by Rebelo (1991), Milesi-Ferretti and Roubini (1998), and Turnovsky (2000). The empirical evidence is very much in line with the theoretical predictions (see Easterly and Rebelo, 1993 and Kneller et al., 1999). The third strand of research we refer to is on the interaction between employment and growth (for surveys see Birk, 2001a and Aricó, 2003). If growth comes through creative destruction (Aghion and Howitt, 1992), the flow of workers into the pool of unemployed and thus the equilibrium unemployment rate is increasing in the growth rate of the economy. On the other hand, a higher equilibrium growth rate induces higher future revenues and thus rising vacancies that lead to more employment (capitalization effect, see Bean and Pissarides, 1993). Bräuninger (2000) investigates the feedback impact of unemployment on the rate of growth of the economy via savings and capital accumulation. All these studies come to the same conclusion: the relationship between unemployment and growth is not very robust and thus difficult to sign. The mixed evidence from empirical work by Davis and Haltiwanger (1992), Caballero (1993), Daveri and Tabellini (2000), and Bräuninger and Pannenberg (2002) confirms this conclusion. The present paper relates most closely to the line of research started by Eriksson (1997), and Daveri and Tabellini (2000). These studies discuss the role of tax policies in models where both unemployment and growth are endogenous variables. Eriksson (1997) focuses on the capital income tax and shows that a higher tax rate works as a disincentive to save and thus reduces the equilibrium growth rate. Due to the capitalization effect, labor market conditions worsen. Daveri and Tabellini (2000) develop an overlapping generations endogenous growth model where wages are set by monopolistic trade unions. In their model higher taxes on labor income are met by a higher bargained wage forcing firms to cut employment. This in turn lowers the income of the young and thus savings and growth. The model we set up in the next section frames a closed economy comprised of infinitely-lived consumers and firms producing a homogeneous good with the help of capital and labor. Growth is made possible by positive externalities of the economy's capital stock, the wage bargain takes place at the firm level. In this framework, the payroll tax turns out to be neutral with respect to the producer real wage, employment and growth. This neutrality result is good news for policymakers given the expected increase in social security contributions due to population aging. To go a step further, the payroll tax should be used to finance a cut in other taxes. We will show that a higher payroll tax combined with (i) a cut in the capital income tax increases both equilibrium employment and the growth rate, combined with (ii) a cut in the capital input tax boosts growth but has an ambiguous effect on employment, and combined with (iii) a cut in the wage tax is neutral for growth and good for employment. Against this background we will draw some conclusions concerning the recommendation or dismission of specific tax reforms discussed in a number of OECD countries. But, of course, we have to emphasize that due to the single nature of our model, our conclusions can at best be tentative. Three components of the model are of crucial importance for our results, namely the modelling of the savings decision, the assumed endogenous growth mechanism, and the level of the wage bargain. A prominent approach to model savings is the overlapping generations framework. In such an economy aggregate savings are a fraction of labor income accrued to the young. The old generation earns capital income and sells their capital, the proceeds are entirely consumed. But this scenario is at odds with facts. Empirical evidence suggests that the old generation indeed accumulates wealth, about 70 % of the elderly are interested in leaving bequests (Kopczuk and Lupton, 2005). And since an OLG-model with an operative bequest motive perfectly mimics a model with infinitely lived individuals, we prefer the Ramsey approach. In this scenario (the change in) labor income looses its central role for the development of aggregate savings and capital accumulation. Concerning the growth mechanism we follow Romer (1986) and introduce endogenous growth by assuming positive learning and knowledge spillover working through the economy's capital stock per worker. An alternative approach would be the implementation of human capital as driving force of the economy, but to keep analytical tractability we employ the simplest mechanism. Concerning the level at which wage bargaining takes place our analysis primarily applies to countries where the bargain is done at the firm or at the national level. In these settings the payroll tax is neutral to labor costs. If, instead, negotiations take place at the industry level, the flexibility of the gross wage is absent, an adaptation of our framework would be necessary. The rest of paper is organized as follows. 2 and 3 present the model and the analysis of the steady state solution, respectively. The tax reform analysis is performed in Section 4, Section 5 concludes.

In this paper we have analyzed the employment and growth effects of different revenue-neutral tax reforms. The main results are stated in both the abstract and the introduction, so there is no need to repeat them here. Our analysis suggests that there is a “first-best tax reform” in terms of higher employment and growth: cut the capital income tax and – in order to fulfill the budget constraint – increase the payroll tax. It is interesting to note that this policy recommendation is a turn-around to the recommendation given by Daveri and Tabellini (2000), who state that a cut in labor taxes financed by higher capital taxes is good for growth and employment. The reason for these conflicting results remains cloudy, since the models differ in the assumed driving force of growth, the modeling of labor market imperfections and the saving decision of households. Further research is needed in order to identify the crucial assumptions and critical parameters, which determine not only the magnitude but even the sign of the employment and growth effects of tax reforms. Lastly, let us mention two limitations of our framework. We do not have a criterion which allows us to analyze meaningfully the welfare implications of alternative policies. In particular, if the employment and growth effects show different signs, an unambiguous ranking of the tax instruments is not possible and thus the policy conclusions are only vague. A related point is concerned with our focus on analytical results. The method of log-linearization restricts us to small changes in the policy parameters. In order to evaluate large policy shocks and/or to get a numerical assessment of the employment and growth effects, a calibration of the model would be necessary.