Our first guest article, by Julien Messias of Uncia Asset Management, takes a break from covering high-level trends in asset management & equity research to focus on a rather specific topic: portfolio management around earnings release periods. Such periods can be defined by the fact that it is impossible to forecast market reaction.
By focusing on market reaction to the release rather than the fundamental earnings data itself, it is possible to generate profits from this period of inefficiency.
EARNINGS RELEASES – A LOTTERY
As difficult as it may be for our equity analyst friends to admit (editor's note: the author is an asset manager), all available empirical data shows that it is impossible to predict market reaction following an earnings release. We thus need to distinguish the fundamental component of the reaction which is less unpredictable (related to turnover, EBITDA and other hard data) from the "price signal" component. The latter has always been impossible to predict, even if we take into account the released fundamental data.
From a statistical perspective, the specific movement linked an the earnings release is on average null, as can be seen from the highly leptokurtic distribution of the movement. For an asset manager seeking to optimize their Sharpe Ratio, it is therefore not worth maintaining a position in the equity over the release period (assuming transaction & liquidity fees to be marginal)
Consider the following (translated) quote from the book L'économie des Extrêmes:
"Admitting that we had as many up days as down days on the New York Stock Exchange (NYSE) between 1983 and 1992, a hypothesis often verifiable on the other stock exchanges, including Paris, the main rises of the index occurred during roughly 3% of the business days, or 8 days per year. … For the practitioner, this phenomenon is an anxiety factor, because it means that a few extreme daily variations count for far more towards the yearly performance than the very numerous small variations." – Daniel Zajdenweber 
In other words, the performance of an asset manager or a financial asset over the year is essentially driven by 8–10 key trading days, the remaining trading days of the year accounting for very little. When faced with such periods of high uncertainty, the manager must therefore maintain a very low profile. But one may wonder–how then do we turn earnings release periods from a source of risk into a source of alpha?
POST EARNINGS ANNOUNCEMENT DRIFT
PEAD is a behavioural bias studied since 1968 in the original article by Ball & Brown, An Empirical Evaluation of Accounting Income Numbers. In the nearly 50 years that have followed, many researchers have studied this effect, especially in US markets. Although many interpretations exist, the underlying philosophy is quite simple, yet counterintuitive. The method is to buy the stock after a spike in earnings, or sell the stock after a dip in earnings, then follow the trend by maintaining the position over a period of weeks or months. This is the opposite of the "intuitive" approach to managing a portfolio through the upsides, which remains largely a mean-reverting approach. Hence executing the PEAD strategy requires a lot of discipline because it runs counter to what seems natural, and investors are only human beings. Many papers by Bouchaud, Potters et al. (e.g., Two Centuries of Trend-Following) make the case for this approach.
The success of the PEAD strategy can be explained by the time required by the market influencers to digest and interpret the earnings information post-release. Firstly, the larger financial institutions often have a very strict investment process in-house. This significantly increases their reaction time (e.g., the investment committee must first be convened) Secondly, let us look at equity analysts, who in some cases are known to be "market movers" and have a significant impact on many stocks they cover. This is especially true for the analysts from the bulge bracket banks.
The Equity Analysts' Lag is a well-known phenomenon that has been empirically demonstrated to exist. When significant new information is revealed, causing a very sharp move of the underlying asset, the analyst first has to re-examine the inputs of their model, then write down or update the investment case, and finally release the update. All of this naturally does not happen instantaneously. This is why the impact of analysts on stock prices generally only occurs some weeks after the release of new information.
Furthermore, in the event of a very sharp upwards move, analysts rarely downgrade the stock, or rarely upgrade following a sharp downwards move. This is because analysts usually forecast the continuation of the movement and feed the trend-following aspect of the market reaction.
The lag therefore explains why the PEAD Strategy works, though it is actually quite complex to put into practice. It needs very rigorous follow-up, and perfectly fits the so-called "momentum" universes, such as the NASDAQ 100 & NASDAQ Composite in the US. Whereas in Europe, this strategy is very difficult to implement because earnings information is released in two stages– first the "Earnings Release" before market opening, covering the past accounting data, and then the "Earnings Call" during market hours, when company management discloses its guidance, i.e. future information. It is thus very common to observe very large market moves from the Release to after the Call. In the US, the information is released simultaneously, with both the Release and the Call occuring after market close with a 30-minute lag. The next trading day's opening price can therefore take both into account.
The momentum investment universe is one of the main fields of expertise here at Uncia Asset Management, the french "high growth" specialist. This approach to managing assets is very different from the traditional value approach. In fact, applying the value recipes to the momentum universe may lead to catastrophic results. In addition to the fundamental advantages of our rigorous stock-picking process – that remains our DNA – we also optimize our market entry & exit for picked stocks by taking the PEAD effect into account. This is the result of our deep research into the 4 main US indexes – Standard & Poor's 500, Russell 2000, NASDAQ 100, and NASDAQ Composite. The latter two are the best proxies for our style. Our research study ran from 2003 to 2015 and therefore covered several different market conditions. The principal conclusion is that the strategy is robust with respect to the various checks performed, the underlying statistical methodology, and the sample used (more than 200 000 earnings release publications)
For instance, the historical composition of US indexes has been used to avoid any survivorship or birth biases and liquidity related issues in the sample used (stocks with market caps over 1 billion USD) The PEAD produces abnormal statistically positive returns, even if the positive signals appear more robust than the negative signals (see the charts below for the NASDAQ Composite & NASDAQ 100)
Systematic use of this strategy can be a predictor of the corresponding index. Investor over-reaction after results are published is an indicator of their mood and risk aversion. If investors over-react more on good news rather than on bad news, we could conclude that their mood is more optimistic than pessimistic. The strategy will therefore create an implied net positive exposure.
The chart below shows that when applied to the NASDAQ Composite, the non-market neutral strategy (with a net exposure related to signals from the PEAD) significantly outperforms the market neutral strategy, particularly in 2008 (sharp drop of the market) and 2013 (strong rise of the market).
To summarize, many investors are skittish and reticent to invest in the growth style, given the outperformance of the value style as illustrated by Fama and French in 1997 in their famous paper, Value vs Growth: The International Evidence. However, according to Sebastian Lancetti of UBS, this underperformance of the growth style over a long period of time is the direct consequence of the investor’s overreaction to negative releases for growth shares which is very significant, whereas value stocks do not suffer as much.
Controlling for this effect, both styles demonstrate the same performance over long periods of time.
Welcome in the PEAD and its Smart Earnings universe!