Seeking Alpha

Early Retiree's  Instablog

Early Retiree
  • on Retirement
Send Message
Having always been a learning machine, I speak five languages, have worked as a sales agent, project manager, translator, computer consultant, software engineer, built a house with my own hands, published books and essays on literature, philosophy and art, have written for magazines of various... More
My blog:
Early Retiree's Instablog
  • Short-Term Trading Makes Markets Inefficient. Here's The Proof.

    The Game

    In 1995 Rosemarie Nagel conducted a study that thereafter was used as the starting point for a game which was played several times with different sets of participants; most of the times however the participants were professional fund managers. The fund managers received the following (or similar) email:

    "Please consider the following question and email your reply to us. The results will be published in a forthcoming weekly. However, we won't disclose none of your individual responses. As an incentive, we are offering a bottle of champagne to the winner.

    You are taking part in a competition with the other readers of our strategy weeklies. The aim of the game is to pick a real number between 0-100 ([0,100]). The winner will be the respondent who chooses the number closest to 2/3rds of the average number chosen.

    Many thanks in advance for your participation."

    Before you continue reading this article, please stop and think about what would be your own answer.

    The Solution

    The first step is to recognize that the average number chosen between 0 and 100 should likely be 50. Yet you can win only if you pick a number close to 2/3rds of the average number chosen. Should you pick 33? (50 x 2/3 = 33,33) And what if many of the other participants go for the same solution? Should you pick 22 instead? Or 15? Or 10?

    There is only one rational solution to the game: zero. Yet it is almost certain that many of the other participants won't find it, raising the winning number to something above zero. But how much above zero?

    The Parallel

    Probably you've already recognized some similarities to the behaviour of short-term traders at the stock market. When company XY issues a dismal earnings report which raises the guidance for the next quarter, a trader must think about what to do: Shall he sell, wait, buy more? His thinking involves assumptions about the behaviour of the other market participants: What will they probably do? The trader himself might think that XY's future prospects are bright (the guidance was increased), but that some of his colleagues will react badly to the dismal results of the past quarter. So he could sell and buy back later on the cheap, pocketing a small profit. But what if everybody does what he plans to do? The stock would drop like a rock. (Ironically it would drop because traders believe in XY's future prospects!) - You see that, just like in the game described above, traders simply can't be right (if not by pure luck). Or, to be more precise: They could only be right by choosing the "zero" solution, i.e. by not thinking at all about what others might do. Because thinking about others might do inevitably leads on the wrong track. And markets get inefficient.

    However, not thinking at all about what others might do is not the game of the short-term trader and as long as short-term trading is the vast majority, markets are condemned to inefficiency (at least in the short-term).

    How did the game participants answer?

    In one edition of the game, among the fund managers who participated, there were even some who chose numbers above 67, meaning that they did not understand the game or were just irrational. One participant answered: "100... I'm not a rational investor, my favourite stocks are Amazon and EBAY." The average number selected was 26, giving a 2/3rds average of 17.4.

    Hence, on average the participants did not go far above first-level thinking (i.e. choice of 33). The numbers that were chosen most frequently on a relative basis in this specific edition of the game were 50 (implying probably a below first-level thinking, as 50 is simply the average of the available range of numbers, but does not factor in the 2/3rds calculation), 33 (first level), 22 (second level) and zero (correct solution). Interestingly, a rather high number of participants chose a number slightly above zero, which I would call third-level thinking, because these people guessed that not all, but almost all participants would choose zero. However, they were too smart and too optimistic at the same time.

    Only 2.5% of the participants chose 17, thus won the competition probably by pure luck.

    (For the geeks among you, more results and analysis can be found here, starting on page 57.)

    Nov 21 4:39 PM | Link | 1 Comment
Full index of posts »
Latest Followers


More »

Latest Comments

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.