by Laurence Aurbach Jr, December 1999

Table of Contents

...Literature Review
Nationwide Geographic Patterns
...Figures 1a-c: Immigration in the 1990's
...Source Regions
...Figure 2: Sending Regions
Segregation and Assimilation
Neighborhood Quality
...Figure 3: Poverty Rate by Source Region
...Figure 4: Legal Immigration by Admission Category
Selected Sources


Immigration is a phenomenon that brings great change to nations. Americaís Third Wave of immigration, from 1880-1914, created bitter strife but also sparked a multitude of innovations. The country is now experiencing a Fourth Wave of immigration that is comparable in size and cultural impact. A key feature of the current wave is its revolutionary change in racial and ethnic composition. For most of its history, the nation has been essentially biracial with a Black population ranging from 12 to 19 percent of the total. When certain changes were made to immigration laws in 1965, huge numbers of Latinos and Asians began settling in the United States. The long-term effects of the current wave are uncertain and the debate is marked by contentious disagreement.

The geographic pattern of immigration strongly affects the social impact. Because immigrants concentrate in a few large cities, their impact is localized and disproportionate to their total numbers. The average annual immigration flow in the 1990ís was 1.1 million, up about 13% from the 1980ís (US Census, 1999). The foreign-born population made up 10% of the population in 1998; it is at a record high in absolute numbers.

Because immigrant fertility rates are higher than native rates, they contribute disproportionately to population growth. The foreign-born population is growing four times faster than the native-born (Suro 1999). Over the next 50 years, immigrants will account for two-thirds of U.S. population growth (National Research Council, 1997). In 2050, Hispanics will make up 25% of the population and Whites will become a minority. The Black proportion is projected to grow only slightly, and Blacks will be unseated as the largest minority group in 2005. The Asian proportion will grow from four to eight. The projected 2050 U.S. population of 387 million will have considerable consequences for social welfare policy (such as affirmative action), city finances and environmental quality.

Literature Review

The national debate on immigration is driven by a few fundamental questions. Does immigration have a positive impact? What level of annual immigration best for the nation and the immigrants themselves? Most of the literature addresses these concerns either directly or indirectly through socioeconomic research.

There are three types of literature. The first is large overviews of trends aggregated at the national level; these are published as major books or reports. Typical examples are studies released by the National Research Council (NRC), the U.S. Council on Immigration Reform and the Urban Institute. These studies generally find that immigration has a net benefit to the country, that current immigration levels are reasonable and should be continued indefinitely, and that assimilation will proceed as well as it has in the past. Studies aggregated nationally see no negatives impact on labor, but sometimes acknowledge localized impacts. The fiscal effects that are found depend on the time frame used for analysis.

The second type of literature is articles in academic journals. Studies of specific cities or regions suggest major, often negative impacts. As Waldinger (1997) points out, the benefits of immigration accrue to owners of capital and the professional class, but not to workers. Combined with skill deficiencies, the result is that immigrants are poorer than natives and use more social services. Certain groups exhibit a high degree of spatial isolation, often in low-quality neighborhoods. The finances of major cities receive an extra burden, creating an inability to address native-born poverty.

Academic studies perform statistical analyses on demographic data. The most common source of data is the Census Bureau. It provides the decennial censuses, the Public Use Microdata Samples (which are the smallest aggregations of individual-level data) and the annual Current Population Survey which estimates the size and characteristics of the nationís foreign-born population. The Census Bureau provides data on population size and growth; social, demographic and economic characteristics; also land use classifications and street patterns.

Researchers also use metro area databases such as the New York City Housing and Vacancy Survey (HVS) or New Yorkís Early Warning Information System (EWIS) database on buildings. These are analyzed using statistical calculations in much the same way as Census data.

The statistical procedures used by these studies are correlation and multivariate regression. Also used but less common is the index of dissimilarity and factor analysis. The overview studies like the NRC use historical summaries and create population models using various assumptions of growth and intermarriage rates.

The third type of literature is popular, produced by think tanks, magazines and interest groups. These groups aim to influence the political debate and their writings cite statistics from the academic literature to support their positions. This creates what the NRC calls a "war of the experts." Public opinion polls are cited and historical summaries are used extensively. The popular literature relies heavily on emotional appeals and arguments. This is a necessary tactic in a field that is so complex and full of divergent information.

