tion to them, for purposes of offsetting high per inmate costs of incarceration. These statistics often are used to make the point that the numbers of immigrants in prison are substantial, and that a large proportion of immigrants who are in prison for drug charges are from Mexico. However, it is also the case that increasing numbers of immigrants have been coming to the United States from Mexico to work for employers who are eager to have them as employees (Calavita, 1994). These immigrants are disproportionately young and male, and therefore of an age and gender for whom crime is a relatively common and frequently transient experience, regardless of citizenship. So a first concern is to take into account not only the increased immigration that is occurring from Mexico, but also the age and sex distribution of these immigrants in considering their numbers in prison. When we do this we find that Mexican immigrants are found in state prisons at an adjusted rate that is not strikingly different from U.S. citizens. Of course, this does not mean we should be unconcerned about the growing numbers of Mexican immigrants in prison. The costs of this imprisonment are high and a concern with the absolute volume of crime involved is understandable. The adjustments for gender and age that we have introduced are intended to provide a context for understanding the numbers and costs involved.

At the same time, it is also important to take into account that immigrants from Mexico and elsewhere may be subject to differential treatment in the criminal justice system. For example, we have demonstrated that immigrants in El Paso and San Diego are at greater risk of being detained prior to trial, and that this results in their increased likelihood of being convicted and imprisoned. Also, we have shown that immigrants in these cities who are charged with drug offenses are at an elevated risk of being sentenced to prison. The overall implications of these findings will not be certain, however, until more is learned about potentially offsetting practices that result from diverting immigrants from the criminal justice system and deporting them from the country.

Another source of public concern is that immigrants, and especially illegal immigrants, are a source of drug problems in the United States. However, arrest records in cities such as El Paso and San Diego suggest that illegal immigrants are less likely than citizens to be involved in drug crime, and instead that they are most distinctively involved in property crime. This kind of petty property offense activity is consistent with the picture of offending that Freeman (1996) has suggested in his foraging model of crime. That is, young male illegal immigrants may be most likely to become involved in petty property crime as they attempt to satisfy basic subsistence needs while moving through the early stages of seeking, finding, losing, and regaining employment.

Overall, we did not find consistent evidence in macro- or micro-level data that immigrants are much more likely than citizens of similar ages and gender to be involved in crime. In particular, we have found that the image presented in prison statistics of the largest group of current immigrants to the United States,