Immigrants and jobs.

In "Comprehensively Awful" (June 11), John O'Sullivan calls it "well known that immigration redistributes income from the poor to the rich as immigrants are willing to work for lower wages than native-born Americans." Like so many "well known" economic notions, this smacks of an abstract deductive model, which is a basic tool of economic analysis, yet must be tested against reality in each practical case.

Such a case was offered by the Mariel boatlift of 1980, when Castro allowed some 125,000 Cubans to leave the country. Of those who fled, about half settled in or around Miami, raising the labor force of the city by 7 percent. Where the abstract model of academic economics had it that the Marielitos would take jobs from blacks and depress their wages, Prof. David Card of UC-Berkeley showed that the contrary had happened.

The sudden infusion of labor afforded Professor Card what he called a "natural experiment." He checked what had happened in Miami against four "control cities" with large black populations and stable labor markets. Other economists jumped in to examine other "natural experiments," and found like results.

Whether Mr. O'Sullivan's model will prove true we can't yet say, and it will depend partly on elements such as progress or stagnation in education and religion, which are unquantifiable for the most part and can be tested only by history. The imperative point is to distrust theory until the character and culture of the people involved in each case are factored in, and their role revealed by history.

Patrick G. D. Riley

Wauwatosa, Wis.