Sunday, 27 July 2014

Do Immigrants Import Bad Institutions?

One of the best arguments against laxer or no immigration restrictions is that third-world institutions may rub off their flavour on first-world policies. For instance, a person may be interested in how the presence of more immigrants as a share of a state's population impacts that state's general level of "freedom", such as defined by researchers William P. Ruger and Jason Sorens in their "Freedom in the 50 States" project. From their website and from the Migration Policy Institute, I collected data to make a small graph depicting the percentage-point change in immigrants' share of each state's population between 2000 and 2012, and the same state's change in the freedom index between 2001 and 2011 (the years do not match exactly due to limited data availability).

Before proceeding, I should point out that, while data on freedom may not be of much concern to everyone, they should help inform the present case, since changes will reflect what has happened to policy in general. Thus, if nothing jumps out from the data, immigration might not impact policy much. So, without further ado, this is what I found (click to enlarge):
In case the quality is too poor for the graph to be informative, there is a slight tendency for states which have declined in terms of freedom to have had an increase in immigrants' share of total population. But this tendency is slight indeed; Nevada, New Jersey and Maryland are the three points farthest to the right which have seen declines in freedom. Remove them and no trend is visible at all. However, these states also happen to be the ones which have had the greatest increase in immigrants as a share of their populations, so one should not speak with too great certainty.
One might suspect that the origin of the increased share of foreign-borns matters. In Maryland, persons born in Latin America were 40.1 per cent of all foreign-borns in 2012, up from 34 per cent in 2000. The corresponding numbers for Europeans and Asians are 10.5 and 16.8, and 32.5 and 35, respectively (i.e. decreases for both groups). African-born individuals were the only major group to join the Latin Americans among those increasing their shares, being 15.4 per cent in 2012 and 12.1 in 2000. Perhaps there is something fishy about Latin American and African institutions?

However, the share of Latin American-borns in Nevada declined over the same twelve-year period (from 61.4 to 58.2), and Africans' share increased only a little bit (from 1.6 to 2.3). Asian-borns were the big relative gainers, up from 22.9 to 29.1 per cent. New Jersey, the third state to have had much immigration and a sizable decline in freedom, offers further heterogeneity; its biggest changes were a decline in the Europeans' share from 23.9 to 15.9 per cent, and an increase in the Asians' share from 27.8 to 32.1 per cent. The other groups did not change much. More detailed data are available here.
Which leads me to conclude that,  in the main, the thing that is really conspicuous about my graph is that nothing stands out. Now I would be a fool to make any grandiose claims based on this unsophisticated little exercise, but it is some evidence that immigrants' impact is not too great. Increases in shares of the population by several percentage points over just twelve years are not associated with any general changes in Ruger and Sorens' Freedom Index, so maybe fears of imported third-world institutions are overblown?

No comments:

Post a Comment