Internal migration has become a fact of life in the United States. We are the most geographically mobile society in the history of the world because we have so many opportunities given our high level of education and wealth. But not everyone is equally mobile in American society. Mobility requires resources: money to move, information about opportunities elsewhere, and the credentials to take advantage of those opportunities. Because money, information, and credentials are not evenly distributed across the population, some people have more opportunities than others. The poor are less mobile than the wealthy, racial minorities are less mobile than Anglo-whites, and new immigrants are less mobile than natives.
Analysis of settlement patterns and migration flows is important because socioeconomic mobility and geographic mobility are linked. People who have the means to move out of impoverished neighborhoods usually do so: Moving up the economic ladder entails moving out. The relative immobility of the poor helps us to understand why poverty is geographically concentrated in certain neighborhoods and cities, as opposed to evenly dispersed across the nation.
One focus of current demographic research is the study of native and immigrant migration flows and settlement patterns. We have learned from numerous studies that native migration flows do not closely parallel those of recent immigrants. Asian and Latino immigrants are usually drawn to areas of co-ethnic settlement in just a few states, and particular areas (counties, cities, neighborhoods) within those states. Sometimes the areas of coethnic settlement provide social networks that help new immigrants gain a foothold, but these networks are often situated within a context of urban poverty, violence, bad schools, and fierce competition for scarce jobs and housing with rival groups. Hundreds of thousands of immigrants appear not to have much by way of a social network and wind up in dead-end jobs paying the lowest of wages and with no benefits, or worse still on welfare. Some immigrants wind up being exploited precisely because established immigrant networks subjugate and intimidate them into working in slave-labor conditions, such as the Fuzhounese immigrants (some of them illegal) in New York’s Chinatown. Poverty is persistent in immigrant communities and it limits geographic mobility as recent immigrants remain stuck in some of the worst labor markets in the country.
Where immigrants do succeed in moving up — and out — to the suburbs, often it is because conditions in suburbia have declined. This lowers rents and creates pockets of suburban poverty such as Chamblee, Georgia — called "Chambodia" by Atlanta locals for the large number of southeast Asian immigrants who have found their way there. Low-income immigrant enclaves can be found in once-exclusive suburban counties such as Orange County, California; DuPage County, Illinois; and Nassau County, New York, but that does not mean that these immigrants have "made it" in America. The term "suburban" is not synonymous with wealth and exclusivity as it might have been at mid-century. Today, many suburbs are just marginally better living environments than older central cities.
Contrast the limited geographic mobility of immigrants with the considerable mobility that most natives enjoy. Native migrants are far more capable of avoiding areas of high unemployment than immigrants. Because they have more money, they can afford to be choosy and discriminating in deciding where to live. They can afford to commute long distances from home to work and often choose to live in towns and neighborhoods that are out of the financial reach of most immigrants. The economic differences between natives and immigrants balkanize states by producing racially homogeneous regions, counties, towns and neighborhoods.
Table 1 (below) shows that the percentage of foreign-born residents residing in the three most urban counties of Texas, Florida, California and New York has increased between 1970 and 1990. In other words, the most urban areas of these states have a larger proportion of the total statewide immigrant population in the 1990s than they did around 1970. Not only has the immigrant population grown larger, it has also become more concentrated in the most urban areas of these four states. But internal migrants, defined here as the percentage of the population born in some other state, have generally become a much smaller share of the total statewide population of interstate migrants in these same urbanized areas. Only New York is an exception, but this is because wealthy internal migrants are attracted to Manhattan (New York County). The general picture is clear: Internal migrants and immigrants are not moving to the same destinations. Because the recent wave of immigrants has come predominantly from Asian and Latin American countries, whereas the majority of natives are white, rival patterns of mobility produce racial residential segregation within states and across metropolitan areas.
Findings from my recent book, Separate Destinations: Migration, Immigration and the Politics of Places (Ann Arbor: University of Michigan Press, 1999) show that geographically separate but ethnically homogeneous communities are growing larger in immigrant-receiving states. In California and Florida, between 1980 and 1990, Asian, Mexican, Central American and South American settlements all became larger and more concentrated, rather than smaller and more integrated. Native-born internal migrants, Canadian immigrants, and European immigrants showed no similar tendency to concentrate in areas where they had previously established a presence — an indication of their higher level of mobility relative to the impoverished immigrant groups from less developed nations.
