[This review was published in the Winter 2011 issue of The Journal of Social, Political and Economic Studies, pp. 506-510.]
Steven D. Levitt, Stephen J. Dubner
Harper Perennial, 2011
They’ve done it again! With SuperFreakonomics, Levitt and Dubner have written a sequel to their enormously successful Freakonomics, published by William Morrow in 2005. This one is every bit as intriguing and full of surprises as the first one, the product of six years of additional mining of data to find unexpected, often startling, results.
Eminently readable and entertaining, the books certainly provide titillation for those who approach them as light reading. This does not detract in the least from the fact that Levitt and Dubner provide real substance. Some of their topics are of high importance, even though others are trivial. What is perhaps equally significant as the topics themselves is the authors’ methodology. It provides a preeminent example of honest scholarship, diligently and openly pursued.
The authors explain that their first book “did have a unifying theme” [even though the disparate topics gave a contrary impression]. It was that “people respond to incentives… although not necessarily in ways that are predictable or manifest.” They add that their continued purpose has been “simply to show that things are different deep down than they might appear on the surface.” Sometimes this involves “taking on controversial topics in unusually objective ways.” All of that leads to the unexpected, much like coming to the denouement of a “whodunit.” But the surprises aren’t the entire purpose, since the topics are pursued honestly and well even without that aspect: “We have done our best to tell stories that rely on accumulated data rather than on individual anecdotes, glaring anomalies, personal opinions, emotional outbursts, or moral leanings.”
SuperFreakonomics’ fascinating nuggets should provide the grist for much good conversation. There’s much more, but here’s a sample:
. We would think that driving while intoxicated would lead to frequent arrest; instead, Levitt and Dubner tell us that “drunk drivers are rarely caught.” The data show that “there is just one arrest for every 27,000 miles driven while drunk.” Moreover, it is far more dangerous to the intoxicated person to walk while drunk than to drive in that condition (although, of course, the driving is more dangerous to other people). Accordingly, the authors suggest that the slogan “friends don’t let friends drive drunk” should be supplemented by an additional slogan: “Friends don’t let friends walk drunk.”
. In the year 1900, the city of New York was home to 200,000 horses, each of which produced about 24 pounds of manure every day. The manure so overflowed things that “in vacant lots, manure was piled as high as sixty feet,” and the streets were lined with large banks of it. This was an environmental disaster, of course – one that was overcome by the advent of automobiles and electric trolleys. (Thus, by looking back into history for even so short a time, we gain a new perspective on the environmental problems that cars themselves have introduced.)
. The “feminist revolution,” with women taking on executive careers in the world of business, has had the effect of drastically reducing the quality of American education. Why? Because most employed women of high intelligence in the past were teachers. When those women have gone into other fields, the average intelligence of teachers has declined. (We’ve mentioned that the books are “honest.” An important part of that is that Levitt and Dubner aren’t afraid to point out consequences that today’s “political correctness” would prefer be ignored.)
. One of the more entertaining topics is the authors’ discussion of the economics of prostitution, where they show that prostitutes’ incomes have fallen in an age when there is so much sex given free. We also find that “a U.S.-born boy is roughly 50 percent more likely to make the majors leagues in baseball if he is born in August instead of July.” The reason is that those born in July are able to get into youth leagues, most of which have a July 31 cut-off date, when they are almost a year older than boys born in August. Those born a month later will wind up being the youngest in the age category and are less able to do well.
Although some of these subjects may be considered “mere nuggets,” others are on topics of major public importance. One of the latter deals with “global warming” and its alleged causes. SuperFreakonomics has a long and fact-ladened chapter that is well worth the reading. In summary, the authors tend to be quite skeptical about it, while at the same time telling how, if there is a genuine problem, there are simple and inexpensive solutions for it. It is interesting that “the world’s ruminants [cows, mainly] are responsible for about 50 percent more greenhouse gas than the entire transportation sector.” There are problems with the computer climate-prediction models, since they “do not know how to handle water vapor,” which is “the major greenhouse gas.” The authors point out that “ice-cap evidence shows that over the past several thousand years, carbon dioxide levels have risen after a rise in temperature, not the other way around” [their italics]. They criticize Al Gore, one of the world’s leaders in concern over global warming Nevertheless, Levitt and Dubner aren’t content to leave it at that. The chapter concludes with several pages of discussion of experts’ suggested solutions to warming [on the supposition it is occurring], each of them simple and inexpensive.
The book makes several points about methodology. Some are obvious (especially to those working in the social sciences), while others are not. This reviewer taught for many years in a college of business, where surveys based on student populations [are you saying based on survey designed/conducted by students?]Note to Roger: No; the students were the people surveyed were given high standing. It is refreshing to find Levitt and Dubner explain how data from surveys are dubious. Among other things, they point to the difference between “declared preferences” [what people say they prefer] and “revealed preferences” [what the people actually do]. The obvious points include one about averages: that, taken by themselves, they can be misleading, as illustrated by the fact that “on average, the typical adult human being has one breast and one testicle.” Several factors make lab studies questionable: “selection bias” [an unrepresentative sample], “scrutiny” [the impact of someone’s knowing he is being watched]; and “context” [the artificial nature of the lab context may change incentives, such as by creating a willingness to assist the investigator]. Another point is one that goes to the heart of many errors in social science today: the too-easy conflating of correlation and causation. It is often forgotten that the correlation of two things has an heuristic value, by suggesting the possibility of a causal link, but that more is needed than simply the correlation.
Arguably the most helpful methodological point comes when the authors point out that much media hysteria distorts information. Great media attention is given to shark attacks, for example, but the data show that many more people are killed by elephants than by sharks. SuperFreakonomics devotes several pages to the Kitty Genovese murder case, where the New York Times and other media made quite a sensation about how 38 people witnessed the murder without the slightest concern for her plight. Levitt and Dubner’s review of the facts shows that the story was almost entirely false. This sensationalizing of certain facts while minimizing others plays a large role in our misunderstanding of historical events. It is one of propaganda’s salient features, helping create the myths that touch almost every aspect of our thinking. One might hope that Levitt and Dubner will devote their next book to those propaganda myths.
Rarely is so much edification found in books that are so much fun to read.
Dwight D. Murphey