From an NBER study of data on border patrol hours invested and sectoral wages in border areas:

...for the United States, only the lumber industry in California and Texas is affected positively, and only slightly, by border enforcement. When enforcement is high, wages rise slightly. For all other industries they examine, there is no impact on the wages of poorly educated males in U.S. border regions.

In Mexican border regions, though, wages fall for poorly educated males in Tijuana when U.S. border enforcement is high. Tijuana is where a large number of immigrants attempted to enter the United States illegally over the authors' sample period.

Using monthly and quarterly data on wages, person-hours logged by the U.S. Border Patrol, and the number of apprehensions made, the authors conclude that stricter enforcement does deter illegal immigration. But illegal immigration has only a minimal effect on labor markets in U.S. border regions, probably for two reasons. First, given continuing illegal immigration, U.S. natives may leave border regions or be deterred from moving to those regions. Second, the economies of border regions may gradually shift to industries that make intensive use of immigrants' skills. The authors thus conclude that the perceived impact of illegal immigration on wages has been exaggerated.

The paper has been published (84 RevEconStat 73-92 (2002)).