تعرفه های تفاضلی، رشد و رفاه در یک اقتصاد کوچک باز

This paper analyzes the effects of consumption and investment tarrifs on growth and welfare. With endogenous laber supply, consumption tariffs are not growth-neutral. Instead, an increase in either tariff reduces both the short-run growth rates of key economic variables such as GDP, consumption, and foreign debt, and their common long-run equilibrium growth rate. Numerical simulations suggest that the investment tariff has a more adverse effect on growth rates and welfare than does a comparable consumption tariff. Accordingly, a revenue-neutral substitution of a consumption tariff for an investment tariff is both growth-enhancing and welfare-improving. The second-best and first-best optimal tariffs are characterized and shown to involve the heavy subsidization of investment.

The impact of tariffs on employment, the rate of capital accumulation, and growth are key issues in macroeconomic policy-making in open economies2. Recent approaches to this issue have been based on Ramsey-type models of capital accumulation. An early paper to develop such an analysis is Sen and Turnovsky (1989), who introduce a consumption tariff in a one-sector production model and find that it leads to a contraction in employment both in the short-run and in the long-run, accompanied by a uniform decumulation of capital.3 Multi-sector analyses of tariffs paint a somewhat different picture. Brock and Turnovsky (1993) adopt a two-sector factor-specific model of trade in which the import-competing sector uses capital while the export sector employs land, both in conjunction with intersectorally mobile labor. Their focus is on the differential effects of a consumption tariff vs. an investment tariff and the implications of this for programs of tariff reform. They find that a uniform increase in both tariffs leads to an unambiguous expansion in the long-run capital stock. This is because the substitution effect of the tariff induces labor to switch to the import-competing sector, thereby increasing the productivity of capital. A tariff on investment alone reduces capital, as one would expect, but paradoxically, the welfare effects of a consumption tariff can be at least as harmful as those of an investment tariff. Consequently, policy recommendations based on this type of model stress the need to lower consumption tariffs at the expense of higher investment tariffs. The fact that the long-run growth rate is exogenous in the Ramsey model makes it inappropriate for analyzing the impact of tariffs on the long-run growth rate. For this purpose, the endogenous growth framework is more suited, and a recent paper by Osang and Pereira (1996) takes up this issue using a two-sector endogenous growth model of physical and human capital accumulation. With labor supplied inelastically, they find that long-run growth rates are unaffected by changes in the tariff on consumption goods, while higher tariffs on investment goods have a negative impact on long-run growth rates.4 These results suggest that governments interested in stimulating long-term growth should implement a tariff reform, which under the condition of revenue-neutrality would reduce tariffs on investment good imports at the expense of higher tariffs on consumption goods. Provided that households care enough about future consumption levels, such a reform should also increase the overall welfare of the economy. This view of an optimal tariff structure is at sharp variance to that implied by the stationary Brock and Turnovsky (1993) model.5 There are several reasons for the striking differences in implications and corresponding policy recommendations between these two types of models, the most important of which is the following. While the first model type ignores the distortionary effects of any tax on the long-run savings/investment decisions by household and firms and thus on the long-run growth rate, the second type ignores the distortionary effects of taxation on the labor–leisure choice, both in the short- and the long-run. Therefore, each type of model tends to downplay the negative impact of a particular tariff; the first, the investment tariff, the second, the consumption tariff.6 Empirical evidence exists to suggest that the long-run growth rate is generally adversely affected by trade restrictions, including tariffs on consumption (Harrison, 1996). Thus, the purpose of the present paper is to develop an endogenous growth model in which both consumption tariffs and investment tariffs impact on the long-run growth rate. We achieve this by considering a one-sector economy, which includes several key features. First, labor is supplied elastically through an endogenously determined labor–leisure choice. Since the tariff on consumption is a pure demand shock, the endogeneity of labor supply is crucial for its impact on the equilibrium growth rate, just as it is for the equilibrium capital stock in the Sen–Turnovsky analysis based on the Ramsey model. The second key feature is that the economy faces restricted access to the world capital market in the form of an upward-sloping supply of debt, according to which the country's cost of borrowing depends upon its debt position, relative to its capital stock, the latter serving as a measure of its debt-servicing ability. The introduction of this constraint suggests that the analysis should be viewed as being particularly applicable to developing countries. To the extent that atomistic agents ignore the constraint in their individual decision-making, it introduces an important externality. It is important to stress that it is the interaction between these two features of the model that introduces the role for tariffs, particularly the consumption tariffs, to influence the growth rate. In the pure benchmark AK growth model of a small open economy with inelastic labor supply, such as developed by Turnovsky (1996), only an investment tariff would have any effect on the equilibrium growth rate. 7 If labor is supplied endogenously but the international capital market is perfectly competitive, the equilibrium growth rate in this model is determined by consumption preferences in conjunction with the fixed foreign interest rate (see Turnovsky, 1999). In that case, neither form of tariff would have any influence on the long-run growth rate; instead, all adjustment is borne along the labor–leisure margin. It is only when the elastic labor supply interacts with the imperfect capital market does both forms of tariff influence the long-run growth rate. Moreover, the capital market imperfection introduces transitional dynamics so that one can contrast the short-run and long-run effects of tariffs as in the earlier literature. The main results of the paper are as follows. First, we show that both a higher investment tariff and a higher consumption tariff have similar qualitative effects on the long-run macroeconomic equilibrium of the economy. They both induce a substitution toward leisure, thereby reducing the productivity of capital and the long-run equilibrium growth rate. Numerical simulations enhance our intuitive understanding of these results and highlight both the similarities and the contrasts between the effects of the two types of tariffs. Thus, while both forms of tariff have comparable modest impacts on the labor–leisure choice, the investment tariff has much more powerful effects on growth, debt, and borrowing costs, and thus, a more significant (negative) impact on welfare. Second, we show that the two forms of tariff are likely to have contrasting effects on short-run employment. A consumption tariff tends to lead to a substitution toward leisure and thus to reduce short-run employment; an investment tariff tends to lead to a substitution from capital to labor in production and thus to increase employment. Third, both forms of tariffs reduce the short-run growth rates of capital and foreign debt below their long-run effects, causing them to rise during the transition, while the effect on the consumption growth rate is more gradual. The growth-reducing impact of the investment tariff on debt accumulation is particularly intense and may cause debt to decline in the short-run. Fourth, since these tariff changes have differential effects on tariff revenues, we analyze the growth and welfare implications of a revenue-neutral tariff reform. Starting from the benchmark, we show that substituting a consumption tariff for an investment tariff both increases the growth rate and improves intertemporal welfare, despite the fact that it is associated with short-run welfare losses. Finally, we characterize the second- and first-best optimal tariff policies. The second-best policy, one that maintains a constant tariff revenue, imposes a substantial tariff on the imported consumption good and subsidizes the imports of investment. The investment subsidy corrects for two positive externalities associated with capital. The first is due to the upward-sloping supply curve of debt and the assumption that a higher aggregate capital stock reduces the costs of foreign debt. The second is introduced through a third feature of the model, namely the positive spillover associated with the production technology, which is the source of ongoing growth. The tariff on consumption is thus necessary to finance the subsidy to investment. We also show that tariff policy alone is incapable of sustaining the first-best optimum of the centrally planned economy. But such an equilibrium can be sustained by taxing debt at the time-varying rate equal to the elasticity of the supply curve of debt, while heavily subsidizing the imported investment good, but leaving imported consumption untaxed. The simulation-based policy recommendations for a welfare-improving tariff reform in this model, namely to decrease investment and raise consumption tariffs, are in sharp contrast to those obtained in the stationary Ramsey model, such as Brock and Turnovsky (1993), and more along the lines found in the two-sector endogenous growth literature, such as Osang and Pereira (1996). While from a purely theoretical standpoint, the model is capable of generating situations in which the consumption tariff has the more adverse impact on the growth rate, for realistic parameter values the distortionary effect of the investment tariff dominates. There are several reasons for this. First, by directly impinging on the rate of capital accumulation, the investment tariff directly retards the growth process. In contrast, the consumption tariff influences growth only indirectly, through the labor–leisure substitution and the upward-sloping supply curve of debt. For plausible parameters, these two channels turn out to be relatively modest in magnitude, although certainly not negligible. Finally, the investment tariff reduces the strong positive production spillover effect associated with capital accumulation in this model, while there is no impact on any kind of externality associated with the consumption tariff.