Nationwide Geographic Patterns


To get an idea of immigrationís impact on U.S. cities, an overview will be helpful to place the discussion in context. That can be done by mapping the locations where recent immigrants have settled in the U.S. It is also useful to know the origins of recent immigrants, since that will suggest their racial and ethnic characteristics.

Figures 1a and 1b are derived from the U.S. Census Bureauís Population Estimates Division. Figure 1a shows international migration during 1990-98 as a percentage of the 1998 population of counties. Net international migration is defined as in-migration less out-migration. About one-fifth of the counties have significant percentages of recent immigrants. The map shows the biggest increases in the Southwestern states, the Northeastern seaboard, Florida and the region surrounding Boise, ID. Few counties experienced net out-migration; these are mostly scattered in the Northern Plains states.

Figure 1b shows the density of net migration from 1990-98. The spatial pattern is far more concentrated than Figure 1a. Dense populations of recent immigrants exist only in certain large metropolitan areas, primarily on the east and west coasts. Lesser concentrations exist throughout Southern California and Florida; and are sprinkled throughout the Northeast and Midwest in medium-sized cities.

Figure 1c shows the change in immigration rates from the 1980ís to the 1990ís. The map shows the Census Bureauís hierarchy of geographic classifications as used in the 1998 Current Population Survey (CPS). The nation is divided into four regions and within those regions the major destination states are delineated. Within those states, the five "primary gateway" cities are listed: Chicago, Los Angeles, Miami, New York and San Francisco. Combined, these cities receive about half of all immigrants to the U.S.

The CPS statistics reveal that most of the country experienced increases in average annual immigration rates from the 1980-89 period to 1990-97. The mean increase for the entire country was 13 percent. The Midwest region had the largest percentage increase, while some states saw significant decreases. California and Floridaís rates decreased 19 and 7 percent respectively. The declines are more dramatic when specific metro areas are examined. Los Angelesís rate declined 23% and San Francisco declined 28%. Bucking the trend, Chicagoís rates increased by 83% (US Census 1999).

Other cities receive significant immigrant flows but the Census Bureau does not estimate those numbers. The flows are smaller than the above five cities, so they are termed "secondary" gateway cities. As identified by Frey (1996), they are Washington, D.C., Houston, San Diego, Boston, and Dallas. These cities, combined with the primary gateways, absorbed two-thirds of the nation's immigrant flow in the 1990ís.

Source Regions

Figure 2 shows the source regions of the U.S. foreign-born population. On average, immigrants arriving in 1990-97 comprise 29% of the foreign-born population. Mexico and Asia were each the source of one-third of the total immigrant flow. The 1990ís flow has contributed 40 percent to the African-born population, 30 percent to the Asian population and 30 percent to the Mexican population.

The European stream is small in absolute numbers and percentage, but it shows the largest increase since the 1980ís. The increase in North American immigration is nearly as great. The Mexican increase is identical to the overall average at 13 percent. Central American rates show big declines, possibly because refugees are less plentiful.

Total Foreign-Born Population in U.S. (000's) 1990-97 Flow (000's) 1990-97 Flow in Percent 1990-97 Flow as % of Source Region Pop. 1980-89 Flow (000's)

Average Annual Rate:
% Change
1980's to






















South America







Northern America




























Other Central America







Region not reported














Figure 2: Sending regions, numbers in thousands. Source: Authorís analysis of Census Bureau CPS 1999 data.


Segregation and Assimilation


In many cities, domestic out-migration has been at least partially offset by immigration. And in Los Angeles, San Francisco, Boston and Philadelphia, Frey (1995) found that domestic losses in were completely outweighed by immigration. The out-migration exaggerates the impact of immigration, causing greater turnover in the population and larger percentages of the foreign-born. Some out-migration may in fact be a response to high immigrant flows. Out-migrants from high immigrant cities tend to be low-income and less-skilled. Whites out-migrate from gateway cities at higher rates than from other cities. In effect, poor and unskilled immigrants are displacing their White counterparts.

