Download this Backgrounder as a pdf
During the 1990s, the nation’s immigrant population grew by 11.3 million — faster than at any other time in our history. Using newly released data from the 2000 Census, this report examines the changing distribution of the nation’s immigrant population by country of origin at the state level. The findings show that in one sense, today’s immigration is more diverse than ever because people now arrive from every corner of the world. In another sense, however, diversity among the foreign born has actually declined significantly. One country — Mexico — and one region — Spanish-speaking Latin America — came to dominate U.S. immigration during the decade. The report also found that immigrants from some countries became more spread out in the 1990s, while the dispersion of others changed little.
Among the report’s findings:
- The dramatic growth in the nation’s immigrant population has been accompanied by a significant decline in diversity. In 1990, immigrants from the top sending country — Mexico — accounted for 22 percent of the total foreign born. By 2000, Mexican immigrants accounted for 30 percent of the total.
- In fact, Mexico alone accounted for 43 percent of the growth in the foreign-born population between 1990 and 2000.
- In 39 states the share of the immigrant population accounted for by the top sending country increased. The decline in diversity was most dramatic in Arkansas, North Carolina, Georgia, Indiana, Tennessee, Utah, Nebraska, and Alabama.
- Even those states with little diversity among immigrants in 1990 experienced a continued decline in diversity between 1990 and 2000. In Arizona, for example, immigrants from Mexico grew from 55 percent to 67 percent of the foreign born and in Texas, Mexicans increased from 59 to 65 percent of the total.
- Looking at diversity as measured by the share of immigrants from just one region of the world also shows a significant decline in diversity. Nationally, immigrants from Spanish-speaking Latin American countries increased from 37 percent to 46 percent of the total foreign-born population during the 1990s.
- Immigrants from Spanish-speaking Latin America accounted for more than 60 percent of the growth in the foreign-born population nationally in the 1990s.
- In 2000, there were 33 states (including the District of Columbia) in which immigrants from Spanish-speaking Latin American countries were the largest single group. Europeans were the largest group in 11 states, East Asian immigrants were the largest in four states and Canadian immigrants were the largest in three states.
- Declining diversity was mainly due to very uneven growth in the size of different immigrant groups. For example, the number of immigrants from Spanish-speaking Latin America increased by seven million and those from East Asia rose by over two million. In contrast, the number of immigrants from Europe increased by less than 700,000 and those from Sub-Saharan Africa increased by about 400,000.
- Immigrants from some countries became much more dispersed during the decade. For example, the percentage of immigrants from Mexico, the Dominican Republic, and El Salvador concentrated in only one state fell significantly during the decade.
- In contrast, immigrants from Cuba became more concentrated, while the share of immigrants from such countries as Iran, Columbia, Jamaica, and Haiti concentrated in one state remained virtually unchanged in the 1990s.
This report is based on newly available 2000 Census long form data, which was released for public use in June of this year.1 One in six households receives the long form questionnaire, which includes questions on whether someone is an immigrant and in what country they were born. This report compares the results from the 1990 and 2000 Census long forms. The definitions of "immigrant" and "foreign born" in this study are the same as that used by the Census Bureau. The foreign born are persons living in the United States who were not U.S. citizens at birth. This includes naturalized American citizens, legal permanent residents (green card holders), illegal aliens, and those on long-term temporary visas such as students or guest workers. Analysis done by the Census Bureau, INS, and others indicates that seven to eight million illegal aliens and one million persons on long-term temporary visas, such as students and temporary workers, responded to the 2000 Census.
Because all children born in the United States to immigrants are by definition natives, the sole reason for the dramatic increase in the immigrant population at the national level is new immigration. At the state level, growth in the immigrant population can be caused both by new immigration from abroad and by the arrival of immigrants from other states. While some immigrants die and others return home, the granting of permanent residency and the settlement of hundreds of thousands of illegal aliens greatly exceeds deaths and out-migration for immigrants from most countries.
