James Surowiecki, who writes a column on finance for the New Yorker, has put together an engaging, well-documented book revealing the fact that decisions made by crowds are better than those made by individuals alone – at least under certain circumstances.
In this counterpoint to Charles McKay’s famous book of 1841, “Extraordinary Popular Delusions and the Madness of Crowds,” Surowiecki offers insight into free markets in particular but also into other means by which people interact, such as committees and corporations. He explains why committees don’t usually work very well and why the cult of the CEDis largely a mistake. And he explains the periodic madness of crowds.
The book starts with a historical anecdote about the Victorian polymath Francis Galton, and how, in 1906, he studied the results of a contest at a county fair in Plymouth, England. A prize was offered for the person who most accurately guessed the weight of an ox on display. When the contest was over, Galton analyzed the 787 guesses, hoping to show that the average guess – the “wisdom of the crowd” – was faulty. He found just the opposite. The average guess was 1,197 pounds, and the actual weight was 1,198 pounds.
The Galton story is one of a rich stream of anecdotes from many academic disciplines and popular sources that illustrate the theme of the book: more information, if properly coordinated, leads to better choices.
Markets, of course, are the granddaddy of mechanisms that coordinate information, and Surowiecki spends a good deal of time exploring them. “Collective decisions,” he writes, “are most likely to be good ones when they’re made by people with diverse opinions reaching independent conclusions, relying primarily on their private infor-
The NASA team failed to recognize the seriousness of the potential damage to the shuttle – or else chose to ignore its seriousness, on the ground that nothing could be done about it anyway.
mation” (p. 57). Market results usually reflect this diversity. The madness of crowds, such as we see in booms and busts in the stock market, is excited when people stop deliberating on their own and simply buy or sell because others are buying or selling.
Surowiecki draws on many kinds of research on decision making, including that of the free-market economists Vernon Smith, Ronald Coase, and Friedrich Hayek. (If you want to find where he discusses the work of these individuals, however, you will have trouble because there is no index. This is usually the sign of an unprofessional book, but perhaps the publisher wants us to use the 1/search inside” function that is available for this book on amazon.com.) And the author goes well beyond the study of traditional markets.
One of the interesting applications of market-based thought is the proposal made by members of the intelligence community after Sept. 11, 2001, for a Policy Analysis Market focused on the Middle East. The proposed market would be open to the public and would be patterned after the Iowa Electronic Market, which allows people to buy and sell futures contracts on the outcome of elections. This one, to be funded by the Defense Advanced Research Projects Agency, ‘iVould have done the same for expectations about events in the Middle East. Unfortunately, it didn’t pass the political. palatability test; the proposal was dropped. As Surowiecki explains, “it seemed to offer the possibility of ordinary people profiting from terrible things happening” (80).
I am already a fan of voluntary decision-making through markets, so for me the most valuable information in the book came elsewhere, in the chapter on committees. If more input is better, as Surowiecki argues, why do committees do such a bad job? To answer the question, he dissects the meetings of the NASA Mission Management Team (MMT) that supervised the ill-fated flight of the space shuttle Columbia in January 2003. He calls the committee’s actions “an object lesson in how not to run a small group”(175).
The committee members knew that a large piece of foam had broken and hit the shuttle, but they didn’t know how much damage had occurred. From the beginning, they failed to recognize the seriousness of the potential damage to the shuttle – or else chose to ignore its seriousness, on the ground that nothing could be done about it anyway. (The transcript of a key meeting suggests that both factors were involved.)
An ad hoc team of engineers studied the possible effect of what had happened, deliberating on the problem between Jan. 16, the day of the launch, and Jan. 21, when they reported their findings to a member of the MMT committee just before the committee met. Even though this was the first opportunity for the full committee to discuss what was known about the potential damage, the topic wasn’t brought up until two-thirds of the meeting was over, and then it received cursory treatment.
The problem with the team, summarizes Surowiecki, was that it blocked crucial information. One technical ad- viser· was outspoken in ·c1aiming that no damage had been done, and this
Surowiecki wants to show that democracy· fits the requirements for good decisions. Yet he understands that voters are abysmally ignorant.
influenced the committee. “[T]heir conviction that nothing was wrong limited discussion and made them discount evidence to the contrary,” says Surowiecki (177). In addition, the head of the committee thought that if damage had occurred, nothing could be done about it. (Surowiecki questions this view.) The leader asked questions that had the answers “built into” them, and there was an “utter absence of debate and minority opinions” (181).
Perhaps the committee’s biggest error was to tum down the request by the Damage Assessment Team to solicit the Deparhneni: of Defense for pictures of the shuttle in orbit, to see whether there were signs of damage. One reason for rejecting the request, says Surowiecki, was that the committee thought “the resolution of the images would not be good enough to detect the small area where the foam struck.” But no one on the team actually knew this for a fact: “[N]one of the managers had the necessary security clearances to know how good the resolution of the photos would be, nor did any of them ask the Department of Defense – which would have taken the pictures – about picture quality”(178). In sum, this example offers a lot of material illustrating the dangers of small groups.
Clearl~Surowiecki has a good grasp of the need for information, and a good sense of which coordination mechanisms work and which don’t (markets, committees, and corporate structures having different degrees of success). He does slip up, however, when it comes to collective political decisions, which are the subject of his last chapter. Even though he cites the work of James Buchanan and Gordon Tullock, either he doesn’t understand public choice the- o~ which analyzes political decision making in the way in which economics typically analyzes market decision making, or he doesn’t want to give it much credence.
In fairness, perhaps fatigue took over. Surowiecki mentions that it was hard to write the book, and he observes that his editor at Doubleday has the “patience of Job.” Whatever the reason, he fails to treat public choice with the care and attention that he gives to other approaches to decision making.
Surowiecki wants to show that de- mocra~ with its decentralized information and coordination through elections, fits the requirements for good decisions. Yet· he understands that voters are abysmally ignorant. (This fact is a fundamental tenet of public choice theory, which speaks of “rational ignorance.”) For example, he cites a University of Maryland poll showing that voters believe that the United States government devotes 24% of its budget to foreign aid. The actual figure is under 1%.
In spite of such popular delusions, he .still struggles to justify democracy. “It is not a way of solving cognition problems or a mechanism for revealing the public interest,” he acknowledges.
What Surowiecki doesn’t know, or is unwilling to state, is that democracy can be a good form of government only if the scope of government is narrow and limited.
“But it is a way of dealing with … the most fundamental problem of cooperation and coordination: How do we live together? How can living together work to our mutual benefit? Democracy helps people answer those questions because the democratic experience is an experience of not getting everything you want” (271).
To me, this is lame: vapid rhetoric, quite unlike most of the book. What Surowiecki doesn’t know or is unwill- ing to state is that democracy can be a good form of government only if the scope of government is narrow and limited. Voters will always be ignorant of most political details because their periodic votes have virtually no effect on the outcome; this inevitable ignorance gives special interests – which do have a direct impact through campaign funds and contributions – extensive power, undermining many great hopes for democracy. But accepting the limitations of democracy is a conclusion that scares modem liberals. Perhaps it scared even this very intelligent and perceptive columnist for the New Yorker, weakening what is otherwise an extremely valuable and engrossing book.