Measuring the Performance of Households’ Investments

Laurent Calvet, Professor of Finance - December 15th, 2009
Mesurer la performance des ménages - Calvet

Key Ideas

• On the whole, households have rational investment strategies.

• They tend to remain stable despite market fluctuations.

• An index combining three types of error can be used to rank the financial adeptness of households according to characteristics such as income or education level.

• Experience with finance itself has only a limited positive impact on the way that households manage their investments.

Laurent E. Calvet ©HEC Paris

Laurent E. Calvet is a professor of finance at EDHEC. He was research professor at HEC Paris from 2004 to 2016. A graduate of École Polytechnique, Calvet holds a Ph.D. in (...)

See CV

In a series of three articles, Laurent Calvet, John Campbell and Paolo Sodini investigate the financial “sophistication” of households. They evaluate, first of all, the nature of households' financial investment portfolios. Analyzing a Swedish database (provided by the Swedish national statistics agency, and whose utility resides in the fact that it compiles data from all Swedish households), they find that households generally hold few stocks directly, relying instead mainly on investment funds. The authors then measured the performance of households’ portfolios of liquid financial assets, which include cash (in the form of bank deposits), investment funds (similar to SICAVs or mutual funds), and directly held stocks. The authors evaluate three types of errors at the level of individual households: underdiversification of the portfolio of “cash,” stocks and investment funds, inertia in risk management, and a tendency to sell winning stocks and sell losing ones (the “disposition effect”).

THE FINANCIAL PERFORMANCE OF HOUSEHOLDS

Calvet and his co-authors observe that three factors positively influence households’ financial performance: level of education, income, and having multiple children. This better level of performance manifests itself in a more diverse investment portfolio. The researchers also studied the dynamic of household portfolios: that is, the behavior of households in the face of market fluctuations. Do they continue to pursue their initial strategy, or do they follow the market and alter their behavior? They show that households have a tendency to counterbalance the effects of the market and thus stabilize the degree of risk in their portfolio. The researchers also observed households’ tendency to sell stocks whose value is rising and to hold on to those whose value is dwindling with the fluctuation of markets. The authors establish a link between this “disposition effect” and manage­ment of portfolio risk.

RATIONAL INVESTMENTS

Through observation of error criteria, the authors show that households are relatively rational invest­ors, judging by their generally stable behavior during market fluctuations and their good strategic decision-making on investments. They add that the investments of wealthier and more educated households systematically perform better. With one exception: entrepreneurs tend to make more errors in their choices, “probably because of lack of time or their high self-confidence, which are characteristics of people who run their own businesses,” explains Calvet. While investment sophistication varies according to the three criteria listed above, the authors show that experience in finance has only a limited positive impact on the way that households manage their investments.

FURTHERING THE ANALYSIS: MEASURING FINANCIAL SOPHISTICATION

The authors also developed an index to measure the risk of errors in terms of underdiversification, inertia with risky assets, and disposition effect, attributing coefficients to each of these factors drawn from the database of the Swedish statistics institute. This index provides information about investment performance and portfolio management in relation to each of the three criteria. Thus, while the authors identified only three character­istics with a significant influence on the financial sophistication of households, other data sources could help to identify other criteria.

Based on an interview with Laurent E. Calvet, and on his article “Measuring the Financial Sophistication of Households” (American Economic Review , vol. 99, no. 2, pp. 393–398, 2009) co-authored with John Campbell, professor in the Department of Economics at the Littauer Center, Harvard University, and Paolo Sodini, professor in the Department of Finance at the Stockholm School of Economics.

APPLICATIONS
APPLICATIONS

Generating information on the financial sophistication of a client base could certainly be of interest to banks and asset management firms. This would make it possible to offer tailor-made products to new clients, as well as to recent and prospective clients. Obtaining personal information from clients on different criteria (such as income, their expected level of investment, their degree of education, etc.) could be used to establish their degree of sophistication. The company could then also use its database to revive sales campaigns and better gain from potential clients, by verifying that a client’s portfolio corresponds to their financial sophistication. Calvet adds that the tool could also be used to analyze the portfolios of French households, if the government were to make the data of the Ministry of Finance available to researchers whilst ensuring that information on individual savers remained confidential, as is the case in Sweden.

Methodology
Methodology

The authors focus on the management strategies of individual investors. They evaluate three types of errors at the individual household level: underdiversification, inertia in portfolio management, and the disposition effect. Underdiversification is measured by comparing the Sharpe ratio (i.e., the performance-risk ratio) of the portfolio of financial assets to that of a stock-market reference index. Inertia is measured by the fluctuation of the proportion of risky assets in the same portfolio. Finally, the disposition effect is measured as the difference between the proportion of stock gains versus the proportion of stock losses realized by the household on its stocks and holdings in funds.