In the days around earnings announcements, stock prices usually rise.
It has long been observed that when firms announce their quarterly earnings, as they are required to do, considerable price volatility and increases in trading volume are evident. In addition, in the days around earnings announcements, stock prices usually rise. In The Earnings Announcement Premium and Trading Volume (NBER Working Paper No.13090 ), Owen Lamont and Andrea Frazzini explore why these phenomena occur. They hypothesize that the predictable rise in stock prices is driven by the predictable rise in volume generated by earnings announcements. They go on to show that the premium is strongly correlated with the concentration of trading activity around previous earnings announcements, and that stocks with high volume around earnings announcements in particular subsequently have both high premiums and high imputed buying by individual investors. This suggests that, at least for some stocks, prices are boosted around announcement dates by demand from individual buyers
In general, of course, stocks tend to rise on high volume and to decline on low volume, but Lamont and Frazzini say that whether this happens because of the interpretation of the announcements or because of irrational or random traders is uncertain. What may well be in play is that certain earnings announcements simply "grab attention," with the result that individual investors are motivated to buy in.
The researchers focus on this "attention-grabbing" hypothesis, because stocks that make news - whether good, bad, or neutral -- have both high volume and high net buying by individuals. Lamont and Frazzini note that arbitrageurs might be expected to eliminate this anomaly, but this would require substantially increased trading activity, which is costly. In addition, the highly idiosyncratic volatility around earnings announcements could deter traders who, for whatever reason, cannot sufficiently diversify. If idiosyncratic risk is somehow preventing arbitrage activity, then in this limited sense the premium may be viewed as a reward for bearing risk. Lamont and Frazzini see evidence that arbitrageurs in fact do trade to eliminate the premium. Prior to the announcement, there are high, imputed buys from large investors. This suggests that arbitrageurs are trading on the anomaly, but simply have not yet eliminated it. Whatever the case, because earnings announcements occur frequently and regularly and generate substantial volume, they provide a good opportunity for testing whether volume drives returns, and especially whether predictable volume generates predictable returns.
Lamont and Frazzini correlate earnings announcement dates, as compiled by Compustat, with data on all common stocks recorded by the Center for Research in Security Prices between January 1972 and December 2004. (Data are incomplete for some stocks in some years.) Lamont and Frazzini consider monthly data and expected announcement months in order to have wide windows on buying activity. This also means that issues related to the timing of the announcements or changes in liquidity around the announcement dates are unlikely to be of significance in the analysis.
The researchers demonstrate that the strategy of buying every stock expected to announce within the coming month and shorting every stock not expected to announce yields a return of over 60 basis points per month. The announcement premium is thus substantial, particularly among large cap securities, lasts about four weeks, and is evident in samples going back to 1927. At the same time, stocks with the largest predicted volume increases in announcement months, as forecast by a high concentration of past trading activity around earning announcements, tend to have higher subsequent premiums. These stocks also tend to have the highest imputed buying around announcement dates by small investors.
Lamont and Frazzini add that the evidence increasingly shows that individual investors seem to make uninformed trading decisions. In line with the attention-grabbing hypothesis, whereby individual investors are likely to buy stocks that seize their attention on the strength of earnings announcements, small-investor buys (as proxied by small buyer-initiated trades) soar on announcement day. This is especially so for securities where most of the past trading activity was concentrated around announcements. One explanation for these phenomena is that some securities attract small attention-constrained investors around earnings announcement dates. Since such investors rarely sell short, the predictable rise in volume boosts prices around announcement dates, thus generating a seasonal component in the stock's expected return.
These results fit with the broader research on the connection between trading activity and prices. Elements such as liquidity, information flow, heterogeneous beliefs, and short sale constraints arguably are all important in understanding this connection. Lamont and Frazzini's findings impose an additional requirement on any theory attempting to connect volume and prices. Any hypothesis, the researchers assert, must now explain why highly predictable volume leads to highly predictable returns. Their likely explanation is uninformed or irrational demand by individual investors, coupled with imperfect arbitrage by informed traders.
-- Matt Nesvisky