It is the corporations and professional classes who remain in gateway cities. They employ immigrant labor to support their profitability and upscale lifestyles (Waldinger 1997). This pattern becomes increasingly prevalent as manufacturing jobs move overseas and the service sector expands. Thus the gateway cities have increasing disparities of income, as well as more economic opportunities.

Frey (1994) lists several motivations for the White flight: Increased job competition, rising taxes earmarked for immigrant services, overcrowded schools, language barriers and xenophobia. The White out-migrants settle in the West outside of California and the Midwestern heartland states. These areas thus become older, less dynamic and more prone to anti-immigrant opinion. Frey characterizes the net effect of these movements as a "balkanization." While the nation overall is getting more diverse, he asserts, it is also becoming more regionally segregated.


The most commonly cited theory related to immigrant assimilation was developed by Massey (1985). This spatial-assimilation model states that immigrants locate first in older, poor enclaves near the city center. As they acculturate and earn higher incomes, they disperse outward to the suburbs. In their search for greater prestige and amenities they integrate into neighborhoods where native populations predominate. This process is affected by the housing stock and market, economic conditions, and size of the immigrant population. Where immigrant communities are large, institutional supports can develop that encourage segregation. The physical distance between groups reflects the social, cultural and economic distances between them.

The spatial-assimilation model has been confirmed by a number of studies but important exceptions have also been found (Allen and Turner, 1996). Length of residence has little correlation with assimilation. Among Asians, income is the main factor determining proximity to Whites. Many upper-income immigrants choose to live in ethnic enclaves even when they could take their pick of house locations.

Allen and Turner studied Los Angeles and found that accessibility affects segregation to some degree. With cars and telephones available, immigrants are relying less on proximity for social support: A majority of recent immigrants live outside ethnic enclaves. A location near family and friends is becoming more important. However, most immigrants are more concentrated than their native-born counterparts. When immigrant concentrations do form, they are often located in older suburbs. The spatial assimilation gradient still exists but its scale is enlarged; immigrant neighborhoods are clustered in the center of a 5-county metro area and immigrants living at the suburban fringe are more assimilated.

The exception to the model was Mexican communities. Many were located in exurban areas, in old farm labor communities that had been engulfed by urban growth. Mexican enclaves do not contain higher proportions of newcomers, possibly because they are already overcrowded. Also, the culture within these enclaves may slow or prevent assimilation.

Alba et al (1999) has in general confirmed the spatial-assimilation theory, but has found significant discrepancies as well. Recent immigrants are able to enter the suburbs more easily and in greater numbers than previously. By 1990, new immigrants were settling in the suburbs at rates that were similar to their overall immigrant group (new and long-term combined) in 1980. Asians and long-resident Cubans move to the suburbs at the highest rates. New Chinese and Korean immigrants in particular live in suburban neighborhoods with less linguistic assimilation than ever before. Alba suggests that an important topic for year 2000 Census studies will be the extent to which suburbanized immigrants are concentrating into enclaves or dispersing into the native population.

The model is more applicable for certain groups. Non-Hispanic Whites follow the predicted pattern. Vietnamese immigrants had high levels of suburban residence in 1980 because they were placed by voluntary resettlement agencies. However, the Vietnamese have since become more spatially graded and those with less language ability are more concentrated in city centers.

In the 1980ís, the percentage of Mexicans immigrants entering the suburbs showed little change, most likely because of their low incomes. Financial status plays a key role in suburbanization rates, and for Asians family characteristics (such as number of children) are influential as well. Dominicans, Afro-Caribbeans and other Black immigrants show low levels of suburbanization, which demonstrates that skin color is a critical factor. In fact, immigrant segregation is more a function of socio-economics and race than oneís status as an immigrant (Glaster et al 1999a). Thus for certain low-income and/or ethnic groups, residence in nonassimilating enclaves has increased.


Suro (1999) adds a further wrinkle to the picture. He finds evidence of increased racial and ethnic intermarriage in the 1990ís. The number of mixed-race or mixed-ethnicity marriages has doubled since 1980 to about five percent of all marriages. He labels this phenomenon "the beginning of the blend," meaning that for the first time Hispanics and Asians are becoming completely assimilated in large numbers. The intermarriage rates are highest among well-educated, high-income Hispanics. Asian women intermarry at high rates, especially in predominately native-born regions. Eight percent of Whites intermarry in high-immigration regions. By contrast, Black intermarriage rates remain low in all locations.

