Just do what you did last month
If you are a professional investor covering US or European stocks, you are about to come to the end of another busy earnings season. If you want to have a cheat sheet for the next couple of months in the markets, so you can relax a bit, today’s post may be for you.
Jessica Wachter from UPenn and Hongye Guo from the University of Hong Kong found that US and global stock market returns, as well as industry returns, are predictable around the quarterly earnings season.
Obviously, during earnings season, professional investors are overwhelmed with information and news about companies. In today’s markets, this information overload is a real problem because it is physically impossible to follow the news on every company in a diversified portfolio (if you have a very concentrated portfolio, that may not be a problem). Hence, we can’t really digest all the information companies are putting out, which gives rise to post-announcement earnings drift.
But beyond that, there seems to be another effect. When a company is in the news in the month after their earnings results, it tends to be the same story as during the results announcement. There isn’t enough time for things to materially change. But investors still seem to be surprised by this news, and share prices react to the ‘news’.
The result is that there is a statistically significant positive correlation between the market return in the first month of a quarter and the second month of a quarter. The chart below shows that on average, for every percentage point of excess performance by the US market in the first month of a quarter (in our case, April), the market shows an excess performance of 0.28% in the second month (i.e., May). So, by just following last month’s price action, you got an inbuilt advantage in the second month of a quarter.
By the way, there is also a reversal effect, which is only statistically significant in the US, but not globally. In the first month of each quarter, stock market excess returns tend to reverse the excess return of the previous month (the third month of the quarter). This may simply reflect the fact that with every new earnings release, there tends to be enough new information to override the trend of the previous quarter.
Second month of the quarter correlates with the first month
Source: Guo and Wachter (2025)
Really interesting paper and the framing helps explain something I've been thinking about for a while. What this is actually measuring isnt a price pattern, its a narrative half-life. Earnings releases create a story about a company and that story has roughly a 60 day lifespan before the next set of results overwrites it. In month one the story is being formed. In month two its being reinforced because analysts, journalists and investors are all anchored to the same narrative and every subsequent data point gets interpreted through that lens. The 0.28% coefficient is essentially the market's narrative persistence rate expressed as a return.
The reversal in the first month of the next quarter is the more revealing finding though because it tells you the old trend doesnt decay gradually, it gets replaced. New earnings arrive, a new narrative forms, and the previous story dies almost immediately. That means the market isnt processing information continuously the way efficent market theory assumes. Its processing it in quarterly pulses with predictable reinforcement and replacement cycles. The market is a step function synchronised to the earnings calendar not a continuous one.
This has an implication for volatility that I think is underexplored. If returns follow this quarterly narrative cycle then realised vol should be structurally higher in month one of each quarter when narratives are being contested and structuraly lower in month two when they're being reinforced. Would be curious whether the VIX term structure around earnings season confirms that, because if it does then the Guo and Wachter finding isnt just an alpha signal, its a volatility regime map.
The chart also shows that the effect was weaker in 1974-2023 as compared to 1926-1973. The markets became more efficient. Unfortunately we dont know what it will look like for 2024-2073.