Income Risk, Precautionary Saving, and Loss Aversion – An Empirical Test

Abstract

This paper empirically examines the behavioral precautionary saving hypothesis by Kőszegi and Rabin (2009) stating that uncertainty about future income triggers saving because of loss aversion. We extend their theoretical analysis to also consider the internal margin, i.e., the strength, of loss aversion, and empirically study the relation between income risk, experimentally elicited loss aversion and precautionary savings. We do so using a sample of 640 individuals from the low-income population of Bogotá, characterized by limited financial education and subject to substantial income risk. In line with the theoretical predictions, we find that an increase in income risk is associated with higher savings for loss-averse individuals, and that this increase in savings grows with the degree of loss aversion. Thus, as suggested by Kőszegi and Rabin (2009), but contrarily to common assumptions, our findings establish that loss aversion is not necessarily an obstacle to saving, and thus identify new approaches of increasing saving among individuals with low financial education.

Precautionary Savings, Loss Aversion, and Risk: Theory and Evidence

“…As far as we know, we are the first to test this prediction in an explicitly dynamic model, where individual-specific loss aversion is directly measured. 25 As noted in Corollary 1, a direct implication is that the precautionary savings of loss averse subjects (λ > 1) are likely to be lower than the precautionary savings of loss tolerant subjects (λ < 1). We test these predictions in this section.…”

“…As described above, this led to our prediction (which we successfully tested) that loss aversion reduces savings. However, several papers with a dynamic structure ignore the effect of loss aversion at time t = 1 and take account of only the effect of loss aversion in mitigating the effect of the bad income state in the future at time t = 2 (Aizenman, 1998;Siegmann, 2002;Koszegi and Rabin, 2009;Park, 2016;Pagel, 2017;Ibanez and Schneider, 2023).…”

“…1 Loss aversion is a central concept in behavioral economics and underpins some of the most successful applications of behavioral economics in humans and animals. 2 The literature has also attempted to derive a link between loss aversion and savings (Aizenman, 1998;Bowman et al, 1999;Siegmann, 2002;Koszegi and Rabin, 2009;Park, 2016;Pagel, 2017;Ibanez and Schneider, 2023); we note our differences from this literature below.…”

“…The only exception that we are aware of isIbanez and Schneider (2023) who use observational data. However, in the introduction, we have noted the extensive differences of our paper from theirs 26.…”

An analysis of precautionary savings practices of professional football players

“…Other studies reveal that high perceived income risks lead to a postponement or rejection of retirement savings (Pfeiffer et al, 2007). Studies that examine income uncertainties in conjunction with individual risk aversion show an increase in precautionary behaviour (Ibanez and Schneider, 2021). Kennickell and Lusardi (2004), on the other hand, demonstrate that lower income groups that face high income risks exhibit low precautionary saving behaviour.…”