1 The survey is considered such an accurate source of information on the foreign‑born because, unlike the decennial census, each household in the CPS receives an in‑person interview from a Census Bureau employee. The 211,000 persons in the Survey, almost 24,000 of whom are foreign-born, are weighted to reflect the actual size of the total U.S. population. However, it must be remembered that some percentage of the foreign-born (especially illegal aliens) are missed by government surveys of this kind. Thus, the actual size of this population is almost certainly larger. Of course, this was also true in past years as well.
2 This includes naturalized American citizens, legal permanent residents (green card holders), illegal aliens, and people on long-term temporary visas such as students or guest workers, but not those born abroad of American parents or those born in outlying territories such as Puerto Rico.
3 Figure 1 reports the number of immigrants living in the country in 1995 through 2005 from the March CPS. The data for 1995 to 1999 were originally weighted based on the results of the 1990 Census carried forward. This was also true for the March 2000 and 2001 CPS. After the 2000 Census, which was conducted in April, the Census Bureau re-weighted the March 2000 and 2001 CPSs based on the results from the 2000 Census. This had the effect of increasing the size of the foreign-born population in the March 2000 CPS by 5.659 percent. While the Census Bureau has not re-weighted the 1995 through 1999 CPSs, it is reasonable to assume that the undercount was similar in those years. If we adjust the 1995-1999 March CPS by the same amount it produces the results found in Figure 1.
4 If the original weights (based on the 1990 census) are used for the 1996 through 2000 data, then the foreign-born population grew from 23 million in 1995 to 28.38 million in 2000 -- 5.38 million. This is less than the 5.7 million growth reported for this time period shown in Figure 1.
5 Unlike deaths, out-migration may or may not rise with the size of the immigrant population. Also, unlike deaths, it can fluctuate from year to year. While the potential pool of return migrants obviously grows as the immigrant population grows, this does not necessarily mean that more will chose to go home, or in the case of illegals be forced to do so. Put simply, out-migration usually is voluntary and can fluctuate; deaths, on the other hand, are not voluntary and therefore occur at a predictable rate. This does not mean that out-migration cannot be estimated. See http://www.census.gov/population/documentation twps0051/twps0051.pdf
6 In order to preserve anonymity, the Census Bureau groups several different years of arrival together in the public use CPS. In the March 2000 CPS, those who arrived from 1998 through March 2000 are one group, those who arrived in 1996 and 1997 are another group, and those who arrived in 1994 and 1995 are still another. We split those respondents who arrived 1994 and 1995 to estimate the number of immigrants who came into the country from 1995 to 2000 so that we can compare two different five year periods -- 1995 to 2000 versus 2000 to 2005. But even if we did not do this, the total number who entered in the six years 1994 to 2000 is still less than number for the five years 2000 through 2005.
7 The report by the Pew Hispanic Center, "Rise, Peak, and Decline: Trends in U.S. Immigration 1992-2004," is at www.pewhispanic.org/reports/report.php?ReportID=53 .
8 This quotation is from a September 26, 2005, email from Jeff Passel who is the lead author of the Pew report. The report does say this less clearly on page 23, but most readers, including some in the media who have reported on his study, mistakenly think the numbers in his tables show levels and not just trends. But Dr. Passel has been very clear in several personal conversations that the actual level of immigration is higher than that shown in the tables of his study. This decision to report figures that are trends and not actual numbers of new arrivals makes the study more difficult to evaluate. For example, it is not possible to directly compare all the numbers in "Rise, Peak and Decline" to a study Pew published just three months earlier entitled, "Unauthorized Migrants: Numbers and Characteristics," even for the same years.
9 The weighting of data is a complex procedure. At a basic level, each person in the survey is assigned a weight, which is suppose to be the actual number of people with the same demographic characteristics the individual represents in the total population. These characteristics include things like race, age, gender, and whether someone is Hispanic. The variables used to weight the data are often correlated with being foreign-born, thus it is not so surprising that the CPS or ACS will show roughly the same level of immigration. It should be noted that being an immigrant is not one of the variables used to weight the data.
10 The number of inconsistent responses are most pronounced in the years with the largest variation in new arrivals. The reason for this inconsistency is not clear. Another problem with this question is that the level of immigration it implies is far too low. On average, the "where did you live last year" question shows that about 1.1 million immigrants came to the United States each year from 2000 to 2005. Not only is this completely inconsistent with the 1.5 million implied by the year-of-arrival question from the March 2005 CPS, (and other March CPSs), but it is also completely incompatible with the growth in the foreign born -- 5.2 million from 2000 to 2005. As was discussed earlier, at least 500,000 immigrants die or go home each year. Thus, if only 1.1 million entered, it would not be possible for the population to grow by over 1 million a year. In contrast, if 1.5 million new immigrants enter each year, as the year-of-arrival question indicates, then a growth of 1 million makes perfect sense.
