Effects of Perceived Scarcity on Mental Health, Time and Risk Preferences, and Decision-Making During and After COVID-19 Lockdown: Quasi-Natural Experimental Study

Abstract

Background: The COVID-19 lockdowns led to significant resource constraints, potentially impacting mental health and decision-making behaviors. Understanding the psychological and behavioral consequences could inform designing interventions to mitigate the negative impacts of episodic scarcity during crises like pandemics.

Objective: To investigate the effects of perceived scarcity on mental health (stress and fear), cognitive functioning, time and risk preferences (present bias and risk aversion), and trade-offs between groceries, health, and temptation goods during and after the COVID-19 lockdown in Shanghai.

Methods: A quasi-natural experiment was conducted in Shanghai during and after the COVID-19 lockdown. Web-based surveys were administered in May 2022 (during lockdown) and September 2022 (post-lockdown). Propensity score matching was used to balance demographic factors between the groups (During: n=332; After: n=339). Data were analyzed using regression analyses, controlling for potential confounders and applying propensity score matching weights.

Results: Perceived scarcity was significantly higher during the lockdown (mean 7.97 (SD 2.1)) than after (mean 4.35 (SD 2.27); P<.001). Higher perceived scarcity was associated with increased stress levels both during (standardized β coefficient=.62, P<.001) and after the lockdown (standardized β coefficient=.65, P<.001). Perceived scarcity also predicted greater fear of COVID-19 after lockdown (standardized β coefficient =.38, P<.001), though not during lockdown. Cognitive functioning remained stable, possibly due to a ceiling effect from high education levels. Monetary risk aversion increased under prolonged scarcity during lockdown (scarcity×during-lockdown interaction standardized β coefficient=4.68, P<.001). Present bias (tendency to choose immediate rewards) showed no significant overall change between groups, in line with recent evidence of stable time preferences during the pandemic. During lockdown, participants allocated more budget to groceries (standardized β coefficient=.67, P=.01) and less to health items (standardized β coefficient=−.61, P=.02), compared to post-lockdown, reflecting shifted priorities on pressing needs under scarcity. Subgroup analyses indicated stratified heterogeneity. Women increased their grocery spending (standardized β coefficient =.16, P=.04) and reduced spending on health items (standardized β coefficient = –.15, P=.05). Lower education participants exhibited more risk-averse attitudes (standardized β coefficient =.80, P=.01) under scarcity, whereas age and income did not significantly moderate these effects.

Conclusions: The study highlights that perceived scarcity during lockdown intensified stress and altered decision-making behaviors, including increased monetary risk aversion and shifts in spending priorities. Theoretically, this study advances the understanding of perceived scarcity by exploring its domain-specific effects on mental health and decision-making. Practically, these findings emphasize the need for public health strategies that mitigate the psychological impact of scarcity during crises, ensure access to essential goods, and support adaptive decision-making behaviors.