“Men, it has been well said, think in herds; it will be seen that they go mad in herds, while they only recover their senses slowly, and one by one.” (Charles Mackay)
The Wisdom of Crowds by James Surowiecki is often cited for the proposition “the large crowd is smarter than the expert few”. The proposition has at least four prerequisites, however: a diverse, decentralized and independent thinking crowd and a means of measuring the collective verdict. Now that’s a tall order to get the benefit of collective wisdom. Without those stars in alignment, the exceptions overcome the rule. The author has said that the crowd can come up with “remarkably stupid decisions.” and a good chunk of his book shows how crowds can “go mad” and “succumb to delusions”. To understand more about crowds see, for starters, tulip mania, information cascade, bandwagon effect and the herding instinct. Just because the lines are long at McDonald’s does not make it the healthiest or tastiest food. Other examples: AOL & Internet Explorer.
Surowiecki was quoted as saying:
Essentially, any time most of the people in a group are biased in the same direction, it’s probably not going to make good decisions. So when diverse opinions are either frozen out or squelched when they’re voiced, groups tend to be dumb. And when people start paying too much attention to what others in the group think, that usually spells disaster, too. For instance, that’s how we get stock-market bubbles, which are a classic example of group stupidity: instead of worrying about how much a company is really worth, investors start worrying about how much other people will think the company is worth. The paradox of the wisdom of crowds is that the best group decisions come from lots of independent individual decisions. (emphasis added)
Since I will always question the wisdom of crowds, you can put me in the camp of Charles MacKay & John Kennedy Toole. I do not presuppose a wise crowd, especially if I can’t identify the individuls who make up the crowd.
Postscript: There are two Suroweicki examples I often hear to support the truth of “the large crowd is wiser than the few experts” mantra which I’d like to poke some holes in (just for fun).
1. The lost submarine. It seems the ones questioned were all “experts” in fields which had a useful relevance, i.e., engineering, physics, oceanography, ballistics. Had they asked the “man on the street” crowd (certainly a more diverse crowd), that sub would still be missing. So pick your large crowd wisely. Conversely, guessing the weight of a pig takes no real expertise. It’s more a function of being above a certain age & life experience. Ask a large crowd of grade school kids & that average pig weight won’t be close. I’ll take that wager any day. And I’ll wager a group of local real estate experts comes closer to guessing the weight value of my house than a diverse group of real estate experts from out-of-town, not to mention a crowd of statisticians with no real estate experience in my town and a software program using public tax assessment data. I would not rely on that confederacy for the wisdom I need.
2. The “Poll The Audience” lifeline on Who Wants to Be a Millionaire gives more correct answers than “Call a Friend”. For one, the poll is done for easier questions & call a friend for the really hard ones. Secondly, the friend called is not usually an expert on the question asked. But even the Poll can be wrong in finding Truth. Do a “Poll the Audience” for the question “Is the earth flat or round*?. The audience would overwhelmingly say “flat” before Columbus sailed.
So, think before following the crowd & always use experts. Just get more than one opinion. That’s just common sense. Happy Columbus Day.
* the earth is currently believed to be an oblate spheroid.
Given the stigma of the word “Crowds,” perhaps a better (albeit unwiedly) title for Surowiecki’s book would have been “Collecting, Analyzing, and Summarizing the Collective Predictive Wisdom of Large, Diverse, Decentralized and Independent (LDDI) Groups of People With At Least Some Knowledge of the Subject Matter, and Circumstances Under Which Such Wisdom May Turn into Madness.”
I’d certainly agree with you that the collective wisdom of in-town Realtors about the price of a local home will usually trump that of out-of-towners.
To be provocative and truly test Surowiecki’s theories, how about this experiment: For that same local home, let’s pit the collective wisdom of an LDDI group of local Realtors vs. that of an LDDI group of local homeowners. At the risk of being heretical, I’d posit the homeowners would give the Realtors a run for their money.
To be even more provocative, how about this one: Pit the Realtors against the homeowners, but this time the prediction we’re looking for is the median price of homes in area X in one year. In this case, I’d bet the homeowners would be more accurate. Why? It has to do with the first “D” in “LDDI.” Homeowners probably have a higher diversity of opinion about the future of real estate than do Realtors, who tend, understandably, towards the bullish side.
