PsyFi Search

Friday 9 March 2018

Cursed By Momentum

Edgy Investors

Although most investors have no edge on the market there's a proportion of them that persist in trading actively, the main effect of which is to enrich their brokers. There are various explanations of why this occurs, but it seems to come down to some combination of inherent overconfidence and a perverse refusal to take account of negative information. 

This is particularly dangerous in calm periods such as those we’ve been experiencing in markets over the past few years.  In such times momentum strategies are particularly effective – and serve to supercharge the behavior of naturally overconfident individuals. The end results are usually not pretty..

Snow Business

We’ve covered the basic research into overconfidence here a number of times. Odean and Barber (see O is for Overconfidence) showed that the more investors trade, the more they tend to lose – high turnover investors spend more on fees and end up losing money relative to less active traders. In fact only the least active accountholders were able to generate a return in excess of an S&P500 index tracker.

In 2009 Mark Grinblatt and Matti Keloharju studied a bunch of Finnish traders in Sensation Seeking, Overconfidence and Trading Activity. In Finland all males have to take an Army psychometric test at age 20 – quite what the Finnish military is worried about isn’t clear, other than a tendency for adult male Finns to drink a lot, play chess and roll about naked in the snow, often all at the same time. The researchers then matched the results of these tests and driving violations to frequency of trading and – no surprise – found that overconfident, sensation seeking individuals were the most frequent traders.

Wrong But Right

As ever, the behavior of investors is conditioned by the times that they find themselves in. People tend to extrapolate the recent past into the indefinite future, which is fine when you’re dealing in the movements of tectonic plates, but less smart if you’re trading on the edge of an active volcano.  Given this it’s no surprise that momentum strategies – essentially buying shares that have gone up in price – tend to work extremely well in low-volatility markets and can fail dramatically when volatility increases and markets go in reverse.

The fact momentum strategies work at all looks like a behavioral effect. In theory all known information should be priced into all stocks at all times, but in practice news seems to only slowly be assimilated in share prices.  Unsurprisingly, peoples’ levels of confidence tends to rise when they get good results and fall when they bad ones, but does so in an asymmetric fashion due to something called self-serving bias – essentially if things go well I’m smart, but if things go wrong I was unlucky (see: Where Two Strangers Never Meet: Self-Serving Bias). Hence investors following momentum strategies in low volatility markets will gain in confidence and will be slow to revise their high opinions of themselves when markets go into reverse.

Cursed

Erik Esyter, Matthew Rabin and Dimitri Vayanos have expanded on this idea in Financial Markets Where Traders Neglect The Informational Content of Prices to develop the concept of “cursedness”. A cursed investor is one that overestimates their own knowledge and simultaneously underestimates the knowledge of others:
“Markets comprised entirely of cursed traders generate more trade than those comprised entirely of rationals; mixed markets can generate even more trade, as rationals employ momentum-trading strategies to exploit cursed traders”
The name comes from the concept of the Winner’s Curse – the finding that the winner of a sealed bid auction will very often overpay to win, such that winning is a curse (see W is for Winner's Curse). Overconfidence and momentum effects often lead to cursed investors overpaying, while smarter ones sit on the sidelines and wait for better opportunities.

Pay Attention

To add to this melange of misbehavior there’s quite a lot of research suggesting that simple lack of investor attention is the main cause of mispricing. In essence, a lot of traders actually fail to analyse or assimilate news properly and carry on trading regardless, presumably based on nothing other than gut feel and price movements driven by other similarly inattentive investors. 

In fact, when investors do pay attention to stocks they tend to drive the price upwards. In Search of Attention, research by Zhi Da, Joseph Engelberg and Pengjie Gao looked at the correlation between Google Searches for specific stocks and the subsequent price action and found that the prices spiked up in the short-term and then fell back over longer periods – prima facia evidence that where attention is given it makes a difference:
“We find that an increase in SVI for Russell 3000 stocks predicts higher stock prices in the next 2 weeks and an eventual price reversal within the year. SVI also contributes to the large first-day return and long-run underperformance for a sample of IPO stocks.”
 (SVI = Search Volume Index)

Neglect is Value

This being the case you'd expect that where stocks are flying beneath the radar then news is much slower to be assimilated – which is exactly what Roger Loh found in “Investor Inattention and theUnderreaction to Stock Recommendations”:
“...the reaction to recommendation changes is much smaller for low prior turnover stocks than it is for high prior turnover stocks. Consequently, the recommendation drift is more pronounced for low turnover firms.”
Basically, hot stocks get all the attention, inattentive investors are slow to adjust to news on less popular shares.  So news only gets assimilated slowly, leading to an emerging momentum effect – which then feeds back into the behavior of overconfident individuals.

 Reversals

The question of why momentum strategies work is a long running theme in finance research, and the obvious answer - that companies that have done well in the past are good companies that will continue to do well in the future, and vice versa - appears not to be true. If that's the case the most likely alternative explanation - that it's down to investor irrationality - implies that reversals in momentum should be expected. 

And that's what happens, a number of researchers have shown stocks that performed badly in the past are more likely to perform better over the next 3 to 5 years - see Narasimhan Jegadeesh and Sheridan Titman's paper Profitability of Momentum Strategies: An Evaluation of Alternative Explanations, for example.

The point is, of course, that momentum strategies can work if you know what you're doing. But if you don't then overconfidence generated by the simple good fortune of holding high flying companies will ensure you are well and truly cursed.

Cursedness added to the Big List of Behavioral Biases


1 comment:

  1. As far as I know the academic papers do not control for the “momentum of actual firm performance”. All stocks are sorted according to their last 12 months performance, then they rank the stocks and form portfolios of past winners and past losers. If the past winner portfolio continues to outperform over the next 12 months, this is dubbed “momentum effect”.

    The problem is that the winners may just be great businesses. For example Facebook`s earnings increased by factor 10 over the last 4 years. I mean – who would have expected that? So with every earnings surprise, the stock continues to go up (for the right reasons). With a trailing 12 month PE of 30 including stock based compensation, before taking account of cash and the tax savings this year, it is hard to argue the stock is overvalued relative to the wider market. There are a number of other large tech companies with similar developments, which just continued to surprise on the upside.

    I would go as far as saying that the complete pace of digitization since the last financial crisis was much quicker and more pronounced than the vast majority of investors had expected. Had this been anticipated, tech stocks should have been priced extremely highly already 10 years ago (and show normalized returns since then). In contrast, we observed a continuous massive outperformance of the tech sector, due to a long string of positive earnings surprises. Technically, this shows up as momentum, but it is actually the recognition of higher fundamental value due to increased earnings power.

    Of course some tech stocks appear overpriced today (but not only tech stocks). But even if, for example, Amazon and Netflix would only trade at half of today`s prices, we would read similar articles today.

    ReplyDelete