Precautionary saving in Spain during the Great Recession: evidence from a panel of uncertainty indicators

The aim of this paper is to study empirically the effect of uncertainty on private consumption using a sample of Spanish households, and to check whether the appropriate measure of uncertainty varies with the macroeconomic context. Using data provided by the Spanish Survey of Household Finances (EFF) and the Labour Force Survey (LFS) we construct several uncertainty measures commonly used in the literature and an additional indicator based on job insecurity data and estimate different econometric models under the life-cycle/permanent income hypothesis, using these measures of uncertainty. Our results are twofold: first, we find evidence in favour of the precautionary saving hypothesis. Secondly, we find that the sources of uncertainty vary with the business cycle: the job insecurity indicator is an appropriate variable to approximate income uncertainty in any macroeconomic context, especially when the unemployment rate is low. When unemployment soars, however, it becomes the main uncertainty source for households, together with the degree of instability at the current job.