Mexican and Central American immigrant communities in the boom and bust housing markets of the 2000s

Abstract

Classic immigrant enclaves, largely voluntary and temporary, have historically served as a strong platform for integration in United States metropolitan areas. However, trends in the early 2000s, including new destinations and skyrocketing housing costs, may have reshaped the landscape, particularly for numerically dominant immigrant groups with very low socioeconomic status. In this article, I use data for 56 metropolitan areas and 31,563 census tracts from the 2000 Census and the 2005–2009 American Community Survey to examine the relationship between the change in concentration of Mexican and Central American immigrants and the characteristics of neighborhoods in 2000. The analyses suggest that, across metropolitan areas, these immigrant communities consolidated in neighborhoods with low home values, adequate but overcrowded housing, relatively small shares of their own immigrant group, and large shares of other Hispanics. Demographic dynamics may have weakened immigrant support networks, leaving Mexican and Central American immigrants vulnerable, particularly in a subset of metropolitan areas where this population grew most substantially outside the central cities.

Keywords:

• immigration
• housing
• enclaves
• integration
• segmented assimilation

Acknowledgments

The author would like to thank Dr Rolf Pendall for his guidance and support conceiving this manuscript, as well as Rebecca Grace for her work assembling the data for analyses; Juan Collazos, Priya Saxena, and Alma Acosta for their work reviewing the literature; and Jessica Luna for her assistance preparing the final manuscript.

Notes

1. Migration Policy Institute tabulation of data from the U.S. Census Bureau’s 2012, 2011, and 2010 American Community Surveys and 1970, 1990, and 2000 Decennial Census data. All other data are from Gibson and Lennon (1999).

2. After the introduction of the American Community Survey (ACS), detailed tables on the origins of foreign-born were no longer available from the Decennial Census. The ACS has considerably greater margins of error, but is the only source of data equipped to capture the population on which these analyses focus.

3. Data on the Mexican and Central American population for census tracts in 2000 come from the PCT019 table in the SF3 file for the 2000 Decennial Census. The corresponding numbers at the later time point were extracted from table B05006 from the 2005–2009 American Community Survey five-year estimates.

4. A total of 6,120,769 of these immigrants were identified across the tracts in these 50 metros including 4,917,935 Mexicans; 48,114 Costa Ricans; 480,370 Salvadorans; 240,435 Guatemalans; 194,475 Hondurans; 168,292 Nicaraguans; and 71,148 Panamanians.

5. The Census defines crowded households as those with more than 1.01 persons per room.

6. Complete plumbing facilities are defined as hot and cold piped water, a bathtub or shower, and a flush toilet.

7. This article employs 2005 definitions of metropolitan areas. As a result, there are no tables from the 2000 Census displaying the median value with this geography. As a result, the median value for each metropolitan area was derived by calculating the median of all tracts’ median values, as weighted by the number of housing units in each tract. For consistency, the same technique was utilized to calculate the metro-level median value for the time period 2005–2009.

8. Issues with large margins of error in the American Community Survey are well documented, particularly with small geographies like census tracts. This error results from relatively small sample sizes, the mixed-mode sample design in the ACS (which uses sampling in the in-person follow-up), and a lack of population controls for smaller geographies, which makes the calibration of the weights difficult.

Additional information

Funding

This work was supported by the John D. and Katherine T. MacArthur Foundation [10-97329-000HCD].