Immigrant Settlement Across States
Table 1 (click here) reports the top 15 countries of birth for immigrants in 2000 for each state plus the District of Columbia. In each state, the countries are ranked based on the top sending nation in 2000. (It is important to note that some of the countries that were lower ranked in 1990 are no longer in the top 15 by 2000. Conversely, some of the top 15 countries in 2000 were not among the top 15 in 1990.) Table 2 (click here) shows the percentage of the total foreign-born population accounted for by the top sending country in each state at the start and end of the decade. The first column in Table 2 shows the share of each state’s immigrant population accounted for by the top sending country in 2000; the second column shows the top country’s share in 1990. The third column shows the percentage point change, and the fourth column shows the percentage change in the immigrant population’s diversity. (A percentage point change reflects the increase or decrease in the share of the state’s foreign-born population represented by the top sending country. In contrast, the percentage change reports the size of the change relative to the level of diversity in 1990. In North Carolina, for example, the top country accounted for 10 percent of the immigrant population in 1990 and 41 percent in 2000 — a 31 percentage point increase. But this change also can be expressed as a 295 percent change. A positive percentage point or percentage change indicates that there was a decline in the diversity of the state’s immigrant population.
Decline in Diversity Is Widespread. Tables 1 and 2 show that in most states the top sending country accounted for a much larger share of the total in 2000 than in 1990. Overall, there were 39 states where diversity decreased. In 24 of these states the top country grew as a share of the total foreign born by at least 10 percentage points. In contrast, there were only 11 states in which the top country represented a smaller share of the foreign born in 2000 than in 1990 — that is, where there was an increase in diversity. Of the 11 that increased in diversity, there was only one in which the top country fell as a share of the total foreign born by more than 10 percentage points. Thus increases in diversity were relatively rare and modest compared to decreases in diversity.
In many cases the decline was due to the top sending country in 1990 increasing its share of the total by 2000. However, in many states the top country changed during the 1990s. In 15 of the 39 states where diversity declined, the top sending country changed during the decade. Thus, in some cases there was a shift in the leading sending country, while in other states there was an acceleration of an already existing pattern. The most dramatic declines in diversity can be found in Arkansas, North Carolina, Georgia, Indiana, Tennessee, Utah, Nebraska, and Alabama. But even in some states that had little diversity in 1990, the situation become more pronounced during the decade. In Arizona, for example, immigrants from Mexico grew from 55 percent to 67 percent of the foreign born and in Texas, Mexicans increased from 59 to 65 percent of the total. Even in California, a state synonymous with immigrant diversity, Mexican immigrants increased as a share of the total foreign born from 38 to 44 percent.
The decline in diversity among immigrants is widespread and is not confined to a few states or even one part of the country. Of the 24 states where the top country’s share increased by 10 percentage points or more, eight are in the South, seven are in the Midwest and nine are in the West. Only the Northeast didn’t experience a significant decline in diversity. But even in that region, there were only a few states where diversity actually increased. In New York and New Jersey, the states with the largest immigrant populations in that part of the country, diversity actually declined slightly in the 1990s.
Diversity Decline and Differential Growth Rates. The decline in diversity reflects the very different rate of growth among immigrant groups. Table 3 (click here) reports the growth during the 1990s for the largest sending countries in 2000. The table shows that the rate of increase varied significantly by country.2 For example, the number of immigrants living in the United States from Italy, Germany, Ireland, and Greece actually declined during the decade. The number of immigrants from such countries as Laos, Canada, Portugal, and the United Kingdom remained roughly constant. In sharp contrast, the number from Mexico and most countries in the Western Hemisphere increased significantly during the decade. The number of immigrants from the former Soviet Union, Pakistan, and India also increased dramatically during the decade. Immigrant populations from East Asian countries such as China, the Philippines, and Vietnam also grew very significantly, though not as dramatically as the numbers from the Western Hemisphere or South Asia. Despite very significant growth among countries other than Mexico, that country still accounted for 43 percent of the total increase in the foreign-born population. This means that eventually if the trend continues, Mexico will come to account for 43 percent of the total foreign born, perhaps within the next 15 years, assuming there is no change in immigration policy.