11 This problem can been seen when we examine the growth that occurred between 2002 and 2003. The Pew study found that over the 1995 to 2005 period, 2002 and 2003 had the lowest number of new arrivals. Yet as Figure 1 shows, growth in the foreign-born between 2002 and 2003 was one million, higher than some of the years in the 1990s and an amount consistent with the average growth over the entire 1995 to 2005 period. There are also some years in the 1990s when growth does not seem to match the flow numbers found in the Pew study. This almost certainly reflects the sampling variability that occurs in any survey, which is why it is so hard to come to a firm conclusion about changes in the flow of immigrants.
12 Comparisons between ACS and CPS results are difficult because the ACS asks respondents when they came to "live" in the United States, while the CPS asks when they came to "stay." Partly as a result of question wording, the number of new arrivals from the ACS and CPS do not always match. Pew ignores the wording difference and averages the results together to get its flow estimates. Other issues include the advertising associated with the 2000 Census, which almost certainly increased response rates to the CPS and ACS among hard-to-count groups, such as the foreign-born. This may create the illusion of a spike in immigration around 2000 in the surveys. Another issue with both surveys is the well-known tendency of respondents to give a round number, such as the year 2000, when asked a question like, "when did you came to the United States?" Researchers often refer to this problem as "clumping," and this too can create the illusion of a spike in 2000. There is evidence in both surveys of this problem. One advantage with the ACS is that it does not combine answers to the year-of-arrival question into multiple-year groups as is done in the CPS, so it is possible to look at the number of immigrants who arrived in an individual year. Individual year analysis does show evidence of a higher level of immigration from 1999 to 2001 and a fall-off in 2002 and 2003. (Data from the ACS for all of 2004 are not yet available.) But the individual year-of-arrival data also show a lot of variation. For example, in the 2002 ACS 1.56 million said they arrived in the year 2000, but in the 2003 ACS 1.75 million said they came in 2000. The number should go down over time as immigrants who came in that year die or go home and theoretically it should never go up, and certainly not by nearly 200,000.
13 The September 11 attacks may have slowed illegal immigration because prospective illegal aliens may have mistakenly thought immigration laws were about to be enforced. Or maybe it made them less willing to respond to the survey. Moreover, the immigration service itself has acknowledged that they processed fewer applications for legal status in the years immediately after the attacks. Thus, even if immigration slowed, it may have had nothing to do with the economy.
14 One way we know this is that the March 2005 CPS showed that 4.3 million immigrants entered in 2002, 2003, 2004, and the first part of 2005. In contrast, the March 2004 CPS showed 2.6 million arrivals in 2002, 2003, and the first part of 2004. To get an implied immigration rate, we subtract the 2004 number from the 2005 number to get 1.6 million
15 To determine who are legal and illegal immigrants in the survey, this report uses citizenship status, year of arrival in the United States, age, country of birth, educational attainment, sex, receipt of welfare programs, receipt of Social Security, veteran status, and marital status. We use these variables to assign probabilities to each respondent. Those individuals who have a cumulative probability of 1 or higher are assumed to be illegal aliens. The probabilities are assigned so that both the total number of illegal aliens and the characteristics of the illegal population closely match other research in the field, particularly the estimates developed by the Urban Institute. This method is based on some very well-established facts about the characteristics of the illegal population. For example, it is well known that illegals are disproportionately young, male, unmarried, under age 40, and have few years of schooling. Thus, we assign probabilities to these and other factors in order to select the likely illegal population. In some cases we assume that there is no probability that an individual is an illegal alien.
16 The INS report estimating seven million illegals in 2000 with an annual increase of about 500,000 can be found at www.immigration.gov/graphics/aboutus/statistics/Ill_Report_1211.pdf . The Census Bureau estimate of eight million illegals in 2000 can be found at www.census.gov/dmd/www/ReportRec2.htm (Appendix A of Report 1 contains the estimates).
