End-of-the-Year Economic Growth and Time-Varying Expected Returns (Digest Summary)

Most financial economists would agree that economic growth should matter for expected returns. The authors’ central hypothesis is that the relationship between economic growth and expected returns is stronger at infrequent points—for example, at the end of the year (fourth quarter). Their key point is that the growth rates of macroeconomic variables during the other quarters of the year are not significant predictors of excess returns.

The authors methodically demonstrate that end-of-year macroeconomic growth rates (growth in real consumption, real GDP, industrial production, employment, capacity utilization, real labor income, and so forth) contain a large amount of information about expected returns over the next year for stocks and bonds, in sample and out of sample, and in the United States and internationally. Overall, fourth-quarter economic growth contains more information about expected returns in subsamples since the 1940s than such commonly used information variables as the dividend-to-price ratio.

To explain their findings, the authors study the empirical relationship between the surplus consumption ratio—a theoretically well-founded measure of risk aversion—and expected returns. The empirical results show that expected returns relate to movements in the surplus consumption ratio more in the fourth quarter than during the other quarters of the year. The authors also demonstrate that fourth-quarter growth in consumer confidence, which they view as a proxy for risk aversion, is a strong predictor of returns compared with growth in confidence during the rest of the year.

The robust relationship between economic growth, risk aversion, and expected returns at the end of the year provides support for models based on infrequent portfolio adjustments, which should be of particular interest to investment practitioners. The authors’ findings highlight the special role of end-of-year economic growth for the time-series properties of asset prices, which opens up a number of possible links between finance and macroeconomics that warrant further research.

The business cycle variables used for the research are the quarterly growth rates of seasonally adjusted industrial production, real GDP, and real per capita consumption of services and nondurable goods. The authors’ source for the business cycle variables is the Federal Reserve Economic Data database from the Federal Reserve Bank of St. Louis. The first observation is from 1948, and the last observation is from the fourth quarter of 2009. For their main results, the authors use the CRSP value-weighted index, including the NYSE, AMEX, and NASDAQ, to calculate returns on stocks. Excess returns on stocks are derived by subtracting a short Treasury bill rate from stock returns. The dividend-to-price ratio and Lettau and Ludvigson’s (Journal of Finance 2001) cay^ ratio are compared with the predictive power of fourth-quarter economic growth and are sourced from CRSP and Amit Goyal’s website, respectively.

The authors use predictive regressions to provide evidence that the fourth-quarter growth rates of the studied macroeconomic variables capture movements in expected returns in the United States and internationally, across different assets, and in and out of sample. They demonstrate why the fourth-quarter growth rate is such a strong predictor by showing, via regression analysis, that the surplus consumption ratio and consumer confidence influence expected returns more strongly during the fourth quarter than in the rest of the year.

The authors’ research, which shows strong return predictability by pure macroeconomic variables when focusing on the fourth-quarter growth rate, substantially builds on the existing return-predictability literature. Investment practitioners should be particularly interested in the conclusions that negative (positive) movements in economic growth during the fourth quarter strongly predict high (low) future excess returns and that growth rates during the other quarters of the year are not significant predictors of excess returns.