This paper addresses the question of when is unemployment insurance likely to be most valuable. Bailey (1978) showed that the extent of individual savings affects the value of UI. More recently, a literature has emerged that studies explicitly how well households can smooth consumption against transitory fluctuations in income, particularly those fluctuations that result from unemployment: there is limited value from UI providing risk,pooling across individuals if unemployment is transitory. However, we show in this paper that UI can still play a role in smoothing consumption when a lack of liquidity means it is costly for individuals to smooth consumption over time. In particular, the value of UI depends on how difficult it is for the unemployed to access credit markets and on the cost of saving. We provide empirical evidence that indicates substantial heterogeneity both in access to credit markets and in the cost of saving. The aim of this paper is to show the implications of these differences in the cost of self insurance for the effects of UI, and to provide a framework for showing how these differences can affect optimal replacement rates.
Differences in the cost of accumulating liquid assets arise for many reasons. First, some individuals are already accumulating illiquid saving for retirement, paying a retirement with, olding tax, and this makes liquid saving more costly. This highlights the interaction between different social insurance programs: forced illiquid saving for retirement may make the liqudity provided by UI more beneficial. Second, more generally, consumption needs differ over the life cycle, reflecting for example the presence of children, and this leads to differences in the cost of forgoing consumption in order to save. Third, individuals differ in their expected income growth, and particularly for those who expect the fastest growth, a need to save for precautionary reasons operates against the desire to smooth consumption. Finally, individuals may differ in their impatience or willingness to be exposed to consumption fluctuations. These sources of heterogeneity in the cost of saving, and also differences in individuals access to credit, will lead to differences in the amount of savings, and hence in the value of UI.