Many of these age profiles for specific programs are interesting in their own right, showing striking differences across generations. We present 5 age schedules representing aggregates of 12 of the 25 individual programs: OASDHI (Social Security benefits including Medicare), Medicaid (including costs for chronic care), SSI, public assistance (AFDC, general assistance, food stamps, and earned income tax credit), and public education (including elementary school, secondary school, and colleges and universities, and bilingual education). Disaggregation by time in the United States (not shown) adds further interesting detail. For example, an immigrant aged 70 years who has been in the United States for 30 years is most likely to be receiving benefits from Social Security and Medicare but is less likely to receive SSI and Medicaid. An immigrant aged 70 years who has been in the United States for only 10 years will probably not qualify for Social Security or Medicare, and so will be more likely draw benefits from SSI and Medicaid.

In addition to these benefits, all of which can be allocated by the age of the recipient, there are other benefits that immigrants receive that are not allocable by age. These are discussed later.

Age Profiles for Illustrative Individual Program Benefits

Figure 7.2 shows the Social Security and Medicare benefits (OASDHI) by age and immigrant generation. Note that here and throughout, third generation refers to the entire population other than immigrants and the children of immigrants. We see that the age schedules are very similar, but that immigrants receive $1,000 to $2,000 less per year in benefits than do second or third generations. This difference presumably reflects lower average earnings throughout their lives, shorter earnings histories in the United States, and, for some immigrants arriving late in life, a failure to qualify at all for benefits.11

Figure 7.3 shows that immigrants tend to receive higher Medicaid benefits up to age 80 or so, and lower benefits thereafter. But these differences are surprisingly small, on the order of a couple of hundred dollars. The age profile reflects separate treatment of Medicaid for the noninstitutionalized population, for which information is available from the CPS, and the institutionalized population, for which we made estimates using PUMS.12 Our analysis indicates that


For both Medicare and Medicaid, the CPS indicates whether an individual is eligible for benefits, but it does not indicate whether benefits were received and, if so, their cost. We have assigned average levels of benefits to all those eligible, conditional on age.


The Public Use Micro Sample from the decennial census contains information on the foreign-born population in institutions, including both nursing homes and prisons. Above age 60, we assume that the institutionalized population is in nursing homes; below age 60, we assume that it is in prison. A logistic regression of the 1990 1 percent PUMS was used to determine proportions in institutions for the population by native-born versus foreign-born status, time since arrival for immigrants, and education. The institutionalization rates for elderly immigrants were about 60 percent as high as those for elderly others. PUMS does not give expenditures per institutionalized person, so we assumed these were the same for immigrants and nonimmigrants.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement