A study by the Center for Immigration Studies indicates wide variations in immigrant families’ propensities to receive welfare, according to their country of origin. Unfortunately, the study was published in 2011. Based on the Center’s appearances on television, it’s fair to say that it’s mostly anti-immigration although I am sure this is an oversimplification.
The score for Mexicans is 57%, 33% for Russians, 19% for Chinese, 14% for Indians, and only 12.5% for (the ever-saintly) Canadians. But the report of these seemingly clear differences may harm rational decision making in the way I warn against above. There are three reasons.
First, by emphasizing country of origin, the table seems to assume that different national groups have been in the US for the same length of time. In fact, nearly all immigrants arrive poor. Now, suppose that the average time in country for Mexicans is one year, and ten years for the Chinese. The 38 percentage point gap might vanish if the Chinese immigrants captured by the table had also been in the country for only one year. Obviously, those are made up figures. I don’t know what we would find if proper control for length of stay in country had been applied. It would have to be an average length of stay which complicates both data gathering and interpretation again although it can be done.
Second, immigrants from different countries probably belong to systematically different classes. This would affect their propensity to go on welfare, irrespective of national culture of origin. Suppose that all Indian immigrants are medical doctors or engineers, and all Mexicans casual laborers. This difference would suffice to account for the 43 point welfare gap between the two groups. The statement, “Indians are much less likely to go on welfare than Mexicans,” in this case, may be more about doctors and laborers than about Indians and Mexicans. That’s irrespective of time in-country. Note that I am not arguing that the two groups are equally desirable as immigrants but that their respective desirability may have nothing to do with the national propensity to go on welfare by Mexicans, specifically. I can think of arguments in favor of admitting more doctors and also arguments in favor of more laborers, factoring in the cost of their possible landing on welfare for a while. Note that the gap in two categories’ propensity to go on welfare may have no ethical meaning associated with nationality of origin. After all, we don’t know what the welfare participation of Mexican immigrants would be if they were at all doctors.
Third, we have to look at immigrants contributions, positive and negative, across several generations, as we do in connection with Latino youth gangs, for example. In fact, immigrants of the same national origin may be objectionable today while their children may be desirable for American society tomorrow. An example: Suppose all Mexican immigrants are young, married, unschooled laborers from rural backgrounds. Such people tend to have many children – more than, say, unmarried, older men from the best engineering schools in India. The first group will be more likely to go on welfare than the second and it will supply America society in the next generation with more contributors to the Social Security fund. In this scenario, paradoxically, the large number of Mexican immigrants may compensate for the likely lower income of their children compared to both the native-born and other immigrants’ children.
[Editor’s note: In case you missed it, here is the link to Part 3]