Different rates of increase reflect many factors: the arrival of new legal immigrants, new illegal immigration, rates of return migration, and deaths. For example, the dramatic decline in the number of Italian immigrants was due to low levels of new immigration from that country, coupled with very high death rates among the group because so many are long time residents and are now quite old. America’s immigration policy is primarily based on family relationships, therefore those groups that had the most young immigrants in 1990, who might want to bring in their relatives, tended to send the most immigrants in the 1990s.
Mexico Top Sending Country in Most States. Table 2 shows the top sending country for each state (the same information can be found in Table 1). Nationally, Mexican immigrants increased their share of the foreign-born population from 22 percent of the total in 1990 to 30 percent by 2000. This continues a long-term trend: In 1980 Mexico, already the leading sending country, accounted for 16 percent of the foreign-born population. The trend of declining diversity goes back even farther; in 1970 the top sending country was Italy, and it represented only 10 percent of the foreign born. At the state level, Mexico was the largest sending country in 18 states in 1990; by 2000 it was the top sending country in 30 states. Table 2 shows that in most places where it was not the leading country in 1990 but became so during the decade, it displaced Germany as the top sending country. The Mexican immigrant population is growing so rapidly because it is not only the leading sending country for legal immigration, but also because of the enormous growth in illegal immigration from that country. The Immigration and Nationalization Service (INS), before it was absorbed into the Department of Homeland Security, estimated that the illegal alien population from Mexico grew by nearly 2.8 million between 1990 and 2000, accounting for 80 percent of the total increase in the illegal population.3 In fact, the INS estimates indicate that in 2000 roughly half of the Mexican-born population in the United States was illegal. Because Mexico dominates both legal and illegal immigration to the United States, it represents a large and rapidly growing share of the total foreign-born population.
While mostly related to immigration from Mexico, the decline in diversity was not only associated with that country. In Alaska, for example, Filipino immigrants went from 21 to 28 percent of the total and in Hawaii they went from 45 to 49 percent of the total. In Montana, immigrants from Canada went from 29 to 40 percent of the total and in New York, Dominican immigrants went from 8 to 11 percent of the total foreign born. While some of these declines in diversity are not very large, it does suggest that declining diversity can occur even in the absence of large-scale immigration from Mexico.
Diversity Based on Region of Origin. Diversity among the foreign-born can be measured in many ways. While tables 1 and 2 examined diversity by country, Table 4 (click here) shows the growth in the immigrant population at the national level based on the region of the world from which they came. The table attempts to categorize immigrants by region in a way that reflects the cultural or linguistic similarities between immigrants from different countries. The Census Bureau typically groups countries only by the continent from which they came. For example, immigrants from Turkey, India, and China are all grouped together by the Bureau as simply "Asian," even though these countries share little in common. In contrast, Table 4 divides Asia into East Asia, South Asia, and the Middle East (the Middle East includes North Africa). Moreover, instead of treating all immigrants from the Western Hemisphere, except Canada, as "Latin American," as the Census Bureau does, Table 4 groups the Spanish Speaking countries of the Western Hemisphere into one group. The Anglophone, Francophone, and Lusophone countries of the region are grouped together as a separate region and Canada is treated as its own region. Africa is also divided between the north, which is part of the Middle East, and the Sub-Saharan region. Grouping countries in this way provides a more accurate picture of immigrants by region of the world, allowing for more meaningful comparisons between immigrant groups than is possible if they were simply categorized by continent.
Measuring diversity by region of birth reveals a similar picture to that found in Tables 1 and 2. Although about 5 percent of immigrants did not indicate their country of birth in 1990, compared to about 1 percent in 2000, it is unlikely this would significantly change the results in Table 4. The results for race and Hispanicity for those immigrants that did not provide their country of birth indicate that their distribution across regions is very similar to those immigrants who did report where they were born.
