امنیت اجتماعی با مصرف کنندگان منطقی و اغراق آمیز
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|24316||2008||20 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Review of Economic Dynamics, Volume 11, Issue 4, October 2008, Pages 884–903
The present paper simulates the privatization of social security in an economy populated by overlapping generations of individuals that have time-consistent or time-inconsistent preferences, face mortality and individual income risk as well as borrowing constraints. We compute the transition path and compensate households in order to isolate the efficiency effects of the reforms. The model is calibrated to the German economy where the social security system offers little income insurance. Nevertheless, we find a positive role for social security due to the insurance provision against mortality risk and the provision of a commitment technology for present-biased consumers. In economies with rational consumers the elimination of social security yields an efficiency loss of roughly 0.6 percent of initial resources, while in economies with hyperbolic consumers the efficiency loss increases to 1.8 percent.
The welfare and efficiency consequences of pay-as-you-go financed social security are discussed extensively in the eco- nomic literature. Since the ground-breaking work of Auerbach and Kotlikoff (1987) various general equilibrium models have been developed which simulate the elimination (or “privatization”) of social security. On the one hand, studies without income and lifespan uncertainty compute the transition path and quantify the intergenerational welfare effects of such a reform. Typically, these models also provide some aggregate efficiency calculations by compensating the income effects of existing generations. Most of these studies demonstrate that social security privatization would generate significant growth effects, improve the welfare of future generations and enhance aggregate efficiency due to the reduction of labor supply dis- tortions. On the other hand, recent advances with general equilibrium simulation models have included idiosyncratic income and lifespan uncertainty as well as liquidity constraints in order to study social security reforms. Their results indicate that, at least in the long run, generations are still better off from privatization even when social security provides insurance in an uncertain world. Consequently, it is not surprising that many economists in industrialized countries propose the partial or complete privatization of social security in order to improve economic efficiency and growth. However, the simulation approaches of the past have some specific deficiencies that impair their central results. On the one side, the welfare and efficiency effects computed with deterministic models do not include insurance and liquidity effects of social security. On the other side, the computed long-run welfare gains in models which include idiosyncratic income and lifespan uncertainty are much less convincing if they are simply due to intergenerational redistribution effects.Therefore, it is useful to compute also in models with idiosyncratic uncertainty the transition path with compensating transfers in order to separate efficiency consequences of social security privatization from pure distributional effects. This is where the present study steps in. We simulate the elimination of the German social security system in a general equilibrium model featuring income and life-span uncertainty as well as household liquidity constraints. After computing the transition path to the new long-run equilibrium and the intergenerational welfare consequences, we isolate aggregate efficiency effects of such a reform in a separate simulation. The German social security system offers a special case for the analysis of privatization policies. In contrast to the US system, benefits in the German system are strongly linked to former contributions. This feature dampens the labor supply distortions of the German system while it reduces the insurance provision against income shocks at the same time. Consequently, the present study complements a recent analysis by Nishiyama and Smetters (2007) who simulate the partial privatization of the US social security system. While our general approach follows this previous analysis, we also offer a number of innovations. First, we isolate self-control problems by comparing aggregate efficiency in economies either populated by rational or hyperbolic individuals. Our approach extends the analysis of ̇ Imrohoro ̆ glu et al. (2003) by simulating the transition path and offering an alternative way to quantify the commitment effect which social security provides for hyperbolic consumers. Second, while the considered pension system provides little income insurance due to the strong tax- benefit linkage, we demonstrate how to separate and quantify its longevity insurance, liquidity and tax distortion effects. Third, in the sensitivity analysis we use a recursive preference approach which allows us to separate the impact of risk aversion and the intertemporal substitution elasticity. Our results indicate that the German social security system clearly enhances aggregate efficiency although it offers little income insurance. In our preferred parametrization, privatization would induce an aggregate efficiency loss of roughly 0.6 percent of resources in economies with rational consumers and a loss of roughly 1.8 percent with hyperbolic consumers. We show that these results are robust for alternative tax structures, budget adjustment assumptions and preference parameters. Finally, we show that a partial privatization would also yield a clear efficiency loss in our model. The next section reviews the previous literature. Then we discuss how we model preferences as well as the tax and benefit system. Section 4 explains the calibration and simulation approach. Finally, Section 5 presents the simulation results and Section 6 offers some concluding remarks.
نتیجه گیری انگلیسی
The results of this paper strongly suggest that the existing social security system has a positive impact on economic efficiency in Germany. For our central parametrization we find that the elimination of social security would induce an aggregate efficiency loss which amounts to roughly 0.6 and 1.8 percent of aggregate resources in the model with rational and hyperbolic consumers, respectively. We show that efficiency losses are mainly due to the implicit longevity insurance which is provided by social security. They are robust for various combinations of preference parameters and tax structures. Even a partial privatization of social security would reduce economic efficiency. Consequently, our results are in contrast with most of the previous quantitative general equilibrium studies on pension privatization. This is due to the fact that most deterministic models that apply our aggregate efficiency approach do not take into account insurance and liquidity effects of privatization. Most stochastic simulation models, on the other hand, only compare steady states or compute aggregate efficiency effects differently. To our knowledge, only Nishiyama and Smetters (2007) apply the same approach as we do. They find significant ag- gregate efficiency losses when they scale down the US social security system. In contrast to the German pension system, the US system is progressive and provides a direct income insurance which is not included in the above calculations. This indicates that social security might play an even stronger positive role in countries such as the US or United Kingdom where it clearly redistributes within generations. In a companion paper (Fehr and Habermann, in press) we also find that a more progressive social security system would increase aggregate efficiency in Germany. Therefore, the future reform debate should focus more on intra- and inter-generational insurance aspects of social security and less on the privatization issue.