Reconsidering Immigrant Entrepreneurship

An Examination of Self-Employment Among Natives and the Foreign-Born

By Steven A. Camarota on January 1, 2000

View PDF Version


Table of Contents

Executive Summary
Introduction
Self-Employment over the Last Four Decades
Full- and Part-Time Self-Employment Rates
Full- and Part-Time Self-Employment Income
Number of Employees
Socio-Economic Characteristics
Self-Employment by Geographic Area
Self-Employment by Region and Country of Origin
Illegal Immigrants
What Explains the Myth of Immigrant Entrepreneurship?
Conclusion
End Notes and References
Other Recent Publications


Executive Summary

One of the most common perceptions of immigrants is that of businesses owner: Stories of immigrant businesses revitalizing neighborhoods or playing a leading role in a particular industry have become a staple of immigration news coverage; advocates often argue for high levels of immigration on the grounds that immigrants infuse the country with an entrepreneurial spirit unmatched by natives; and Stephen Moore of the Cato Institute, one of the nation's leading immigration advocates, argues that immigration should be increased further because one of the distinguishing characteristics of immigrants is their "very high levels of entrepreneurship." Conservative commentator Joel Kotkin observes, "Today's immigrants be they Asian, European, or Latino are refueling our entrepreneurial economy." Unfortunately, generalizations of this kind often have been based on anecdotal evidence or an examination of only one immigrant group or immigrants in only one industry.

Drawing conclusions about overall levels of immigrant entrepreneurship from limited samples and unrepresentative populations is like trying to determine the level of criminality among immigrants by looking at the prison population in a high-immigration state like California. While immigrants make up a very large share of inmates in that state, it would be unfair to draw the conclusion that immigrants are disproportionately criminals. One would need to know what share of the total population immigrants account for in that state before making generalizations about immigration and crime. The same is true when examining entrepreneurship.

Unfortunately, most of the media coverage of immigrant entrepreneurship fails to place it into any kind of demographic context. In an effort to correct this situation, this reports looks at the self-employment patterns of all immigrants in order to determine the actual impact of immigration policy on the overall level of entrepreneurship in the United States. Using data collected by the Census Bureau, the report compares immigrant and native full- and part-time self-employment rates, self-employment income, the number of workers employed by immigrant and native businesses, and other characteristics. The findings indicate that while immigrants were once significantly more entrepreneurial than natives, this is no longer the case. Since 1980, immigrants and natives exhibit remarkably similar levels of entrepreneurship.

Findings

  • The non-farm self-employment rate of immigrants has fallen from 13.8 percent in 1960 to 11.3 percent in 1997. Over the same period, the self-employment rate for natives increased from 9.6 to 11.8 percent (see Figure 1 on page 8 ). As result, natives are now slightly more likely to be self-employed than immigrants, a reversal of the historic pattern. Natives are also slightly more likely to be self-employed on a part-time basis.
  • If self-employed farmers are included, the native advantage is even larger. In 1997, 12.3 percent of natives (including farmers) where self-employed, compared to only 11.4 percent of immigrants.
  • The presence of immigrants has no effect on the overall level of entrepreneurship in the United States. In 1997, 11.7 percent of all adults (including immigrants) were self-employed. Without immigrants, the rate would have been 11.8 percent.
  • The average annual income of self-employed immigrants is $20,710, somewhat higher than the average native self-employment income of $18,509. However, the majority of both immigrant and native entrepreneurs have very modest incomes, with more than half of both groups either losing money or earning less than $10,000 a year from their businesses.
  • There is no significant difference in the number of workers employed by immigrant and native entrepreneurs, with 86 percent of native- and 87 percent of immigrant- operated businesses employing fewer than 10 workers.
  • Even in the central cities of the nation's metropolitan areas immigrants are not more entrepreneurial than natives. The self-employment rate of natives and immigrants in central cities is exactly the same 10.3 percent.
  • There is almost no difference in the self-employment rate of immigrants and natives in the nation's major immigrant-receiving states and metropolitan areas.
  • The presence of immigrants as entrepreneurs or their availability as workers has no obvious effect on entrepreneurship activity in a metropolitan area. The total self-employment rate (native and immigrant) in the top immigrant-receiving cities is 11.4 percent. In contrast, the total self-employment rate in the rest of the country is 11.8 percent.
  • Immigrant entrepreneurs do not significantly affect the mix of businesses in the American economy. Based on the distribution of self-employed immigrants across industries, the types of businesses started by immigrants and natives are very similar.
  • While there is little difference between the entrepreneurship of natives and immigrants as a group, there are substantial differences between immigrants from different parts of the world. Immigrants from the Middle East, Europe, and East Asia have the highest self-employment rates and incomes, while immigrants from Mexico, Central America, and the Caribbean have the lowest.
  • Difference in education levels and duration of stay in the United states explain some difference among immigrant groups' level of self-employment. However, culture and other factors also play an important role in entrepreneurship.

Graph: Percentage of Natives and Immigrants Self-Employed, 1960-1997

Why Study Immigrant Entrepreneurship?

The entrance of over one million legal and illegal immigrants each year in the 1990s has wide ranging effects on the United States. In recent years there has been growing concern over high poverty rates and welfare use among immigrants, as well as their impact on schools, public coffers, and job competition with natives for low-wage jobs. What's more, because immigration has become the primary factor driving population growth in the United States, many worry that immigration worsens the problems of pollution, congestion, and sprawl. However, a number of authors have argued that high rates of entrepreneurship among immigrants offset, at least in part, the costs associated with immigration. If immigrants were significantly more entrepreneurial than natives, it certainly would be relevant to the immigration debate because it would mean that immigrants have a positive effect on the economy well beyond their numbers. By starting businesses at higher rates than natives or businesses that are very different from those operated by natives, immigrants could create new employment opportunities for native-born workers and more choices for consumers.

The results of this report, however, indicate that immigrants are not more entrepreneurial than natives. The presence of immigrants has no appreciable effect on the overall level of entrepreneurship in the United States. Instead, immigration increases the number of business owners creating jobs at about the same rate as it increases the number of people looking for jobs. The findings of the report also indicate that immigrants do not operate businesses much different from those operated by natives: The distribution of immigrant and native business across industries is very similar. There is also no indication that immigrants in the central cities of the nation's metropolitan areas are more entrepreneurial than natives the two groups have exactly the same self-employment rates. All the evidence examined in this report indicates that immigration is an unlikely vehicle for increasing entrepreneurship in the United States. Therefore, the debate over immigration instead should focus on other areas where immigrants actually do create real benefits and impose real costs on the United States. Entrepreneurship is simply not one of those areas.

Data and Methodology

Estimates of immigrant entrepreneurship in this paper are based on data from the last four Decennial Censuses as well as a combined sample of the March 1997 and 1998 Current Population Surveys (CPS) collected by the Census Bureau. All persons not born in the United States, one of its outlying territories, or of American parents living abroad are considered immigrants. All persons born in the United States, including the children of illegal aliens, are considered natives. While the Census Bureau refers to individuals born outside of the United States as "foreign-born", in this report foreign-born and immigrant are used synonymously. The March CPS is used because it contains an extra-large sample of Hispanics and is considered the best source of information collected on immigrants between the Censuses.

