Health and random shocks: self-perceived health in Spain and in some European countries during the financial crisis

Abstract

We report in this Policy Paper on our results from two research papers, either published already or in the process to do it. In those papers we evaluate the association between the variations in income and wealth (both aggregate and split between real estate and financial wealth), and self-perceived health in Spain and in some other European countries using a longitudinal sample of individuals before and after the financial crisis. We estimated generalized linear mixed models, with a binomial response and a logistic link, for three/four waves of the European Survey of Household Finances (before and after the crisis), adjusting for variables at the family and individual levels. We also controlled for familial and individual heterogeneity and for temporal trends. In the case of Spain (2005, 2011, 2015, 2017), while an increase in wealth greatly increases the probability of younger individuals reporting better health, this is not the case for older individuals. Decreases in gross wealth are associated with decreases in the probability of declaring good/very good health only in families whose reference person is over 44 years old. We find moreover that: (i) not just income but net wealth effects impact on the consequences of income fluctuations on consumption and health assessed, (ii) the composition of individuals’ net wealth may also matter, since they are differently affected by the shocks in the economic crisis, (iii) age plays a significant role and, finally, (iv) individual reactions in terms of consumption and savings, given any level of income and wealth, according to the risk aversions for precautionary idiosyncratic motives, may also need to be considered in order to complete the picture. For the whole set of European countries we estimate three waves of the Eurosystem Household Finance and Consumption Survey (HFCS) (2011, 2015, 2017), adjusting for family and individual heterogeneity and for temporal trends. We observe that overall variations in income have a positive and significant impact on changes in self-perceived health during the financial crisis, but not after 2015 (expansion period). As a result, changes in income are important in protecting health during crisis periods. We do not observe an effect of changes in wealth. When comparing the results by country, our findings still hold for most of them, with income being the main driving force behind better self-perceived health. These results are important because changes in individuals’ income seem to be more important than net wealth, and are differently affected by the shocks in the economic crisis. Minor effects in changes of net wealth can be found just for the Netherlands and Germany. In brief we conclude that despite the fact that the financial crisis affected European countries differently, in most of them, income, rather than wealth, played an important role in protecting health. However, changes in income were not relevant in the expansion period. Wealth variation in all the specification types proves to have minor or no effect.