Gladwell’s Outliers

After flipping through Malcolm Gladwell’s Outliers: The Story of Success late last year, I have finally read the book (nothing like over 30 hours of travel to get through a few).

Having heard a few podcasts involving Gladwell (such as this), I knew largely what to expect. Gladwell is strongly on the nurture side of the nature-nurture debate and is dismissive of explanations involving the individual or their inherent traits. While he does (at times) concede that nature might play a role, he suggests this is uninteresting and that we pull this explanation out too often. I think he is right that it may be pulled out too often in explaining the success of a particular person, but there is a large gap between giving nature the right level of focus and ignoring it altogether as Gladwell suggests.

So, rather than reviewing the book (as has been done plenty of times in the blogosphere already), there are a few specific parts of the book that I feel are worth mentioning.

First, the main points of agreement. I don’t doubt that for the most extreme of outliers – take Bill Gates or the Beatles – that luck played a large part. If Bill Gates had been born in any other country or if computers were not available to him at school, he would not have founded Microsoft and become the richest man in the world. Similarly, I don’t doubt that an ice hockey player is more likely to become a star if they have the good fortune to be born at the right time of the year. Nassim Taleb makes many similar points in The Black Swan on the role of luck.

The other side of this point, however, is that there is still plenty of room for nature to play a part. Why did Bill Gates and Paul Allen, of all the students at their school, take advantage of this opportunity? The January born ice hockey stars are still a very small sample of those born in January, so what distinguishes those January born stars from the others born in January? And the December born players who make it despite their disadvantage?

In some ways, Gladwell’s focus on the most extreme of outliers for much of the book is what gives luck such an important role. Take the example he makes of the little benefit to having an IQ above 120. Even if that were true (I am not sure it is – in many sciences those extra IQ points are still worth a bit), 90 per cent of the population has an IQ below 120. IQ is a strong predictor of income, status, health and a raft of other factors. While someone with an IQ of 140 might be as likely as someone with an IQ of 180 to win a Nobel prize, Nobel prizes are not the measure of success for most of us. If Gladwell had been examining success in the way most of us think of (or experience) it, IQ and other inherent abilities cannot be ignored. Or to put it another way, the absence of difference in outcomes between someone at the 99.9th percentile and 99.99th percentile of IQ does not mean it is unimportant for everyone else.

This brings me to Gladwell’s strong focus on IQ, as opposed to other heritable characteristics. In Gladwell’s discussion on the link between hard work and maths results, he refers to the Trends in International Mathematics and Science Study (TIMSS) test. It was found that the ranking of countries in this test corresponded to the country ranking for the number of questions answered in the accompanying (and 120 question long) questionnaire. Children who did better at the maths test also filled out more information on their family, education and a raft of other background issues. Gladwell points to this as evidence of the link between hard work and mathematical achievement, as it takes patience to work hard and learn mathematics or to answer the questions. He suggests that any IQ (or inherent quality) based explanation is flawed.

Ignoring that the ability to fill out a long questionnaire at a young age is probably influenced by IQ (answering questions on family education requires some cognitive skills), Gladwell lines up his attack on IQ but does not question whether a broader suite of heritable traits might be at play. If it is not IQ that is relevant, Gladwell suggests it must be environmental factors. Take time preference (patience), which has a heritable component and would undoubtedly influence competency at mathematics and willingness to complete the survey. Might that play a role? Gladwell draws a similar conclusion on the lack of success of Christopher Langan (with an IQ approaching 200) and suggests that compared to Robert Oppenheimer, he lacked the skills required to navigate the world. Once he claims to have eliminated IQ, Gladwell pins it to environmental causes, despite the possibility that these other skills have a genetic component. (Gladwell’s approach reminded me of a paper by Samuel Bowles and Herbert Gintis on the inheritance of inequality. The attempt to pin down the heritability of the level of income to IQ showed that other genetic factors would need to be considered. I’ll blog about this paper in the next couple of days.)

Two other anecdotes stuck out. First was Gladwell’s example of Jewish immigrants coming into New York with a wealth of tailoring skills at just the right time. He suggests that their children went on to become highly successful (usually as lawyers and doctors) after they saw the hard work of their parents in the home – hard work that those parents “lucked into” by having a skill that was suddenly in huge demand. It is a nice story, but it does not explain the success of Jewish immigrants in field after field where high cognitive ability is an advantage – be that banking, law, research (plenty of Nobel prizes there), medicine, and the list goes on. The success occurred on such a broad scale despite the varied (and often disadvantaged) family histories. Should Gladwell be looking further back in time for an explanation?

He does that in the other anecdote I found most interesting, which was Gladwell’s discussion of some intractable family disputes in some parts of the United States (think the Hatfields and McCoys). Gladwell suggests that their ancestors originally came from marginally fertile areas where herding dominated and there was a need to establish a strong reputation to protect their herds. This resulted in “cultures of honour” forming, in which misdeeds needed to be punished quickly and brutally to ward off future attacks. When they migrated to the United States, Gladwell suggested that these people brought their culture with them and it has persisted through generations, despite the shift in countries and for many people, significant changes in wealth and status. It is an interesting explanation, but it is one part of the book where innate traits are crying out to be examined. Gladwell did refer to some interesting studies on this issue, which is something that I will definitely be following up.

