Financial scarcity and cognitive performance: A meta-analysis

• Published in: Journal of economic psychology 2024-03, Vol.101, p.102702, Article 102702
• Main Authors: de Almeida, Filipa, Scott, Ian J., Soro, Jerônimo C., Fernandes, Daniel, Amaral, André R., Catarino, Mafalda L., Arêde, André, Ferreira, Mário B.
• Format: Article
• Language: English
• Subjects: Cognitive functioning; Education; Financial scarcity; Poverty; Procedural processes; Working memory
• ISSN: 0167-4870; 1872-7719
• DOI: 10.1016/j.joep.2024.102702

Whereas several studies find that financial scarcity has a detrimental impact on cognitive functioning, some studies find no relationship and others even report beneficial effects. To shed light on this issue we conducted a meta-analysis on the relationship between financial scarcity and cognitive functioning. We went beyond testing the direct relationship between these two concepts and looked at potential moderators, namely education, the moment of scarcity, the severity of scarcity, the type of tasks used to assess cognitive functioning, and the type of study. Our findings suggest that scarcity does have a detrimental effect on cognitive functioning. Across 256 effect sizes from 29 datasets involving 111,852 respondents, we found a detrimental total effect of scarcity on cognitive performance of Hedge’s g = -0.43. We then estimated a meta-regression model of the drivers of the effect of scarcity on cognition. Education strongly explained this relationship, reducing the effect size by 60 % (partial effect of scarcity on cognitive performance is Hedge’s g = -0.15, when accounting for education), to a small effect size. The moment and the severity of scarcity also contribute to this relationship, by moderating the effect, such that lifetime and adulthood scarcity have a larger effect than childhood scarcity, and more extreme levels of scarcity lead to higher cognitive dysfunction. The type of task used to assess cognitive functioning did not moderate the effect. And when controlling for education, higher effect sizes were found for non-correlational designs. We discuss these findings and their implications in light of existing research and theories.