Immigration Beneficial or Not? -- Two Book Reviews

By Donald Huddle
Volume 1, Number 2 (Winter 1990-1991)
Issue theme: "Uses and misuses of history in the debate over immigration reform"

The publication of a Julian Simon book is a cause for rejoicing among the advocates of laissez-faire and open-border immigration. According to Dr. Simon, who teaches business administration at the University of Maryland and is an adjunct scholar at the Heritage Foundation and the Cato Institute, all immigrants and refugees, no matter how many or in what form, are good news for the American economy. His latest opus will not disappoint his followers, but it adds little if any substance to the real-world immigration policy debate.

The much smaller volume by University of California economist George Borjas is a valuable contribution to the debate on immigration -- it is a penetrating, scholarly work incorporating state-of-the-art economic research and is very accessible to the noneconomist. In contrast to Simon's replay of the same old message, that the United States needs many more immigrants, Borjas finds that recent immigrants are much more likely to live below the poverty line, to be unskilled and unemployed, and to go on welfare. Borjas concludes that the United States must dramatically upgrade the quality of immigrants if we are to avoid the very large costs of the past fifteen to twenty years.

I begin with Simon's notions because they are both oversimplified and extreme. Those aspects of Simonism that are not contradicted by Simon himself are contradicted, for the most part, by Borjas later on.

Simon claims substantial scientific documentation to support his controversial thesis increased immigration of at least one million a year is a foolproof way for the U.S. to advance every major national goal and ensure our economic success. Some of Simon's key contentions are that immigrants

-- work harder, save more, and are more innovative than are natives;

-- do not displace native workers, not even unskilled minority workers;

--actually create new jobs on net balance by increasing the purchasing power of goods and services and by starting new businesses;

-- use fewer welfare services and more than pay for those they do use in taxes;

-- are typically as well-educated occupationally as natives upon arrival immigrants earn less than natives, but within five years they catch up with and then earn more than native workers.

To Simon the popular belief -- documented in opinion polls -- that immigrants are harmful to the U.S. economy is the result of misinformation from the media. According to Simon, cultural homogeneity is just the contemporary codeword for racist opposition to immigration. Those who differ with him are either badly informed or have a hidden special-interest political agenda.

Simon claims to be the first economist to quantify the costs and benefits of immigration because he has brought together a scientific approach and a solid economic-statistical base for determining the social loss from keeping out non-white foreigners. The reader will be appalled to find that his analysis consists largely of pseudo-scientific method, overstatement, hyperbole, and contradiction.

What then are Simon's so-called scientific and economic statistical bases that prove' that immigrants, legal and illegal, bring untold economic benefits to American shores? The best way to illustrate Simon's methods and proofs is to offer several typical quotes from his work

'Immigrants have a high propensity to start their own businesses; this seems obvious to the casual observer. For example, the proportion of taxi drivers (often self-employed) who have foreign accents seems high from New York to Melbourne to Molmo, Sweden.'

'There does not seem to exist a reliable study of immigrant business openings for the U.S. (though it is badly needed).'

Simon uses a similar method in proving that immigrants save more, work harder, and are more innovative than natives

'...regrettably this research has not yet been done. Therefore, I will be able to do no more than present some scraps of unrefined evidence from other countries' experience.'

In effect, Simon offers absolutely no systematic evidence for the majority of his assertions. He is very critical throughout his book of his detractors who use anecdotal accounts and causal empiricism as evidence. But what is sauce for the goose is not sauce for the gander for Simon himself freely uses newspaper accounts and anecdotes and engages in causal empiricism from cover to cover.

Simon also shows his inconsistency when he turns to the all-important question of whether immigrants displace native workers. He concludes that immigrants do not take jobs; they make jobs. Along the way. however, he cites a number of empirical studies that conclude that immigrants do displace native workers. For instance, he relies heavily on a study by Larry C. Morgan and Bruce L. Gardner (1982), which shows that the introduction of 210,000 Mexican braceros led to a fall in migrant farm labor wages of 9 percent as well as a reduction of 51,000 jobs previously held by natives. According to my simple math, this is a displacement rate of 24.3 percent. Yet Simon later concludes that empirical studies reviewed in this chapter suggest that general immigration causes little or no unemployment.

The Urban Institute's study of Hispanicized urban labor markets in Los Angeles by Thomas Muller (1984) is also frequently cited by Simon as providing proof positive that immigrants are a windfall benefit. Muller seems to say that the very abundance of cheap Mexican and Central American labor helps retain low-standard, assembly-line industries in Los Angeles County, whereas otherwise, such industries (including clothing sweat shops) could migrate abroad, presumably to Mexico's cut-rate border zone, to Taiwan, Korea, or Haiti. Thus, jobs are saved.

Yet Simon overlooks Muller's empirical findings that make a strong case against laissez-faire and open-border immigration. True, low cost labor does subsidize the economy of L.A. County, but there are numerous adverse social effects. As Muller himself emphasized, Mexican-Americans and unskilled immigrants already in Southern California had to absorb much of the adverse impact, such as depressed wages and housing shortages.