Both the Census and CPS ask respondents if they are self-employed. This question is used to determine self-employment rates for immigrants and natives. For the CPS, self-employment is based on whether the respondent was self-employed in the previous year of the survey. Annual self-employment income (revenue minus expenses) is also based on the previous year. Therefore, self-employment rates and incomes from the March 1997 and 1998 CPSs are for 1996 and 1997 respectively. Unless otherwise indicated, analysis in this study is confined only to employed persons 25 years of age and over who are not farmers.

What Explains the Myth of Immigrant Entrepreneurship?

If immigrants are not more entrepreneurial than natives, why does the stereotype continue to exist? It is certainly common for public perceptions to be incorrect about economic and demographic facts. For example, a number of surveys show that the public thinks teenage mothers make up the majority of the people on welfare, when in fact, administrative and other data clearly show that this is not the case. Another example of mistaken public perceptions is that most Americans, including minorities, overestimate by at least a factor of two the percentage African-Americans represent in U.S. population. We should therefore not be surprised that the perception of immigrant entrepreneurship also turns out to be incorrect. Only with a large, scientifically collected survey like the CPS or a national census is it possible to get an accurate economic or demographic picture of the nation.

While mistaken impressions may be common, it does not explain why the impression of immigrants as highly entrepreneurial exists in the first place. Three factors probably account for the stereotype: First, the fact that immigrants were once much more likely to be self-employed than natives may explain why the perception of immigrants as highly entrepreneurial exists. Many observers may simply be ascribing the traits of past immigrants to today's immigrants. Second, high levels of entrepreneurship among some immigrant groups may also account for the perception of immigrants as uniquely entrepreneurial. Elevated rates of entrepreneurship for Koreans and Middle Easterners, for example, may be falsely attributed to all immigrants. Third, policy makers, the public, and even reporters are more likely to come into contact with immigrant entrepreneurs than with immigrant workers. The immigrant restaurant owner who greets customers is much more likely to be remembered than are the immigrant cooks and dishwashers, whom the patron never sees. Most Americans have much more personal contact in their daily lives with self-employed immigrant street vendors or kiosk operators than with immigrant farm labors or construction workers. Since most people make generalizations based on their own experience, it is not surprising that they see immigrants as particularly entrepreneurial.

Conclusion

Like the negative stereotype of immigrants as predisposed toward crime, the positive stereotype of immigrants as super-entrepreneurs turns out to be incorrect. Both Census and CPS data indicate that a higher rate of entrepreneurship is not a distinguishing characteristic of immigrants. In terms of self-employment rate, income, and number of workers they employ, immigrants and natives are remarkably similar. As a result, immigration has no discernable effect on the overall level of entrepreneurship in the United States. Therefore, by itself, immigrant entrepreneurship cannot be used to justify or oppose current immigration policy. One must look elsewhere to make a case for or against immigration. There is no question that immigration has real effects on the United States, but increasing the level of entrepreneurship is not one of them. As immigration continues to be the subject of intense national debate, the focus should be on those areas where immigration creates real costs and benefits.

Introduction

A brief walk through an immigrant neighborhood in any of the nation's cities leaves an impression of intense entrepreneurial activity. Street vendors sell everything from produce to pajamas, while small shops and restaurants advertise in the community's native language. Stories of immigrant businesses revitalizing neighborhoods or playing a leading role in a particular industry have become a staple of news coverage on immigration, and many commentators use immigrant entrepreneurship to make the case for continued high levels of immigration.

Stephen Moore of the Cato Institute, one of the nation's leading immigration advocates, argues that "immigrants are highly entrepreneurial," and that immigration should be increased further because one of the distinguishing characteristics of immigrants is their, "very high levels of entrepreneurship" (Briggs and Moore, 1994, 108,122). "Today's immigrants be they Asian, European, or Latino are refueling our entrepreneurial economy," observes conservative commentator Joel Kotkin (Kotkin, 1991). California businessman and politician Ron Unz goes so far as to say that, "The overwhelming majority of California's immigrants are self-reliant and entrepreneurial..." (Unz 1994, 46). And Julian Simon, one of the country's strongest immigration advocates until his death in 1998, argued that one of the reasons immigration should be increased is that, "immigrants clearly have a higher propensity for self-employment than do natives." (Simon, 1989, 72). 1

Unfortunately, such observations are often based only on anecdotal evidence or studies of immigrants from one particular country, or only in one city or industry. Drawing conclusions about the overall level of immigrant entrepreneurship from limited samples and unrepresentative populations is not unlike trying to determine the level of criminality among immigrants as a group by looking only at San Quentin prison in California. While it is true that immigrants make up a very large share of inmates in that institution, it would be unfair to draw the conclusion that immigrants are disproportionately criminals; one would need to place incarceration figures from one particular prison into context before making generalizations about immigration and crime. The same is true when examining entrepreneurship. Unfortunately, most media coverage and many commentators fail to place immigrant entrepreneurship into any kind of demographic context.

In order to better inform the public discourse, this report examines immigrant entrepreneurship in a systemic fashion by using the most representative data available collected by the Census Bureau. The purpose of this report is to determine whether immigrants are significantly more entrepreneurial than natives. That is, are immigrants more likely to be self-employed than natives? Do their businesses generate more income than those of natives? And, how do immigrant and native businesses compare in terms of number of employees? In addition, this report compares immigrant and native entrepreneurs across localities and industries as well as differences between immigrants from various parts of the world. The findings indicate that while immigrants were once significantly more entrepreneurial than natives, this is no longer the case. Since 1980, immigrants and natives exhibit remarkably similar levels of entrepreneurship.

This report is unlike much of the recent work on immigrant entrepreneurship because it does not focus on differences between immigrant groups or factors hindering or facilitating immigrant self-employment (Razin and Light 1998; Fernandez and Kim 1998; Sander and Nee 1996; Yuengert 1995). Instead, it compares immigrants and natives in order to determine immigration's impact on the overall level of entrepreneurship in the United States.

Why Study Immigrant Entrepreneurship?

The entrance of more than one million legal and illegal immigrants each year has wide-ranging effects on the United States. In recent years, there has been growing concern over high poverty rates and welfare use among immigrants, as well as their impact on public schools and job competition with less-skilled natives (Camarota, 1999; Edmonston and Smith 1997; Borjas1999, Freeman and Katz 1997; Borjas and Hilton, 1996). What's more, because immigration has become the primary factor driving population growth in the United States, some have worried that immigration is worsening the problems of pollution, congestion, and sprawl (Beck, 1996; Bouvier and Grant, 1994). However, if immigrants are significantly more entrepreneurial than natives, then this might offset, at least in part, some of the costs associated with immigration. After studying immigrants in California's Silicon Valley, Annalee Saxenian, a professor at U.C. Berkeley, has observed that, "The [immigration] debate must be widened to reflect the new role of immigrant entrepreneurs as creators of jobs, wealth, and global linkages." (Quoted in Van Slambrouck 1999). A Business Week editorial in 1996 conceded that while there are costs to immigration, it "also boosts the country's entrepreneurial energy" (Oct 14, 1996, 142). Professor Ivan Light of UCLA points out that, "Immigrants have had a higher self-employment rate than natives for at least 100 years." High rates of entrepreneurship, he argues, mean that immigrants create more jobs than they take, and this should be taken into account by people who argue against immigration on the basis that it reduces opportunities for natives (Shaughnessy 1995). Saxenian, Light, and many others argue in effect that a full accounting of the costs and benefits of immigration needs to consider the contribution of immigrant entrepreneurs. If immigrants are more entrepreneurial than natives, it certainly suggests that immigrants have a significant positive effect on the economy well beyond their numbers because they are creating new employment opportunities for native workers and more choices for consumers.

