Self-Insurance and Consumption Risk-Sharing between Birth-Year Cohorts in Turkey

Self-Insurance and Consumption Risk-Sharing between Birth-Year Cohorts in Turkey

17/01

January 2017

This paper tests the empirical validity of consumption risk-sharing hypothesis across urban and rural regions in Turkey. For this purpose, I analyze twelve consecutive waves of the TURKSTAT Household Budget Surveys from 2003 to 2014 and prepare a pseudo-panel data set for birth-year cohorts following Deaton (1985). There are three important findings of this paper. First, our empirical analysis shows that there is imperfect consumption risk-sharing between birth-year cohorts in Turkey. We observe that the growth of cohort consumption is positively and significantly associated with both the growth of cohort income and the growth of aggregate consumption. Second, the need for risk-sharing is analyzed using the variance of the growth of cohort consumption as a proxy variable for future labor income uncertainty. Empirical findings reveal that urban households and mature working-age households are more sensitive to future labor income uncertainty. Finally, cohorts would have sacrificed modest amounts to insure their consumption.

Consumption risk-sharing, Income and consumption smoothing, Cohort, Pseudo-panel