Right data - immigrant income versus that of US natives

National Review, Oct 28, 1996 by Ed Rubenstein

UNTIL recently, most economists viewed immigration as a net boon to the United States. Arriving relatively poor, the newest Americans supposedly acquired the job and language skills needed to get ahead. But a new study by the Rand Corporation finds that this rosy scenario doesn't apply to all immigrant groups.

In 1970, for example, Chinese, Korean, and Japanese male immigrants earned 12 per cent less than the average native-born American male; in 1990 they earned 10 per cent more. Europeans and Canadians held their own, with wages 10 to 30 per cent higher than natives'. By contrast, Mexican immigrants' income slipped from 66 per cent to 56 per cent of natives'. Because Mexicans and Central Americans constitute an increasing fraction of all immigrant workers (20.8 per cent in 1990, up from 6.5 per cent in 1970), they've dragged the overall average down.

Education explains a lot here. In 1960 newly arrived immigrants had, on average, about six months more schooling than U.S. natives. Then, in 1965, the immigration laws were amended. Instead of promoting the immigration of workers needed to fill specific labor needs, the new law emphasized ''family reunification.'' Education levels fell. By 1990 the newest immigrants had 1.3 fewer years of schooling than natives.

All unskilled workers -- native and immigrant alike -- have suffered income declines relative to the technologically proficient. But even among the unskilled, immigrants are faring much worse: the earnings of foreign-born in the bottom tenth of the American workforce fell to 63 per cent from 94 per cent of the earnings of natives in the same group.

During the high-growth years of the 1980s, academic studies touted immigration as a good investment. Over time even less-educated immigrants were expected to pay more in taxes than they received in public benefits. Unfortunately, it hasn't worked out that way. Economist George Borjas reports that 9.1 per cent of newly arrived immigrant households received public assistance in 1990, compared to 7.4 per cent of native households. By contrast, immigrant households in 1970 were less likely to be on welfare than native households. Still more troubling is Borjas's finding that the likelihood an immigrant will be on public assistance rises the longer he resides in the United States. This reflects a steady rise in income thresholds for welfare eligibility coupled with declining relative incomes for immigrants.

Perhaps one-quarter to one-half of the increase in poverty since the early 1970s reflects immigration. Between 1973 and 1995 the number of individuals with incomes below the official poverty line rose by 13.5 million. Of those, 6.2 million, or 46 per cent, were Hispanics.

Although children born to immigrants will fare better, history suggests lingering disadvantages. Borjas finds, for example, that a 30 per cent wage differential between ethnically distinct immigrant groups in 1910 translated to a 15 per cent differential among their children in 1940, and a 7.5 per cent differential among their grandchildren in 1980. Given the sorry state of public education, it's likely that descendants of the current wave will take even longer to catch up.

IMPORTING POVERTY: IMMIGRANTS FALL BEHIND
                  Immigrant Income as %       of All Immigrant  %
                  of U.S. Average              Workers
Country of Origin    1970       1990           1970        1990
Mexico                66         56             5.1        15.6
Central America       79         62             1.4         5.2
Caribb., S. America   80         86            17.1        27.4
Japan, Korea, China   88        110             3.3         8.5
Europe               110        115            46.8        19.5
U.K., Canada         113        130            20.7         7.8
All Immigrants        99         89           100.0       100.0

Source: Rand Corporation,1996