tudies of the effect of potential unemployment benefits on the duration of unemployment usually focus on the wage-benefit replacement ratio. Since personal labour history and past earnings are used to calculate benefits, however, it is very difficult to disentangle the independent influences of labour-market history and the generosity of benefits. This has led to the application of ‘quasi-natural experiments’, where it is easier to observe exogenous and sometimes selective changes in the benefit system.

In Discussion Paper No. 1534, Rudolf Winter-Ebmer applied this methodology to Austria in 1988, when the period for which elderly people could claim unemployment benefits was extended from one to four years in specific regions of the country. The paper confirms first that the anticipated effect of increased unemployment for the group did materialize and then concentrates on the incentive effects of potential benefit on unemployment duration. Previous work for Germany and the United States has indicated that increased potential duration does lead to slightly longer actual unemployment spells. Although the period over which benefits can be claimed is one of the main differences between the European and US benefit systems, the author argues that this feature can explain only a small part of the difference in long-term unemployment rates. The empirical analysis shows that men react significantly to the benefit duration whereas women generally do not. The magnitude of this reaction is smaller in comparison to Germany and the United States. Furthermore, the impact of extended benefit duration is differentiated for short and long spells. Whereas for long spells a higher impact for men and for women is found, no unemployment-prolonging effects could be detected, for short spells.

Potential Unemployment Benefit Duration and Spell Length: Lessons from a Quasi-experiment in Austria
Rudolf Winter-Ebmer