research

Financial Literacy and Vulnerability: Lessons from Actual Investment Decisions

A TFI Research Challenge project by Milo Bianchi
Posted on March 23, 2018

The research project "Financial Literacy and Vulnerability: Lessons from Actual Investment Decisions" is one of the eight projects selected last year for the TFI Research Challenge.

In this report, Dr Milo Bianchi combines data on household portfolios with measures of financial literacy to explain why less literate households experience lower risk-adjusted returns. This study provides insights for a better understanding of how much financial literacy can help households to make better investment choices. 

Summary

Understanding financial vulnerability is a topic of increasing interest for academics and policy makers. Households are required to take more and more (possibly complex) decisions in various financial domains, which result in an increasing need for sophisticated tools to make efficient investment choices. Not all households are well equipped. It is now established that household investment portfolios vary considerably  and some portfolios appear inefficient. It is also well known that households vary substantially in their financial literacy levels; some households seem unfamiliar with basic financial principles.

Understanding whether and how less sophisticated households − those with lower levels of advanced financial literacy − tend to hold suboptimal portfolios would allow for a better assessment of  whether households are truly making mistakes. Moreover, if mistakes are being made, it is important to get a sense of how costly they are and of how much financial literacy can help. This would inform the wider debate on whether it is worth investing in financial education.  

The mechanisms behind the relationship between lack of financial sophistication and investment performance are not well understood. Part of the challenge is empirical. It is difficult to find data that combine detailed information on household portfolios with measures of household sophistication. Administrative data from financial firms typically lack direct measures of financial sophistication. Survey data typically lack the details and the panel structure necessary to observe the portfolio choices of the same individuals over time. An important dimension of heterogeneity may arise from how households rebalance their portfolios over time in response to market conditions or to their own returns.

In this report, the author reveals several novel patterns that contribute to a better understanding of the relationship between financial literacy and suboptimal investment choices.

This report summarizes the results of a study in which I was able to exploit a financial firm’s panel data on portfolio choices matched with survey measures of financial literacy. These data are particularly interesting for these purposes. First, they provide a very rare opportunity to combine detailed information on actual investment choices with survey measures on investors' preferences, beliefs and sophistication. This allows for the assessment of the effects of financial sophistication controlling for other possibly confounding behavioral factors. Second, they are representative of the (French) population and they focus on a widespread investment product, called assurance vie, which typically represents a substantial fraction of investors' financial wealth. Households' main choice is how to allocate their wealth between relatively safe and relatively risky funds, and how to change their allocations over time. Third, these data follow investors over time. We record clients' portfolio of contracts at a monthly frequency from September 2002 to April 2011.

I document several novel patterns that contribute to a better understanding of the relationship between financial literacy and suboptimal investment choices. I first classify households on an index of financial literacy, based on their ability to correctly answer basic questions on household finance. The measure covers various aspects of financial sophistication: ability to compute compound interest, knowledge of financial products, information about market trends, math ability.

Financial literacy can then be related to demographics and to broad financial behaviors. In our sample, financial literacy appears positively correlated to education, income and wealth and it is negatively correlated with being married and female. We also observe that investors with higher financial literacy display larger stock market participation in the sense that they are more likely to hold stocks (either directly or indirectly) and to hold a larger variety of financial products. These correlations are consistent with other findings in the literature.

Turning to our portfolio data, we start by showing that, even controlling for various demographic and portfolio characteristics, less sophisticated households experience lower risk-adjusted returns in their investments. Specifically, controlling for various measures of portfolio risk, those with the highest level of financial literacy experience about 0.4% higher returns than those with the lowest level of literacy, relative to an average return of 4.3%.

Our key contribution is to show that this relationship between low literacy and low returns can only be explained by looking at specific portfolio rebalancing behaviors, as opposed to standard cross-sectional variations in stock market participation or risk taking. In particular, I show that less sophisticated investors hold a lower exposure to risk (a lower fraction of risky assets in their portfolio) precisely at times when risky funds are expected to offer higher returns. These results suggest that one way in which more sophisticated households experience higher returns is by holding a larger exposure to risk when market returns are higher.

I also show that less literate households are less likely to act as contrarian. Rather, they move their wealth from funds that have experienced relatively lower returns to funds which have experienced relatively higher returns in the recent past, hence chasing trends. They are less likely to keep their risky share relatively constant over time and they are more likely to display portfolio inertia, implying that their actual risk exposure is further away from their desired risk exposure.

Less sophisticated investors hold a lower exposure to risk precisely at times when risky funds are expected to offer higher returns.

These results suggest that rebalancing behaviors are an important determinant of portfolio returns. These behaviors are key to understanding their lower performance and they suggest that less sophisticated investors naively try to time the market. Moreover, the returns experienced by less sophisticated households tend to fall short of those that they would have earned without rebalancing their portfolios.

Our results have clear implications for financial institutions. European regulation, for example, requires financial institutions to gather information about their clients' objectives and preferences before selling them financial products. These results show that clients' financial literacy should be carefully taken into account when advising individual investors. Financial literacy can be measured in a meaningful way. Together with a series of other papers, the present work provides a simple method to elicit investors' degree of financial sophistication. Moreover, it shows that lack of financial literacy can make investors exposed to significant mistakes. The extent of these mistakes is severely under-estimated in traditional studies focusing only on static choices and on cross-sectional variations in stock market participation or risk taking. These mistakes are predictable and sometimes large, and they seem of first order importance when assessing households' financial vulnerability.   

Dr Milo Bianchi is an Associate professor at Toulouse School of Economics, TSM. University of Toulouse Capitole, Toulouse, France. E-mail: milo.bianchi@tse-fr.eu