ادغام بیمه بیکاری با بیمه بازنشستگی

This paper analyzes a social insurance system that integrates unemployment insurance with a pension program, allowing workers to borrow against their future wage income to finance consumption during an unemployment episode and thus improving search incentives while reducing the risks arising from unemployment. This paper identifies the conditions under which integration improves welfare and the factors which determine the optimal degree of integration. We show that when the duration of unemployment is very short compared to the period of employment or retirement, the optimal system involves exclusive reliance on pension-funded self-insurance. This system imposes a negligible risk burden for workers while avoiding attenuating search incentives. We also argue that joint integration of several social insurance programs with a pension program through an individual account is desirable unless the risks are perfectly correlated with each other.

The East Asian crisis brought home to much of the developing world a lesson that the Great Depression brought home to more advanced countries 70 years ago—the importance of a safety net. But as countries like Korea go about constructing their safety nets, they are cognizant of the complaints that have been raised against unemployment insurance systems: they attenuate incentives. To be sure, there are adverse incentive effects (or, as they are today generally referred to, moral hazard effects) in all insurance programs. What worries critics is that the risk reduction benefits might, on the face of it, be outweighed by the adverse incentive effects. For most individuals, a typical spell of unemployment is less than 6 months (and that spell would presumably be shorter, possibly much shorter in the absence of unemployment insurance.) Over a working time of, say, 45 years, an individual with three such spells would lose perhaps 4% of his lifetime income—a risk which presumably the individual could easily absorb if he had sufficient savings or could borrow against future earnings. With the bulk of savings used for retirement, and mostly dedicated to social security programs, the amount individuals have to buffer themselves against these income shocks is limited; and well-documented limitations in capital markets make it difficult for individuals to borrow much against future earnings. Thus compulsory old age public pension programs, while they help resolve one problem–the tendency of individuals not to save enough for their old age, because of the possibility of public “bail-outs”2–exacerbate another. This naturally leads to the suggestion of an integrated unemployment and pension program, which we will call the integrated unemployment insurance (UI) system. Such integration makes particular sense with individual accounts, which are increasingly forming the basis of even public pension programs. 3 In such programs, benefits are related to contributions by simple formulae; in the simplest form, there is no redistribution. Such programs are like defined contribution pension programs, though some of the contributions can be used to “purchase” insurance (e.g. against inflation or interest rate fluctuations) which is not available on the market. Although it would be simple to impose redistributions on top, for simplicity, this paper ignores the redistributive components. 4 Under the integrated UI program, an individual who is unemployed can have his unemployment payments taken out of his individual account. Thus, the individual obtains the liquidity to maintain his standard of living; the compulsory and universal nature of the contributions provides, in effect, perfect collateral, so that early on in his life, his account balance could become negative. Because normally the risk is small, the individual can bear this risk—when it is spread out over his entire life; and since the individual bears the risk, there is no attenuation of job search incentives.5 If, however, the loss from unemployment is substantial, it is optimal to have some true unemployment insurance—the individual should not bear the cost even over his lifetime. In general, individuals should not rely exclusively on the pension-funded self-insurance provided by an integrated UI system. (In this paper we take this ‘lifetime’ unemployment insurance to be tax-funded, but the results would be identical with a mandatory private insurance program with competitively determined premia.6) We characterize in this paper the optimal mix of the two types of unemployment benefits—tax-funded UI benefit and pension-funded borrowing (a form of self-insurance7), and to identify the circumstances in which integration is relevant (that is, welfare enhancing). The lower the risk-aversion, or the greater the elasticity of reemployment with respect to the insurance benefit, the less reliance should be placed on tax-funded insurance as opposed to (what might be viewed as implicit) pension-funded self-insurance. In an extreme case, if a worker is risk-neutral, then there should be no need for tax-funded unemployment insurance, and if there is no incentive problem associated with unemployment insurance, there is no need to rely on pension-funded self-insurance. Not surprisingly, the larger the risk, which in turn is related to the length of the period of unemployment relative to the working period, the greater the need for tax-funded insurance. In the limit, if the period of unemployment is vanishingly minute, then the individual can bear all the risk through pension-funded self-insurance with no welfare loss. The fact that unemployment risk is small compared to one's lifetime payoff may suggest a heavy reliance upon self-insurance. Even if individuals must rely on self-insurance, it may be possible to finance the required amount of self-insurance by savings made prior to unemployment. Thus, whether integration would enhance welfare depends upon the amount of pre-unemployment savings an individual makes, as well as upon the amount of “required” self-insurance. We identify in this paper the circumstances under which integration is welfare enhancing. Integration is more likely to be welfare enhancing as the optimal mix of unemployment benefits entails a smaller amount of UI benefit, i.e., in the case of lower risk-aversion, higher search elasticity and lower unemployment risk. The effects of self-insurance through savings and borrowings upon optimal unempljoyment insurance have recently been discussed in the optimal UI literature.8 Earlier literature focusing on the effect of UI on search intensity had discussed how benefits and wage tax should vary over time. For simplicity, these models had, however, ignored the possibility of savings and borrowing, which clearly can affect search intensity.9 These authors, however, do not address the central questions which are of concern here: when can the first and second best solutions be decentralized, when does self-insurance make any unemployment insurance unnecessary, when does private savings make any borrowing against pension savings unnecessary, and, more generally, when is there a social gain from allowing borrowing against such pension savings. The one set of papers that touch more closely on these issues recognizes that improving capital markets reduces the need for UI. Hansen and Imrohoroglu (1992) and Hamermesh (1982) empirically examined the welfare effect of a borrowing constraint for an unemployed worker by estimating how much UI increases his welfare, while Flemming (1978) showed by simulation that the optimal replacement rate of UI would be reduced in the presence of a perfect capital market. 10 Unemployment is, of course, only one of many risks that individuals face. There are, for instance, the risks of disability and health, as well as unemployment. The idea of integration can be applied to each of these risks, leading us to consider what we call an integrated lifetime insurance program. With integrated lifetime insurance, pension savings can be used to provide cover for these risks. The Provident Funds of Malaysia and Singapore provide prototypes of such an integrated program. The problem is that while the loss from any one of these risks may be small, there is some chance the individual may experience all of these losses. In that case, the funds available to an individual at the time of retirement may be reduced to an unacceptably low level. If the government has to provide lifetime insurance to cover such contingencies–in effect, bailing out the individual account because it would provide an unacceptably low level of pension–the borrowing by an individual to smooth consumption to cover any one of these risks would have an adverse disincentive effect much as the tax-funded benefit does. These considerations might appear to diminish the relevance of joint integration. We argue in this paper, however, that so long as the risks are not perfectly correlated then it always pays to integrate all the social insurance programs with the pension program rather than to have separate insurance programs covering each risk. The key point is that the integrated lifetime insurance system allows a given amount of pension savings of an individual to be used for benefits under all the risks. This benefit of joint integration–having a common pool from which to draw upon–gets larger as the correlation gets smaller. In the next section, a basic model is presented to characterize the constrained optimum and to examine its properties, and Section 3 shows how the integration of UI with an (optimally designed) pension program can support the optimum. Section 4 identifies the circumstances under which the constrained optimum entails no unemployment insurance and under which an integrated UI program improves welfare. Section 5 extends and generalizes the analysis of the previous sections in several directions, and an informal presentation of an integrated lifetime insurance system incorporating multiple risks is provided in Section 6.

