Media and Google: The Impact of Information Supply and Demand on Stock Returns

This paper gives the first empirical examination of the joint effect of information supply and demand on stock returns. Unlike previous studies, I examine the relationship between cross-sectional stock returns and "pairs" of shifts in information supply and demand. I use the number of news articles and Google search volume (for a company) as proxies for information supply and demand respectively. The results show that only an upward shift in both information supply and demand predicts future returns among shift pairs. A monthly rebalanced portfolio of buying stocks with this shift “pair” and short selling the other stocks generates an abnormal return between 16% and 22% per year with the Sharpe ratio between 0.85 and 0.9 (compared with a Sharpe ratio of 0.049 for the S&P500 during the same period). The abnormal return increases to between 23% and 34% per year in a subsample of small stocks. These findings imply that an econometric model with only information supply (demand) under-estimates the price impact of an increase in information supply (demand) when information demand (supply) increases, and overestimates the impact when information demand (supply) decreases. The results are consistent with the hypothesis that an increase in information supply drives stock prices up only when an increase in information demand confirms that information supply succeeds in raising new investors’ awareness and existing investors’ additional learning effort. The good news due to ostrich effect can also partially explain the abnormal return, but it only explains a small component of the total abnormal return. The empirical findings affirm the importance of incorporating both information supply and demand in predicting stock returns.