Are Presidential Candidates Outliers?

There’s the idea that those who grow up being relatively older than their classmates or peers have better life outcomes.

Kevin M. Kniffin and Graham Dietz, Cornell University investigate whether this is also true for US presidential candidates.

Malcolm Gladwell’s book Outliers (2008) introduced millions of people to the concept of Relative Age Effects (RAE) using entertaining anecdotes. In one of the book’s best, Gladwell recounts how Canadian junior hockey players who are relatively old among their same-birth-year teammates tend to have significantly better outcomes later in life.

In this story, players who were born in January, February, or March tend to do better than those born in October, November, or December because – thanks to rules that establish that cohorts are based uniformly on birth year – the former become accustomed to being older, more dominant, and more successful when compared with their same-age peers.  This rationale appeals to sensible and good intuition but also immediately invites the question:

Why does this matter for those of us who are neither a junior hockey player in Canada nor a parent or relative to one?

Gladwell, to his credit, uses the interesting birth-year rule for Canadian junior hockey as a model for thinking about comparable age-restricted social environments.  Far from the hockey rink, Gladwell argues in Outliers that elementary schools should be broken into half-grades. This would minimize the RAE that appears to help relatively old members of a cohort while artificially and unnecessarily limiting the development of a cohort’s younger members.

The problem with the hockey story that Gladwell popularized – which is based on original research by Barnsley and co-authors (1985, 1993) – is that his outcome variable is a poor measure for actual success as professionals.  Specifically, the hockey players who Gladwell recognizes as winners were selected disproportionately high in the National Hockey League (NHL) entry draft but that doesn’t necessarily mean they will actually perform better in the league.  In fact, an article by Gibbs, Jarvis, and Dufur from 2011 shows the reverse to be true. They found that relatively younger Canadian hockey players with birthdates in October, November, and December have longer professional careers, are elected to more National or Olympic teams, and more NHL All-Star teams.

Generally speaking, Gladwell’s focus on the importance of an individual’s “ecological environment” on their success (or failure) makes sense and his use of hockey is definitely clever.  Unfortunately, though, Gibbs, Jarvis, and Dufur’s work suggests the Outliers story is wrong or, at least, overblown – in relation to Canadian junior hockey players.

To venture outside the hockey rink, we conducted an informal review of birthdates for current and recent US Presidential candidates to see if there are any strong Relative Age patterns.  Since school cut-off dates are different across US states with respect to which birthdates are relatively old or young, we took that variation into account when classifying each of the candidates.  We also assumed that the candidates had not been “redshirted” or kept back a full extra year in relation to their classmates.

We agree with Gladwell that people, such as candidates for the White House, have had their careers largely shaped by the social environments in which they’ve grown and lived.  When it comes to supporting Gladwell’s point about Relative Age Effects, though, we would expect to see that almost all of the (most successful) candidates are relatively old amongst their peers.

In the Figure that we include here, which categorizes ten top Republican candidates and two Democrat candidates for 2016 along with the same parties’ final candidates from 1976 onwards, you can see that there are, proportionally, many more relatively young finalists (fourth Quartile or Q4 in the parlance of RAE researchers) when compared with the Q1 or relatively old finalists.

RAE Presidential Candidates

While there are certainly some Q1 members in the sample (Reagan, Mondale, and Fiorina), the general pattern – in this admittedly small sample – fits Gibbs et al.’s article rather than Gladwell’s Outliers.  The Q4 (or relatively young) members in the sample are Presidents Carter and Bill Clinton along with Hillary Clinton, Ben Carson, Mike Huckabee, and the Independent candidate Ross Perot.  We included Vice President Biden in our sample as well – he’s also from the 4th quartile – since he was a potential contender at the time that we surveyed relevant online sources (InsideGov, Bio.com, and the History Channel).

In a much more systematic study of RAE in a domain outside of elite hockey players, one of us (Kniffin) along with Drew Hanks of Ohio State found that – analyzing a sample of 14,535 people who earned the PhD in 2010 – there is no substantive effect of Relative Age upon individual performance when earning the PhD.  Published in the current issue of Contemporary Economic Policy, that article reports that whether or not someone was relatively old or relatively young among their childhood classmates turned out to have no significant effect on whether or not they pursued and gained a doctoral degree. Prior research has also looked at the relevance of RAE across a wide range of educational and employment outcomes – generally finding that the effect subsides at approximately 3rd grade.

But, you might wonder, what about Donald Trump?  Or Jeb Bush?

Trump grew up in the 3rd quartile and Jeb Bush in the 2nd.

Trump, Bush and the others who have made it close to – or inside – the White House are obviously outliers in many other respects but being relatively old (or young) in one’s childhood cohort does not, thankfully, seem to generate lifelong effects.

References

Allen, J., and R. Barnsley. “Streams and Tiers: The interaction of Ability, Maturity, and Training in Systems with Age-Dependent Recursive Selection.” Journal of Human Resources, 28, 1993, 649–59.

Barnsley, R. H., A. H. Thompson, and P. E. Barnsley. “Hockey Success and Birthdate: The Relative Age Effect.” Canadian Association of Health, Physical Education and Recreation (CAHPER) Journal, 51, 1985, 23–28.

Gibbs, B. G., J. A. Jarvis, and M. J. Dufur. “The Rise of the Underdog? The Relative Age Effect Reversal Among Canadian-born NHL Hockey Players: A Reply to Nolan and Howell.” International Review for the Sociology of Sport47, 2011, 644–49.

Gladwell, M. Outliers: The Story of Success. New York: Little, Brown, and Co., 2008.

Kniffin, K. M., and A. S. Hanks.  2016.  Revisiting Gladwell’s Hockey Players: Influence of Relative Age Effects upon earning the PhDContemporary Economic Policy, 34, 21-36 DOI:10.1111/coep.12114

Image courtesy of FRE Lens.