Muller labeled the almost one million immigrants who settled in the L.A. metro area in the 1970s as a success story. But how about the one million other people who left the county in the meantime? Workers who were unwilling to accept lower wages tended to leave the region. Other people, who could not adjust to the spreading Third World immigrant subcultures and the Hispanization of public schools (where currently over 50 percent of the students are now of Mexican or Hispanic origin), left as part of the white flight. Muller himself was seriously concerned about the future socio-economic implications of the high dropout rate of school-age Hispanics of rural peasant background.

Simon has also claimed that immigrants pay more in taxes than they use in welfare services. He ignores refugees, each of whom costs the American taxpayer at least $7,000 per year. His rationale refugees do not enter the U.S. under normal immigration quotas. Hence they are not immigrants. Even so, he is not able empirically to prove his case. The one extensive study he does depend on, by Gilberto Cardenas and Sidney Weintraub, is only for Texas data, and they ignore such large costs as food stamps, welfare outlays, and unemployment of displaced legal workers.

Borjas' major findings deal a death blow not only to Simon's claim that immigrants benefit the public coffers, but to nearly every one of Simon's major hypotheses. Borjas is able to show that the quality of recent immigration waves to the United States has greatly deteriorated, with less schooling, weaker labor market attachment, higher unemployment rates, lower wage rates, higher poverty rates, and higher rates of welfare use than the waves that arrived in the 1950s and 60s. Nor have recent immigrants performed as well as natives in U.S. labor markets. Borjas finds that this decline in quality costs American society at least $6 billion per year. In present value terms, the immigrants of the late 1970s alone are costing us $110 billion with a $27.5 billion corresponding loss in tax revenue.

Borjas overturns Simon's entire equation by reversing Simon's assertion that immigrants are more productive than natives. Borjas concludes that average immigrant earnings never catch up with native earnings. Hence immigrants will also pay less in taxes than natives. Moreover, immigrants on average use more, not less, welfare than natives. According to Borjas, the longer female-headed immigrant households live in the United States, the more their welfare usage increases, until after thirty years it exceeds that of the natives by 10 percentage points.

Why did Simon find continuously improving immigrant productivity and upward mobility whereas Borjas found the opposite? The answer is simple. Simon's data are 20 years old and he did not bother to update them, despite the fact that he has read a 1985 study in which Borjas originally documented the decline in immigrant quality. Simon has dismissed the study -- saying that it would take years to arrive at a satisfactory understanding of the phenomenon.

Borjas also reverses Simon's generalization that immigrants make rather than take jobs. According to Borjas, a doubling of the number of immigrants in the local labor market reduces the native wage by about 2 percent; a 10 percent increase in the number of immigrants reduces the labor force participation of white natives by .1 percent and reduces the number of weeks worked by .3 percent, not affecting the native unemployment rate.

The major problem with Borjas' analysis of labor market effects is that it focuses on legal and not illegal immigrants. Simon has always blurred such differences. In my opinion they should be treated independently, if only because they have differential labor market effects. Legal immigrants are much less likely to displace natives than are illegal alien workers, because it is the illegals who often provide unfair competition for natives and immigrants by working for cash off the books, either paying no taxes or underpaying taxes by getting excess deductions for dependents living abroad, and by undercutting legal wage and safety standards. My own 1982 and 1985 studies of displacement in the Houston metropolitan area indicate that for every 100 illegal alien workers, 70 legal workers are displaced or not able to obtain employment.

Despite his somewhat cavalier treatment of the illegal alien problem, Borjas has brought a number of other important insights to current immigration policy, by far the most important of which is significant decline in the quality of immigrants since the 1950s and 60s, including, of course, the three million immigrants legalized recently by the 1986 Immigration and Reform Act. It is hard to argue with Borjas' conclusion that these recent immigrant waves are placing a much greater burden on the American welfare state.

Borjas believes it is imperative, therefore, for the United States to raise the quality of future immigrants. Simon, surprisingly enough, agrees with Borjas on this point. As good neoclassical economists, both Borjas and Simon want to sell scarce U.S. greencards to the highest bidder. Borjas realizes, however, that the goal is to bring in high quality immigrants, and this may be inconsistent with selling to the highest bidders, who, once they are here, may turn out to be unproductive. Borjas rightly admits that any visa allocation system is inherently discriminatory. For instance, the 1965 Immigration Act emphasized family reunification (and it was largely responsible for falling immigrant quality). It discriminated against the educated and skilled who had no family connections here.

What I find puzzling, particularly about Borjas' final conclusions, is that he does not seem to recognize the major alternative option -- reduce the number of low quality immigrants who are so costly to the United States by cutting immigration quotas below the current 750,000 a year level. Do this by limiting family reunification to spouses and children rather than, as now, inclusion of parents and brothers and sisters. Then determine solely by labor-market criteria just how many skilled immigrants, if any, will productively benefit the U.S. economy without harming the millions of potentially upwardly mobile American workers currently working at low level jobs who would benefit from tight labor markets. From past experience, we know that only when there are shortages of labor do American employers, contractors and recruiters break through the barrier of structural unemployment to hire and help train minority workers.

About the author

Donald Huddle, Ph.D., is a professor of economics at Rice University in Houston, Texas.