On the other hand, if immigrants are less than or equal to natives in entrepreneurship, this would imply that immigration is not a vehicle for increasing the overall level of entrepreneurship in the United States. It would mean, for example, that immigration increases the number of business owners creating jobs at about the same rate as it increases the number of people looking for jobs.2 If this is the case, then the presence of immigrants has no discernable effect on the overall level of entrepreneurship in the United States. As a result, Saxenian, Light, Moore, and others are not correct and the debate over immigration should not take into account immigrant entrepreneurship, but instead, should focus on other areas where immigrants actually do create real benefits and impose real costs on the United States.

Data and Methodology

This report draws on data from the last four Decennial Censuses3 as well as a combined sample of the March 1997 and 1998 Current Population Surveys4 (CPS). The Censuses are used because they are virtually the only historical data available that identifies persons born outside of the United States referred to as "foreign-born" by the Census Bureau.5 (For the purposes of this report, foreign-born and immigrant are used synonymously.) The March CPS is used because it contains an extra-large sample of Hispanics and is considered the best source of information collected on immigrants between the Censuses. Both the Census and CPS ask respondents if they are self-employed. This question is used to determine self-employment rates for immigrants and natives.6 Unless otherwise indicated, analysis in this study is confined only to employed persons 25 years of age and over. In order to avoid confusing the decline in family farms that has taken place over the last half century with other trends in self-employment, self-employed farmers are excluded from the tables in this analysis.

Self-Employment over the Last Four Decades

Declining Immigrant Self-Employment

Figure 1 compares the percentage of immigrants and natives who are self-employed full-time7 from 1960 to 1997. The figure indicates that while at one time immigrants were significantly more entrepreneurial than natives, this is no longer the case. In 1960, 13.8 percent of immigrants were self-employed compared to only 9.6 percent of natives a 4.2 percentage point immigrant advantage. By 1970, the immigrant advantage in self-employment had declined to 2.8 percentage points. Since 1980 the self-employed rates of immigrants and natives have been very similar, with immigrants having only slightly higher self-employment rates in both 1980 and 1990. By 1997, even this very small advantage had entirely disappeared. In fact, natives are now slightly more likely to be self-employed than immigrants, a reversal of what had been true a few decades earlier. Figure 1 indicates that although once true, the conventional wisdom that immigrants are significantly more entrepreneurial than native-born Americans is not correct. At the very least, it is clear that self-employment is not a distinguishing characteristic of immigrants, nor has it been for the past two decades.

What Explains the Decline in Immigrant Entrepreneurship?

Three factors probably account for most of the decline in immigrant entrepreneurship found in Figure 1. First, in recent years a good deal of research has concluded that the average education level of immigrants has fallen relative to that of natives and the needs of the U.S. economy (Borjas 1999; Edmonston and Smith 1997). As will be seen in Table 4, education is positively correlated with self-employment. Therefore, falling relative levels of education are likely to account for some of the decline in immigrant self-employment.

Graph: Percentage of Natives and Immigrants Self-Employed, 1960-1997

Second, substantial changes in immigration law made in the 1960s, along with other factors, have resulted in a significant shift in the immigrant sending-countries. Prior to the 1960s, immigrants to the United States came overwhelming from Europe. By the 1970s, Latin American and Asian countries had come to dominate the immigrant flow. Because significant differences exist in entrepreneurship by region and country of origin (see Table 9), the changing composition of the immigrant population has likely played a role in the decline of immigrant self-employment.

Third, immigration has been increasing steadily since the 1960s. As result, a larger share of immigrants were new arrivals in 1997 than was the case in 1960. (Since 1980, however, there hass been almost no change in the proportion of immigrants who are new arrivals.) Because newly arrived immigrants are less likely to be self-employed than more established immigrants (see Table 1), the recency of the immigrant population must account for some of the decline in self-employment. All three factors taken together probably explain the decline in immigrant self-employment found in Figure 1. But whatever the reasons for the drop in immigrant entrepreneurship, it is clear that they are no longer more likely to be self-employed than natives.

Full- and Part-Time Self-Employment Rates

Full-Time Self-Employment

Table 1 (below) provides more detailed information on the self-employment patterns of immigrants and natives in 1998. The top portion of the table reports self-employment information for all immigrants and natives, while the bottom part reports it for immigrants by year of entry in 1997. The first column shows the average age for immigrants and natives. The second column reports the percentage of immigrants and natives who are self-employed full-time.8 As we have already seen, the non-farm native self-employment rate of 11.8 in 1997 was slightly higher than the 11.3 percent for immigrants.9 If self-employed farmers are included, the native advantage in self-employment increases. In 1997, the total self-employment rate of natives (including farmers) was 12.3 percent, almost one percentage point higher than the 11.4 percent for immigrants when farmers are included. Since farmers are overwhelming native-born, this result is certainly not surprising.

While natives do seem to be slightly more likely to be entrepreneurs than immigrants, even when self-employed farmers are included the native advantage still is small.10 Instead, what is striking about the full-time self-employment rates is the fact that the two groups are so similar and have been for almost 20 years. This similarity means that the presence of immigrants has no effect on the overall level of entrepreneurship in the United States. In 1997, the total self-employment rate (including immigrants) in the United States was 11.7 percent. If immigrants are excluded, the rate would have been 11.8. Put another way, immigrants make up 11.8 percent of the workforce11 and they comprise 11.4 percent of the nation's entrepreneurs. Thus, immigration has no discernible impact on the overall level of entrepreneurship in the United States.

Table: Self Employment Rates and Average Incomes for Immigrants and Natives in 1997

Part-Time Self-Employment

Some entrepreneurs continue to work at their regular jobs in addition to having their own businesses on the side. This is especially likely to be the case when they are trying to get a new business started. The third column in Table 1 shows the percentage of individuals who report having some self-employment income, but are not self-employed at their primary job. As is the case with full-time self-employment, natives are somewhat more likely than immigrants to be part-time entrepreneurs 2.3 percent compared to 1.4 percent of immigrants. While both groups have low part-time self-employment rates, the slightly higher rate for natives makes it clear that part-time self-employment is not a distinguishing characteristic of immigrants. Combining the full- and part-time self-employment rates of immigrants and natives indicates that 14.1 percent of natives compared to 12.7 percent of immigrants are entrepreneurs. Although the 1.4 percentage points separating the combined full and part-time self-employment rates of immigrants and with those of natives is still small, it is large enough to be statistically significant.12 However, even though the difference is statistically significant, it would be a mistake to characterize this difference as large. What is important about the self-employment rates in Table 1 is not their statistical significance, but instead the fact that like many stereotypes, the perception that immigrants are uniquely entrepreneurial turns out to be incorrect. If anything, natives are slightly more entrepreneurial than immigrants.

