Is the Level of Unemployment Insurance Self-Perpetuating?

The generosity of unemployment insurance systems differs substantially across countries. According to a summary measure provided by the OECD, during the last decade unemployment benefits in Western Europe (with the exception of Italy and the UK) have been about three times as large as those in Japan and the US. A growing literature has argued that the level of benefits is a major factor in explaining the large differences in unemployment rates and income inequality observed in Europe and the US. Yet the majority of this work treats unemployment insurance as an exogenous variable and very few authors have attempted to explain the existence of the different levels. So why do countries choose such dramatically different levels of unemployment insurance?

One possible answer is that agents simply have different preferences in different countries. Or, more subtly, agents have different perceptions of the effects of different institutions on the economic performance of their country. However, a Discussion Paper by John Hassler, José V Rodríguez Mora, Kjetil Storesletten and Fabrizio Zilibotti argues that different societies populated with identical rational agents who differ only in their initial distribution of human capital may still choose very different unemployment insurance levels that can be sustained as stable equilibria.

The process through which unemployment benefits affect search behaviour in the labour market is well documented. The main contribution of Hassler et al's paper is that it explains how search behaviour can alter society's preferences towards unemployment benefits, thus giving rise to multiple steady-state equilibria with different levels of unemployment insurance. In particular, a European-type steady state with high unemployment, low employment turnover and high unemployment insurance can coexist with a US-type steady state with low unemployment, high employment turnover and low unemployment insurance. The authors present a calibrated version of their model, featuring two distinct steady-state equilibria with unemployment levels and duration rates resembling those of Europe and the US.

The paper develops a dynamic general equilibrium model that is characterized by search frictions in the labour market and populated by a continuum of overlapping generations of non-altruistic agents. Workers acquire sector-specific skills through learning-by-doing on the job. Job destruction is stochastic and the probability of losing a job depends on the worker's human capital in the sector where they are working. Agents are risk averse and can self-insure through precautionary saving.

Depending on their current labour market conditions, some agents attach more value than others to unemployment insurance and this results in divergent political views about the amount of income taxation used for financing unemployment benefits. The unemployed generally prefer more generous levels of benefits, but their political influence is limited due to their relatively small numbers. But preferences over unemployment insurance also differ across groups of employed workers. In particular, and central to the paper's results, more 'specialized' workers (i.e. those with a pronounced comparative advantage for working in a particular activity) will tend to value insurance more highly than workers whose skills are of a more general nature.   

There is a large body of empirical evidence documenting how wages change when displaced workers are forced to switch industries. Workers switching industries after losing their previous job usually suffer much larger losses than equivalent workers remaining in the same industry. This evidence supports the view that there is a significant accumulation of human capital on the job and part of this is lost if a worker switches industries. Hence, when a specialized worker is displaced, they face a trade-off between accepting any job and suffering a wage cut with respect to their pre-displacement wage, or waiting for a job offer where they have a comparative advantage, which implies a longer unemployment spell.

Specialized workers, therefore, tend to pursue relatively more selective search strategies, which entail a greater risk of long durations of unemployment. In order to hedge this risk, they prefer more generous unemployment insurance. And the more generous level of unemployment insurance reinforces the degree of specialization among workers. Hence, it is this reinforcing interaction between specialization and preference for insurance that can give rise to multiple steady-state equilibria. In particular, two economies with small or even no differences in preferences or technology may exhibit very different political choices over unemployment insurance and therefore large differences in their economic performance.

The central mechanism of this theory is that workers who suffer large wage losses when accepting certain job offers would reject these offers if unemployment benefits were more generous. It is therefore a key empirical prediction that post-displacement wage losses should, in equilibrium, be lower in Europe than in the US. This implication is confirmed by the data: a number of empirical studies suggest that displacement leads to wage losses of 10–25% in the US and 0–4% in Europe.

It would seem that the accumulation of sector-specific skills can generate a two-way causality where social insurance affects economic behaviour, which in turn feeds back to preferences for social insurance. But the notion of specialization goes beyond human capital accumulation. The schooling system could be an alternative channel: when unemployment benefits are high a specific (risky) educational system – such as the European vocational schools or college degrees aimed at a specific profession – becomes more attractive. If a large number of workers have acquired these skills then the willingness to pay for unemployment insurance is likely to be high. Geographical mobility, which is lower in Europe than in the US, is another potential channel since buying a house serves as a region-specific capital investment.

The authors extend the paper by introducing the concept of hysteresis – i.e. the human capital of specialized individuals depreciates during spells of unemployment. For reasonable rates of loss of human capital the initial result of multiple steady-state equilibria remains, although the range of parameter values for which this is possible is reduced.  Finally, the paper shows that a calibrated version of the model has two sustainable steady-state equilibria: a 'European' equilibrium with an unemployment rate of 12.7%, an average duration of unemployment of 23 months and a replacement ratio of 76%; and a 'US' equilibrium with an unemployment rate of 6.4%, an average duration of unemployment of 4.5 months and a 24% replacement ratio.

It is therefore possible to explain the large differences in institutions and economic performance observed in Europe and the US without resorting to exogenous structural differences, other than different initial distributions of agents. The paper is able to justify the strong political support for generous unemployment insurance in Europe, despite a growing consensus that it causes high unemployment. It would seem that Europeans are willing to accept a high-unemployment, high-unemployment-insurance scenario, as this may best represent Europe's high proportion of specialized workers.

A general conclusion from the results is that strong inertia in changing social institutions may emerge endogenously, even if no exogenous cost of change is involved. However, although policy reform may be met by strong initial opposition, the political support for this should fade over time as the new levels of unemployment insurance change the distribution of the labour force. This would seem to give some credence to the idea of pushing ahead with welfare reforms despite their unpopularity. But the paper draws a slightly different conclusion. The results from the social welfare maximization case suggest that it may be socially optimal for Europe and the US to retain their respective levels of unemployment insurance. Since their labour market institutions have been sustained over a long period of time, they have led to distributions of voters where many would lose from changes in the status quo.

Discussion Paper No. 2126: 'Equilibrium Unemployment Insurance' by John Hassler (Stockholm University and CEPR), José V Rodríguez Mora (Universitat Pompeu Fabra, Barcelona), Kjetil Storesletten (Institute for International Economic Studies, Stockholm University, and CEPR) and Fabrizio Zilibotti (Institute for International Economic Studies, Stockholm University, and CEPR).