The failure of markets to provide adequate insurance against certain risks has long been recognized.68 This, combined with the fact that social norms do not allow individuals in their old age to suffer from insufficient income, whether their misfortune arises partly because they have chosen to save insufficiently (either because of excessive myopia or because of rational exploitation of these social norms), or because they have undertaken excessive risks provides a rationale for a compulsory public pension program. This paper has developed a further advantage to such a program; it allows for the collateralization of future wage income in a way which is not easily possible otherwise, thus allowing individuals in effect to self-insure against a variety of risks. This paper has addressed two related issues. The possibility of pension-funded self-insurance does not eliminate the desirability of some tax-funded unemployment insurance. We have identified the factors on which the optimal degree of pension-funded self-insurance depends. Although our analysis suggests a heavy reliance on pension-funded self-insurance and argues for the relevance of integration, there are some limiting circumstances under which integration is not welfare enhancing. This paper also identifies the conditions under which this is the case. Regardless of the motivation for public pension programs, they have become a part of the basic policy frameworks of all advanced industrial countries, and increasingly of countries in the developing world. The political processes which shape those programs are complex, and there is no reason to believe that the resulting social security programs are necessarily optimal, in any sense. This paper can be looked at in another way: it suggests that a simple reform, allowing for the limited borrowing against public pension savings, may be welfare enhancing.69 When there are multiple risks (including the risk of multiple bouts of unemployment), again some reliance on pension-funded self-insurance is in general desirable. The integrated lifetime insurance system can always generate a welfare gain from allowing a common pool of pension savings to be shared to finance the benefits for those facing various shocks. The general principle naturally leads to the suggestion of a fully integrated lifetime insurance system through a joint individual account, an extension of the Provident Fund of Singapore and Malaysia, where major risks including disability and health are integrated with the public pensions program.