Consumption measurement and the limits of insurance: Evidence from Australia

Abstract

This study constructs disaggregated consumption measures using household- and individual-level panel data from the Household, Income and Labour Dynamics in Australia (HILDA) survey to examine the full consumption insurance hypothesis. The results reject full insurance, showing that individual consumption responds to idiosyncratic income shocks despite partial within-household risk-sharing. To strengthen empirical identification, new proxies for idiosyncratic risk are introduced, including involuntary job loss and financial stress. Overall, the evidence suggests that incomplete insurance and heterogeneous risk exposure undermine representative-agent assumptions, reinforcing the need for asset pricing tests based on micro-level consumption data.