**As a personal side note, I am spending the next three months in Zurich at ETHZ’s Institute of Integrative Biology. This is to tighten up the evolutionary biology side of my research (I am an economist after all). From the perspective of this blog, being away from the day job will hopefully allow me to post on a more regular basis.

Comments

  1. says

    There is a fair reading of Gladwell that is a bit more nuanced than this post suggest.

    Part of his message is not so much that nature doesn’t matter as it is that there is a disconnect between ability and social rewards for ability that is capricous. This is a counterpoint to the oversimplified social darwinism of Econ 101 that suggests that the economy naturally matches productivity, which by inference flows from ability, to compensation, without giving fair credit to the degree to which the correspondence is merely a rough one. To cite an example that he doesn’t explore in much detail, one of the reasons that we have such exhorbitantly paid CEOs is that we have an economy where the economies of scale dicatate that most of the economy is most efficiently conducted by a few thousand big businesses. A few thousand people will end up at the top of each of these enterprises and find themselves in a position to reward themselves richly. The greater the disconnect between the number of big business chief positions the economy dictates and the number of people qualified to fill those positions becomes, the more luck plays a part. Similarly, our national media economy has a fairly fixed number of slots of athletic and entertainment celebrities — somebody is going to secure those slots and get the attendant riches, but immense luck goes into who that will be. The less individualized point he makes that whole generations of people entering an industry at a particular time may be exceptionally successful relative to those who come before and after them is particularly apropo.

    I do think that his “ten thousand hours” discussion is exceptionally weak, as it fails to look at what traits make ten thousand hours of effort possible to contemplate (from an ability perspective) and as it fails to look at why so many people who devote ten thousand hours to something are not exceptional.

    Also a key theme in Outliers is that luck isn’t random. A vision of economic and social phenomena as things that come in waves that have to be ridden at just the right times and places, rather than as true random chance or merit based selection, is important. Biologists who study population genetics are very familiar with the notion that traits irrelevant to fitness routinely get carried along when being in the right place at the right time, or being part of the right group, or having one trait that matters for fitness purposes over all other predominates. But, people who think about economics and social justice don’t. Most people who are successful see their sucess as a validation of their merit to a much greater extent that is warranted, even in seemingly meritocratic situations like sports leagues.

    Even when you disagree with him, his point of view suggests better ways to directly measure the impact of nature v. nuture than the standard twin study model. For example, his book has suggested to me that it would be very interesting to look at situations where an entire ruling class is swept from power (e.g. the Chinese Cultural Revolution) to see what happens to those people and their descendants once swept from their context (something personally relevant to me, as my wife’s family experienced just such an event about half a century ago and I am now in a position to see the results).

    • says

      I am happy to concede that Gladwell’s book might be more nuanced than my post suggests – I was simply picking what I considered to be the more interesting points.

      I also don’t disagree with Gladwell’s assessment that the link between ability and rewards has a capricious element, nor with some of Gladwell’s suggestions to increase the link between ability and outcome – such as splitting hockey players into smaller age cohorts (say, separate leagues for people born in the first and second halves of the year). On many of his other points, I don’t believe there is any suitable response, apart from moderating the stories we tell about why Bill Gates is the richest man in the world or certain lawyers founded the top New York law firms. Some of Gladwell’s other suggestions, however, such as picking students for gifted programs via a lottery would be counter-productive if you accept there are relevant inherent traits.

      What I do find interesting is the idea that social justice is achieved through equal opportunity. Once everyone’s environment is the same, people will be assorted by inherent traits. Is a genetic lottery socially just?

  2. says

    “What I do find interesting is the idea that social justice is achieved through equal opportunity. Once everyone’s environment is the same, people will be assorted by inherent traits. Is a genetic lottery socially just?”

    It might be more helpful in practice, since opportunity obviously isn’t equal, to at least attempt to distinguish between results achieved as a result of unequal opportunities, and those that are the result of inherent traits, as a means of distinguishing what you have earned (for surely inherent traits are part of what is “you”) and what you have been received (or not received) as a result of unequal opportunity – with the notion being that the latter ought to be more amenable to just redistribution. For example, it might be more just morally to redistribute inherited wealth than earned income.

    Equal opportunity is a plausible standard against which to measure social justice even if it is merely hypothetical. It is one that courts have experience redressing in the context of discrimination lawsuits. The mere notion that there is a definable standard against which social justice ought to be measured other than that of Candide’s professor (“this is the best of all possible worlds.”) , itself is sign of progress.

  3. JL says

    The idea that additional IQ points after 120 do not matter was debunked in the Study of Mathematically Precocious Youth. It showed that those who scored in the top 1 percent were less successful as adults than those who scored in the top 0.5 percent, who in turn were less successful than those in the top 0.01 percent.

Comments, thoughts, suggestions?