Self-Employment by Year of Entry

The lower portion of Table 1 presents the self-employment rate of immigrants based on when they entered the country. Examining immigrants by entering cohort is important because it is one way to determine how immigrants perform over time. The table shows that the longer immigrants are in the country, the more likely they are to be self-employed. In 1998, 17 percent of immigrants who entered prior to 1970 were self-employed on a full-time basis, 5.2 percentage points higher than that of all natives. The high self-employment rate of pre-1970 immigrants is not surprising because, as we will see in Table 3, age is positively correlated with self-employment. Because the average age of pre-1970 immigrants is 51 years, nine years older than the average age for natives, the self-employment rate of pre-1970 immigrants should be correspondingly higher. Natives in their fifties have very similar self-employment rates. The high self-employment rate of pre-1970 immigrants may also reflect the fact that most of these immigrants were admitted under the system that existed prior to 1965, which tended to produce a more educated flow of immigrants.

Table 1 also illustrates that the self-employment rate for immigrants who arrived in 1970 or after is 10.1 1.7 percentage points lower than that of natives. Looking at immigrants who arrived in each of the last three decades since 1970 shows rising rates of self-employment over time. The self-employment rate of 6.1 percent for 1990s immigrants is the lowest of any decade. Like the high self-employment rate for pre-1970 immigrants, the low self-employment rate for 1990s immigrants is also not surprising. 1990s immigrants are on average 36 years old, five years younger than the average native. Moreover, because 1990s immigrants have been in the country for only a few years, they are less familiar with the economy, language, and culture of their new home.

The self-employment rate for 1980s immigrants of 10.8 percent and the 12.7 percent for 1970s immigrants indicates that, as immigrants become more familiar with their new country, they have self-employment rates that are very similar to those of natives. Moreover, these immigrants are about the same age as natives 1970s immigrants are one year older on average and 1980s immigrants are three years younger than natives. Looking at immigrants who arrived in the 1970s and the 1980s indicates that, after a few years, immigrants are about as likely as natives to be entrepreneurs, but their rate of self-employment does not exceed that of natives.

Only when immigrants are much older than natives, as are pre-1970s immigrants, are their self-employment rates much higher than those of natives. But even then, as we will see, their self-employment rates do not exceed those of natives who are the same age. Table 1 also illustrates that the same basic pattern of rising self-employment rates exists when we look at part-time self-employment. However, unlike full-time self-employment, the part-time self-employment rate of immigrants never reaches the 2.3 percent for natives, even when immigrants have been in the country for many years. Of immigrants who arrived prior to 1970, 1.7 percent are self-employed part-time; for 1970s immigrants the figure is 1.4 percent; for 1980s immigrants 1.3 percent are self-employed part-time; and 1.2 percent of the most recent arrivals are part-time entrepreneurs. While natives do have higher part-time self-employment rates than immigrants, regardless of year of entry, part-time self-employment is rare for both groups. Few workers, immigrant or native, seem to be able to hold a job while also operating their own business on the side.

Full- and Part-Time Self-Employment Income

Unfortunately, there is no way to measure directly the total revenue of immigrant and native businesses. The CPS does, however, ask respondents how much self-employment income (revenue minus expenses) they earned in the previous year. Thus, it is possible to determine how much money immigrant and native entrepreneurs are generating for themselves. Income is important because it is one way to measure the size and success of a business.

Full-Time Self-Employment Income

The fifth column in Table 1 shows the average full-time self-employment income of immigrants and natives. While natives are slightly more likely to be self-employed, immigrants who are self-employed have slightly higher incomes. In 1997, the annual income of self-employed natives was $18,447, compared to $20,710 for immigrants. The self-employment income figures suggest that immigrant entrepreneurs are slightly more successful than native entrepreneurs, at least in terms of generating income for themselves. However, this difference is not large enough to be statistically significant.13 Moreover, self-employment is certainly no guarantee of success for either group. Forty-two percent of natives and 40 percent of immigrant entrepreneurs either lose money or only break even. An additional 18 percent of self-employed natives and 16 percent of self-employed immigrants reported that their self-employment income was less than $10,000 a year. Clearly, self-employment provides only a modest income for natives and immigrants alike. It should be noted, however, that it is common for self-employed persons to deduct as many expenses as possible in order to reduce their tax liability. This may explain the seemingly low self-employment income for both groups.

Part-Time Self-Employment Income

Part-time self-employment income is reported in the sixth column of Table 1.14 As is the case with full-time income figures, immigrants have somewhat higher part-time self-employment income. In 1998, the average part-time self-employment income for natives was $6,718, compared to $8,423 for immigrants. This is not a statically significant difference. Perhaps more important than statistical significance, the average part-time self-employment income figures show that both groups make less than $8,500 annually. It is clear that being self-employed on a part-time basis generates very modest incomes for the relatively few workers (immigrant and native) who are able to operate a business in addition to holding down another job.

Self-Employment Income by Year of Entry

The lower portion of Table 1 shows the average self-employment income for immigrants based on when they entered the country.15 The year of entry data indicate that while the income of immigrant entrepreneurs grows over time, the increase is not dramatic. The income of self-employed immigrants who arrived in the 1990s ($18,627) is equal to 90 percent of the average self-employment income for immigrants overall. Thus immigrants are more likely to become entrepreneurs over time, but the income of those who are self-employed does not change very much with duration of stay in the United States.

Number of Employees

In addition to income, the CPS also asks respondents how many employees work for their employer at all locations. For self-employed individuals, this question should provide a rough idea of the size of immigrant and native businesses. Table 2 (below) reports the number of employees working in native and immigrant businesses in 1997. The table shows that businesses run by both groups tend to be small, with roughly 87 percent of immigrant and 86 percent of native entrepreneurs having fewer than 10 workers. Overall, the two groups are almost identical in terms of the number of employees. This is also true when we examine the year of entry data located at the bottom of Table 2.

The year of entry data also show that immigrant businesses do not seem to grow significantly in size over time. For example, 87 percent of business operated by 1990s immigrants had fewer than 10 employees. In comparison, 86.9 percent of businesses operated by pre-1970 immigrants and 85.7 percent of businesses operated by 1970s immigrants have fewer than 10 employees.

It should be pointed out that the CPS is not designed to measure differences in the number of employees working for small business. The lowest value recorded by the survey is nine or fewer employees. However, if businesses operated by immigrants created significantly more jobs than those operated by natives, some differences in the number of employees should still show up in this data. To the extent that there is any difference in the number of employees between immigrant and native businesses, the results in Table 2 very slightly favor native over immigrant businesses. Overall, it seems that, like those of natives, the vast majority of immigrant businesses are small and remain so even when immigrants have been in the country for many years. Like the other measures of entrepreneurship examined so far, what stands out when we compare immigrant and native businesses in terms of employees is the similarities between the two groups, not the differences.

