Using the UK Household Longitudinal Study (UKHLS), a large-scale panel survey representative of the UK population, we show that the increase in aggregate saving rates following the outbreak of the COVID-19 pandemic is disproportionately driven by homeowners with mortgages compared to outright homeowners. However, these effects are moderated by certain personality traits (extraversion and neuroticism) among the mortgagors. Our difference-in-differences (DID) estimate allows us to control for contemporaneous events that affect both the treatment and control groups (e.g., the shutdown of much of the economy during the COVID-19 pandemic, which may result in “forced savings”), as well as for demographic and economic variables such as pre-pandemic household long-run income. A crucial difference between the treatment group (homeowners with a mortgage) and the control group (outright homeowners) is the composition of their balance sheets. Specifically, mortgagors hold sizeable illiquid assets but little liquid wealth compared to outright homeowners, and they have mortgage payment commitments. Our first contribution to the literature is to quantify the causal impact of liquidity constraint on individual precautionary savings. Our findings show that mortgage debt may be a useful predictor of cyclical fluctuations of saving rates. Our empirical results also reveal a heterogeneous response of saving rates to the COVID-19 pandemic among mortgagors with different personality traits, as measured by the Big Five. This complements the growing literature relating personality traits to economic decision-making by linking personality traits to preferences and subjective beliefs. Our paper shows the importance of both household economic circumstances and persistent psychological characteristics in determining the savings dynamics facing external shocks. These differential changes in precautionary asset accumulation have important implications for the propagation of macroeconomic shocks and the subsequent recovery of consumer spending.

Co-authors

Dr Lingxiang Zhang, Beijing Institute of Technology, China