comprise the generosity and the duration of unemployment benefits, labor income taxation, the
cost of retraining and welfare benefits. The socio-economic characteristics of the unemployed
taken into account are their age, skill level, and time preference. The effects of labor market
performance are captured by exogenous shocks to contemporaneous and expected earnings in
the labor market, and these shocks might be temporary or permanent.
Our analysis yields a large number of results that are in line with empirical findings, and
some of our results are qualitatively similar to those obtained by Ljungqvist and Sargent (1998).
However, the main difference to their results is that we find a generous welfare state, in partic-
ular high taxes on labor income and generous welfare assistance, to be important determinants
of long-term unemployment, while they identify unemployment insurance as the main culprit.
These results are qualitatively similar in that these factors all reduce the attractiveness of work-
ing relative to unemployment, but practical policy implications might still differ.
We believe that our approach has several advantages. Our model is rich in that it incorporates
a large number of determinants of unemployment; it is simple in that the formal reasoning is
tractable and can be graphically displayed in a single diagram, it is flexible in that the model
can easily be extended to account for present-biased preferences, and it is accurate in the sense
that most of the predictions are in line with stylized empirical facts. These advantages come
at the cost of some limitations. Our account is partial as it focuses on the supply-side and
is not embedded in a general equilibrium framework; if abstracts from search frictions, and
macroeconomic fluctuations are modelled in a highly stylized fashion. On balance, we believe
that our model provides an attractive complement to more complex models of unemployment.
The paper proceeds as follows. The model presented in Section 2 builds on the standard
assumptions that the unemployed make their decisions to retrain and on how long to remain
unemployed exclusively according to monetary criteria and they are assumed to make these
decisions in a fully rational and consistent way. Section 3 investigates comparative-statics of
labor market institutions, socio-economic characteristics, and labor market conditions. Section
4 relaxes the standard assumptions by allowing for time-inconsistent preferences. This opens
up the possibility that agents get stuck in an unemployment trap and remain on the dole
permanently, against their best intentions.