Retail investors and earnings surprises
On average, retail investors aren’t the smart money in the market. They tend to trade on noise and in some cases, they may trade against fundamental information, creating systematic market mispricing and opportunities for ‘smarter’ investors.
One such systematic opportunity that is made worse by retail investors seems to be the post earnings announcement drift. A new study by researchers from Harvard, the Chicago Fed, and Farallon Capital Management presents evidence that retail investors on average act as short-term contrarian investors for momentum and earnings surprises, thus reducing the efficiency of markets in incorporating fundamental news.
To recall, the post earnings announcement drift is the observation that earnings surprises don’t get fully incorporated into share prices on the day of the announcement. Instead, companies with positive earnings surprises tend to see their share prices drift higher in the weeks and months after the announcement especially when they are trading close to their 52-week highs. Similarly, companies with negative earnings surprises tend to see their share prices drift lower for weeks and months after the announcement as the bad news gets incorporated into the share price only slowly.
The new study confirms this effect once again and traces it back to retail investor flows. To do this, the study looked at the net retail flows relative to the previous quarter's average in and out of stocks and split them between stocks with high and low earnings surprises as well as high and low price momentum.
My first chart shows the net retail flows into stocks with different price momentum, with ‘1’ denoting the stocks with the weakest price momentum and ‘5’ denoting the stocks with the strongest price momentum. The chart on the left shows the stocks with the lowest earnings surprise while the chart on the right shows the stocks with the highest earnings surprise.
Net retail flows by earnings surprise and price momentum
Source: Luo et al. (2023). Note: Price momentum sorted from 1 = low to 5 = high. Left chart = lowest tercile by earnings surprise, middle chart = middle tercile by earnings surprise, right chart = highest tercile by earnings surprise.
Note that, on average, retail investors are buying into stocks with low price momentum and selling stocks with high price momentum, but this effect gets stronger, the larger the earnings surprise is. In other words, retail investors tend to buy predominantly stocks with low price momentum and large earnings surprises (i.e. they are hoping that the positive surprise indicates a reversal of fortunes for the company) and tend to sell stocks with strong price momentum and large earnings surprises (i.e. they sell on good news for stocks that have gone up in price a lot).
This means that shares with large earnings surprises and weak price momentum underreact to negative earnings surprises but overreact to positive earnings surprises. Meanwhile, shares with strong price momentum underreact to positive earnings surprises and overreact to negative earnings surprises.
That is exactly the opposite of what should happen in a rational market. The result is that negative earnings surprises do not get incorporated fully for underperforming stocks and neither do positive earnings surprises for outperforming stocks. In the weeks and months that follow the first price reaction, the market then ‘corrects’ this mispricing which leads to stocks with positive earnings surprises and strong price momentum drifting higher while stocks with negative earnings surprises and weak price momentum drifting lower.
That retail investors are a key driver of this post earnings announcement drift can be seen in the chart below which shows the drift for companies with high and low institutional ownership.
Stocks with low institutional ownership, where retail investors exert greater influence on the share price, show a significant downtrend on stocks with negative surprises in the days and weeks after the announcement and a small but measurable uptrend for stocks with high earnings surprises. Meanwhile, stocks with high institutional ownership show much less post earnings announcement drift. In particular, the stocks with the lowest earnings surprise and high institutional ownership show no negative post earnings announcement drift.
Level of institutional ownership and post earnings announcement drift
Source: Luo et al. (2023). Note: SUE = surprise earnings.
Stock prices are set by the last marginal seller or buyer; 2% of stocks trade each day, but the other 98% mark their portfolios to market based on the last trade. It might be interesting to somehow adjust for differences in holding periods.
I would guess that retail investors tend to hold longer. I remember someone once joking after a stock he held went down a lot "well, that 'trade' just turned into an 'investment'!". I also know from personal experience that sometimes retail trades are delayed in order to turn the capital gain or loss from short-term to long-term (in the US the latter is taxed at a much lower rate than the former), or apply tax loss carry-forwards. If a US retail investor has a $100 short-term gain, and a 45% marginal income tax bracket, if he sells right away his after-tax profit is $55. If he waits until he's held it for a year, his long-term capital gains rate is 15%, so his after-tax profit is $85. If he has a long-term tax loss carry-forward from a previous loss of $100, and applies this toward the gain, his after-tax profit is $100. I've often heard people say "I'd love to take profitsin this name, but I need to find a loser to sell first so I have an offsetting tax loss". Warren Buffett once said something along the lines of "never make any investment decision based upon tax considerations, only fundamentals", but people routinely ignore that ... including himself when it came to Apple stock.
In Germany, all interest, dividends, and capital gains are taxed at a flat 25%, so I wonder if retail traders there would show more favorable results. I was in institutional investment for almost 40 years, and never *once* heard anyone care at all about tax considerations, which might explain why they may appear more adroit by comparison. I don't necessarily think they're collectively much smarter ;-)
This last graph “Level of institutional ownership and post earnings announcement drift” suggests the following:
- Either institutional investors pick “better” stocks than retail investors, leading to higher returns,
- Or(/and) retail investors react much slower than institutional investors to earnings surprises.