One last experiment: Homeowners vs. the regular readers of a “Real Estate Bubble” blog about prices in area X in one year. Hands down, the homeowners would win — again, because they would have more diversity of opinions than the typical doom-and-gloom of the Bubblistas.
Now those are reasonable experiments. Excellent choice of groups. Let’s also throw zillow in for educational purposes. How do you suppose zillow would do? I’d love to see that outcome.
I’d agree the LDDI homeowers would give the realtors a good run but I’d bet the local Realtors would best them on the value of a local home, just by virtue of experience in selling in the current market. It would not surprise me if Zillow came in dead last. And that would be further proof that they are not the best starting point. I always said that asking a neighbor was as good a starting point as zillow.
As to median home value in area X over 1 year, I don’t know. I just don’t know whether the realtor consensus would be bullish & whether that would matter if indeed the market did rebound—it’s more a lucky guess on the future market isn’t it? But what value is knowing the median value now or a year from now? How would that knowledge help a buyer or realtor?
Since future events are affected by many factors, I could not say homeowners would better the bubblelistas, despite the bias. The bubble meisters fit the classic mold of a mad crowd, but in predicting future events none of that matters—–any vote on the future of a real estate market is a pure guess to me. Hey, a broken watch is right twice a day.
I am not a believer in any group’s ability to accurately predict the future since there are factors at work which are beyond the control or knowledge of anyone. A natural disaster, terrorist attack, war, whatever, is the ultimate spoiler to any groups’ educated guess.
Who was it again who said, “The problem with the future is that it’s so difficult to predict”?
Responding to:
I just don’t know whether the realtor consensus would be bullish & whether that would matter if indeed the market did rebound—it’s more a lucky guess on the future market isn’t it?
That’s the whole point of a prediction market — it tells you the overall consensus of the LDDI group. Any individual’s prediction is indeed a guess, and that guess is called “lucky” if that individual’s prediction happens to be near to the truth. Collectivize those opinions, however, and you end up with something that is really a lot more than a guess — in fact, it’s the “Wisdom” that Surowiecki refers to in his book.
There’s an analogy to Google here. My understanding is that a lot of G’s search algorithm is based on the number and quality of the links pointing to a certain page. If an LDDI group of people collectively point more frequently to page X than to page Y in reference to a particular query, then, all other things being equal, page X will outrank page Y — another example of “Wisdom.”
Responding to:
But what value is knowing the median value now or a year from now? How would that knowledge help a buyer or realtor?
I think it would be immensely valuable. If I were thinking of buying or selling a home, or I were advising a client thinking of buying or selling a home, that LDDI prediction would be another data point to consider in trying to figure out the direction of the market.
I know the whole “Wisdom of the Crowds” thing sounds dubious, but, properly formed, it does appear that LDDI predictions can be pretty good.
If the realtors were bullish and the market did rebound, the realtors would win.
If the homeowners were less bullish and the market continued to fall, the homeowners would win.
I don’t see the wisdom here—–To me, the market decided the winner, not the group. In hindsight, one group can be said to have predicted correctly but I don’t think that proves that a collective wisdom was at work to put either in the correct camp.
The reason I say this is because the market can be affected by many forces, including a natural disaster. This spoiler, or a similar one, cannot bestow wisdom on the group’s prediction.
I know nothing about Google’s search mechanism so can’t disagree or agree.
Re: Knowing the median now.
The median is only a middle number. One will usually buy above or below the median. How much? Who can tell? So, if the median price for homes is $250,000 & you showed me a house for $300,000 would you advise me not to buy it because it was $50,000 above the median? Likewise, should I put my house up for sale at $300K if the median is $250K?
Knowing the number in the future.
When will you know if the predicted median of $250,000 is correct? Or the margin of error? Don’t you have to wait for the future event? Or will you feel comfortable advising me now?
My prediction on how to make a small fortune in the stock market—-
start with a large fortune.
I do not say the proposition is dubious. As the author admits, it requires certain key factors (your LDDI). In my opinion, these key factors are not present in every group dynamic. Thus, the crowd is just as likely to be mad as wise.