Can you Help Someone Become Financially Capable? A Meta-analysis of the Literature

Main findings

Headline Findings: a summary statement

Financial education is shown to have a positive impact on savings behaviours and may have a positive impact on record keeping—but not on retirement savings or loan default rates.

Evidence Base

This review includes a total of 188 papers, including 140 from the USA, 12 from sub-Saharan Africa, 12 from South Asia, 12 from Central and South America, seven from South-East Asia and Australia, four from Europe, and one from the Middle East. Of these, 43 percent assessed financial-education interventions providing training on a combination of financial topics, 30 percent looked at interventions providing information on savings/retirement, 9 percent were on credit counselling, 7 percent were on business management, 5 percent were on mortgages and 6 percent were on other financial issues. Around 29 percent of included studies were randomised controlled trials (RCTs), while the rest were either natural experiments or non-experimental regression analyses. The majority of interventions (54 percent) involved taught courses in a classroom or workplace, while 9 percent involved individual counselling and 5 percent were delivered through mass media. The findings of the review are based on 19 of the 188 studies that were included in the meta-analysis.

Implications for policy and practice

• A meta-analysis of six studies (all RCTs) from low- and middle-income countries (LMICs) examining the impact of financial education on savings behaviour indicated a marginally statistically significant positive impact of financial education on the probability of saving (Effect Size [ES]: 0.03, 95 percent confidence interval [CI]: 0.00, 0.06).
• A meta-analysis of seven studies from both LMICs and developed countries examining the impact of financial education on record keeping (both personal finance and microenterprise or business records) indicated a positive and statistically significant impact on the probability of record-keeping behaviours (ES: 0.23, CI: 0.10, 0.37). A meta-regression analysis found that the length of exposure to the intervention was positively associated with improved record-keeping behaviour.
• A meta-analysis of six studies from the USA indicated a positive but not statistically significant impact of financial education on the probability of retirement-savings behaviours (ES: 0.10, CI: −0.00, 0.20)
• A meta-analysis of five studies from both developed countries and LMICs indicated no effect of financial education on the probability of loan defaults (ES: 0.01, CI: −0.08, 0.09)
• Meta-regression analysis of the impact of financial education on the outcomes of interest did not produce statistically significant results.

Implications for further research

The authors call for further research to assess the use of outreach to support increased savings and also assess the costs and benefits of financial-education interventions. Further research should also employ rigorous evaluation methods, particularly randomised controlled trials, and report complete statistical data. To make comparison of studies easier, common terms to define outcomes variables should be developed, along with measurement criteria. The authors also call for on-going research to be registered on evaluation registries such as those of the American Economic Association or 3ie.

Background

The global financial crisis of 2008 highlighted the importance of general financial knowledge among consumers, with studies showing that consumers in the USA did not fully understand the subprime housing market. Further research has confirmed that the general public has difficulty understanding basic financial concepts. Financial knowledge is important not only for avoiding risks, but also for financial gains—for example, when opening bank accounts, borrowing money, buying a house or planning retirement. In a recent global poll of country officials and financial experts, respondents cited financial education as the most effective policy to improve access to finance for low-income borrowers. Financial literacy is now a top policy priority for governments, international organisations and private financial institutions. This paper looks at evidence on the impact of financial-education interventions on consumer behaviour.

Research objectives

The authors aimed to assess the effect of financial-education interventions on financial knowledge and financial behaviours.

Methodology

The authors included papers evaluating the impact of interventions designed to strengthen the financial knowledge, attitudes and/or behaviours of individuals. They included studies discussing any programme, educational outreach, media intervention or other type of communication or training for consumers. The authors searched EconLit for peer-reviewed papers published in journals between January 2000 and September 2013, literature reviews published from 2007 onwards, recent World Bank studies and the following websites: OECD, World Bank, Global Partnership for Financial Inclusion and Alliance for Financial Inclusion (AFI). The authors used a random-effects model to undertake meta-analysis of included studies.

Quality assessment

The review has identified a large volume of relevant evidence and makes a good attempt at providing key relevant information about these studies and meta-analysing the results. However, the review has several major limitations. The search strategy is not sufficiently comprehensive – the authors search relevant limited number of databases and have inconsistent rules about the inclusion of grey literature studies. As a result, the review may have missed potentially relevant studies. The authors do not explicitly state all inclusion criteria and do not report procedures for selecting articles for inclusion, so it is not clear that they avoided bias. The review does not sufficiently describe the types of evidence included in the review and does not assess the risk of bias in the included studies. This is of particular concern, as the review includes a number of study designs with potentially high risk of bias in attributing effects to the intervention. Moreover, the authors do not report how they calculated the effect sizes, making it difficult to interpret their findings. The review includes 188 studies but includes only about 10 percent of these in the meta-analysis, even though it would have been possible to include a larger number of studies in the meta-analysis by standardising effect sizes and meta-analysing categories of interventions and outcomes with less than five studies. The authors do not use narrative approaches to synthesise the findings of the included studies that they excluded from the meta-analyses, potentially missing a large amount of relevant information that could inform policymaking.

Source link

http://www-wds.worldbank.org/external/default/WDSContentServer/IW3P/IB/2014/01/16/000158349_20140116085445/Rendered/PDF/WPS6745.pdf