Seeking Alpha

Erik van Dijk's  Instablog

Erik van Dijk
Send Message
Hi! I am Erik L. van Dijk, principal at LMG Emerge. LMG Emerge is an internationally-operating institutional investment consultant with offices in the Netherlands and at Mauritius. Our clients are pension plans and other institutional investors, family offices and HNW individuals. In close... More
My company:
LMG Emerge
My blog:
LMG Emerge - New Economies
My book:
Asset Pricing and Risk in Emerging and Frontier Markets
  • The Link between Behavioral Finance, the Adaptive Market Hypothesis and Today's Financial Market Crisis 3 comments
    Aug 13, 2011 11:25 AM

    Introduction; Interesting Times

    We are living in interesting times. Within academia top level professors like Andrew Lo of the Massachusetts Institute of Technology (MIT) present innovative work that links Economics to Psychology and Neuro-Physiology to explain why our traditional, rationality-based treatment of financial markets doesn't lead to solutions that work.

     

    Prof Andrew Lo - MIT ''Adaptive Markets Hypothesis''

     

     

    In-and-of-itself we already know that the so-called Efficient Market Hypothesis that played - and still plays - a central role in mainstream Economics and Finance is under siege. Behavioral Science literature (both within and outside Economics) provided us with too many strange examples of situations where people acted anything but rational. A few examples:

     

    Overconfidence

    Russo & Schoemaker (1989) performed a fascinating experiment. They asked a group of test candidates 10 general knowledge questions. But instead of asking them for their 'best guess' answer, they asked for a 90% confidence interval. In other words: 'Answer the 10 questions in such a way, that you expect to score 9 out of 10'. Outcome: only 1% of the participants succeeded!! The actual number of mistakes (i.e. answers falling outside of the confidence window) was 3-4 out of 10! This is indicative of decision takers suffering from a tendency to overestimate their abilities. Juggling around with numbers in today's Financial Crisis it is quite clear that it is not unlikely that quite a few decision takers that are at the helm in the Developed Economies of Europe and the US are suffering from this behavioral bias.

    The unwarranted appeal of details

    Kahnemann & Tversky (1982) created a nice experiment that indicates what happens with us when provided with more details. They told their test candidates about a lady called Linda and painted a picture of her: she was a bright, socially-engaged, sympathetic lady, active in her community and assertive. Through the various pieces of information that the test candidates received about her, they did more or less get the feeling that they knew Linda. Then, Kahnemann & Tversky asked them to provide them with the most likely answer to the question 'What do you think Linda is?':


    a) A bank teller; or

    b) A bank teller and woman activist


    The funny thing is that 90 percent of the respondents chose b!. But obviously the best answer is a) since - with more details - the likelihood of b) being correct is smaller. This strange phenomenon of feeling more confident when knowing more details is also at work in today's Crisis, where you see that problems in Debtor nations in the Developed World does also translate into the phenomenon that a lot of funds are withdrawn from markets that do not suffer from the same fundamental problems. Emerging Markets went often even more down than the developed markets. Reason: in general Western investors know far less about these markets than they do about their own markets and they translate that into the 'certainty' that those markets will therefore be ''more risky''.


    Loss Aversion, Risk Aversion and Gambling Behavior

    When asked a choice between a certain profit of USD 240,000 or an uncertain gamble with a 25 percent probability of a USD 1 million gain and a 75 percent chance of leaving with nothing, most people will opt for the first option: the certain profit. Of course, when calculating expected gains, the second one will score USD 10,000 higher (0.25 x 1 million plus 0.75 x 0). But obviously, we have to subtract a risk penalty > 0 in the second case. Result: only the least risk averse or even risk-seeking (positive reward to risk) persons will choose the uncertain gamble.


    But now, to make things a bit more relevant in today's Crisis, let's slightly change the example. Suppose we can now choose between a certain loss of USD 240,000 on the one hand and another uncertain gamble with a 25 percent probablility of a USD 1 million loss and a 75 percent probability of breaking even.  Applying the standard expected result mathematics we end up deriving that the certain example leads to a USD 240,000 loss of course and the uncertain example a USD 250,000 loss. In other words: the uncertain gamble is USD 10,000 less good already before subtracting the risk penalty!


    Kahnemann & Tversky (1979) showed however that whereas the bulk of people opted for certainty when faced with a chance to generate profit, but with the uncertain gamble when faced with a loss chance: risk aversion was replaced by loss aversion. Not rational, but still it happens. And here we are in the Financial Crisis. Developed nations are nations that - most of the time - are led by politicians who have to be democratically re-elected. They will for sure take the risk in an uncertain gamble when the choice is between 'bad' (certain bet) and 'very bad but with a chance that results will be good' (uncertain bet). The bulk of the difference between negative and very negative outcomes will be visible after the next election date anyway, so better not to play it too safe when it is about negative things. In that case all will pin this event's tail on your government's donkey. Result: the likelihood of representatives of those parties finding a solution together that will reduce global risk is not too big. On the contrary: they will lavish themselves in the risky game with a lot of praying, postponing of tough decisions et cetera.