In this paper, we have studied the growth, employment, and welfare effects of consumption and investment tariffs in a small growing open economy. The key features characterizing the model include: (i) the endogeneity of labor supply, (ii) the presence of imperfections in the world capital market, and (iii) spillover effects in domestic production. The interaction of (i) and (ii) introduces an important role for tariffs. In particular, we show that because of the endogeneity of the labor–leisure decision, the consumption tariff is no longer growth-neutral. Like the investment tariff, a higher consumption tariff induces a substitution toward more leisure, thereby reducing the productivity of capital and the long-run equilibrium growth rate of the economy. Both (ii) and (iii) are associated with spillover effects, which are crucial for the specific characterization of the second- and first-best policies. We have calibrated the model to match a number of indicators for a typical low-income LDC. The numerical results indicate that a given increase in the investment tariff leads to a far more negative decline in the growth rate and therefore, a more adverse effect on welfare than does a comparable increase in the consumption tariff. A doubling of the investment tariff from its benchmark value of 0.16 to 0.32 leads to more than a 4% reduction in welfare, whereas a comparable change in the consumption tariff has less than a 1% effect on welfare. Since these tariff changes have differential effects on tariff revenues, we analyze the growth and welfare implications of a revenue-neutral tariff reform. We show that starting from the benchmark that substituting a consumption tariff for an investment tariff both increases the growth rate and is welfare-improving. Interestingly, the policy recommendations for a welfare-improving tariff reform in this model, namely to decrease investment and raise consumption tariffs, are in sharp contrast to those found in the Ramsey models on tariff reform in small open economies, such as Brock and Turnovsky (1993). Finally, we have provided a simple intuitive characterization of the role and limitations of tariff policy in correcting for the externalities impinging on the economy. We conclude with two observations. First, an interesting aspect of our analysis is that despite the fact that the model is a straightforward extension of existing models, it yields very different implications from the underlying components and offers significant new insights. This is a consequence of the interaction between two of the key elements, the endogeneity of labor supply and the upward-sloping supply curve of debt. Finally, it is worthwhile to note that the policy recommendations for a welfare-improving tariff reform in this model, particularly the ones related to the first-best optimum, seem to be quite relevant for a number of trade policy issues currently debated by policy-makers. Optimal policy in this model has the following structure: (i) there are no tariffs on traded commodities, a result that is one of the long-term goals of the WTO and a number of regional trade agreements (e.g. NAFTA); (ii) there is a subsidy on capital accumulation, a result suited to take advantage of production externalities which, according to a number of empirical studies De Long and Summers, 1991 and Rauch, 1993, are an important element of the production process; (iii) there is a tax on the inflow of foreign capital, an idea that has attracted a lot of attention in the context of the recent economic crisis in Asia and has already been implemented by a number of developing countries such as Chile and Malaysia.