Things Men Do in the Absence of Women

Fact One: In the last half of the 1970's, China informally introduced a one-child policy, increasing the share of male births as families, for a number of entangled cultural and economic reasons, increased their output of boys relative to girls.

Fact Two: Between 1992 and 2004 crime rates in China almost doubled.

Ratio of men to women:

Crime rate:

Various economists and political scientists have long suggested that when the relative size of males to females, known as the sex ratio, favors the y-chromosome, then the likelihood of antisocial behavior is bound to increase. There might also be a marriage-effect where men's wilder tendencies are tamed by women. Empirical research in this area, however, has been scarce. Monetary policy this is not.

Last week, I attended a seminar at New York University where Columbia University economist Lena Edlund presented new research into this phenomenon. Edlund and her coauthors (which include Hongbin Li, whose work I reviewed here) estimate that between 28 percent and 38 percent of the documented rise in crime can be attributed to China's one-child policy.

The thing about the one-child policy is that it wasn't implemented in all Chinese provinces at the same time. This situation allowed the researchers to see how crime rates fared as the sex-ratio in the provinces took different paths. After controlling for factors that might drive up crime (income inequality, employment levels, population density, etc.), they found that the sex-ratio did indeed affect the crime rates of 16-25 year-olds, the ones most likely to commit crimes.

To test the strength of their findings, the researchers also looked at a type of crime that shouldn't necessarily be affected by increasing the portion of men in society. Corruption fits that description nicely since it's more likely to be carried out by wealthier man, who, because of their higher status in society, should be immune from the competition for women. They found no increase in corruption rates as a result of the skewed sex ratio.

What's most surprising is that although one might suspect sex crime rates to increase because there are relatively more men than women, they actually began to decline, implying that the types of criminal activity affected by the increased sex-ratio were violent and property crimes.

(Here is the paper, Endlund told me an updated version will be posted on her web site soon.)