26 Jul, 2019

Investing in Equities: Lottery Effect

Imagine how much harder physics would be if electrons had emotions

Professor Richard Feynman

So why do investors buy (especially retail investors) high beta names especially with negative news around them?

This is because of a phenomenon called the Lottery Effect.

For emotionally significant events, the size of the probability simply doesn’t matter. What matters is the possibility of winning.

Daniel Kahneman, the Nobel Prize–winning psychologist

In their book Quantitative Momentum, Gray and Vogel, cite a paper “A Model of Casino Gambling” by Nick Barberis which specifically addresses why people go to casinos and how they behave there. Setting aside the behavioral utility benefits (for fun) of gambling, the key assumption in the paper is that there is something additional at play: “The inability of of human beings to properly weigh their chances of success in low probability events”. A similar point to what Kahneman makes above. Here the overoptimism of the human race, which is essential to get out of bed everyday in the face of drowning pessimism, goes against them.

In another paper “Maxing Out: Stocks as Lotteries and the Cross Section of Expected Returns” by Bali, Cakici and Whitelaw, the authors examine how this behavior plays out in stock exchanges. Their central hypothesis is that investors irrationally overpay for lottery-like gambles, assume their odds are higher than in reality,and, thus, stocks with lottery-like characteristics will under perform on a risk-adjusted basis.

Lottery bias, they go on to state may also help explain the beta anomaly I mentioned above. But why do investors get attracted to these highly volatile stocks in the first place. That is explained by the frequent news flow and event around the stock which creates an environment where the stock is highly discussed in chat groups, various Warren Buffett quotes and examples are cited. Some recent examples of these are stocks in the table below. They have been high beta stocks. And they are often in the news with respect to regulatory issues or other issues. And as their issues continue to play out in the media, retail holding in these stocks has gone up.

This is also due to anchoring bias. Many have seen these stocks earlier at much higher levels. And their system one thought process or heuristic is that these stocks will bounce back to these levels or somewhere close to them. Our experience says stocks that usually decline 70-80% (in the absence of a general market crisis like 2008 or 2000), do not bounce back. There are exceptions but the odds are not in their favor.

Low Volatility Investing

There are a number of anomalies in finance and the one that has caught on quite well after AQR put out it paper called Buffet’s Alpha and a ten year period where volatility has been relatively low – is the low volatility factor. That doesnt mean it is a recent phenomenon. This has worked even before 2008. Basically the TLDR version of paper above is that Buffett, circa 1980s started buying these consumer stocks which were low vol in nature as they had predictable earning flows. And he levered his capital 1.7x using float from his insurance and reinsurance businesses. Simple in hindsight, not so much back then. This also is known as Betting-against-Beta.

Betting against Beta or Low volatility is simply a way of buying stocks which rank low on standard deviation or volatility over a a look back period of 6-12 months. Stocks that would rank high on this factor would be stocks where most investors under-react to information being fed to them on a regular basis. This is under-reaction is also cited as on the behavioral reasons why momentum investing works. And by momentum we don’t mean punting or growth investing – but that is for another post.

Now let us look at the other side of low volatility. Conventional CAPM model tells us that high beta stocks should have high “expected returns”. That is the key word – “expected”. But actually they don’t. NSE has a set of various strategy indices and they have a High Beta 50 Index and a Low Volatility Index. And as per conventional finance theory over  long period high beta stocks should have done much better than low volatility and low beta stocks (btw volatility and beta are not the same thing but to keep things simple we are assuming them to be the same). If we look at the performance of high beta index and low volatility index it shows clearly that low vol has beaten high beta by a mile. It has even done better than the Nifty 50 Index. ICICI Prudential has a Nifty Low Vol 30 Index ETF but with very low AUM. Still not a bad choice if you want low vol exposure. We also use Low Vol as a factor in our stock portfolio – AlphaBets

10 year Returns of Nifty Low Vol 50, Nifty High Beta 50 and Nifty 50

As it is truly said: Finance is not the study of money, it is more the study of how people behave with money. More on managing risk if you get caught in the itch to buy a lottery stock in my next blog.