Immigration numbers games

According to several reports, Senators David Perdue (R – Arkansas) and Tom Cotton (R – Georgia) have introduced legislation to limit legal immigration, partly by reducing the number of green cards issued each year from 1 million to 500,000. The senators argue that the lower number reflects “historical norms” for immigration. The senators are speaking in counts.

In criticizing the bill, David J. Bier brings in evidence from a report in Cato. Bier talks about a rate: The count of new legal permanent residents each year, divided by the count of people in the United States. “From 1820 to 2017, the immigration rate averaged 0.45 percent of the population annually. In 2017, that rate was 0.32 percent.” The Cato article reports that the rate has varied greatly across our history: The low was .0.02% in 1933, 1934, 1943, 1944, and 1945. The high was 1.5% or slightly more in 1850, 1851, 1854, and 1882.

In 2016, The National Academy of Sciences (NAS) issued a report on integration of immigrants into American society, and another on economic and fiscal consequences of immigration. Those reports and a 2015 report from the Pew Research Center offered a third kind of number—the proportion of persons who live in the United States, who also were born abroad to parents who were not US citizens. In those reports, the proportion of us born abroad was about 14% in 2014. In the big immigrations of the 19th century, that number peaked at about 15%. The low for that number was a little less than 5%, around 1970.

There’s no big disagreement here about the extent of immigration: 1 million immigrants (Perdue and Cotton’s number) entering a country of 310 million would be 0.32 percent (Bier’s number). One million entering every year for the last 35 years or so would be 35 million or so, or about 11.3 percent of 310 million, in the ballpark of the number provided by NAS and Pew.

The choice of the kind of number reflects the authors’ purposes. Perdue and Cotton want immigration to look large, so they point to the count of immigrants for a recent year. They want “historical norms” to be restrictive, so they want the count of immigrants now to match the count when we were a much smaller country.

Bier wants immigration to look small, so he wants to use the rate per capita per year. He wants to expose the Senators’ ploy regarding “historical norms,” so he wants to examine that rate over time.

The NAS and Pew, I suggest, want to inform us about the dimensions and effects of immigration, so they point to the proportion of us that are immigrants at a given time. That proportion is likely to be related, for example, to the probabilities of meeting an immigrant in a grocery store, hiring or being hired by an immigrant, competing with an immigrant for a job, walking into a business started by an immigrant, marrying an immigrant, and so on.

Without hesitation, I choose the numbers supplied by the NAS and Pew.