Getting the Number

The estimate of unauthorized workers is derived by using a variant of a basic “residual” method. The method relies on exclusion. In other words, the unauthorized population consists of persons and groups not included in the authorized population. To reach that number, the first step is to develop an estimate for the legal, foreign-born population. That estimate is based on admissions into the country provided by the Department of Homeland Security (DHS) and its predecessor, the Immigration and Naturalization Service (INS), as well on the number of refugees admitted and the number of asylum applications granted.

After allowing for legal temporary migrants and for legal immigrants missed in the Census or the yearly March Supplement to the CPS, the initial estimate of the unauthorized population covered in the survey is derived by subtracting the estimated legal population counted from the Census or CPS figure for the foreign-born population. The initial number of unauthorized migrants is then adjusted upward to account for omissions. All calculations are done by country or region of birth, age, sex, period of arrival and state or region of residence.

The residual method has been used for several decades to measure unauthorized migration to the United States. Specifically, some of the first sound empirical estimates (Passel 1986 and Warren and Passel 1987) were arrived at by applying residual methodology to the 1980 Census. Variants of the residual method have been used or discussed by the Census Bureau, the Panel on Immigration Studies, the Bi-National (U.S.-Mexico) Study, the Commission on Immigration Reform, the Immigration and Naturalization Service, and several other organizations and researchers. Recently, the DHS has released estimates using a similar methodology (Warren 2003; also, Warren 2001 for earlier variants).

Although the residual method has been applied to previous March CPSs (e.g., 2004—Passel 2005; 2003—Capps et al. 2005; 2002—Passel, Capps, and Fix, 2004), the statistical properties of the residual estimates make it very difficult to use estimates from successive CPSs to measure year-to-year changes. Specifically, most of the sampling variance of the total foreign-born population is part of the variance of the residual estimate of unauthorized migrants. This property means that the variance of the measure of annual change is quite large. Nonetheless, the series of annual estimates provides indications of overall trends and the difference in unauthorized population estimates separated by several years can provide information on average annual change. In this report, we use a residual estimate based on Census 2000 together with the CPS-based estimate for March 2005 to assess change over the five-year interval.

  • Persons who arrived before 1980 are all assumed to be legal by 2000.
  • Persons who are already in legal status in the US and “adjust” to legal permanent residence. These are added to the legal population in the year they receive their permanent residency (unless they have already been counted in other groups listed here); however, they are counted as “arriving” in the U.S. in the earlier when they arrived in their previous legal status.
  • “New Arrivals,” or persons getting green cards as they enter the U.S. These are counted in the year they arrive (unless they are already counted in other groups listed here)
  • Persons who acquired legal status under the Immigration Reform and Control Act of 1986 (IRCA) are included as legal when they obtain their legal permanent residency; their year of arrival is based on survey data of this group.
  • Cubans, Haitians and other entrants, Amerasians and various groups of parolees are included as legal when approved and not when they received their legal permanent residency. The same applies to those who receive asylum
  • Refugees are counted in the year they arrived in the U.S. and not when they receive legal permanent residency.

See Passel, Van Hook, and Bean 2004, 2006 for a more detailed explanation of the methodology.

Who is Unauthorized

Virtually all unauthorized migrants fall into two categories: those who overstayed their visas or those classified by the government as “entries without inspection,” or EWIs. Visa overstayers likely represent between 25% and 40% of the unauthorized migrants.

There are also various administrative categories that are not included in the estimated legal foreign-born population. Many people are classified as having Employment Authorization Documents, or EADs, that are issued by the Department of Homeland Security and thus could be considered “authorized” in some sense.

There are an estimated 1 million to 1.5 million people who fall into the unauthorized category but are under several “quasi-legal” categories. Some examples include persons with Temporary Protective Status, or TPS, and those with Extended Voluntary Departure, or EVD, as well as those applying under these statutes. Together, these migrants may account for 300,000 to 400,000 unauthorized persons.

Another 250,000 persons have applied for asylum but have not had their cases adjudicated. Although some may have their applications approved and may eventually receive legal permanent residency, many and perhaps most will not. A third group not included in the estimate of legal residents is made up of persons who have employment authorization and are waiting for “green cards” or LPR status, but whose status is not yet final. There are about 600,000 people in the U.S. who have applied for legal permanent residency but are waiting for them to be issued. And finally, there are about 100,000 persons who are immediate relatives or fiancées of legal residents who are awaiting their legal permanent residency. Most in these groups will eventually receive permanent legal status.

The inclusion of the quasi-legal group with the unauthorized or their exclusion as legal depends in part on the availability of data to estimate their numbers and in part on the ultimate use of the estimates. The data for these groups are problematic. In particular, the numbers cited above are approximations; the required demographic detail is generally lacking; and the various categories overlap to some unknown (and possibly substantial) degree. For use in assessing potential programs for dealing with the unauthorized population, it seems appropriate to treat the quasi-legal group as part of the unauthorized program since a significant share of this group would probably be eligible to participate in any program that might lead to regularization of their status, such as a temporary worker program or an earned legalization program. (For an opposite view, see Martin 2005.)

Augmentation of the CPS

The CPS data presented here are largely based on tabulations of augmented files from the March 2005 CPS and Census 2000 5-percent PUMS. The augmentation encompasses assignment of probably legal status to individual migrant cases in the survey and adjustment for omissions. The data and techniques employed were developed initially at the Urban Institute by Passel and Clark (Passel and Clark 1998 and Passel, Van Hook, and Bean 2006 for description of the various methods used.)

The methods involve estimating the number of unauthorized migrants with the techniques described previously. In this process, the CPS data are first corrected for over-reporting of naturalized citizenship on the part of aliens. Then, persons entering the U.S. as refugees and individuals holding certain kinds of temporary visas (including students, diplomats, and “high-tech guest workers”) are identified in the survey and assigned an immigration status using information on country of birth, date of entry, occupation, education, and various family characteristics. Individuals that are definitely legal and those that are potentially unauthorized are identified in the CPS (based on state of residence, age, sex, occupation, country of birth, and date of entry). Next, using probabilistic methods, enough of the potentially unauthorized are selected and assigned to be unauthorized so as to hit the estimated numbers of legal and unauthorized migrants included in the survey. This last step, which assigns more than three-quarters of the potentially unauthorized as unauthorized, involves a consistency edit to ensure that the family structures of both legal and unauthorized populations “make sense.” The whole process requires several iterations to produce estimates that agree with the demographically-derived population totals. Finally, the survey weights for the foreign-born are adjusted upward so that the tabulated figures agree with the analytic, demographic estimates of the total number of legal and unauthorized migrants developed in the very first step.