July 22, 2009

Mexican Immigrants: How Many Come? How Many Leave?

III. Fewer Mexican Immigrants Are Arriving

Immigration from Mexico to the U.S., especially unauthorized immigration, began to drop off in mid-2006, and that pattern has continued into 2009, according to population surveys in both countries and U.S. enforcement data.

By the period spanning March 2008 to March 20091, the estimated annual inflow of immigrants from Mexico—about 175,000 as estimated from the Current Population Survey—was lower than at any point during the decade and only about half of the average for the previous two years (Table 2).

Annual immigration from Mexico has risen and fallen several times during the decade, according to CPS estimates. For example, immigration dropped by about one-third, from 570,000 for March 2000-March 2001 to an estimated 397,000 for March 2002-March 2003. For the three-year period of March 2003-March 2006, Mexican inflows nearly regained their previous levels and averaged about 550,000.2 Since then, immigration from Mexico has decreased substantially, dropping almost 40% to an annual average of about 350,000 for March 2006-March 2008 and continuing with the sharp decline noted for the most recent year.

Mexico’s National Survey of Employment and Occupation (ENOE), which asks questions of each household in its sample about people who departed for other countries, shows a similar pattern. The flow out of Mexico, more than a million for February 2006-February 2007, declined by more than 20% to about 814,000 for the same period in 2007-2008. It decreased by another 20% to about 636,000 for the same period in 2008-2009 (Table 1).

Although the changes over time are similar to those shown in the CPS, the flow levels reported by the Mexican survey are quite a bit higher because the two surveys are not measuring the same group of migrants. The CPS is designed to measure people whose principal residence is in the U.S. and who are settled on a long-term or permanent basis. The Mexican survey, meanwhile, provides estimates for a broader group of migrants. They include Mexicans who come to the U.S. for short periods and may return home within weeks or months. Some people counted as leaving Mexico in one quarter may be included in the count of returnees in a subsequent quarter. Those “circular” migrants may not appear in the CPS.

ENOE, the Mexican survey, measures migration flows on a quarterly basis, but because migration to and from the U.S. is highly seasonal, it is not appropriate to track quarter-to-quarter changes in the number of Mexicans leaving for the U.S. However, annual data show that number of people leaving Mexico has declined each quarter on a year-to-year basis.

Apprehensions by the U.S. Border Patrol of Mexicans attempting to enter the United States illegally show a pattern very similar to that in the CPS and ENOE data. The number of apprehensions declined by about one-sixth from fiscal 2006 to fiscal 2007 followed by a similar percentage decline in 2008. By fiscal 2008, the number of Mexicans apprehended by the Border Patrol—662,000—was 40% below the mid-decade peak of 1.1 million in 2004.

The total number of apprehensions in 2008—724,000—was at the lowest level since 1973. More than 90% of people detained by the Border Patrol are Mexican.

Apprehensions by the Border Patrol are not a direct measure of immigration for a number of reasons. First, apprehensions include an unknown number of people detained more than once. Second, they represent only the people prevented from entering and not those who are successful. Finally, to some degree the number of apprehensions is a function of how many agents the Border Patrol places at the border and how successful they are at apprehending clandestine border crossers. The Department of Homeland Security, which oversees the Border Patrol, cautions that “the relationship between the number of border apprehensions to either the number of attempted illegal entries or the number of successful illegal entries is not known.”

The Border Patrol attempts to stop illegal entries to the U.S. and does not generally apprehend unauthorized immigrants who are leaving the country, so the data address only inflows to the U.S. Nonetheless, the apprehensions data provide an indicator of the magnitude of the flow across the border that tends to rise and fall with the number of successful entries and with immigration levels. The record year for apprehensions was 1986, just before enactment of the Immigration Reform and Control Act, which allowed several million unauthorized immigrants to legalize their status and instituted stricter enforcement.

These data do not address the reasons for the drop in apprehensions. A Department of Homeland Security fact sheet suggests that the decrease could be due to factors including the weakened U.S. economy as well as stepped-up border enforcement. The threat of being caught could discourage some would-be immigrants from attempting to enter the U.S. Some scholars suggest that stepped-up enforcement also could discourage unauthorized migrants from leaving the U.S. for home visits, because they would risk capture when they tried to re-enter.

  1. numoffset=”3″ Measures reported here are based on differences in recent arrivals from Mexico as measured by CPS averages of January-April from one year to the next. So the results from January-April 2009 CPS measure immigration from roughly March 1, 2008, to March 1, 2009.
  2. Because of the range of error for individual flow estimates calculated from population differences, the three annual flow estimates for 2003-2004, 2004-2005 and 2005-2006 are not significantly different from one another. Thus, we report based on their average.