    The Carol Effect

    Game theory does also report about the so-called Carol Effect which implies that the prettiest, most wanted girl in town might actually not be the one being courted by most guys. Guys will analyze their chances using three potential scenarios: first, I approach her and it works. Most guys will immediately think: if it would be that easy, what about the other guys? They would then compare their chance of success with the chance of failure and embarrassment. With the latter obviously being much larger than the first, except maybe for the most overconfident macho's. The embarrassment scenario with all its negative repercussions would get a zero or even negative utility score. Of course: winning the heart of Carol would be fantastic, but when multiplied with a very low probability it is not that likely that this score will really outweigh the multiplication of a much more probable non-successful event multiplied by the zero or even negative score. And: if we add to that the option that we don't approach Carol in the first place and continue to do what we are doing right now (i.e. ignore her), then it is very unlikely that Carol will be asked to have a drink with us.

     

    Uma Thurman: Experienced the Carol Effect in Real Life

     Actress Uma Thurman has confirmed that the Carol Effect does exist. When asked about it, she told interviewers that she did always have far less men approaching her than other girls did....

     

    You might think: What is the relevant of all this for today's Financial Crisis? Well, when thinking of Carol think of China, Russia and Arab nations with big Wealth Funds filled with oil and gas dollars and you might understand what we mean when thinking of the macho guys as Western government leaders struggling alone with their Debts. Only when the Debt levels reach such levels that they have no choice anymore, will they go after Carol and ask her for help.....but by that time Carol might be old and not so pretty anymore. The latter could even be the result of us not going after her: because the Crisis will also affect her financial beauty.

    Of course: there will be new Carols around, and new guys. But in democracies the new guys are probably new government leaders and the whole game will start again!

     

    One thing that worries us though, is that part of the Carol effect - when applying it to the Financial Crisis - is not about the guys not being brave enough to approach the rich Carols of this world - but about Western leaders being too overconfident and cocky: 'We don't need Carol! We can do it alone!'.

     

    Efficient or Adaptive Markets Hypothesis

    The Efficient Market Hypothesis states that all publicly available information has been incorporated in prices in an unbiased way. This is the result of the interplay of rational decision takers and arbitrageurs. Whenever they find a mispricing their buying and selling behavior will automatically imply that prices will quickly adjust until equilibrium is re-established and prices reach their fair value again. This neo-classical doctrine has been the basis of important investment theories like Mean-Variance, the Capital Asset Pricing Theory, Black-Scholes option model et cetera. On the other hand the behavioral flaws mentioned above where presented as evidence that markets are not efficient. But most of the time the proof was presented in an ad-hoc manner.

     

    Over the last decade top-level scholars like MIT's Andrew Lo have tried to structure the anecdotal, fragmented behavioral evidence in a way that would contribute to the derivation of a more structured alternative to the Efficient Markets Hypothesis (EMH). Lo's paper on the Alternative Markets Hypothesis (NYSE:AMH) seems to be a good effort to do so. It links standard economic thinking - the bulk of which was the basis for mainstraim EMH style theory - to psychological and neuro-physiological ways of thinking. The latter is very much linked to evolutionary thoughts about how men's brains can be seen as a combination of a more reptillian emotional core, combined with a mammalian mid-layer and a humanoid upper-layer. The lower levels are more emotional, less deep and less logical, more spontaneous and less 'controlled' by us. Result: especially in situations of fear and/or greed these layers tend to gain in importance unless we really establish mechanisms to control them. Strange? Not really. Our history as mankind was one in which in the end the only thing that matters was 'survival'. But 'survival of the fittest' and 'preservation of the species' are not the same as 'survival of the richest' in financial markets. But a lot of our behavioral biases are related to us not being capable yet of controlling our lower level behaviors sufficiently when faced with new risks in financial markets. Compare it to the strange behavior of fish when out of the water. They make similar movements as the ones they would make when in the water. Difference: under water it would help them swim away from the danger or predator. On the land it is useless, and just tiring them.

     

    Behavior of market participants in financial markets can therefore remain irrational and illogical for quite some time to go, and - when the emotional/irrational actors have sufficient market power / financial resources - this could go on long enough to kill the rational ones that try to arbitrage these mistakes away. In a New World Order in which Emerging Markets are here to stay, with their own wealth, different cultures, different actors, different political systems it is more than likely that the old tricks of what has literally become a collection of one trick pony's (Western financial institutions) might not necessarily be the best way forward anymore!

