We’ve become fans of talking about financial literacy here at the Financial Access Initiative, so I was excited to see the May issue of the American Economic Review with selected papers from the American Economic Association’s Annual Meeting. There are four papers that address financial-literacy questions. I’ll offer some big-picture thoughts further down, but first, the punch lines of the four papers:
- In “How Financial Literacy Affects Household Wealth Accumulation,” Jere R. Behrman, Olivia S. Mitchell, Cindy K. Soo, and David Bravo use data from the Chilean Social Protection Survey to see if financial literacy affects a person’s wealth holdings, including total net wealth, pension wealth, and net housing wealth. The researchers define financial literacy as an ability to correctly answer questions on topics such as compound interest, risk diversification, and the Chilean pension system. They use a statistical manipulation (an instrumental variables approach, for the econometricians in the crowd) and find that financial literacy leads to greater wealth. When the researchers control for years of schooling, the effect of financial literacy drops by about half. Part of the “financial literacy” effect comes from good, old-fashioned education—but a big chunk of the effect remains attributable to financial literacy itself.
- In “Financial Education and Timely Decision Support: Lessons from Junior Achievement,” Bruce Ian Carlin and David T. Robinson examine how advice in the moment of financial decision-making changes behavior for people who have had financial education and for those who haven’t. The researchers study the decisions of Los Angeles high-school students participating in a Junior Achievement budget simulation. Some had gone through a financial literacy curriculum that taught lessons such as “be wary of the costs of credit” and “plan for the future,” while others hadn’t. Additionally, in the simulation, certain scenarios came with specific tips—e.g., amortizing loans more quickly helps avoid higher lifetime interest costs. Such advice had more of an effect on students who had received the classroom financial literacy training. In general, those students were more likely to delay gratification to maximize overall wealth. Yet interestingly, sometimes the financially educated students deployed heuristics intended for other sorts of situations—and actually made decisions the researchers considered less-than-optimal.
- In “Financial Knowledge and Financial Literacy at the Household Level,” Alan L. Gustman, Thomas L. Steinmeier, and Nahid Tabatabai start with a well-established connection between numeracy—e.g., being able to calculate fractions, percentages, and compound interest—and wealth, and test whether financial knowledge sits in the middle of the relationship. They do this with data from the Health and Retirement Study, which captures individuals’ numeracy, wealth, and knowledge of financial topics such as pensions and Social Security. To their surprise, the researchers find no intermediary role of financial knowledge. People who are numerate do not necessarily have a better understanding of their pensions or Social Security. The relationship between numeracy and wealth is also not affected by such financial knowledge. The researchers caution that simply improving people’s numerical abilities may not lead to greater wealth accumulation.
- Finally, in “Effectiveness of Employer-Provided Financial Information: Hiring to Retiring,” Robert L. Clark, Melinda Sandler Morrill, and Steven G. Allen describe how they worked with a number of companies to see if providing information about things such as Social Security and pension plans could change employee behavior. In one controlled experiment, the researchers distributed flyers explaining how small 401 (k) contributions might add up over a long career. While there was no broad-based difference in retirement savings rates between workers who did and did not receive flyers, the researchers were able to find an effect among younger employees (those aged 18 to 24), who tended to save more after receiving the flyer. At another company, the researchers tested to see if workers remembered the lessons of retirement planning seminars a year later, and found that participants retained much of what they learned on topics such as pensions, Social Security, and Medicare.
Taken as a group, these papers do some things quite nicely. Even careful evaluations of financial literacy programs often treat the process of moving from knowledge to action as a black box. These papers strive to be clearer about causal chains—whether that means disaggregating financial literacy from education (Behrman et al.), financial decision-making from numeracy (Gustman et al.), or financial knowledge from a supportive environment for making decisions (Carlin et al. and Clark et al.). The paper by Carlin and Robinson is particularly noteworthy in this regard. By considering the broader context in which people make financial decisions—sometimes you get useful advice from an insurance agent and sometimes you don’t—the researchers are able to reach more nuanced conclusions about the efficacy of financial literacy education and to make more robust policy recommendations.
What these papers generally fail to do is present rigorous thinking about why certain financial outcomes are the “right” ones. For instance, more retirement saving is almost always taken to be axiomatically better than less. But people lead complicated financial lives and can easily have resources—such as a house they will inherit, or family members who will care for them in old age—that are invisible to researchers. Even more importantly, these four papers, and the research tradition from which they follow, are stingy on talking to people about what financial outcomes they would like to see in their lives. Perhaps saving for college is more important than saving for retirement; maybe an investment with a predictable pay-off is more desirable than an investment with a potentially large one. When researchers work with pre-existing data sets, as some here do, it is largely infeasible to get inside the heads of respondents in this way, although there are still plenty of chances to make fewer normative claims, such as Gustman et al.’s statement that people would be better off if they knew more about their retirement savings so that they could rebalance their portfolios more often. Many financial advisors would disagree.
To end on a note of self-promotion: a better understanding of individuals’ financial aspirations is exactly one of the insights we are hoping to gain through the U.S. Financial Diaries. That project revolves around the collection of incredibly rich real-time cash-flow data, but we’ll also ask respondents about their plans for the future, as well as the relative importance of their various financial goals. Hopefully, that data will help inform future conversations about why financial literacy education is important in the first place.