The (In)efficient Market


“Marry the Optimal Wife [or Husband].  Get Divorced when the Utility of Being Married Turns Negative.”

Yes, that is absurd, and in no way represents how the majority of people think about important life decisions.  It’s hilarious though when framed in those terms.   We have Prof. Richard Thaler to thank for that comment, a perfectly constructed, wry jab at much of traditional academic economics.  If humans are utility-maximizing automatons, doing calculations in their head constantly, how can anyone argue with his quip?

At Greyfeather, we were very pleased to see Prof. Thaler (University of Chicago) get the recognition this past week that he so rightly deserves – the Nobel Prize in Economics.  Thaler, in his own words, said that his contribution, “[..] was the recognition that economic agents are human, and that economic models have to incorporate that.” In academia for much of the 20th c., the idea that humans couldn’t be modeled as perfectly rational optimizers was pure heresy.  Ironically, most of the academics at the institution where Thaler has spent much of his career (U Chicago) disagreed with the spirit as well as the content of his work, including Eugene Fama (father of factor models, efficient market hypothesis, etc.). 

Thaler did far more than come up with a few counterexamples to existing models, he rigorously developed evidence to support hypotheses that were motivated by research findings concerning how people behave.  Thaler has always been someone whom I respect deeply as a scholar and his insights cannot be ignored if you trade/invest in the financial markets.  For those who are less familiar with his ideas, we wanted to highlight some of them (albeit a very small portion):

1. De Bondt, Werner F.M. and Richard H. Thaler. "Does the Stock Market Overreact?" Journal of Finance 40, (1985): 793-805.

Thaler knew that behavioral psychologists had shown that people don’t really follow Baye’s rule, and are very bad at updating probability assessments after information changes.  There is a distinct tendency for people to overweight recent information vis-à-vis past data, which can cause an overreaction effect.  Since markets are composed of human participants reacting to information, It stands to reason that there should be evidence of this effect.  That is what Thaler looked for and he certainly found it. The data showed overreaction; he found that long portfolios built from the worst performing stocks went on to outperform the market and that long portfolios built from the best performing stocks underperformed.  Put simply, he showed that mean reversion exists in security prices and is evidence of overreaction.  The critics don’t dispute that mean reversion is in the data, they just believe that it is rational and reflects available information, and as the information changes, then the price follows.  Thaler’s chart is below:

Screen Shot 2017-10-16 at 2.00.58 PM.png

Perhaps the most extreme over-reaction has been demonstrated in the recent asset price bubbles and crashes in recent memory: the bubble and subsequent collapse, real estate prices pre-2008 and the ensuing financial crisis, flash crashes, 1987, etc. etc.  Yes, some of those were affected by computerized trading systems, portfolio insurance, etc. – however, all the code was written by humans, and the end result of it was irrational. Participants, whether human or automated were reacting to other participants and not information regarding the future capacity of companies to make money and pay dividends.  Thus, in our view, the point still stands. 

2. De Bondt, Werner F.M. and Richard H. Thaler. "Do Security Analysts Overreact?" American Economic Review 80(2), (1990): 52-57.

In the earlier paper, Thaler showed that the market as a whole tends to overreact, but he was left open to the criticism that perhaps professionals don’t overreact, maybe they defy the herd and make the right call most of the time.  In “Do Security Analysts Overreact?,” he examined whether an over-reaction effect could be found in the assessments of professional market analysts.  These are professionals and one would expect them to have better performance, right?  Here is the conclusion, in Thaler’s own words:

“The conclusion we reach from our examination of analysts’ forecasts is that they are decidedly human.  The same pattern of over-reaction found in the predictions of naive undergraduates is replicated in the predictions of stock market professionals.  Forecasted changes are simply too extreme to be considered rational.  The fact that the same pattern is observed in the economists' forecasts of changes in exchange rates and macroeconomic variables adds force to the conclusion that generalized over-reaction can pervade even the most professional of predictions.”

It’s not that humans are fools, it is just that we tend to get too excited from recent developments.  We’re all guilty of doing it in various contexts, and that is one of the main reasons for using a systematic approach to investing, one that is underpinned by data and repeatable processes. 

3. The CUBA Fund.

Thaler has an exceptional ability to state complex ideas in very straightforward, easy to understand language.  He has said that the efficient markets hypothesis boils down to two components – (1) “No Free Lunch” and (2) “The Price is Right”.  No free lunch means that it is difficult to beat the market and that most active managers fail to do it net of fees.  Thaler doesn’t push back on (1).  But he has a lot to say about (2), i.e. that market prices are correct, or that they accurately reflect all known information.  Here is what he said during his Presidential Address to the American Economic Association in 2016:

“So it turns out there is a closed-end mutual fund that happens to have the ticker symbol CUBA (Herzfeld Caribbean Basin Fund, NASDAQ: CUBA).  Now, it invests in the Caribbean, it has never, of course, invested in the country of Cuba for two good reasons: (1) it would be illegal, and (2) there are no securities to buy. So it has never owned any shares in Cuba, and it still doesn’t. Historically, this fund traded at a discount of 15% to Net Asset Value (NAV).  That in and of itself is a well known anomaly known in behavioral finance that I have written about and Benjamin Graham wrote about in the 1930s.  Here is a plot of the price of the CUBA fund, you notice there is a little spike there:

Screen Shot 2017-10-16 at 9.05.38 PM.png

You can guess what happened.  That was the day that President Obama announced his intention to relax relations with Cuba.  You can see from the yellow line which is the NAV this announcement had no effect on the Caribbean economy or the portfolio of stocks owned by this fund, but investors were quite excited about the prospect of buying shares of the CUBA fund after this announcement.  And the fund price remained above NAV for over a year.  This picture sure looks like a bubble to me.”

The above examples are only a select few.  His work is timeless though and we encourage everyone who plays in markets - of any kind - to read Thaler thoroughly - it could potentially spare you from the next (ir)rational market cycle.



Evaluating AI-Based Hedge Fund Strategies

On 28 Mar 2017, I gave a presentation at the Battle of the Quants Conference in Frankfurt, Germany on how investors should assess AI-based investment strategies. The contents of the talk formed the basis for the white paper, "Artificial Intelligence Simplified: A Framework for Assessing AI-based Strategies."

This white paper presents a framework for assessing investing/trading strategies that are driven by Artificial Intelligence (AI).

Considerable hype surrounds the technology and that is to be expected given not only its capability today but what might be expected of AI in the future. This environment poses difficulties for investors and allocators who are trying to assess AI-based strategies in the present. Distinguishing hype from reality is the starting point. Then, investors must focus not on the technical aspects of the technology in all its various forms, but on what particular mispricing or inefficiency the strategy exploits, and the way in which AI enables the discovery and capture of such alpha.

Read the white paper here.