Table 4 shows very different growth rates for immigrants by region of the world. For example, the number immigrants from Spanish-speaking Latin America increased by nearly seven million and the number from East Asia rose by over two million. In contrast, the number from Europe increased by less than 700,000 and the number from Sub-Saharan Africa increased by fewer than 400,000. Table 5 (click here) reports the top sending region in each state in 2000 and 1990. The table indicates that in 2000, Spanish-speaking countries from Latin America were the top-sending countries in 33 states, up from 12 states in 1990. In 11 of the 12 states where Spanish-speaking immigrants were already the largest group in 1990, they increased their share of the foreign born. In addition, in 21 other states immigrants from Spanish-speaking Latin America displaced another part of the world as the leading sending region. While in general states that declined in diversity went from having Europe as the top-sending region to Latin America, this was not true in every state. In Oklahoma, Georgia, Oregon, and Tennessee, East Asia was the leading sending region in 1990, but by 2000 Spanish-speaking Latin America was the leader.
Ways to Measure Diversity. Of course, there are limits to how well the regions used in Tables 4 and 5 actually measure diversity. The countries of some regions may share more in common with each other than those in other regions. For example, while Europe is a region of great linguistic, religious, and economic diversity, the Middle East has less religious diversity, though it is still diverse in other ways. This may be especially true for those individuals who actually emigrate to the United States. While there are certainly differences between Latin American countries, Spanish-speaking Latin America is probably the most homogenous of the world’s regions as defined here. Despite differences within regions, Tables 4 and 5 suggest that when immigrants are grouped into regions, there has been a decline in diversity at least when measured by the share represented by the top sending region. This decline took place both at the national level and in many states.
Immigrants by State
So far we have only examined the settlement of immigrants by state, Tables 6, 7, and 8 look at this question from the other direction. Tables 6 (click here) and 7 (click here) show the top seven states of settlement for the 100 largest countries in 1990 and 2000. Table 8 (click here) reports the share of immigrants from the 40 largest countries living in only one state. Examining immigration in this way is important because it creates a better understanding of the distribution of immigrant groups across the United States, and how this changed in the 1990s.
Concentrations by Country. The tables show great variation between immigrant groups. Of the countries listed in Table 8, there were 11 in which 20 percent or less lived in one state in 2000. Immigrants from these countries are spread throughout the United States. Among the most diffuse immigrants are those from Germany, Nigeria, Canada, the former Yugoslavia, India, Peru, Brazil, and the United Kingdom. Of course, not all immigrants exhibit defuse settlement; there were 14 nations in which more than 40 percent of the immigrants lived in just one state. Those from Cuba, Guyana, the Dominican Republic, Iran, Trinidad & Tobago, Philippines, El Salvador, Ecuador, Guatemala, and Mexico tend to be the most concentrated. In general, European and South Asian immigrants tend to be the most dispersed, while those from the Western Hemisphere tend to be the most concentrated. Those from East Asia tend to fall in the middle of the distribution.
Changing Concentrations in the 1990s. In terms of the changing distribution of immigrants by country during the 1990s, we also see great variation between countries. In Table 8, about half of the countries became more dispersed during the decade, while the concentration of the other half remained about the same. The largest increase in dispersion was among immigrants from Guatemala, Mexico, Cambodia, El Salvador, the Dominican Republic, and Laos. Most of the countries that became more evenly spread out across the country were those that tended to be the most concentrated in 1990. Thus, it should not be too much of a surprise that these highly concentrated immigrant groups tended to become more dispersed over the last 10 years. Even so, many of these countries remain among the more concentrated in 2000, despite an increase in dispersion.
Table 8 also shows that not all immigrants became more dispersed. Immigrants from Cuba, Poland, and Brazil actually became somewhat more concentrated between 1990 and 2000. In general immigrants did become more dispersed over the last ten years, but this was by no means a universal trend among all immigrant groups.
Using the newly released 2000 Census data, this report has examined the changing settlement patterns of immigrants across America. The data show that along with a historically unprecedented increase in the number of immigrants, there has been a significant decline in the diversity of the nation’s foreign-born population. The decline in diversity occurred not only at the national level, but also in many states. Most states saw the leading sending country increase its share of the total foreign born during the 1990s. When immigrants are grouped by the region of the world from which they came, the same general pattern exists. Mexico, specifically, and Spanish-speaking Latin American countries in general now comprise a larger percentage of the foreign born than any other country or region of the world.
