The strong demand for charlatans

In the improbable event of ever being invited to give a commencement address, my advice to graduates wanting a lucrative career would be: become a charlatan. There has always been a strong demand for witchdoctors, seers, quacks, pundits, mediums, tipsters and forecasters. A nice new paper by Nattavudh Powdthavee and Yohanes Riyanto shows how quickly such demand arises.

They got students in Thailand and Singapore to bet upon a series of five tosses of a fair coin. They were given five numbered envelopes, each of which contained a prediction for the numbered toss. Before the relevant toss, they could pay to see the prediction. After the toss, they could freely see the prediction.

The predictions were organized in such a way that after the first toss half the subjects saw an incorrect prediction and half a correct one, after the second toss a quarter saw two correct predictions, and so on. The set-up is similar to Derren Brown's The System, which gave people randomly-generated tips on horses, with a few people receiving a series of correct tips.

And here's the thing. Subjects who saw just two correct predictions were 15 percentage points more likely to buy a prediction for the third toss than subjects who got a right and wrong prediction in the earlier rounds. Subjects who saw four successive correct tips were 28 percentage points more likely to buy the prediction for the fifth round.

This tells us that even intelligent and numerate people are quick to misperceive randomness and to pay for an expertise that doesn't exist; the subjects included students of sciences, engineering and accounting. The authors say:   

Observations of a short streak of successful predictions of a truly random event are sufficient to generate a significant belief in the hot hand.

It's easy to believe that this happens in real life. For example, the people who are thought to have predicted the financial crisis of 2008 are invested with an expertise which they might not really have.

Of course, there are other reasons why people might want to pay for forecasts; maybe they want a false sense of security of a predictable world, or they want someone to blame if things go wrong. This paper, however, suggests that these are not the only motives. Instead, people are too quick to perceive skill and thus to pay for something that doesn't exist. The demand for forecasters and tipsters substantially exceeds the real ability such pundits actually have.

Comments

[Blissex]

This is a demonstration of the analogy that Warren Buffett used to describe fund management returns contests: a coin-tossing tournament.

[Dabo]

You assume the crash was a random unpredictable event. Witness the gold price. It started to climb in 2005 as smart money cashed out their chips. The crash was going to happen, its just the when that was unpredictable.

[Keith]

I think there is this desire in human nature to seek reassurance. Medical quackery is famous and some times fatal when it leads to sick people refusing effective or reasonable treatments for pleasing flim flam.

[Brythonfa]

This is how, in 2010, Carmarthenshire Council were duped into spending up to £349 per vehicle, attaching magnets to fuel tanks of their vehicle fleet, in the vain attempt to reduce fuel consumption. (Please stop laughing). At the time all the science said this was bunkum, but apparently the sales pitch of the company flogging the 'magic magnets' had found some remote statistic that, somewhere, at some point, it once worked (random). Surprisingly the more senior the council 'expert' the more susceptible to being duped they became. Stuff of snake oil. The next part of this scam, if results don't materialise, is to sell the sucker 'more powerful magnets'.

[s jay]

Ever consider that Buffett himself is an example of the phenomenon? Success is a fantastic cover. the website longorshortcapital.com has an excellent summary in their "how to become a billionaire" post.

[Bryan Simpson]

There were no consequences to buying the useless information. How serous could a participant be if it wasn't his money at stake? The need to see narrative were randomness exists is something we all have in common to varying degrees. It's hard to disagree that useless information is bought every day, but this study doesn't prove anything except that people are careless with house mon

[s jay]

This is exactly what a magician does, he exploits the successful algorithms that evolved as we survived. Being able to recognize, as in the experiment, the pattern that our pattern matching algos can be misled, seems to have kept the worst of the problems with them in check, witness that fact that humans have so far survived.

[Bryan Simpson]

Sorry, my last sentence was cut off. mon=money

[james higham]

Crown Prince being Blair.

[Steve Case]

I saw a show, Maybe "Hitchcock Presnets" or perhaps the "Twilight Zone" decades ago identical to Derren Brown's "System".

[Fab]

By the way, when you say that the "predictions were organized in such a way" that a quarter of students saw two right predictions after two tosses, the word "organized" is a bit misleading.

The coin was not manipulated. The "predictions" could have been the same for all students, or totally random, or some mix, it doesn't matter: About a quarter of students will have seen two predictions come true after two tosses... The remarkable thing is that they were willing to pay more.

[jh]

Raindrops on roses and whickers on kittens-I've always liked "random" - you find folks saying "completely random" and "totally random" - curious!

People use eg. MS. XL and similar to produce series of random numbers - very curious! How many remember books of random number tables which you had to cross off each time you used some to ensure they "remained random" - yet many many people were using the exact same book!

Either things are random and a priori unpredictable or they are not and a priori predicatable at some level of understanding. Of the latter some are quite complicated and thus "diificult to predict".

I once new a professional gambler (horses) - his advice was "never bet on a handicap or 'over the sticks'" (steeplechase). Sound advice just so far as it goes!

jh

[UnlearningEcon]

I think you are unfair to those who predicted the crisis properly. Consider two situations:

"This car is going to crash, it is going to crash, it's going to crash" ad nauseum.

Car crashes.

"This car is going to crash, the axles are too weak and I have measured pressure building up where they are connected to the wheels."

Car crashes, axles snap where they are connected to the wheels.

(I'm aware my examples are probably not sound from an engineering perspective).

[Nick Drew]

"even intelligent and numerate people are quick to misperceive randomness and to pay for an expertise that doesn't exist; the subjects included students of sciences, engineering and accounting."

I would almost say "especially". And economists! Numerate people's propensity to believe (and pay for ...) 'econometric' forecasts knows few limits, despite the entire history of such forecasts demonstrating beyond peradventure that, whatever theory they are based upon, that theory is wrong.

It is related to the near-universal failure to grasp how forward markets behave (especially by economists!). Good forward traders clean up against these suckers, in the manner of sober professional poker players against drunken Saturday-night amateurs.

[gogole]

The demand for charlatans may be high in absolute terms , but compared to the apparently skyrocketing supply , it would seem to me to be trivial. Thus the price for a charlatan , for all but the superstars , must be declining.

I would bet that the charlatan Gini index is at an all-time high.

[wooden beds]

Either things are accidental and a prior capricious or they are not and a prior predictable at some akin of understanding. Of the closing some are absolutely complicated and appropriately "difficult to predict