How to trade the ‘earnings drift’ effect

Are investors good at ‘pricing in’ information in real time? Read any news article commenting on the latest move in a stock, bond or commodity, and the answer would appear to be ‘yes’; something new happens or is published, and the balance of risk and reward is repriced, just as believers of the efficient market hypothesis would suggest.

However, although we often assume that in large markets the dazzlingly wide array of buyers and sellers – both human-directed and automated – is good at interpreting data and applying it to the expected return from an asset, we also instinctively know that things are a bit messier.

That's because the array of buyers and sellers might not be quite so dazzling. Data might be hard to interpret, take a while to interpret, or lack the requisite information to correctly re-price expectations about the future. What’s more, there is memory, anchoring and human bias to deal with. New information should change our views (both individual and collective), but it also must get past myriad pre-conceptions about its validity, and how it might apply to a given situation, company or sector.

Half a year ago, we debuted a new screen – since named the Double Up – which seeks to exploit a quirk in the efficiency with which investors ‘price’ information, and a curious market phenomenon known as the ‘earnings drift’ effect.

Its design is loosely based on a 2023 paper by Tobias Kalsbach and Steffen Windmüller of the Technical University of Munich. In seeking to explain why markets can be slow to digest news with positive implications for earnings, the academics found that this underreaction could be stoked when it involved stocks trading near their 52-week highs. The study’s findings suggest that rather than buy into positive earnings surprises, traders should look for situations where positive earnings surprises combine with price highs to better capture the ‘drift’ that follows.

The screen, which is applied to both the FTSE All-Share and Aim constituent lists, works off two data sources familiar to trawlers of our Ideas Farm section: stocks at or near their one-year highs, and stocks with the largest one-month bumps in broker earnings forecasts. By hunting for names that score highly on both measures, the Double Up restricts itself to a pair of rules:

• A share price within 5 per cent of the 52-week high.
• An upgrade in the next 12 months’ earnings per share forecast of at least 3 per cent, over the past month.

Three months ago, assuming that these two datapoints serve as some kind of proxy to the TUM paper, I guessed that quarterly refreshes would be the best way to exploit the earnings drift. This was based on two further assumptions. First, that when it comes to UK shares, momentum is most powerful over shorter durations (as the evidence from our quarterly momentum screen currently suggests). Second, that the earnings drift effect will eventually decay over longer time horizons.

One quarter further on in this experiment, it's too early to tell whether these hunches are accurate. But judging by the performance of both January’s Double Up selections and those of the previous quarter, there is conflicting evidence that a three-month window is best.

What we can say, regardless of the window, is that Double Up stocks are doing better than the indices from which they are drawn. Of the 15 that made the grade last time around, two-thirds finished in positive territory, while the average total return was 4.9 per cent, one percentage point ahead of a straight split of the All-Share and Aim indices.

However, our returns would have been a fair bit higher had we simply stuck with the nine stocks selected in October, which raced ahead of their respective benchmarks and saw positive total returns in all but one instance.

Should we switch to a biannual rotation, then? Given the lion’s share of the subsequent rise in the October cohort came from two names – oil giants Shell (SHEL) and BP (BP.) – I’m not entirely convinced. Though we might concede that the gradual upward trend in oil and gas prices was still poorly reflected in their January share prices, especially amid clearly escalating geopolitical tensions,  the ability of analysts (or anyone for that matter) to accurately call the trajectory of the world’s primary energy commodities is notoriously weak.

However, this being a largely theoretical exercise rather than a deeply thought-through form of portfolio construction, I’m going to keep a watch on how the past few quarters’ selections ultimately play out. After all, for now, it shouldn’t be too tricky to keep tabs on each.

This month, following another relatively bullish period for UK equities (not that you’d know it from recent market jitters and the general narrative of gloom), the screen has identified a wider range of stocks where the anchoring effect is holding back investors’ optimism. I’ve limited them to 30 names (eight of which hail from the junior market), which can be found in the table below and in more detail in a downloadable spreadsheet in the online version of this article.