    Also because of that, it is best to get the Carol's of this world as quickly as possible to the negotiation table to find a proper solution. And yes, that solution might even involve that Western countries have to accept a sell-off of certain larger firms, other entities or infrastructural projects previously considered ''strategic''. But then again: when it was about the Western nations gaining market share in the colonial countries the definition of ''strategic'' has been implimented much looser as well. With a bit of fantasy - and along the lines of EMH thinking - we could see it as one big, longer-term equilibrating force!

     

     


     

Back To Erik van Dijk's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (3)
Track new comments
  • Amerlafrance
    , contributor
    Comments (538) | Send Message
     
    I find this field very fertile for further exploration, but note a few questions with the risk aversion examples.

     

    In the case of a certain 240k vs a 1/4 chance of 1M, was the subject thinking they had one chance to play, 4 chances, or a hunderd? Avoiding risk if I have one chance to play and electing a certain payoff is the wise move. If I have 4 chances, the certainty of 960k may be much better becasue I am not risking a deviation from the odds at a cost of only 4% of the most probable outcome (of course the deviation could give me 4M or 0). Only at 100 chances do I see the possibility of taking the risk. Considering the market as a whole as hundreds of chances would indicate it is irrational to take the certainty, but viewed as 100 individuals each with one chance, the result of choosing certainty is to be expected. The market is not the mathmatical odds of all possible otcomes, it is the likely action of rational individuals.

     

    Then we get to the individual's postion coming into the game. If I am unemployed and struggling to keep my home, I will chose certainty becasue I am risking 240k; if I am Warren Buffett and don't need the money, it would be rational to take the gamble because the loss does not move the needle. In this example, the 240 is certain, so you are really putting 240k on the table that you would otherwise keep.

     

    The loss aversion example has similar constraints and shows why the bailout of foolish creditors is bad policy (at least at the level of personal incentives and not the system as a whole). If the larger loss is not real (the person is judgment proof or out of office when the loss has to be paid), then the postponment of pain is entirely rational. We need to have people accountable for their choices, and when we remove that from the equation it is no longer a financial problem but a personal benefit analysis with the power, prestige and enjoyment of life while risk is postponed the rational choice.

     

    I enjoy your pieces.
    14 Aug 2011, 01:43 PM Reply Like
  • Erik van Dijk
    , contributor
    Comments (20) | Send Message
     
    Author’s reply » Thanks for your excellent comment.

     

    A few caveats apply:

     

    1) Your note seems to be based on a strong belief in rational outcomes.

     

    2) Although we do believe in rationality ON AVERAGE and in the longer term (exactly what you liked to express concerning the examples!) in the short term behavioral finance teaches us that people tend to be sensitive to behavioral biases and irrationalities.

     

    If we follow the movement of stock markets during a period like the current one (Debt issues in the Euro zone with struggling European leaders and the IMF trying to solve it) we see that more or less random and in-and-of-itself not so important events with clear visibility and easy interpretation (e.g. the images of the killing of Khadafi) seem to be at least as important as the actual progress when it comes to solving the debt crisis. And there are dozens of examples like that.

     

    We believe in a bounded rationality model , also based on academic papers about this topic, and think that - especially when taking into account a multi-factor approach where all investment opportunities are disentangled into combinations of factor sensitivities - an investor who combines solid application of probability theory (his long-term, rational strategy) with a very versatile short-term flexibility to incorporate the irrationality of others (e.g. by using technical analysis indicators to capture the behavioral flaws and market moods) will do best.

     

    In our own work we do so by using Momentum and Overreaction indicators (including seasonal patterns). We also have good experience with the use of the VIX indicator as additional 'climate' indicator.

     

    Thanks for your excellent feedback and I hope that our reaction - albeit a bit later than expected because of a busy period here at LMG - is of use to you.

     

    Check out the Seeking Alpha main pages for a new article in which we also analyse this mixed presence of rationality long-term and behavioral biases short-term.

     

    Kind regards,
    Erik L van Dijk
    LMG Emerge
    24 Oct 2011, 01:36 PM Reply Like
  • Amerlafrance
    , contributor
    Comments (538) | Send Message
     
    As always interesting. I try to use rationality (realizing I have behavioral biases I probably don't realize or have taught myself to ignore) for my investment choices and agree that the market has irrational participants and at times can be given over to irrationality. Most of my mistakes, in hindsight, are faulty analysis on my part either from not drawing the correct conclusions or not having enough data to draw the conclusion. That is a hard enough problem for me to deal with.

     

    I will study "bounded rationality," but my problem is investing based on an assumption that the irrational will occur seems like a losing propsoition because by definition assuming the irrational means you cannot predict the outcome.
    24 Oct 2011, 05:45 PM Reply Like
Full index of posts »
Latest Followers
Posts by Themes
Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.