“Why do they have pictures of knives and forks here, daddy?” My daughter, then six-years old, was gazing out of the window of the airport bus as it brought us from the aircraft to the terminal building of Frankfurt Airport. I looked up briefly from my text messages. “Where do you see knives and forks, sweetheart?” I asked. “Everywhere”, she cried. I peered out of the window more intently this time. I leaned over to align my viewpoint with hers. I looked behind to check the area the bus had just passed. I looked left and right, inside and outside. Nowhere could I see cutlery of any description. “I can’t see any knives or forks”, I said, apologetically. She grew visibly exasperated: “Look! There! And there!” I didn’t want to upset her, so I remained silent as I scrutinized our surroundings. Then I saw them. They were indeed everywhere.
The curse of knowledge
“Oh, that isn’t a knife and fork. It is supposed to be flying bird. That’s its wing stretching out on top. It is the logo of the airline company, Lufthansa.” She cocked her head sideways and studied the logo again. She wasn’t too convinced with my interpretation of the image, but she seemed to accept it. Undoubtedly, she continued to see knives and forks for the remainder of the trip. I did too. Although it was not my first passage through Frankfurt Airport, this was the first time I recognised the logo as being anything other than a stylized representation of a bird in flight. Or was it? The truth is that I cannot recall what I might have naively thought when I first saw the Lufthansa logo. I am convinced, now, that I have always seen a bird, and I can hardly comprehend the naivety of someone who cannot immediately see it too. This is what is known as the curse of knowledge. Once we know something, it is almost impossible to put ourselves in the position we were in before we knew it.
I should not be too harsh on myself, as my daughter too suffered from the curse of knowledge. She was irritated by my inability to see something that was right in front of my eyes. It was inconceivable that someone else could have an alternative interpretation of the same information. In addition – and this is a further consequence of the curse of knowledge – she underestimated the effort required to transfer her seemingly obvious knowledge to me (I would encounter the same problem, several years later, while trying to explain an algebra solution to her). I was a doting father, so I went to some lengths to try to see my six-year old’s knives and forks. Most people, however, are not so indulgent. As a result, differences of judgement that arise because of the curse of knowledge can be greater and more stubborn than we think.
In a competitive environment, like a stock market, participants regularly arrive at differing judgements about the same information. Hence, there are buyers and sellers at practically every
price. Admittedly, some of those market participants might hold information the others do not. However, in the end – the long-run – the price ought to converge on the truth. As information propagates through the market, and everyone reaches the correct judgement about it, the trading behaviour should result in a consensus price. Sadly, this does not always happen. The curse means that knowledge transfer from one person to another is hampered, so the information propagation process is disrupted.
Worthless insider information
Let us imagine for a moment that my daughter and I were bidding for the rights to the Lufthansa logo. If cutlery logos were popular, she could be tempted to bid enthusiastically for these rights. I, on the other hand, recognising that flying bird logos are ‘out’, might not want to bid anything at all. Now let’s assume that the other passengers on this airport bus are not business travellers, but her primary school classmates. They might also see knives and forks, and decide to bid aggressively for the logo rights. If there were enough of them, a veritable bubble in the price of Lufthansa logos might develop. All the while, I would look balefully on, puzzled as to why anyone would want to pay so much for a stupid flying bird. I am right, and they are wrong, of course. I know what the image really represents. I possess information about the designer’s intention that they do not. I am an insider, so to speak. Yet this insider information does not help me if I have underestimated the possibility of errors in everyone else’s judgement, or if I have overestimated the ease with which these errors will be corrected through a transfer of knowledge.
Once we know something, it is almost impossible to put ourselves in the position we were in before we knew it.
The curse of knowledge contradicts one common assumption in economics, namely that more information is better. If I want to understand the sudden price bubble in Lufthansa logos, I would have to be able to put myself in the position of someone who is unable to see flying birds, which is something I struggle to do because of the curse. The knowledge that is supposed to make me wiser is, in this theoretical marketplace, making me poorer. Ironically, any one of my daughter’s classmates would probably be willing to pay for my insider information about the true nature of the logo even though it would make her poorer too. (1)
Can the curse be lifted?
Clearly, making good predictions in the stock market is not just about knowing the truth, but also about guessing what everybody else’s truth is. The economist, John Maynard Keynes, famously likened it to a beauty contest. The optimum strategy, he argued, is not to find the most objectively attractive candidate, but rather the candidate the other judges are likely to find the most attractive. To do this, however, we would have to disregard our own preferences, i.e., to deliberately unknow what we already know.
People’s ability to accurately make predictions based on the expected preferences of others has been the subject of much empirical testing, primarily using an experiment appropriately named a beauty contest. In it, subjects are asked to choose a number between zero and 100. The winner is the person whose choice comes closest to 2/3rds of the average number chosen. For instance, if I believe the average number will be 50, I should choose 33, which is 2/3rds. However, if I think that the other competitors might figure that out, then I should choose 2/3rds of 33 instead, which is 22. But what if others make that step too? I could continue this reasoning until I arrived at the only logical conclusion, which is zero. Indeed, only one number, x, satisfies the equality, x = 2/3x, and that is zero. Yet, this is not the most frequently chosen number.
The beauty contest game has been conducted many times, among graduate students, economists, financial professionals, and even among game theorists. The largest samples have been the readers of major European financial newspapers.(2) Popular responses among this group were 34, 23 (the average response) and zero. However, the occurrence of the number 15, which is 2/3rds of the average choice, was unremarkable. The most frequently chosen number was 1. This came about, not because readers failed to take their iterations to the logical conclusion, but rather because they did. After having found the only rational solution, those respondents fell prey to the curse of knowledge. They overestimated the ability of
everyone else to be able to find it too. Some made an adjustment to their prediction to allow for the irrationality of others, but it was wholly inadequate. In this game, knowledge of the logical answer was a handicap to winning.
Sadly, one cannot be too optimistic about freeing oneself from the curse of knowledge. We cannot un-know what we already know. So help must invariably come from the outside: a fresh set of eyes, or a novice’s opinion. That outside perspective could even come from you, after you have taken a break. For instance, after you have busied yourself with other things, how easy was it for you to regain your former train of thought? The best advice, though, for those who know the right answer (because they are too clever by half, or because they have had some private information) is to be very, very patient. It might take a lot of time and lot of explanation before everyone else catches up.
1. Loewenstein, George and Moore, Don A. and Weber, Roberto A., Paying $1 to Lose $2: Misperceptions of the Value of Information in Predicting the Performance of Others (July 16, 2002). Carnegie Mellon Behavioral Decision Research Working Paper No. 301. Available at SSRN: https://ssrn.com/abstract=323221 or http://dx.doi.org/10.2139/ssrn.323221.
2. The Financial Times in the UK, L’Expansion in France, and Spektrum der Wissenschaft in Germany.
This article was first published by TCAM Asset Management Limited