The Pew Hispanic Center has estimated 10.3 million illegals from the March 2004 CPS. This includes an adjustment for those missed by the survey. The Pew report can be found at www.pewhispanic.org/files/reports/46.pdf . The Urban Institute has also done estimates by legal status. It estimates that in March 2002, 8.3 million illegal aliens were counted in the CPS, with an additional one million being missed. Urban's estimates based on the March 2002 CPS can be found at http://www.urban.org/url.cfm?ID=1000587 . Additional information was provided by Jeffery Passel, now at the Pew Hispanic Center, in a May 24, 2004, telephone interview. Dr. Passel is the lead author of both the Urban Institute and Pew studies.
17 Table C in the INS report on illegal immigration shows the number of non-IRCA legalizations in the 1990s. It can be found at http://uscis.gov/graphics/shared/aboutus/statistics/Ill_Report_1211.pdf
18 It should be noted that the 5.2 million figure compares the March 2000 CPS and March 2005 CPS. The 2000 Census showed 31.1 million immigrants. But that figure includes persons in group quarters, such as prisons and nursing homes, who are not counted in the CPS. It also was conducted in April 2000, not March.
19 See Robert Warren and Ellen Percy Kraly, 1985, "The Elusive Exodus: Emigration form the United States." Occasional Paper #8. Population Reference Bureau, Washington, D.C.
20 So that comparisons can be made between 1995 and 2000 and 2005, we have attempted to adjust figures for 1995 to reflect the results of the 2000 Census. However, the effects of these adjustments are small. For more discussion of weighting prior to and after the 2000 Census see End Note 1.
21 This figures refers to persons aged 18 or older who are in the workforce. That is, they are either employed or actively looking for work.
22 It should be noted that year of arrival data is grouped in the CPS to preserve the anonymity of respondents. Table 7 reports figures in as detailed a manner as possible given this grouping.
23 See for example the Urban Institute study, "Immigration and Immigrants: Setting the Record Straight," which can be found at www.urban.org/publications/305184.html#III
24 Programs included are TANF, SSI, general assistance, Food Stamps, public/rent subsidized housing, WIC, and Medicaid.
25 Of immigrants who did not have a high school degree and had lived in the country 20 or more years, one-fifth lived in poverty and 57 percent lived in or near poverty. Both rates are substantially above that of natives.
26 See page 21 of the Census Bureau's "Methodology and Assumptions for the Population Projections of the United States: 1999 to 2100" The report can be found at www.census.gov/population/www/documentation/twps0038.pdf
27 "Immigration in an Aging Society: Workers, Birth Rates, and Social Security" can be found at www.cis.org/articles/2005/back505.html .
28 We report poverty for children born in the United States who have immigrant mothers and are under the age of 18. This has the effect of counting children who have two parents who are foreign-born or just those whose mother is foreign-born. Those who have only foreign-born fathers are not counted. In this way we avoid double counting. It should be noted if we report figures for children with two foreign-born parents or just an immigrant father the results are very similar.
29 Figures are for children with immigrant mothers. See footnote 16 for more detail.
30 The article is entitled "Health Care Expenditures of Immigrants in the United States: A Nationally Representative Analysis," by Sarita A. Mohanty, Steffie Woolhandler, David U. Himmelstein, Susmita Pati, Olveen Carrasquillo, and David H. Bor in the-American Journal of Public Health. August 2005, August 2005.
31 Figures are for children with immigrant mothers. See footnote 16 for more detail.
32 See for example Figures 20-1, 20-2 and 20-3 in Profiles of the Foreign-born Population in the United States 2000, U.S. Government Printing Office. Dianne A. Schmidley, Series P23-206.
33 The Census Bureau released the 2005 CPS without figures for the EITC, thus the figures for the Credit are from the 2004 CPS.
34 The primary refugee sending countries that can be identified in the CPS are Poland, Yugoslavia, Afghanistan, Cambodia, Iraq, Laos, Vietnam, Nicaragua, Cuba, Ethiopia, and those who said they were from the USSR, Russia, or Ukraine.
35 For a discussion of the decline in immigrant education relative to natives see, "The Slowing Progress of Immigrants: An Examination of Income, Home Ownership, and Citizenship, 1970-2000," at www.cis.org/articles/2001/back401.html.
36 See End Note 9.
37 To estimate poverty for illegals and their U.S.-born children, we calculate poverty for illegal immigrants and then for children born in the United States in illegal families. In total, we estimate that there are three million U.S.-born children of illegals in the March 2005 CPS.
38 In a report published last year, we found that 89 percent of households headed by illegal aliens had at least one worker. The report entitled, "The High Cost of Cheap Labor: Illegal Immigration and the Federal Budget," can be found at www.cis.org/articles/2004/fiscal.html.
Click Here to support Frontpagemag.com.