Population Decline of Unauthorized Immigrants Stalls, May Have Reversed

New Estimate: 11.7 million in 2012

1. Overview

The sharp decline in the U.S. population of unauthorized immigrants that accompanied the 2007-2009 recession has bottomed out, and the number may be rising again. As of March 2012, 11.7 million unauthorized immigrants were living in the United States, according to a new preliminary Pew Research Center estimate based on U.S. government data.

The estimated number of unauthorized immigrants peaked at 12.2 million in 2007 and fell to 11.3 million in 2009, breaking a rising trend that had held for decades. Although there are indications the number of unauthorized immigrants may be rising, the 2012 population estimate is the midpoint of a wide range of possible values and in a statistical sense is no different from the 2009 estimate.

Different trends appear among the six states in which 60% of unauthorized immigrants live—California, Florida, Illinois, New Jersey, New York and Texas. Of these, only Texas had increases but no decrease in its unauthorized immigrant population over the 2007-2011 period. The other five states (and the balance of the country) all experienced peak numbers of unauthorized immigrants in 2007 followed by declines over the next year or two.

In terms of country of origin, the post-2007 population dip was even sharper among Mexicans (who made up 52% of 2012 unauthorized immigrants) than the overall population decrease, although the Mexican decline appears to have stopped after 2010. In 2012, 6.05 million Mexican unauthorized immigrants were in the U.S., a decline of about 900,000 from 2007.

Methods and Cautions

The new Pew Research estimates rely on the center’s residual methodology, in which legal immigration is subtracted from the adjusted total foreign-born population to arrive at a total for unauthorized immigrants. These estimates slightly revise previously published estimates for 2000-2011 (Passel and Cohn, 2012a) to account for new findings from the 2010 Census. This report also presents the initial, direct residual estimates that cover six individual states and aggregate the rest of the country; more detailed estimates for the remaining states and for characteristics of the unauthorized immigrant population are developed from these estimates and will be available in a later report.

These estimates also introduce a new data source for the Pew Research estimates—the U.S. Census Bureau’s American Community Survey (ACS) for 2005 to 2011, replacing the March Current Population Survey (CPS) for those years. The ACS has about 12 times more sample cases than the CPS and thus provides more precise estimates with smaller margins of error.

In analyzing recent year-to-year change, especially at the state level, this report focuses on 2005 to 2011 (especially 2007 to 2011) because the large sample size for ACS-based estimates permits more statistically sound statements about annual differences than was possible with the previously published estimates. The 2012 CPS-based estimates are derived from a much smaller sample, resulting in margins of error generally too large to draw conclusions about change from 2011 to 2012. These 2012 CPS estimates should be treated as preliminary because they will be replaced with ACS-based estimates when the 2012 ACS microdata become available later this year. (See Methodology for more details.)

Comparison with Previous Estimates Estimates presented here of the unauthorized immigrant population replace those previously published by the Pew Research Center for 2000 to 2011 (Passel and Cohn, 2012a). The figures for the foreign-born population used in these new estimates are drawn from two U.S. government surveys—the March Supplement to the Current Population Survey (CPS) and the American Community Survey (ACS). Those surveys are, in turn, tied to the Census Bureau’s estimates of the population during the survey period. At the time the surveys are released by the Census Bureau, they incorporate the most recent official population estimates. After the release of 2010 Census results, the Census Bureau issued a set of new population estimates that incorporated updated information about population change over the previous decade. The same process took place after the 1990 and 2000 censuses. However, the Census Bureau generally does not return to previously released ACS, CPS or other survey data to incorporate updated information from these revised intercensal population estimates. In the interest of greater accuracy, the Pew Research Center developed new survey estimates for the 2005-2009 ACS and the 2000-2010 March CPS that are consistent with the results of both the 2000 and 2010 censuses; for March 2011 consistent with the 2010 Census; and for March 1995-1999 consistent with both the 1990 and 2000 censuses. These new survey estimates are based on the Census Bureau’s revised annual intercensal population estimates, i.e., those that are consistent with both the preceding and subsequent censuses. Those new survey estimates are the basis for the new unauthorized immigrant numbers presented in this report. The new estimates of the unauthorized immigrant population generally show the same trends over time as the previous estimates but at slightly higher levels for some years (notably in the 1990s and second half of the 2000s). Users are cautioned that previous estimates should not be compared with those in this report.

Other Evidence

Additional evidence about trends in unauthorized immigration comes from U.S. government data on border apprehensions.

Driven by unauthorized immigration from Mexico, U.S. Border Patrol apprehensions rose through the 1990s and peaked at 1.7 million in fiscal 2000. After reaching somewhat lower levels for 2001-2007, apprehensions fell dramatically from 2007 to 2011. In 2012, the number of unauthorized immigrants apprehended at the Mexican border rose modestly, to 365,000, but only because of growing apprehensions of non-Mexicans; apprehensions of Mexicans continued to decline.