Of course, diversity could be defined in other ways. Race or language diversity are other possible ways of examining the issue. But these variables are highly correlated with country and region of origin so the results are likely to be very similar. It is unlikely there exists one best way to examine diversity among immigrants. The data show that the top sending country and region increased their share of the total foreign born nationally and in many states over the last decade. This decline in diversity was the result of very different rates of growth among immigrant groups.
There is also the question of the starting point for any comparison. While we compare 1990 to 2000, we could have compared 1980 to 2000. This does not mean the decline in diversity would necessarily be any less dramatic. For example, Mexico, the top sending country in 1980, increased it share of the total from 16 percent in that year to 22 percent by 1990 and 30 percent in 2000. In 1970 the top sending country — Italy — accounted for only 10 percent of the foreign born. Thus there is 30-year decline in diversity, at least as measured by the share represented by one country. It also should be pointed out that when mass immigration was beginning in the mid-nineteenth century, Ireland accounted for an even larger share of the foreign born than Mexico does today. However, Ireland’s standing as the top country was temporary and transitory. It was soon replaced by the nations that became Germany, though it probably makes more sense to see "Germany" as a cultural-linguistic collection of countries through much of this period, the way Spanish-speaking Latin America is today. Germany was later displaced by Italy. For at least the last 120 years, no country has accounted for such a large share of the foreign born as Mexico does today.
We also examined the changing settlement patterns of immigrants by country and found significant variations between countries. While immigrants from some countries tend to be very concentrated, those from other countries tend to be very dispersed. In general, European and South Asian immigrants tend to be the most dispersed, while those from Spanish-speaking Latin America tend to be the most concentrated. We also found that those countries that were the most concentrated in 1990 tended to exhibit the largest relative increase in dispersion, though they often remained among the most concentrated even in 2000. However, increasing dispersion was not a universal trend; the concentration of immigrants from many countries changed little or not at all during the decade.
What of the costs or benefits of the declining diversity or the changing distribution of immigrants by country across the United States? It seems reasonable to assume that the changing nature of immigration must have some implications for American society. While outside of the scope of this study, the most serious potential problem associated with a larger and less diverse immigrant population is that it may hinder the assimilation and integration of immigrants by creating the critical mass necessary to foster linguistic and spatial isolation. In contrast, a more diverse immigrant population may increase incentives to learn English or become familiar with American cultural more generally. The English language and American culture are the means by which diverse groups communicate with each other and the larger society. But if one group dominates in an area, then this could fundamentally reduce the need to Americanize.
On the other hand, there may be benefits to less diversity among immigrants. For example, if most immigrants in a state come from one cultural-linguistic group, then providing welfare or other government services may be easier for government agencies because they will only have to sensitize themselves to the needs of one immigrant community in order to deliver services. In addition, natives might find it easer to live in areas of heavy immigrant settlement if there is one dominant group because they will only have to learn to become familiar with one culture. For example, an American may only have to learn one foreign language rather than several in order to be employable.
It must be pointed out that this report does not address the costs or benefits created by the changing patterns of immigrant settlement. What it does provide is an detailed description of an important change taking place in American society.
Table 1 - State Data
Table 6 (click on each table for larger image)
Table 7 (click on each table for larger image)
1 The 1 percent census data for 1990 and 2000 were provided by the University of Minnesota. Steven Ruggles and Matthew Sobek et al. Integrated Public Use Microdata Series: Version 3.0 Minneapolis: Historical Census Projects, University of Minnesota, 2003. www.ipums.org
2 The public use file of the 1990 Census shows that a much larger share of the foreign born population did not record a country than in the 2000 Census. While this does not have a large impact on the overall results, it may affect the results for small countries.
3 The entire INS report, including figures for Mexico, can be found at
Steven A. Camarota is Director of Research at the Center for Immigration Studies. Nora McArdle was an intern at the Center and is currently a graduate student at Duke University.