Determinants of Household Savings: A Cross-Country Analysis

Abstract

This paper analyses determinants of household savings in a model based on an extension of the disequilibrium savings theory. These extensions follow from the life-cycle and permanent-income theories. Based on panel data for 14 countries spanning the period 2000–2018, fixed-effect least squares and two-stage least squares estimation procedures were used. In line with previous studies, there is strong and robust evidence for the hypotheses of disequilibrium savings theory, specifically, positive effects of unanticipated income changes, unanticipated inflation and the lagged savings rate. There is also robust evidence for the income uncertainty hypothesis that uncertainty has a positive effect on savings. The analysis presents some evidence that social security suppresses savings, but finds no significant effects on the interest rate or old-age dependency ratio. Unexpectedly, the participation rate of the elderly has a significant positive effect in some specifications. These findings contribute to the debate on whether and how governments can influence saving behavior.

Household Savings and Negative Interest Rates

“…In line with disequilibrium-savings theory, unanticipated income, unanticipated inflation and the lagged savings ratio all have a positive and statistically significant effect on the savings ratio. Koskela and Virén (1983) and Fredriksson and Staal (2021) presented results similar to those of Aizenman et al (2019) for unanticipated (5)…”

“…However, Koskela and Virén (1983) did not find a significant effect of social security, but presented evidence supporting the disequilibrium-savings theory. The findings presented in Nicolescu-Aron and Mihăescu (2012), Aizenman et al (2019), andFredriksson andStaal (2021) support the latter theory, indicating that unanticipated income, unanticipated inflation and the perceived optimal savings ratio have positive effects on savings. Finally, El Mekkaoui de Freitas and Martins (2014) and Aizenman et al (2019) provided evidence for the Ricardian equivalence proposition.…”

“…Instead of a comprehensive review of the vast empirical literature on household savings behavior, this literature review is limited to studies based on an approach similar to the one used in this paper. Feldstein (1974Feldstein ( , 1977Feldstein ( , 1980 and Fredriksson and Staal (2021) presented evidence that social security depresses household savings, as predicted by the permanent-income and life-cycle theories. However, Koskela and Virén (1983) did not find a significant effect of social security, but presented evidence supporting the disequilibrium-savings theory.…”

“…With R it either R l it or R s it , and using the constant-and static-expectation hypotheses, respectively, this implies four additional estimations of Eqs. ( 5) and ( 6): and Note that Koskela and Virén (1983), Nicolescu-Aron and Mihăescu (2012), Aizenman et al ( 2019), El Mekkaoui de Freitas and Martins (2014), and Fredriksson and Staal (2021) used similar specifications in their analysis. This enables relating the findings of this paper, especially for Eqs.…”

“…Thus, the lagged savings ratio is expected to affect savings positively. As in Koskela and Virén (1983) and Fredriksson and Staal (2021), the disequilibrium-savings theory is the foundation for the estimated savings function. In addition to the unanticipated income effect due to inflation, as well as the inflation effect on savings and the lagged savings ratio motivated by this theory, the analysis also considers interest rates, income uncertainty, social security spending, the old-age dependency ratio, the participation rate of the elderly, government surplus, and health expenditures as explanatory variables for saving decisions.…”