Immigrant concentration and educational attainment: Evidence from US data

Abstract

Ninety-five percent of the 30 million foreign-born persons living in the UnitedStates reside in a metropolitan area, and more than one half reside in just six gateway cities. We investigate whether less-educated immigrants tend to settle in large urban communities of their compatriots. Based on statistical analysis we find a negative relationship between the level of education and the size of immigrant population for the immigrants born in China, India, Philippines, Korea, El Salvador, Cuba, and Mexico. The relationship is strongest for Chinese, with a decrease of 2.7 years in average educational attainment for each tenfold increase in the size of immigrant community. Generally, this relationship is stronger for the immigrants from Asia and Latin America and nonexistent or mixed for immigrants from Europe and Canada. Des 30 millions de personnes n~es a l'(tranger et maintenant aux Fztats-Unis, quatrevingt-quinze pour cent vivent dans une rdgion m~tropolitaine et plus de la moitid habitent une des six villes portails. L'dtude se penche sur l'hypoth~se que les immigrants moins instruits ont tendance h se retrouver dans Ies mfmes grandes communautds urbaines que leurs compatriotes. Une analyse statistique fait ressortir une corrdlation ndgative entre le rendement scolaire et la taiUe de la population d'immigrants pour ceux qui sont nds en Chine, en Inde, aux Philippines, en Corde, en Salvador, ~ Cuba et au Mexique. La correlation est la plus forte pour les Chinois, qui ddmontrent une baisse moyenne de 2, 7 ans sur Ie plan du rendement scolaire chaque fois que ta taille de la communaut~ d'immigrants d~cuple. De fagon gdndrale, cette correlation estplus forte chez les immigrants d'Asie et d'Am(rique Iatine. Elle est soit inexistante soit mixte pour les immigrants d'Europe et du Canada.

The Location Choice of Employment-based Immigrants among U.S. Metro Areas*

“…More detailed investigation of the location choice of immigrants from different countries has found that the better-educated immigrants tend to disperse more widely than their less-educated compatriots. Presumably higher education In the 1990s, this pattern was not as distinct as in the 1970s and 1980s since relatively more immigrants started to settle in inland cities and states (see, for example, Camarota and Keeley, 2001;Fix, Zimmerman, and Passel, 2001). allows new immigrants to be less dependent upon the safety networks provided by ethnic communities (see Bartel, 1989;Dunlevy, 1991;Kritz and Nogle, 1994;Newbold, 1996Newbold, , 1999Zavodny, 1998;Frey and DeVol, 2000;Jaeger 2000;Izyumov et al 2002). It has been further established that the secondary migration of foreign-born persons does not deviate from this pattern of settlement and, if anything, reinforces it.…”

“…However, the total number of immigrants of the same nationality seem to capture the ''ethnic networks'' effect fairly well. Izyumov et al (2002) show that the number of foreign-born persons in a metro is a good proxy for enclaves, as a majority of new immigrants from most countries live in a small number of postal zip codes of the major gateway cities of the United States. In Chicago, for example, 47 percent of Chinese immigrants reported residence in but 10 of the metros' 381 zip codes.…”

“…In other words, many Canadians may choose to live well-dispersed in a U.S. city and for reasons other than the joint economic security typically ascribed to enclaves. Izyumov et al (2002), show that the majority of new immigrants to the United States live in a small number of postal zip codes of the major gateway cities. However, of the 17 immigrant groups studied, the British and Canadians were the least concentrated.…”

Why Do Low- and High-Skill Workers Migrate? Flow Evidence from France

“…Network effects have also been subject to special scrutiny in the migration context as they are seen as alleviating migration costs (Bartel, 1989, Zimmermann, 1996 or improving employment and/or wage opportunities for newcomers (Gross and Schmitt, 2003). 3 Note that Izyumov et. al.…”