Social Econ Blog

Over at Econball, Tony V. posts an interesting puzzle – the distribution of the zodiac signs of Members of Congress differs radically from what one would expect. There are disproportionate numbers of Geminis, Cancers, Virgos, and Libras in Congress. This finding begs a question – why does the Washington Post report Congressional Votes by Zodiac sign?

Ok, we might also wonder why people born under these signs are more likely to end up in Congress? It is tempting to think that age based cut-off dates might explain this phenomenon. A few weeks ago, the Freakonomics blog discussed how the distribution of World Cup soccer players exhibits a similarly skewed pattern. They argue:

…because the FIFA cutoff date for determining a child’s age for international play is January 1, we would expect that a disproportionate number of the players in the World Cup would be born in the early part of the year. The idea is that these kids will get special treatment and attention when they are teenagers because they will be developmentally almost a full year ahead of kids born in the latter part of the year.

Perhaps more relevant for our purposes, Elizabeth Dhuey and Stephen Lipscomb recently produced a paper which shows that kids who are older within their school cohort are more likely to hold leadership positions in high school:

Economists have identified a substantial adult wage premium attached to high school leadership activity. Unresolved is the extent to which it constitutes human capital acquisition or proxies for an "innate" unobserved skill. We document a determinant of high school leadership activity that is associated purely with school structure, rather than genetics or family background. That determinant is a student's relative age. State-specific school entry cut-offs induce systematic within-grade variation in student maturity, which in turn generate differences in leadership activity. We find that the relatively oldest students are four to eleven percent more likely to be high school leaders.

Can school kindergarten cutoffs also explain the concentration of congressional birth signs in certain months? This would seem like a plausible explanation since we might expect holding leadership positions while in school to positively correlate with winning Congressional elections.

Table 1 in the Dhuey and Lipscomb paper presents the cut-off dates for each state for children entering kindergarten in 1947-49, 1959, and 1967-1969. During 1947-1949, most states had a Jan. 1 or Nov. 1 cut-offs (with a few Oct. 1 or Feb. 1). Only New York was very different with a May 1 cut-off. In the 50s and 60s states moved their cut-off dates around (and some stopped mandating them at the state level altogether), but they were still predominantly concentrated in the fall.

Thus, if the social and physical advantages of being relatively old within one’s grade affect one’s likelihood of winning a Congressional seat, we would expect to observe disproportionate concentrations of Congressional births in the late fall and early winter – at least among the three quarters of congress born during the years around these three periods.

The table below presents the birth month (not zodiac sign) for all members of the 109th Congress separated by birth cohorts (note the total percentages are slightly different then what Tony V. presents because I also include Senators; I also include the actual birth rates from 1978 so people have a better sense of how these percentages differ from what might otherwise be expected). Given that cut-off dates vary over time and across states, focus on the 1940-1945 column which roughly corresponds to a period where we know the cutoff dates for all states, and they are concentrated is a small range. The concentrations in November and December may be explained by states with Nov. 1 cutoffs, but the concentrations in May and June cannot be explained by school cut-offs (there are not enough people born in NY around this time for New York’s May 1 cut off to explain these concentrations).



Ultimately, school cut-offs may explain a small fraction of this pattern, but most of it likely requires some other explanation. I think (in spite of the odds) it may simply be an anomaly. I’d hoped to figure out if the same pattern exists over a longer period using biographical data from a larger set of Congresses, but month of birth is not included in the large datasets covering congressional biographies I found, and I am not ambitious enough to enter that by hand. Ideally, one would add the school cut-off and birth month data to one of the large congressional biographical dataset that includes birth state (a proxy for state where started school) and look carefully identify whether or not members of Congress tended to be very old for their class. This would more conclusively explain if age advantages during school affect the likelihood of ending up in Congress.

Update -- in a comment at Econball, someone asks for the distribution by Chinese zodiac sign -- here it is (I couldn't easily find the US population distribution by birth year, so I, unfortunately, am not able to compare this distribution to the distribution of Chinese zodiac signs in the whole US population):