Labour market regimes and saving behaviour

The article analyzes the influence of labour market regimes in Germany and the USA on private household saving from a comparative perspective. I try to answer the question whether the amount of saving of employees is related to the perception of their individual unemployment risk. The theoretical foundation is a “buffer-stock saving“ model, which assumes that individuals insure against income volatility by the accumulation of wealth i.e. precautionary saving. The focus lies on a very specific form of income volatility, as “(…) the most obvious form of illiquidity arises when involuntary unemployment constrains labour supply at zero“. Explanations on the macro level like the cannot fully account for the differences in saving behaviour. While micro level studies allow to analyze the influence of individual perceptions and other household specific factors, it is done most of the time in a national specific context only. Therefore, a comparative perspective considers the interaction of both levels in different national contexts: lower flexibility of the German labour market – indicated by the average unemployment duration – and consequently higher individual fears of job loss could be part of the explanation why in continental welfare states like Germany, France or Austria the private saving rates are substantially higher than in their Anglo-Saxon counterparts i.e. USA, Canada, Australia or the United Kingdom. To empirically test this hypothesis I use panel data on employees in Germany and the USA from the SOEP and the PSID. This data is then combined with job attitudes obtained from the ISSP via a two-stage auxiliary instrumental variables estimator. The approach allows for insights on the relation between general unemployment risk factors, like employment sectors, and the actual subjective risk perception of the individuals under study. Finally, the model also controls for possible confounders like differing individual consumption preferences and country specific credit market imperfections.