Intermarried couples are typically in the 15 to 24 age cohort; many Hispanics in the 25-34 cohort intermarry also. The rate of mixed-race unions varies spatially depending on which group is examined. White rates are high in the gateway cities where a "critical mass" of minorities promotes interaction. Apparently, in many of these locations familiarity breeds admiration. Asian and Hispanic intermarriage is greatest in low-immigration states where availability of mates is a factor and cultural assimilation may be accelerated.

The blending of races and cultures is a sign that immigrant assimilation may proceed with Asians and Hispanics as it did in the past with White Europeans. What no one can predict is how widespread the trend can become, and what tensions and backlashes may result. A clue might be found in the fact that advertisers in the mass media regard intermarriage as a difficult, "touchy" subject and have avoided its depiction. Some ethnic groups resent intermarriage, regarding it as a form of disloyalty to oneís heritage. And Whites continue to have xenophobic or racist impulses. The evidence so far shows a wide range of immigrant assimilation patterns in America, from intermarriage to nonassimilating enclaves.

Neighborhood Quality


There is a wide variety of immigrant settlement patterns, and there is also a wide differential in socioeconomic achievement. Is there a linkage between neighborhood type and potential for upward mobility? Glaster et al (1999a) has found that neighborhood characteristics do indeed affect economic outcomes for certain groups. The research found a strong correlation between a concentration of oneís own ethnic group and rates of unemployment and poverty. This is contrary to theories which say that ethnic concentrations provide institutionalized support and economic benefits.

In addition, the researchers classified neighborhoods based on education, employment and public-assistance rates. Residence in low-employment neighborhoods led to higher school dropout rates. Residence in low-education neighborhoods led to lower entry into professional occupations. In short, life in a low-opportunity neighborhood has a deleterious effect on newcomers.

Immigrants are following the larger societal trend by splitting into rich and poor groups (Clark 1998). The success they achieve differs according to country of origin and skin color. Figure 3 illustrates the differences in poverty rates among the foreign-born by area of origin.

Figure 3. Source: U.S. Census Bureau CPS 1999 data.


Entrapment in an "underclass" neighborhood is a major problem for certain immigrant groups: Jamaicans, El Salvadorans, Dominicans and Mexicans. While these groups often live in low-status conditions, it should be noted that Black Americans continue to occupy the most disadvantaged neighborhoods (Glaster et al, 1999b).


The neighborhoods in New York where immigrants locate have been investigated on the basis of health and crime. Rosenbaum et al (1999) found that immigrants live in neighborhoods with significantly less access to health care than native-born households. Tuberculosis rates were higher in those locations. Latin Americans and Dominicans in particular lived in sickly districts. Crime rates showed no statistical differences.

When the researchers controlled for household characteristics, the differences between foreign-born and native ethnic groups disappeared. Their interpretation was that immigrants are more represented in low quality neighborhoods because for the most part they are colored. Again, race and ethnicity play a greater role than nativity.


The quality of housing inhabited by immigrants is an emerging topic of study. Foreign-born households have more difficulty affording housing and live in worse conditions than natives according to Schill et al (1999). As mentioned above, the service economy that supports corporations and professionals creates a fundamental mismatch between wages and housing costs. New York housing costs are among the highest in the nation, and newcomers respond by living in overcrowded conditions. Other reasons for living in poor quality housing include language limitations and the desire to live with persons of the same ethnic group. Discrimination limits the choices available, especially to Black immigrants.

Immigrants from all countries are more likely to live in overcrowded conditions (defined as more than one person per room). Asian Indians are most likely to live in overcrowded housing. Hispanic and Black immigrants are more likely to live in unsafe, substandard buildings. Dominicans have the largest proportions living in unsound housing, 11 percent.

