Regional Economic Activity and Stock Returns

This study analyzes the link between regional economic conditions and stock returns. I identify all U.S. states that are economically relevant for a firm through textual analysis of annual reports and construct a novel proxy for future regional economic activity. Using this proxy, I find that forecasts on economic conditions of firm-relevant U.S. regions positively predict stock returns on a monthly basis. This finding is robust to short-term reversal, individual stock momentum, industry momentum, geographic dispersion and a list of standard controls. A zero-cost trading strategy based on this new predictive variable generates a risk-adjusted return of 5.75 (5.30) percent per year using an equal-weighted (value-weighted) portfolio. Additionally, these results indicate that information arising from all relevant states matters over and above the information content of the mere headquarter state. Further evidence indicates that forecasts of regional activity also predict firms' real operations, suggesting that economic conditions of U.S. regions capture an important cash flow component of stock returns. Finally, I show that the gradual diffusion of regional information is slower among difficult-to-arbitrage firms.