Optimal unemployment insurance systems trade?off incentives and insurance. Since unemployment risk is intimately related to the business cycle situation it is to be expected that the value of insurance is business cycle dependent. At the same time it may be conjectured that the distortions from unemployment insurance may be larger in periods with low unemployment, and vice versa. Both of these effects go in the direction of making optimal benefit levels counter cyclical, that is, benefit generosity is high when unemployment is high, and low when unemployment is low.
Some countries have explicit rules linking elements of the unemployment insurance system to the state of the labour market. Probably the most sophisticated scheme is found in Canada where benefit eligibility, levels, and duration depend on the level of unemployment according to predetermined rules . The US has a system of extended benefit duration in high unemployment periods (see Committee on Way and Means (2004)). Other countries have pursued a more discretionary approach in some cases in a semi automatic fashion by adjusting labour market policies to the state of labour market i.e. extending benefits or labour market policies in general in high unemployment periods and tightening the schemes in periods with low unemployment.
The is a large literature on the design of unemployment insurance schemes. Since Baily (1978) it is well?known that the optimal benefit level trades?off insurance and incentives. Recent work has extended these insights in various directions (for a survey see e.g. Frederiksson and Holmlund (2006)). Surprisingly there is neither a large theoretical literature on the effects of state dependent unemployment insurance, nor an empirical literature exploring the state dependencies in the effects of various labour market policies including the benefit level. Kiley (2003) and Sanchez (2008) argue within a search framework that the initial benefit level should be higher and its negative duration dependence weaker in a business cycle downturn compared to an upturn. Both models are partial and rely on the assumption that benefits are more distortionary in a boom. In Andersen and Svarer (2008) it is shown that the optimality of countercyclical benefit levels depends not only on the possibility of using the public budget as a buffer but also whether distortions move procyclically. In this case countercyclical unemployment benefits may also contribute to lower the structural (average)unemployment rate. However, the model is static and does not allow for changes in the business cycle situation.
This paper develops a search model in which the business cycle situation may change between good and bad times. Matching frictions imply a co existence of unemployed persons and vacant jobs, but the underlying job separation rates and job finding rates are business cycle dependent. The unemployment benefit scheme is tax financed and benefits are allowed to be state contingent. Since the main issue in this paper is the trade off between insurance and incentive, the model is cast in such a way that it focusses on how unemployment benefits affect job search incentives. The key task is to work out the implications of state dependent benefit levels on the unemployment rate, and the welfare implications of such a dependency. State dependent unemployment benefits strengthen also automatic stabilizers which may have effects via aggregate demand effects. Such effect do not arise in the present framework which is has focus on the structural consequences of state dependent benefit levels.
It is shown that the possible change in the business cycle situation has an important effect on search behaviour and therefore unemployment and other key variables. The reason is that agents perceive the possibility of a change in the business cycle situation and this affects the search behaviour of unemployed. Clearly this effect depends on both the difference between the two states and the likelihood of a change in the business cycle situation. It is an implication that the incentive effects of a given benefit level or changes herein differ between upturns and downturns, with distortions of search behaviour being largest in upturns. Allowing benefits to depend on the business cycle situation may therefore have important effects on search behaviour and the unemployment rate (both in the different states of nature and on average across the states of nature). It is among other things shown that higher benefits in a downturn and lower benefits in an upturn may increase search effort in both states of nature, and therefore cause a fall in unemployment in both states. This arises if the business cycle situation is not very persistent and agents in a downturn perceive a high probability of a shift to an upturn with a higher job finding rate. Optimal state contingent benefits have higher benefits in downturns than upturns, and this is shown to lower structural (average) unemployment, but it may be achieved at the cost of more variability in the unemployment rate (higher in downturns, and lower in upturns). This shows that state contingent benefits levels may improve insurance without jeopardizing structural concerns.
The paper is organized as follows: The model is set?up in section 2, and as a prelude to the subsequent discussion section 3 briefly considers as a benchmark case the one state version of the model. The main results are given in section 4 exploring both the consequences of state dependent benefit levels and the optimal benefit levels. Concluding remarks are given in section 5.
Download
PDF Ebook Business Cycle Dependent Unemployment Insurance