Table: Number of Employees Working for Immigrant and Native Businesses in 1997

Socio-Economic Characteristics

The comparisons made in this study so far have been for all employed persons 25 and older, with self-employed farmers excluded. This section of the report examines differences based on various socio-economic characteristics.

Self-Employment by Age and Education

Table 3 (below) reports self-employment rates by age for persons 25 and older. The table shows that as they age, both immigrants and natives are increasingly likely to be self-employed. The difference in self-employment rates for the two groups is one percentage point or less in every age group with the exception of persons 55 and over.

But even in this category, the 2.8 percentage-point advantage enjoyed by natives is not very large.

Table 4 (page 24) reports self-employment rates by education. The table shows that the more education a person has, the more likely he or she is to be self-employed. While there is more variation in self-employment rates between immigrants and natives by education than by age, in most educational categories immigrants and natives have very similar self-employment rates. Only among high school dropouts do natives seem to be somewhat more likely to be self-employed, and only among college graduates do immigrants have somewhat higher levels of self-employment. In general, Tables 3 and 4 indicate that there is not much difference between the two groups by education or age.

Self-Employment by Industry

All the evidence examined so far indicates that immigration does not have a significant impact on the overall level of entrepreneurship in the United States. Perhaps, however, immigrants benefit the American economy not by having higher rates of entrepreneurship or by the size of the businesses they create, but instead by the kind of businesses they operate. If immigrants start businesses very different from those started by natives, this might create significant benefits for natives because it could mean that immigrant businesses offer goods and services that might not otherwise be available. This in turn could create more choices for consumers and new opportunities for workers and native-owned businesses.

Table 5 (page 24) presents the percentage of native and immigrant entrepreneurs in each industry. The table reads as follows: In 1997, 6.2 percent of native entrepreneurs worked in manufacturing, compared to 5.4 percent of immigrant entrepreneurs. The table shows that immigrant and native entrepreneurs have a similar distribution across industries. The main exception is that native entrepreneurs are more likely to be in the construction business, while self-employed immigrants are more heavily concentrated in retail sales. It seems unlikely, however, that this difference is enough to significantly change the mix of businesses in the U.S. economy. The higher rate of self-employment in retail sales for immigrants increases the overall number of self-employed people in that industry in the U.S. economy by only 7.7 percent.16 This is not likely to make a significant difference in the overall economy, especially because a very large share of self-employed persons in retail sales are street vendors with particularly modest operations.

 

Table: Self-employment for Immigrants and Natives by Age in 1997

 

Immigrant businesses may be different from those of natives, but the data presented here are too aggregated to discern these differences. Given the sample size of the CPS, it is not possible to further divide self-employed immigrants into more detailed industrial classifications and still obtain reasonable estimates. With a larger sample and greater divisions by industry, significant differences might emerge. Based on this evidence, however, it seems unlikely that immigrants create a significant benefit to the economy by fundamentally changing the mix of businesses operating in the United States. They operate businesses very similar to their native-born counterparts.

Self-Employment by Geographic Area

Although immigrant levels of entrepreneurship are remarkably like those of natives nationally, this may not be the case across the United States. There may be parts of the country where immigrants are significantly more entrepreneurial than natives. The following section examines this question by looking at self-employment by state and metropolitan areas.

Self-Employment by State

Table 6 (below) reports immigrant and native self-employment rates for the top eight immigrant-receiving states. In six of the top eight states, natives have a higher self-employment rate than do immigrants. The largest difference is found in Arizona, where natives enjoy a 2.6 percentage-point advantage in self-employment. The smallest difference is in Florida where native self-employment is only .1 percentage points higher than immigrant self-employment. Illinois and New Jersey are the only states where immigrant self-employment is somewhat higher than that of natives. But even in those states the difference is small only 1.2 percentage points. The results indicate that there is little or no difference in the proportion of natives and immigrants who are entrepreneurs in the top eight immigrant-receiving states. Thus, despite the fact that immigrants differ significantly across states by country of origin and other demographic characteristics, there is not much difference in their self-employment rates in comparison to natives. The table also shows that self-employment is almost identical outside of the top immigrant states, with 11.8 percent of immigrants and 11.6 percent of natives being self-employed in the rest of the country. The fact that there is no significant difference in native self-employment between high- and low-immigrant states implies that the presence of immigrants does not have a significant positive or negative effect on self-employment among natives.

Table: Self-Employment for Immigrant-Receiving States in 1997

Self-Employment by Metropolitan Area

Table 7 (page 26) reports self-employment rates for the Consolidated Metropolitan Statistical Areas (CMSA) that have the largest immigrant populations, with several large central cities broken out separately. In the majority of metropolitan areas, natives either have a sizable advantage, as in the Dallas and Boston CMSAs, or less than two percentage points separate the two groups. This is true even in the San Francisco area, which includes the communities that comprise Silicon Valley. Of course, immigrants do make up a large share of the entrepreneurs in high-immigration CMSAs like San Francisco or Los Angeles. In 1997, for example, 27 percent of self-employed persons in the San Francisco CMSA were immigrants; however, immigrants also accounted for 27 percent of the local work force. This means that if immigrants are not counted, the overall level of self-employment in the San Francisco area (both city and CMSA) would be almost exactly the same. The findings in Table 6 indicate that it would not be correct to describe immigrants in San Francisco or other high-immigrant CMSAs as particularly entrepreneurial.

Table: Self-Employment for Immigrants and Natives in Top Immigrant-Receiving Consolidated Metropolitan Areas (CSMA)

There are significant differences between high-immigration metro areas: Both immigrants and natives are much more likely to be self-employed in the San Francisco and Miami CMSAs than in Chicago or Dallas. But what is important about the findings in Table 7 is that there is little difference between immigrants and natives within the same metro area. With the possible exception of the Washington-Baltimore CMSA, there is no city where the inclusion of immigrants can be said have a significant effect on the overall level of entrepreneurship.17

In addition, like the state data in Table 6, there is no indication that high-immigration areas exhibit higher levels of self-employment than those parts of the country with few immigrants. Nearly two-thirds of the nation's immigrants live in the nine CMSAs shown in Table 7, and immigrants make up 25 percent of the workforce in these CMSAs. The total self-employment rate (immigrant and natives) in the top nine immigrant-receiving CMSAs is 11.4. In contrast, in those parts of the country outside of the top immigrant-receiving metro areas, immigrants make up only 6 percent of the workforce and the total self-employment rate is 11.8. Immigrants appear to have no impact on the overall level of entrepreneurship in a metropolitan area. Therefore, the presence of immigrants as entrepreneurs or their availability as workers has no obvious effect on entrepreneurial activity in a CMSA.

