The economic importance of financial literacy is well documented. Consequently, the implementation of national strategies promoting financial literacy and the design of financial education policies and school mandates have become a high priority for policymakers around the world. However, the effectiveness of financial education is debated in quite fundamental ways, largely stemming from the fact that early rigorous experimental impact evaluations sometimes showed muted effects. This paper sets out to conduct a systematic review of the literature and a meta-analysis of the impact of financial education on both financial literacy and financial behaviour, in order to provide useful guidance for those evaluating future financial education programs.
The study is a meta-analysis of 76 randomized experiments with a total sample size of over 160,000 individuals in 32 countries with varying target groups. The papers selected were as follows:
- Studies reporting the causal effects of educational interventions designed to strengthen the participants’ financial literacy and/or leading to behaviour change.
- Studies using random assignment into treatment and control conditions.
- Studies providing a quantitative assessment of intervention impact that allows researchers to code an effect size estimate and its standard error.
As such the goal of the study is to provide the most comprehensive analysis of the work on financial education by using the most rigorous studies - randomized control trials; by assessing the heterogeneity in the programs such as differences in target groups, quality and the intensity of interventions; by discussing magnitudes of the effects in terms of economic significance per participant costs; and by discussing how to assess the impact of financial education and whether education decays with time.
Financial education has a positive effect: The authors find that financial education has positive causal treatment effects on financial knowledge and financial behaviours. The treatment effects on financial knowledge are similar in magnitude to the average effect sizes realized by educational interventions in other domains, such as math and reading. The effect sizes of financial education on financial behaviours are comparable to those realized in behaviour-change interventions in the health domain, or those aimed at fostering energy conservation.
The effects are larger than previous analysis has suggested: The estimated (weighted average) treatment effect is at least three times as large as the weighted average effect documented in previous analysis (cited) of randomized controlled trials. When accounting for the possibility of cross-study heterogeneity, the results show an estimated effect more than five times as large as reported in the same previous analysis.
The effects are of economic significance. The authors look at effect sizes and cost effectiveness and conclude that the effects of financial education are sufficient to be of economic significance. In contrast to earlier studies, the authors do not find differences in treatment effects for low income individuals and the general population. They don’t find strong evidence for a rapid decay in the realized treatment effects, or for the sustainability of long-run effects.
The review concludes that although consumers are generally behaving rationally to maximize their own utility, some are still shown to misuse their credit cards, make obvious mistakes, suffer from various behavioural biases and sometimes suffer exploitation by the banks. The authors recommend both increasing consumer awareness of the issues, via adverting or educational programmes and increasing regulations on card issuers, such as limiting fees, and making contracts easier to understand.
Points to consider
Methodological strengths/weaknesses: A major challenge in every meta-analysis lies in the heterogeneity of the underlying primary studies and how to account for it. The sample includes programs from a variety of countries, with varying target groups, representing a range of educational interventions from provision of an informational brochure to offering high-intensity classroom instruction; outcomes are also measured at different points in time and with different types of data. The studies also vary in their choice of dependent variables, ranging from a number of financial behaviours to financial knowledge.
- Further limitations include; that it is hard to determine the long-run impacts of interventions as few of the trials studied long-run effects; few studies are able to link their experiments to administrative data, so rely on self-reported survey data; only 20 papers within the 76 studied include a discussion of cost so estimates of cost-effectiveness are limited.
- The authors go into a great deal of detail into the measures – statistical and otherwise – they have taken to ensure the robustness and validity of their findings. For example, they apply various sensitivity checks (detailed in an appendix) to the findings about the average effect of financial education programs; they restrict the sample to studies published in top general interest or top field economics journals; when comparing to previous analysis, they take care to replicate methods previously used (cited and detailed in an appendix); they estimate five alternative models (detailed in an appendix ) including a correction for potential publication selection bias and a consideration of the power of the underlying primary studies.
- The authors conclude that results are robust, irrespective of the model used, when restricting the sample to only those RCTs that have been published in top economics journals, when restricting the sample to only those studies with adequate power to identify small treatment effects, and when employing an econometric method to account for the possibility of publication selection bias favouring the publication of statistically significant results.
Relevance: Assessing the impact and value of financial education is always relevant and topical.
Generalisability/ transferability: The finding that financial education has a positive effect on financial knowledge and behaviour can be applied in any setting, given that the studies come from multiple countries and cover many types of intervention. However the details around effect size and cost effectiveness are limited to this study.
- The authors aim the study at those developing financial education programmes. It would also be useful for those looking at the costs and economic implications of financial education.