For a few months now we have been exploring some of the theories and strategies behind momentum investing. For a concept that only started to receive rigorous academic testing in the 1990s, momentum has emerged as a major source of superior returns, with reams of research to back it up, providing intellectual support for those hardened TA investors who've always maintained that there's something in those price charts after all. But while the buy-high-sell-higher phenomenon is now widely acknowledged, the reason why momentum even exists – and why you can take advantage of it – remains less well understood.
When US academics Jegadeesh and Titman revealed to the investing world their backtested theories about momentum in 1993, they acknowledged that the reasons why the phenomenon existed were unclear. Their findings showed that that stocks that perform the best over a three- to 12-month period tend to continue to perform well over the subsequent three to 12 months. To explain this, they only got as far as suggesting that investor expectations are systematically biased and that under-reaction to news about a stock was likely to be playing a role. Since then, investing professionals and academics have gone to town on trying to figure out what really causes momentum.
It could be down to risk…
Broadly speaking there are two camps on what the cause of the momentum effect really is; the first of which argues that investors balance the risk of a momentum strategy by demanding higher returns. The second puts much more emphasis on the behavioural instincts of investors for influencing the phenomenon.
In terms of risk-based explanations, research is comparatively thin on the ground. It could be the case that momentum returns are just compensation to investors for some unique risk associated with momentum investing but, if so, it's hard to fathom what this risk is. While momentum strategies are known to infrequently ‘crash’ when markets suddenly turn, momentum stocks typically display high Sharpe ratios, which means that they produce higher risk-adjusted returns compared to, say, growth stocks. In Kent Daniel’s 2011 research Momentum Crashes, he concluded that: “By virtue of the high Sharpe-ratios associated with the momentum effect, they are difficult to explain within the standard rational-expectations asset pricing framework.” A subsequent paper by Rachwalski and Wen in 2012 entitled Momentum, Risk, and Underreaction concluded that “a purely risk-based explanation of momentum is insufficient”.
…but it’s more likely to be behavioural
A much more popular explanation is rooted in the way that investors think and behave. Between the 1960s and 1990s Professor Eugene Fama’s Efficient Market Hypothesis was widely acclaimed as the definitive explanation of how the market works and that it instantly priced securities based on the information that was known about them. Over time, Fama’s theory has been discredited, not least because evidence suggests that it can take time for markets to fully price-in the full implications of news. Indeed in 1996 Fama and French acknowledged that their three-factor risk-return model could not explain the short-term returns produced by momentum. For momentum investors, this has all sorts of implications.
We have previously discussed the concept of ‘anchoring’ both in the context of fundamental investing strategies and technical analysis. In finance psychology, anchoring occurs when an investor under-reacts to the implications of good news about a stock. According to research by Hong, Jordan and Liu, anchoring is particularly at work in what’s known as the 52-week high effect. Here, stocks that are trading close to their 52-week highs often go on to produce superior returns. This is believed to be caused by investors using the current price as an ‘anchor’ and only slowly responding to the implications of good news. In many respects, anchoring resembles Post Earnings Announcement Drift, which is a phenomenon that was identified back in the 1960s. That research showed that even after earnings are announced, a stock’s price can drift upwards on good news and downwards on bad news.
News watchers and momentum traders collide
This concept of anchoring and the slow diffusion of news has been tested in several different ways by academics. In 1999, Hong and Stein argued that momentum could be explained by seeing investors in two separate categories: ‘news watchers’ and ‘momentum traders’. News watchers trading on new fundamental information tend to react slowly as the market digests information, causing the momentum to begin. Meanwhile, the momentum traders, who invest on price movement, spot the early signs and enter the market themselves. Together, the two sides drive up the price and eventually the stock becomes overpriced and reverses back to its fundamental value. This potential explanation was put to the test in some 2005 research by US finance professor Seung-Chan Park, which ultimately concluded that anchoring was also a factor in interpreting the relationships between 50-day and 200-day moving averages. Here, Park claimed that investors anchor to the longer term average even when the 50-day price surges on good news, offering an opportunity for others to ride the price up as the new information slowly begins to get priced-in.
Painful losses and overconfidence
Beyond anchoring, another widely accepted potential driver of momentum is a phenomenon known as the disposition effect. The background to this effect is rooted in something called Prospect Theory, which was identified in 1979 by Kahneman and Tversky (One of the pair, Daniel Kahneman wrote the recent bestseller, "Thinking Fast and Slow" - highly recommended!).Their argument went that investors don’t treat stock gains and losses the same way – they feel much more pain from a percentage loss than they feel joy from the same percentage gain. The connection with momentum is that behavioural analysts believe that investors often sell stocks too quickly on good news to lock-in profits while selling too slowly on bad news in the hope that the price may rebound. US economist Toby Moskowitz has described this effect as causing “an artificial headwind”. What happens is that panic selling by investors suppresses the accurate value of the stock, while investors that desperately cling on to shares on bad news are effectively holding up a price that will inevitably fall further.
Elsewhere, other behavioural factors have been credited for causing momentum, including overconfidence and confirmation bias. Here, investors can arguably be guilty of being too confident in their information about a stock – or indeed, look for evidence in new information that confirms their existing views about the stock. Either way, their overconfidence leads to an over-reaction and the stock price runs away from fundamental value, eventually leading to a reversal. This theory was explored in a 1998 paper by Daniel, Hirshleifer and Subrahmanyam.
Jumping on the bandwagon
A final psychological driver of momentum suggested by analysts is the bandwagon effect, which unfolds when investors copy the behaviour of each other - for example, a fund manager might wory about underperforming his/her colleagues and peers and decide to ape them instead. According to AQR Capital, which has conducted substantial research into momentum strategies, the bandwagon effect can occur when short-term investors use a stock’s recent performance as a trading signal and long-term investors look to recent performance to confirm their convictions. When these two sets of investors collide, persistent price run-ups or -downs that can occur.
Taking advantage of the behaviour of others
In the acres of research carried out into momentum investing strategies, it’s not uncommon to find analysts acknowledging that no-one is entirely clear on what causes momentum. While the phenomenon has undergone exhaustive testing and guru investors have made fortunes for themselves and their clients using the strategy, it’s quite refreshing that the precise cause is not known. What analysts do agree on is that conventional risk theories that explain so much about how markets price securities generally don’t apply with momentum. Investor behaviour in one or more forms appears to be the driving force – which is good news for the investor that is prepared to put faith in a buy-high-sell-higher strategy.
Filed Under: Momentum Investing,