Self-Employment in the Central Cities

One argument often made by advocates of high immigration is that immigrants start businesses where they are most needed. That is, immigrants in depressed central cities are particularly entrepreneurial (Muller 1993; Wattenberg and Zinsmeister 1989; Frost 1999). This in turn revitalizes blighted and declining cities by creating businesses where no other entrepreneurs wish to locate. Stories of immigrant businesses in inner-city neighborhoods have been a common theme in news coverage of immigrant entrepreneurship in recent years. In addition to reporting self-employment rates by CMSA, Table 7 also provides the self-employment rate for immigrants and natives in the central cities of four metropolitan areas. Only self-employment within the central city is reported. The cities are chosen because they are the only central cities with an immigrant population that is large enough to measure using the CPS. In all four cities, immigrant self-employment is actually lower, not higher, than that of natives. Thus, at least in these cities, it would appear that immigrants are not more entrepreneurial than natives.

Table 8 (below) examines self-employment for persons living in all of the central cities of the nation's metropolitan areas.18 The table shows that immigrants in central cities throughout the country have exactly the same self-employment rate as natives 10.3 percent. In 1997, immigrants comprised one-fifth of employed persons 25 and over in central cities, and they made up one-fifth of the entrepreneurs. It would seem, therefore, that contrary to the often-heard argument, immigrants are not particularly entrepreneurial in the inner-cities. That is, if there were no immigrants in the nation's central cities, the self-employment rate would be exactly the same.

Looking at the entire central city may not be a perfect measure of immigration's ability to revitalize blighted neighborhoods. It may be that immigrants are much more entrepreneurial in those parts of a city in desperate need of new businesses, while at the same time they are much less entrepreneurial in the rest of the central city. This creates an overall self-employment rate in the core city that is the same as natives. While certainly possible, there is no evidence that this is the case. Immigrant entrepreneurs can be found throughout cities with large immigrant populations. Moreover, one would still expect that if immigrants were highly entrepreneurial in the inner city, then their self-employment rates would be significantly higher than natives in the nation's central cities. Because this is not the case, the argument that immigrant businesses are revitalizing the nation's inner cities by starting businesses at rates unmatched by natives is called into question.

Table: Self-Employment for Immigrants and Natives in Central Cities, Suburban, and Rural Areas, 1997

Self-Employment by Region and Country of Origin

The fact that immigrants as a group are no more entrepreneurial than natives does not mean that this is true for immigrants from all sending countries. This section examines  self-employment by region and country of origin. Given the sample size of immigrants from some countries, the differences between countries should be interpreted with caution. The estimates are not quantified absolute differences, but can be used to make determinations of the relative differences between countries.

Self-Employment by Region of Origin

The first column of Table 9 (below) reports the self-employment rate of immigrants by region and country of origin, and the second column gives the average self-employment income for the self-employed. As we have seen, self-employment rates vary significantly by age and education. Therefore, the third column gives the average age for all immigrants from that region or country, while the fourth and fifth columns respectively show the percentage with less than a high school degree and the percentage with at least four years of college. Table 9 reads as follows: 6.3 percent of Mexican immigrants are self-employed, and self-employed Mexicans have an average annual self-employment income of $13,070. Mexican immigrants are, on average, 38 years old, with 64 percent being dropouts and 5.6 percent having at least a college degree.

Table: Self Employment, Age, and Education by Region of Origin and for Selected Countries, 1997

Table 9 reveals some dramatic differences in entrepreneurial activity among immigrants by region and country of origin. For example, the difference between Middle Eastern immigrants who have a self-employment rate of 28.2 percent, the highest of any region and the self-employment rate of 4.8 for Central Americans, the lowest of any region, is extremely large. By region of origin, immigrants from Europe, the Middle East, and Canada (if it is treated as its own region) have self-employment rates that are four or more percentage points higher than those of natives. In contrast, immigrants from Mexico (if it is treated as its own region) and Central America have self-employment rates that are more than four percentage points lower. The self-employment rate for immigrants from other regions of the world are within a few percentage points of natives.

Although it is not as large as the variation in self-employment rates, there are large variations in income between immigrant groups. Those groups that have the highest self-employment rate also tend to have the highest income. However, groups with dramatically higher self-employment rates do not have incomes that are proportionately as high. For example, the self-employment rate of Middle Eastern immigrants is six times that of immigrants from Central America, but their income is about double that of Central Americans. This suggests that there is far more variation in the propensity to be self-employed among immigrants than there is variation in success once an immigrant has become an entrepreneur.

Self-Employment by Country of Origin

Turning to countries of origin, we see somewhat more variation in entrepreneurship. In particular, Koreans, Cubans, Canadians, and immigrants from the United Kingdom have the highest self-employment rates, while immigrants from El Salvador, Guatemala, Mexico, Haiti, the Dominican Republic, and the Philippines have the lowest rates of self-employment. There is also significant variation within regions. For example, Cuban immigrants are much more entrepreneurial than Haitian or Dominican immigrants, even though they are all from the Caribbean. What's more, among East Asian immigrants, Koreans have much higher self-employment rates than immigrants from the Philippines or China. While the smaller sample size for individual countries, compared to regions, explains some of the variation, it is also clear that immigrants from some countries are substantially more entrepreneurial than those from other countries.

What Explains the Differences Between Immigrant Groups?

While Table 9 is not a fully developed explanatory model of self-employment, it does provide important clues to why there are such large differences between countries and regions. Those groups that are on average older because they have been in the country on average longer, such as Europeans and Canadians, are more likely to be self-employed and have higher self-employment incomes than those groups that tend to be composed of younger and more recent arrivals, such as Mexicans, Central Americans, and immigrants from sub-Saharan Africa. Education also explains some of the differences. A very large share of immigrants from Mexico and Central America lack even a high school education and only a small percentage have a college degree. This partly explains their low self-employment rates and incomes. In contrast, Middle Eastern immigrants are the most highly educated and they have the highest rate of self-employment by region. Clearly, differences in both education levels and age account for a good deal of the variation in entrepreneurship between immigrant groups.

Age and education, however, do not account for all of the differences between immigrant groups. For example, 27.5 percent of college-educated Middle Eastern immigrants between the ages of 35-44 are self-employed. In contrast, college-educated Mexican immigrants in the same age group have a self-employment rate of only 10.5 percent. Thus, much of the difference between the two groups remains even after controlling for these two factors. This suggests that in addition to education and age, other factors facilitate and hinder self-employment. These factors are likely to include: the amount of money immigrants bring to the United States, the existence of networks of co-ethnic business owners who can provide capital and expertise, discrimination by lenders, and cultural attitudes toward entrepreneurship. But whatever the reasons for the differences between immigrant groups, it is clear that current immigration policy does not produce a flow of immigrants that fundamentally alters the overall level of entrepreneurship in the United States.

Illegal Immigrants

Because the Current Population Survey attempts to create a representative sample of the entire U.S. population, a significant number of illegal aliens show up in the survey. It is certainly possible that being in the country illegally creates some impediments to self-employment. Of course, since being self-employed may involve little more than being an unlicenced street vender, it may be that legal status has little or no effect on self-employment rates. Assuming that it does reduce self-employment, the inclusion of illegal aliens in the CPS may understate entrepreneurial activity for legal immigrants. While it is not possible to definitively distinguish legal immigrants from illegal aliens in the CPS, it is possible to look at immigrants who, because of their demographic characteristics, are unlikely to be in the country illegally.

