We examine a simple measure of operating leverage: the ratio of fixed costs (measured by depreciation and amortization plus selling, general, and administrative expenses) to the market (or book) value of assets. We find that this measure of operating leverage positively predicts returns. This operating leverage measure is not explained by common factors and performs better than the traditional measures of operating leverage. Furthermore, an exploratory two-factor model with the operating leverage factor works at least as well as, but does not subsume, the Fama and French five-factor model.

We propose a naïve model averaging (NMA) method that averages the OLS out-of-sample forecasts and the historical means and produces mostly positive out-of-sample R2s for the variables significant in sample in forecasting market returns. Surprisingly, more sophisticated weighting schemes that combine the predictive variable and historical mean do not consistently perform better. With unstable economic relations and a limited sample size, sophisticated methods may lead to overfitting or be subject to more estimation errors. In such situations, our simple methods may work better. Model misspecification, rather than declining return predictability, likely explains the predictive performance of the NMA method.