Overall it appears that the majority of immigrants live in decent housing, although it represents a heavy financial burden. However, some localities are using the pretext of building code violations to pass invasive or discriminatory laws. The town of Brookhaven, NY recently proposed a Neighborhood Preservation Act that would limit the number of people renting a house and allow authorities to enter and inspect homes without a warrant (Cooper, 1999). Anti-immigrant groups are supporting the law as they profess concern for newcomersí safety. Civil rights groups argue that the law would still allow homeowners to crowd as many persons as they wish into their own homes.

Stefan Krieger of the Hofstra Law School asserts, "This is happening in many places." Similar laws have been passed in several other New York towns. In Patchogue, Freeport and Mount Kisco police and building inspectors join forces to make midnight raids. In the Chicago suburb of Addison, an entire neighborhood of Latinos was forcibly removed under eminent domain. Many of them were immigrants (Morales, 1997).

Pader (1994) notes that many immigrants come from cultures where several people sleep to a room and rooms are multifunctional. The one-person-to-a-room standard is an expression of middle-class Anglo preferences rather than a validated safety standard. Conversely, occupancy codes have been effective in preventing the worst excesses of tenement housing.

Suburban jurisdictions want jobs to add to their tax bases, but not the low-income housing that is required to house workers. Burchell (1995) found that all forms of inexpensive housing fail to generate enough taxes to cover school district costs. Immigrants have more children than natives; they have unique bilingual and educational requirements; and they receive more social services than natives. As a result, city officials and long-term residents often view them as a drain on local finances.

In addition, xenophobia is a persistent force in local politics. Discriminatory building and zoning codes are local expressions of nativism. Nativism has been a constant thread in American thought (Johnson 1997), but since immigration policy is made at the federal level, that is where the political activity more commonly occurs. 




Immigration is at record high levels and will be responsible for the majority of Americaís future growth. Generally, assimilation among the newcomers proceeds in a manner similar to previous waves. But in certain cities nonassimilating immigrant enclaves are established and growing. These are usually located in the center city but can be found in the suburbs as well.

The neighborhoods in which immigrants settle have a large effect on the potential for success. Immigrants are more likely to live in low-quality neighborhoods and in low-quality housing. However, the studied effects of nativity are usually outweighed by the effects of race. Dark-skinned persons, immigrant and native alike, live in the worst conditions.

The large increase in poor, under-educated immigrant populations places an extra burden on the social service and educational systems in cities. Together with xenophobia, this has fueled the nativist political movement at the local and federal levels.


The United States should implement policies that will reduce the entry of poor, uneducated immigrants. This will allow the established enclaves time to assimilate and reduce pressure on housing stocks. It will also relieve schools and social service agencies so they can focus on the entrenched poverty and discrimination facing both natives and resident immigrants. The advancement of our resident population should take precedence over the giving of opportunity to people in foreign countries.

Among the major studies addressing policy options, there is a remarkable amount of agreement on one point: The family preference provisions of the 1965 Immigration and Nationality Act Amendments should be revised (NCR, 1997; US CIR 1997). Currently, 30-40% of entries come under the provision which allows siblings, parents and adult children of immigrants the right of automatic entry (A similar proportion are relatives of citizens, Figure 4). That right should be restricted to the nuclear family. This will reduce the backlog of applications, slow the process of chain (network) migration and improve the skill level of the average immigrant.

Figure 4: Legal Immigration by Admission Category Excluding IRCA: Fiscal Years 1988-97
Source: U.S. Immigration and Naturalization Service 1997 Annual Report.

We can also focus on reducing illegal immigration, specifically from Mexico. Massey (1998) found that border enforcement, visa reductions and employer sanctions have been ineffective policies. Instead, he suggests that temporary work visas should be issued for a $300 fee. The proceeds would go toward binational development, insurance, and loans to migrants that will remove the incentives for illegal entry.

These measures represent an initial step toward improving immigration policy and the quality of life for existing U.S. residents. More discussion and action needs to occur in the political arena. The impacts from immigration range from dire to vibrantly beneficial. The effects are differentiated in space, by race, and by socioeconomic factors. Policymakers should target the locations where solutions will be of greatest benefit.




Selected Sources

Alba, Richard D. et al. 1999. "Immigrant Groups in the Suburbs: A Reexamination of Suburbanization and Spatial Assimilation." American Sociological Review 64, p. 446-460.