One way of doing this is to examine 1970s immigrants. Previous research indicates that by 1997 there where very few illegals among this group (Clark and Passel, 1998; Warren, 1999). Moreover, these immigrants have been in the United States for more than 20 years on average, so they are very familiar with the country. Tables 1 and 2 show that 1970s immigrants are very similar to natives in terms of self-employment. Immigrants who arrived in the 1970s do have somewhat higher self-employment rates 12.7 percent compared to 11.8 percent for natives. Even so, this modest immigrant advantage in self-employment may reflect the fact that 1970s immigrants are slightly older than natives on average. If we control for age, then the proportion of natives and 1970s immigrants who are entrepreneurs is almost identical. For example, 1970s immigrants between the ages of 45 and 54 have a self-employment rate of 13.1, almost exactly the same as the 13.3 percent for natives of the same age found in Table 3. In terms of part-time self-employment rates, as we have seen, natives enjoy a small advantage over 1970s immigrants of almost one percentage point. On the other hand, 1970s immigrants do have higher self-employment incomes than do natives. Turning to the number of employees, Table 2 shows that natives and 1970s immigrants are virtually indistinguishable from one another. All of the available evidence indicates that, in general, there are only very modest differences between 1970s immigrants and natives. Therefore, it seems unlikely that the presence of illegal aliens alters the overall findings in this report.

It should also be noted that when evaluating the impact of immigration, it may be better to see legal and illegal immigration as closely linked and not as distinct phenomena; many illegal aliens come to the United States to join friends and relatives who are legal residents. Sociological research indicates that one of the primary factors influencing peoples' decisions to emigrate is whether a family member or person from their home community has already come to United States (Massey and Espinosa 1997; Palloni, Spittel and Ceballos 1999). Communities of recent legal immigrants serve as magnets for illegal immigration by providing housing, jobs, and entree to America. Analysis done by the Immigration and Naturalization Service indicates that one out of four legal immigrants who receives a green card in any given year is in fact an illegal alien already living in the country (INS press release, January 1999). Thus, it is probably more accurate to view illegal immigration as a direct consequence of large-scale legal immigration and not as a separate phenomenon that should be dealt with independently. If this is the correct interpretation, then it makes more sense not to separate out illegal aliens from legal immigrants, but instead to look at the characteristics of all immigrants as reflective of the nation's immigration policy in its totality.

What Explains the Myth of Immigrant Entrepreneurship?

If immigrants are not more entrepreneurial than natives, then why does the stereotype continue to exist? It is certainly common for public perceptions to be incorrect about economic and demographic facts. For example, a number of surveys show that the public believes that teenage mothers comprise the majority of the people on welfare, when in fact, administrative and other data clearly show that this is not the case. Another example of mistaken public perceptions is that most Americans, including minorities, grossly overestimate the percentage African-Americans represent in the U.S. population. We should therefore not be surprised that the perception of immigrants as highly entrepreneurial also turns out to be incorrect. Only with a large scientifically collected survey like the CPS or a national census is it possible to get an accurate economic or demographic picture of the nation.

While mistaken impressions may be common, this phenomenon does not explain why the particular impression of immigrants as highly entrepreneurial exists in the first place. Three factors probably account for the stereotype: First, the fact that immigrants were once much more likely to be self-employed than natives may explain why the perception of immigrants as highly entrepreneurial exists (see Figure 1). Many observers may simply be ascribing the traits of past immigrants to today's immigrants. Second, high rates of entrepreneurship among some immigrant groups, such as Koreans and people from the Middle East, may be falsely attributed to all immigrants. Third, policy makers, the public, and even reporters are probably more likely to come into contact with immigrant entrepreneurs than with immigrant workers. For example, the immigrant restaurant owner who greets customers is much more likely to be remembered than are the immigrant cooks and dishwashers whom the patron never sees. Most Americans have much more personal contact in their daily lives with self-employed immigrant street vendors and cab drivers than with immigrant farm or construction laborers. Because most people make generalizations based on their own experience, it is not surprising that immigrants are seen as particularly entrepreneurial. These three factors taken together probably explain, to a large extent, why immigrants are seen as highly entrepreneurial.

Conclusion

There is an unfortunate tendency among proponents of high immigration on the one hand and those favoring lower levels of immigration on the other hand to exaggerate the attributes of immigrants. In the case of entrepreneurship, the findings of this report indicate that the arguments made by the advocates of high immigration about immigrant entrepreneurship are simply not supported by the available data. While at one time immigrants do seem to have been significantly more entrepreneurial than natives, since 1980 there is little difference between immigrants and natives. In fact, in 1997, the full-time self-employment rate of natives of 11.8 percent was slightly higher than the 11.3 percent for immigrants. Detailed comparisons of immigrants and natives in 1997 show that they are remarkably similar not only in their full-time self-employment rates but also in terms of their part-time self-employment rates, self-employment income, and in the number of workers they employ.

Immigrant and native entrepreneurs are also distributed across industries in a very similar manner. Thus, immigrant entrepreneurs are not likely to change significantly the mix of businesses operating in the country. Equally important, there is no indication that the presence of immigrants has any effect on the overall level of entrepreneurship in the nation's central cities. The self-employment rates of immigrants and natives in the central cities of the nation's metropolitan areas are exactly the same 10.3 percent. Thus, the often-heard argument that immigrants are revitalizing America's inter-cities by starting new businesses at rates unmatched by natives is simply not supported by the evidence examined in this report.

Although immigrants as a group cannot be described as "highly entrepreneurial," immigrants from some countries and regions of the world are more entrepreneurial than do natives. In particular, immigrants from the Middle East, Europe, Korea, Canada, and Cuba have much higher self-employment rates and incomes than natives. High levels of self-employment for immigrants from some countries may also partly explain why the perception exists that immigrants as a group are uniquely entrepreneurial. Elevated levels of entrepreneurship may be falsely attributed to all immigrants.

It is possible that immigrants are significantly more entrepreneurial than natives, but Census Bureau data cannot be used to discern the differences. It is, however, difficult to imagine a scenario in which immigrants are much more entrepreneurial than natives without showing up in significantly higher full- or part-time self-employment rates, self-employment income, or in terms of workers employed. If immigrants were more entrepreneurial than natives, then some differences should emerge from the data. Analysis of Census data is certainly capable of showing such differences. Korea, Canadian, and Middle Eastern immigrants were found to have much higher self-employment rates and incomes than natives. The fact that the data are able to show significant differences should give us some confidence in the overall conclusion of this report that differences in entrepreneurship between immigrants and natives are small.