Segregated settlement patterns often are not highly visible except at the micro-level: across city blocks, neighborhoods, and census tracts. But in states experiencing high rates of population growth, such as California, balkanization may even appear at the county or regional level. Maps 1 and 2 (pages 4 and 5) help to illustrate this point. Map 1 shows that the stunning increase in the foreign-born population since 1970 has been most dramatic in Southern California counties (Riverside, San Bernardino, Orange, San Diego), in specific agricultural regions of the Central Valley (Merced, Stanislaus), and in Northern California (Butte, El Dorado, Solano). Notably, however, the places that have attracted immigrants do not neatly match those that have attracted domestic movers from other states. Map 2 shows the growth in the number of internal migrants over the same period. One should note in comparing the two maps that not only has the growth in the number of internal migrants been much slower, it has not occurred in the exact same places as growth in the foreign-born population. Internal migrants have staked increasing claims in the mountain counties along the Nevada border, in Riverside County in the South, and in several northern California counties (Lake, Trinity, Lassen). Internal migrants have been careful to avoid Los Angeles, San Mateo, and Santa Clara Counties — where their presence has actually shrunk since 1970 — even as the foreign-born population has grown in these locations. Riverside is notable for attracting both immigrants and domestic migrants; but within this large county, the two groups settle in very different locations.
California is only the most obvious example of where rival regions of the state are being reshaped by disparate internal migration and immigration flows. Even places not known for being immigrant ports of entry are experiencing the balkanization associated with the gap between native and immigrant skill, education, and income levels. Take Colorado’s mountain resort towns, for instance. Immigrant workers are valued in places such as Aspen, Vail, and Keystone, as they are elsewhere, because they are willing to work hard for low pay. But immigrant workers find it nearly impossible to reside anywhere near the exclusive mountain communities where they work. The high cost of housing in the ski areas has forced many low income workers to commute twenty or thirty miles, snarling traffic and burdening existing infrastructure. At other locations, balkanization is the result of the preferences of the immigrant groups to sustain their own culture, language, and identity. Through their own preferential and discriminatory practices, immigrants create niches that exclude outsiders. Roger Waldinger and other sociologists have pointed out that since the 1970s, native born blacks have seen a sharp decline in their labor market position while the employment of immigrants has expanded. The result of immigration influx has been the expansion of all-minority ghetto areas in New York City — all-black and black-Hispanic neighborhoods. The preferences of natives and the preferences of immigrants have conspired with economic inequality to create segregated communities.
Segregated settlement patterns have important implications for the style and substance of electoral politics practiced in these places. A large body of social-scientific evidence suggests that the larger the minority population, the more segregated it is likely to be: As immigrant enclaves become larger, they become more isolated and insular and as they become more isolated and insular, communities turn inward rather than outward. This geographic isolation prevents minorities from voicing demands to outsiders so citizenship and interest in participating in the civic life of the host society drops. Hence, we observe the lowest naturalization rates in areas of highest immigrant concentration and among those who are naturalized in these locations we observe the lowest political participation rates.
Of course one reason we observe such low participation rates is that votes in the electoral districts where many new citizens reside often do not matter to the outcomes of city, state, and national legislative races. The places where the recent foreign-born concentrate are usually dominated by Democratic officeholders and where Democrats have an overwhelming registration edge over Republicans and minor parties. Because these districts are so one-sided, individual votes matter little to who wins in November. With little competition to stimulate participation, abstention becomes a highly rational act. Why expend the effort to go to the polls when the outcome is known so far in advance? When they do decide to participate, local political tradition has ensured there is only one real option: the Democratic party. This is why so many newly naturalized immigrants find their way into the Democratic party — it is the only party that exists in the places they live. Democratic loyalty is particularly high among Latinos, less so among Asians, although it varies with the national origin and economic standing of Asian subgroups. Asians, however, remain a much smaller immigrant population than Latinos.
Since legislative representation in the United States is geographically based, the settlement patterns of internal migrants and immigrants have enormous political implications. Because immigrants wind up concentrated in urban areas, and are usually drawn into Democratic party politics, high immigration flows sustain one-party political jurisdictions in urban areas, undermining the type of partisan competition necessary to the exercise of political accountability. Legislators from districts such as these find themselves routinely re-elected by lopsided margins without the scrutiny of their performance in office that would come with serious political opposition.
