End-of-the-year economic growth and time-varying expected returns

We show that macroeconomic growth at the end of the year (fourth-quarter or December) strongly predicts the returns of the aggregate market, small- and large-cap stocks, portfolios sorted on book-to-market and dividend yields, bond returns, and international stock returns, whereas economic growth during the rest of the year does not predict returns. End-of-the-year economic growth rates contain considerably more information about expected returns than standard variables used to predict returns, are robust to the choice of macro variables, and work in-sample, out-of-sample, and in subsamples. To explain these results, we show as the second main fi?nding of our paper that economic growth and growth in economic confidence (consumer con?dence and business con?dence) are strongly correlated during the fourth quarter, but not during the other quarters. In summary, we therefore show that when economic growth is low at the end of the year, confi?dence in the economy is also low such that investors require higher future returns. During the rest of the year, there are no such relations between growth, confi?dence, and returns.