Do Higher Homicide Rates for Central American Countries Increase the Number of Unaccompanied Alien Children coming to the US?

Jul 16, 2014 | Featured

A number of people have looked at how homicide rates have changed over 13 years for the countries that have supplied the vast majority of unaccompanied alien children to the US: focusing on El Salvador, Guatemala, and Honduras.  The claim is that increased violence ushered in the flood of unaccompanied children into the U.S.  In all the years up to 2012, Mexico has accounted for most of the unaccompanied children coming to the US.  For 3 of the 4 Central American countries and Mexico (accounting for almost all the unaccompanied alien children entering US) higher homicide rates are actually more likely associated with fewer unaccompanied children coming to the U.S. than the reverse.

Surprisingly, the homicide rate has only risen sharply for one country — Honduras.  But between 2011 and 2013 the homicide rates fell for El Salvador, Honduras, and Mexico.  They were also flat for Guatemala.  But this is the time that unaccompanied alien children coming to the US soared by over 140 percent.

Screen Shot 2014-07-17 at  Thursday, July 17, 4.15 AM

According to the U.S. Border Patrol data on the number of unaccompanied alien children is overwhelmingly from just four countries (click to make the figure bigger, figure from the New York Times). Screen Shot 2014-07-16 at  Wednesday, July 16, 1.50 AM 1 Obviously there are only a small number of observations here, but we can still see how homicide rates are correlated to the number of children coming to the U.S. from these countries (surprisingly something that no one seems to have done).  And when one does that, three of these four countries show that during the four years from 2009 to 2012, higher homicide rates were associated with fewer unaccompanied children coming across the U.S. border.  Only for Honduras does more homicides produce children coming to the U.S..  However, even though Honduras has by far the highest homicide rate out of all four countries in 2012 (twice as high as El Salvador and over four times higher than Mexico), it accounts for the fewest number of children.  Given Mexico’s proximity to the US, it isn’t surprising that it accounts for by far the most unaccompanied children over this period.  But children from El Salvador and Honduras have to travel virtually the same distance and Honduras has a 25% larger population.  But Honduras still has a much higher homicide rate and yet up until 2013 it had a much lower number of unaccompanied children coming to the U.S..

Screen Shot 2014-07-17 at  Thursday, July 17, 3.57 AM

The graphs here make the results that higher homicide rates are associated with fewer unaccompanied children clear.

Screen Shot 2014-07-17 at  Thursday, July 17, 3.40 AM

The results are inconsistent with the notion that the number of unaccompanied children coming to the U.S. are being driven by increasing homicide rates in these countries.  Across all the observations, each one percentage point increase in homicide rates reduced the number of unaccompanied alien children coming across the border by 164 and the result is statistically significant at better than the one percent level.  There is some justification for using this specification as it picks up the fact that Honduras as a high homicide rate but relatively few of its children trek across the U.S. border unaccompanied.

Accounting for the differences in countries to explain the different number of unaccompanied alien children implies that higher homicide rates still reduce the number of children crossing the border, but the effect can reject the claim that there is a positive relationship at less than the 10 percent level for a one-tailed t-test.  Controlling for differences across countries and years, eliminates any statistical significance.

Screen Shot 2014-07-17 at  Thursday, July 17, 3.42 AM 1. Screen Shot 2014-07-16 at  Wednesday, July 16, 2.53 AM 1

johnrlott

0 Comments

Archives