Internal migrants moving to homogeneously white suburbs are more politically divided than immigrants, but their affluence is generally associated with Republican party identification. What this means is that political balkanization is not just a function of growing immigrant populations but also emerges as a result of internal migration patterns. Indeed, the combination of immigrant and migrant settlement and mobility patterns threatens to create rival communities of interest recognizable mainly by their monolithic racial and ethnic character. With legislative districts drawn such that their most distinguishing trait is the racial and ethnic homogeneity of their population, racial divisions will find a lasting and permanent home in the politics of state and national legislative chambers. This would not be a problem except that we have fought such a long-running battle to heal racial divisions in our polity. It is hard to imagine how these divisions will subside when we tie legislative representation to geographically situated districts with racially distinct populations.
Racial residential segregation is productive of political balkanization because partisanship in the United States has such a strong racial and class component. The Democrats capably maintain the image that they are for the poor and disadvantaged while taking every opportunity to cast their GOP opponents as the partisans of the rich. The Republicans have frequently cast the Democrats as the party of inner city minorities and liberal moral reprobates, running up big margins among anti-government rural voters, white religious conservatives, and affluent suburbanites afraid to venture downtown. Neither party appears to be on the verge of swapping electoral support with the other, suggesting that political balkanization will continue to follow from migration and immigration patterns and will not diminish as the result of changing party stances on issues.
Knowing how the settlement patterns of internal migrants and immigrants have militated against the goals of residential integration and political party competition, what remedies can be offered?
First, we can adopt the goal of maximizing political party competition as we draw legislative district boundaries. People should not be grouped into legislative districts on the basis of their political similarity, but on the basis of their dissimilarity. The guiding criteria for drawing boundaries should be to maximize racial, economic and political diversity within a district, not to minimize it.
Internal migration is difficult to control in a society that values freedom. While blatant prejudice is subject to regulation by appropriate and necessary enforcement of state and federal anti-discrimination laws, we cannot force people to settle down in neighborhoods where they do not feel safe so internal migration will continue to be motivated by a combination of lifestyle choices and prejudices. We should recognize that discriminatory zoning ordinances and exclusionary practices that limit the availability of affordable housing also contribute to residential segregation and undermine the goal of assimilating the foreign-born.
If prejudice were the only cause of racial segregation, we could stop by arguing that stronger civil rights enforcement is the answer. But residential segregation today is determined more by economic differences between (among) groups than by racial ones. Groups are residentially isolated from one another because of vast differences in the kind of housing, amenities, and taxes they can afford. To the extent that immigration widens the gap between rich and poor, it undermines our goals of equality and integration. The good news is that immigration levels are subject to clear policy control. Existing patterns of racial and ethnic segregation are being reinforced by the influx of immigrants whose geographic mobility and capacity to move beyond urban enclaves is limited by their low skill and education levels. A temporary slowdown in the flow of immigrants into the United States would go a long way toward realizing the longstanding goal of residential integration. Restricting entrance to skilled immigrants would lessen the competition for the limited supply of unskilled positions and increase the wages and employment prospects for low-skill natives and previous immigrant arrivals. Placing a higher priority on skills, English literacy, and education would lessen the reliance of the immigrant on the coethnic enclave and greatly facilitate the geographic mobility of the immigrant population. In turn, the prosperity of native minorities and already admitted immigrants would diminish the extent to which economic grievances based on racial inequities became the foundation for political demands.
Conflict between (among) groups with differences on some salient issue is the hallmark of politics in a democratic system. But it is important to distinguish between healthy and unhealthy differences. The unhealthy conflicts are ones rooted in unalterable traits such as race, which divide people based on permanent group memberships. The healthy differences are those that can be resolved more easily by the movement of individuals into and out of penetrable coalitions. It is not inevitable that political conflict in the United States must be based on race and ethnic differences, it only seems that way given the nation’s long struggle with this issue. Recognizing the importance of competitive electoral districts while maintaining efforts to promote minority opportunity and advancement in the economy is essential to creating a political system in which the issues dividing us are more temporary and soluble.