Allen, James P. and Eugene Turner. 1996. "Spatial Pattern of Immigrant Assimilation." Professional Geographer 48(2) p. 140-155.

Branigin, William. 1998. "Immigrants Question Idea of Assimilation." Washington Post, May 25, p. A1.

Brimelow, Peter. 1995. Alien Nation. New York: Random House.

Burchell, Robert W. and David Listokin. 1995. Land, Infrastructure, Housing Costs and Fiscal Impacts Associated with Growth. Cambridge, MA: Lincoln Institute of Land Policy.

Clark, William A. V. 1998. The California Cauldron. New York: The Guilford Press.

Cooper, Michael. 1999. "Laborers Wanted, but Not Living Next Door." The New York Times on the Web <> accessed November 28, 1999.

Fix, Michael and Jeffery Passel. 1994. "Immigration and Immigrants: Setting the Record Straight." Urban Institute <> accessed November 24, 1999.

Frey, William H. 1994. "The New White Flight." American Demographics April p. 40.

Frey, William H. 1995. "Immigration and Internal Migration ĎFlightí from US Metropolitan Areas: Toward a New Demographic Balkanization." Urban Studies 32(4-5) p. 733-757.

Frey, William H. 1996. "Immigrant and Native Migrant Magnets." American Demographics <> accessed December 4, 1999.

Galster, George C. et al. 1999a. "Neighborhood Opportunity Structures and Immigrantsí Socioeconomic Advancement." Journal of Housing Research 10(1) p. 95-127.

Galster, George C. et al. 1999b. "Neighborhood Opportunity Structures of Immigrant Populations, 1980 and 1990." Housing Policy Debate 10(2) p. 395-442.

James, Franklin J. et al. 1998. "The Effects of Immigration on Urban Communities." Cityscape 3(3) p. 171-192.

Johnson, Kevin R. 1996. "The New Nativism: Something Old, Something New, Something Borrowed, Something Blue." In Immigrants Out! The New Nativism and the Anti-Immigrant Impulse in the United States. Juan F. Perea ed. New York: New York University Press.

Kasarda et al. 1997. "Central City and Suburban Migration Patterns." Housing Policy Debate 8(2) p. 307-359.

Massey, Douglas S. 1985. "Ethnic Residential Segregation: A Theoretical Synthesis and Empirical Review." Sociology and Social Research 69(3) p. 315-350.

Massey, Douglas S. 1999. "March of Folly: U.S. Immigration Policy After NAFTA." The American Prospect Online <>, accessed Nov. 26 1999.

Morales, Rebecca. 1997. "The Suburbanization of Immigrants in the Chicago Metropolitan Area." APA Immigration and World Cites Project <> accessed November 9, 1999.

Muller, Thomas. 1993. Immigrants and the American City. New York: New York University Press.

National Research Council. 1997. The New Americans: Economic, Demographic and Fiscal Effects of Immigration. Washington D.C.: National Academy Press.

Pader, Ellen J. 1994. "Spatial Relations and Housing Policy: Regulations that Discriminate Against Mexican-Origin Households." Journal of Planning Education and Research 13(2), p. 119-135.

Rosenbaum, Emily et al. 1999. "Nativity Differences in Neighborhood Quality Among New York City Households." Housing Policy Debate 10(3) p. 625-658.

Schill, Michael H. 1998. "The Housing Conditions of Immigrants in New York City." Journal of Housing Research 9(2) p. 201-235.

Suro, Roberto. 1998. Strangers Among Us: How Latino Immigration is Transforming America. New York: Alfred A Knopf.

Suro, Roberto. 1999. "Mixed Doubles." American Demographics <> accessed December 4, 1999.

U.S. Census Bureau. 1999. "Profile of the Foreign-Born Population in the United States, 1997: Detailed Tables." March 1997 Current Population Survey <> accessed October 23, 1999.

U.S. Commission on Immigration Reform. 1997. U.S. Refugee Policy: Taking Leadership. <> accessed December 7, 1999.

Waldinger, Rodger. 1997. "Immigrant Integration in the PostIndustrial Metropolis: A View from the United States." Paper presented at the Metropolis First International Conference <> accessed November 21, 1999.

Back to Home page