Skip to content

Counterpoint: People Control the Markets

October 22, 2009

A curious fact is that I am actually a former psychology major, and prior to deciding to go to business school and engage in the  requisite finance training, I was on track to entering a doctorate program in cognitive psychology .  Ironically I received cautionary advice from the business school guidance counselor about embarking on a finance major because: “you won’t be able to handle the math!”  and “finance is all about the numbers.” My decision to ignore his advice led me along the path to becoming a “quant.”

 Behavioural finance giants like Kahneman and Tversky were commonplace names for me throughout the years, and I was quite familiar with their work. Funny my surprise when their work became the undoing of much is was  considered to be standard economic theory concerning rational agents. It is quite strange to read economic theories that assume that people act in a cold and rational manner when you have read hundreds of documented studies of people making “irrational” decisions on a wide variety of controlled tests. Furthermore, a person’s IQ was  not a mitigating or significant variable in these studies. Being a rational person myself, I began to realize that the assumption of rationality was designed specifically to make the “math” easier to solve rather than as an actual working theory.

You see the problem is you can’t get away from people and their biases in decision making. It doesn’t even matter if you are a fund that controls a billion dollars and uses state of the art quantitative technology. You could have the best professors and minds around the globe working under your roof.  Ultimately if you have a down period or a down year, or just an unusual period, your investors dictate when you buy and sell things. When they pull the plug en masse, you will be creating a tidal wave of  forced liquidation on your own positions that is completely independent of your view or quantitative outlook of the market. If your fellow quant friends have also been using a lot of the strategies that you have been running even to a small degree, they will have a lot of overlap with your positions. Your liquidation impacts them and other related funds, and this can cause a domino effect of client withdrawals. Suddenly everyone is running for the exits at one time and finally we have a mini-meltdown.

Why has the last two years seen such violent swings? The answer is people.  Fear, greed, decision regret, and depression to name a few have driven the actions of portfolio managers and governments over the last two years.  Both agents (fund managers and governments) no matter how powerful or knowledgeable ultimately answer to the common man with all of his/her cognitive biases. Lately people have been very emotional and have demanded that the government “do something!” to address the current crisis. Blame must be assigned and villains must be villified–history is replete with identical reactions to every similar situation.  Inaction and intelligent study or reflection after such an event is completely unacceptable to the public. The same phenomenon exists after a rare but inevitable plane crash where the public immediately demands new safety regulations. Government and fiduciary (fund manager) actions and  must appeal to what the people want rather than what is optimal. There is no getting around this, and this means that volatility will always produce opportunity. To the quants who look at factors, or to the traders that look at backtests, remember that people create contrarian or momentum opportunities. When you hear about mass withdrawals and bank runs, and billion dollar scandals you can bet that few are behaving rationally. Mean-reversion strategies do best in an environment of uncertainty and fear, and momentum strategies do best in periods of optimism and euphoria or performance chasing (like right now). At all the states in between the market is fairly efficient, as rationality is most likely to dictate decision making.

I would argue that as long as people are responsible for making investment decisions whether directly or indirectly, there will always be a huge impact  of their aggregate bias on the market. For this simple reason—or simplifying assumption—the markets can never by extension be completely efficient.

6 Comments leave one →
  1. Joe Marc permalink
    October 23, 2009 6:50 am

    Hi again David,
    RE: your statement above (“there will be a huge impact of their aggregate bias on the markets”.
    First, again thanks for your insight. Awesome stuff you are working on. I do not want to fan political arguments like last time, and this is purely an objective observation. But I can not get over how surreal the actions from our “leaders” have become and how those actions (it seems to me) will eventually, (probably) lead to extremely manic market actions as people respond to what seem to me to be totally insane actions. I feel like I am at an Air Show with a thousand F-14s in the air and every one of the pilots up there have just escaped from an asylum and they all are on Meth, Cocaine and Booze. Like good grief, run for cover every one, this is going to be wilder than wild. Could be unprecedented.
    Hope I am wrong, but…
    Joe Marc

    • david varadi permalink*
      October 23, 2009 9:54 am

      hi joe, good analogy! should be very interesting times—as long as one has a bomb shelter!
      cheers
      dv

  2. Richard Sotter permalink
    October 23, 2009 9:51 am

    The quant strategies can only work reliably if they are based on sufficient history to account for different sets of market and economic environments. That means more than just a few years of data. A quant analyst has to both understand this and insist on taking the time to build up a large enough database.

    • david varadi permalink*
      October 23, 2009 9:58 am

      richard, you are absolutely correct, a proper backtest history must account for as many varying enviroments as possible. this also brings up the need to do conceptual or logical backtest to account for rare but different events like a “credit crisis”, to “runaway inflation” or even terrorist attacks like Sept 11. Accounting for these different types of situations makes a backtest far more robust.
      cheers
      dv

  3. Alex permalink
    November 2, 2009 7:03 pm

    dont’ remember where i just read this, but the “cumulative bias” is only material and 0, if the population is not DIVERSE. so i’d modify the last paragraph to have this qualification as opposed to people making decisions. the herd creates opportunity. whether they’re “quant” opportunities or “one-off one-time yet history rhymes” type is arguable. either way though, they’re there.

Trackbacks

  1. Counterpoint: People Control the Markets | Financial engineering resource center

Leave a comment