بیمه بیکاری در شیلی: آیااین در چرخه کسب و کارثبات ایجادمی کند؟

We explore the stabilizing effects of unemployment insurance in Chile. A dynamic general equilibrium model is calibrated for the Chilean economy for the 1960–2000 period. We assume that the economy is subject to exogenous technological shocks and that a fraction of the population is liquidity constrained. Our main conclusion is that unemployment insurance has some stabilizing effect on the business cycle, especially on consumption, but that this effect is of the second order of magnitude. We also find that the larger the fraction of the population that is liquidity constrained, the more likely the program is welfare improving. Our results suggest that the objective of stabilizing the business cycle would be more efficiently achieved using alternative instruments.

In October 2002 an unemployment insurance fund was introduced in Chile with the stated aim of protecting workers’ income levels when they become unemployed. This paper considers the potentially unintended business cycle effects of unemployment insurance, in particular the question of whether this insurance has stabilizing effects in terms of making the business cycle less pronounced. Stabilizing effects ensue if liquidity constrained agents are allowed access to unemployment insurance funds when they become unemployed in a recession, allowing them to reduce their consumption by less than they would have done if there were no unemployment insurance system. As contributions to the fund are larger in booms than in recessions, this potentially provides an additional stabilizing effect. We assume non-diversifiable aggregate technological shocks that produce fluctuations in variables such as production, employment and consumption. People are assumed to be liquidity constrained and they do not have perfect access to the capital market. The benefits of reducing business fluctuation have been widely studied in the literature. Lucas (1987) voiced doubts as to the value of these benefits, calculating that the effects on welfare are minimal. He compared his estimate of the benefits of attenuating the volatility of the business cycle with the large welfare benefits that attend economic growth, concluding that the profession would do better to focus on growth rather than on stabilization policies. In the case of unemployment insurance the literature has focused on the stabilization and welfare properties of this insurance when markets are incomplete. Based on a model of unemployment insurance Baily (1977) reports results as to how much insurance should be provided, and in what form. Hamermesh (1982) makes use of a model to determine whether current levels of unemployment insurance (UI) in the US are sufficient to overcome the liquidity constraint faced by the unemployed. He finds that a large portion of UI benefits do little to stabilize the economy, because people consume them as if they were fully expected, reducing their saving behavior when employed. Easley, Kiefer, and Possen (1985) use a two person, two period general equilibrium model with uncertain productivity in the second period. As agents cannot self-insure the introduction of UI implies a potential Pareto welfare improvement. They also make use of a theoretical model to compare the welfare gains of UI vis à vis a negative income tax. Hansen and Imrohoroglu (1992) study the role of unemployment insurance in an economy with liquidity constraints and moral hazard using a quantitative general equilibrium model. They assume that people cannot borrow in the capital market and that agents face exogenous idiosyncratic employment shocks (there are no aggregate shocks). They conclude that if there is no moral hazard the optimal replacement rate may be as high as 0.65 (similar to that found in the US economy) and that the welfare benefits of UI are large. However, if there is moral hazard and the replacement rate is not set at the optimal level, the economy can be worse off with UI than it would have been without it. Imrohoroglu (1989), and Atkenson and Phelan (1994), argue the unemployed bear a disproportionate burden of the cost of employment fluctuations during recessions. Both papers focus on the unemployment risk as the main undiversified risk associated with the business cycle. Nonetheless, their estimates of the welfare gains of curbing business cycle fluctuations are also small1 because the data shows very little time variation in the average duration of US unemployment. Hence, the risk of a long period of unemployment in a recession is relatively small. However, Beaudry and Pages (2001) argue that focusing only on unemployment duration may underestimate the welfare gains of stabilization policies. They conclude that mild variability of aggregate wages may hide important business cycle fluctuations in individual wages and that this source of risk implies substantial welfare costs. They also conclude that attention to the design of unemployment insurance is required if UI is to contribute to diversifying the risk of economic fluctuations. More specifically, they find that unconditional UI can be an inefficient way of reducing the cost of business fluctuations, while a state contingent UI scheme that offers more generous subsidies during recessions than during expansions improves risk sharing and reduces the cost of business cycles. Brown and Ferral (2003) study the interaction of the business cycle, unemployment insurance and the labor market for young men in Canada. They argue that the design of UI is important, proving that in some cases a poorly designed UI scheme can exacerbate recessions. The effect of unemployment insurance on the business cycle has not been studied for the Chilean economy. In this paper we use a dynamic general equilibrium model to study the stabilization properties of the Chilean UI program on the business cycle. Specifically we use a real business cycle model with liquidity constrained agents and an economy subject to exogenous technological shocks. The model captures the effect of the unemployment insurance program on fluctuations of output, consumption, investment, the capital stock and employment. It is important to bear in mind that while unemployment insurance has the effect of reducing the liquidity constraint for people that are laid off, hence reducing the volatility of consumption, the taxes used to finance the program are themselves distortionary. We find that in the case of Chile the unemployment insurance program marginally attenuates business cycle fluctuations. Whether the program is welfare improving is found to depend on the fraction of the population with liquidity constraints. The paper is organized as follows. In Section 2 we describe the unemployment insurance program in Chile. Section 3 presents the model and Section 4 its calibration and simulation. The policy implications of our results are discussed in Section 5. Section 6 concludes.

This paper studies the effects of unemployment insurance on the business cycle. We consider whether the unemployment insurance program that was introduced in Chile in October 2002 has had stabilizing effects on the business cycle, simulating the presence of this insurance over the 1960–2000 period. We use a dynamic general equilibrium model à la Hansen (1985), where the economy is subject to exogenous technological shocks which produce fluctuations in output, investment, consumption and employment. The model also has a fraction of the population that is liquidity constrained: individuals who do not have access to the capital market and hence cannot save, forcing them to consume their income, thus making their consumption more volatile. Such individuals cannot replicate the unemployment insurance program in the capital market, which makes this program potentially welfare improving. Our results show that unemployment insurance reduces the volatility of the macroeconomic variables under consideration, especially the volatility of consumption. However, the effect is rather small. We conclude that the most appropriate justification for the current unemployment insurance program in Chile is that advanced when it was created: that it improves the welfare of the poor when they are unemployed. The possible additional justification explored in this paper – the stabilization of the business cycle – does not seem to be large enough to be considered an important achievement of the program. We also conclude that the larger the population with liquidity constraints the more likely that the unemployment insurance program is welfare improving. This is because on the one hand the program loosens the constraint for those that are liquidity constrained, but on the other it is funded via distortionary taxation. The larger the fraction of the population that is liquidity constrained, the more important the welfare improving first effect. We also find that as the tax rate increases, the distortionary effect becomes more significant and the likelihood that the program will reduce welfare rises. A final consideration relates to the potential instruments that an economy like Chile, which faces large exogenous shocks, has at hand for stabilizing the business cycle. This paper suggests that the unemployment insurance is not an efficient way of attaining this goal. Although not the topic of this paper, it is likely that policies such as Chile's current fiscal structural balance,10 or a price-smoothing fund for the commodities it produces11 are more efficient means of smoothing the business cycle.