These results are not inconsistent with previous scholarly work. In its extensive study of the impact of immigration on the United States, the National Research Council touched only very briefly on immigrant entrepreneurship. After examining the evidence, however, the report did conclude that, "it is difficult to argue that self-employment and any entrepreneurship associated with it mark a critical distinction between immigrants and the native-born." (Edmonston and Smith 1997, 162.) All the evidence examined in this report certainly supports this conclusion.

Policy Implications

Because immigration does not have a significant effect on entrepreneurship in the United States, the implications of this report are clear: Policy makers and those engaged in the immigration debate should focus on other areas where immigration actually does have a significant effect on the United States. For example, research indicates that by increasing the supply of unskilled labor, immigration drives down wages for low-wage natives. This in turn should generate higher returns on investment for owners of capital, and higher wages for more skilled workers who can now be paid more. Concern for the working poor on the one hand or the interests of businesses on other should be one area of focus for policy makers.

Immigration policy has also become the determinate factor in driving population growth in the United States. Census Bureaus projections indicate that immigration at the current level will add 80 million residents to the United States by 2050 (Spencer 1996). Some argue that population growth is always desirable because it creates more opportunities for businesses and consumers, while others are concerned about the congestion, pollution, and suburban sprawl that comes with increases in population. Whatever one's views on population growth, it is clear that immigration has a profound effect on the size of the U.S. population. Immigration also has a significant impact on schools and public coffers as well as important cultural implications. All of these issues as well others should be carefully considered by policy makers and voters when determining the desirable level of immigration. Entrepreneurship, on the other hand, should simply be removed from the current immigration debate because it is one area where immigrants have little or no impact.

End Notes

1 This represents only a very small sample of quotations from advocates and commentators using immigrant entrepreneurship to make a case for high levels of immigration.

2 It is also possible that immigrants are beneficial not because they are more entrepreneurial than natives, but because the businesses they start are very different from those operated by natives or because they locate in areas under-served by native entrepreneurs, such as inter-city neighborhoods. These questions are explored later in this report.

3 For the 1960, 1970, 1980, and 1990 Censuses, the study uses a 65,000-person sample drawn from the Integrated Public Use Microdata Series provided by the University of Minnesota.

4 The CPS is a sample containing 50,000 households conducted each month by the Census Bureau primarily to collect employment and labor force data.

5 All persons not born in the United States, one of its outlying territories, or of U.S. parents living abroad are considered immigrants. All persons born in the United States, including the children of illegal aliens, are considered natives. Both the Census and the CPS include some illegal aliens and a small number of persons on temporary visas, such as students, which allow them to stay for multiple years in the United States.

6 For the March 1997 and 1998 CPS, self-employment is based on whether the respondent was self-employed in the previous year. Annual self-employment income (revenue minus expenses) is also based on the pervious year. Therefore, self-employment rates and incomes from the March 1997 and 1998 CPS are for 1996 and 1997 respectively.

7 Full-time self-employment means that these individuals are self-employed at their primary jobs.

8 Full-time self-employment does not refer to hours worked per week; instead, persons are considered self-employed full-time if they are self-employed at their primary job.

9 This means that 11.2 million out of 95.1 million employed natives age 25 and over were self-employed. For immigrants, a little more than 1.4 million out of 12.7 million employed workers age 25 and over were self-employed.

10 The slightly higher self-employment rate for natives is not statistically significant at the .90 level. For the difference between all natives and all immigrants to be statistically significant, a little more than one full percentage point should separate the two values. Because the sample sizes are much smaller, for self-employment by state, metropolitan area, or for particular immigrant groups, differences would have to be much larger in order to be statistically significant.

11 Workforce is defined as employed persons 25 and over, excluding self-employed farmers.

12 This assumes a 90 percent confidence interval. That is, there is a 90 percent chance that there is an actual difference between immigrants and natives in the population.

13 The statistical significance is based on a 90 percent confidence interval. This difference is not statistically significant because of the relatively small sample size of self-employed immigrants along with the great variation in self-employment incomes among both immigrants and natives. There is also a good deal of variation between the two years. In the March 1998 CPS, the difference between immigrants and natives was only $986, considerably less than the $2,263 when the March 1997 CPS is included.

14 These figures are for persons who report self-employment income, but are not self-employed at their primary occupation. Individuals who are self-employed part-time and have zero income from their business are not identified.

15 Part-time self-employment income by year of entry is not reported because less than two percent of immigrants report part-time self-employment income. Dividing this small percentage by year of entry would make the sample to small to obtain reliable estimates of part-time self-employment income by year of entry.

>16 This figure is derived by comparing the actual number of immigrant entrepreneurs in retail sales with the number who would be employed in that industry (14.4 percent) if immigrants had the same concentration as natives in retail sales.

17 This difference is not statistically significant.

18 The central cities data are based on Metropolitan Statistical Areas (MSA) or Primary Metropolitan Statistical Areas. The central cities are composed of the core urban area of each MSA or PMSA. In many cases this is larger than just the city at the center of the metro area.

References

Beck, Roy. 1996. The Case Against Immigration: The Moral, Economic, Social, and Environmental Reasons for Reducing U.S. Immigration Back to Traditional Levels. New York: W.W. Norton & Company.

Briggs, Vernon M., and Stephen Moore. 1994. Still the Open Door? U.S. Immigration Policy and the American Economy. Washington, DC: American University Press.

Borjas, George J. 1999. Heaven's Door: Immigration Policy and the American Economy.Princeton: Princeton University Press.

Borjas, George J., and Lynette Hilton. 1996. "Immigration and the Welfare State: Immigrant Participation in Means-tested Entitlement Programs." The Quarterly Journal of Economics. May 1996.

Borjas, George J., Richard B. Freeman, and Lawrence F. Katz. 1997. "How Much Do Immigration and Trade Affect Labor Market Outcome?" Brookings Papers on Economic Activity. Washington DC: Brookings Institution Press. Vol 1.

Bouvier, Leon F., and Lindsey Grant. 1994. How Many Americans: Population, Immigration and the Environment. San Francisco: Sierra Club Books.

Camarota, Steven A. 1999. Importing Poverty: Immigration Impact on the Size and Growth of the Poor Population in the United States. Paper # 15. Washington, DC: Center for Immigration Studies.

Clark, Rebecca L., and Jeffrey Passel. 1998. Identifying Legal and Illegal Immigrants in the Census and Current Population Surveys: A New Technique Based on the Occupational Distribution of Illegal Aliens. Washington DC: Urban Institute.

Edmonston, Barry, and James Smith Eds. 1997. The New Americans: Economic, Demographic, and Fiscal Effects of Immigration. Washington, DC: National Academy Press.

Fernandez, Marilyn, and Kwang Chung Kim. 1998. "Self-Employment Rates of Asian Immigrant Groups: An Analysis of Intragroup and Intergroup Differences." International Migration Review. Vol 32, No 3: 0654-0681.

Frost, Martin. January 22, 1999. "Where Freedom-Seekers Yearn to be Free: Halting Immigration Deprives America of Fresh Talent that Keeps it on Top." Knight Ridder/Tribune News Service.

Kotkin, Joel. September 1991. "The American Way." Inc. p. 96.

"Legal Immigration, Fiscal Year 1997." January 1999. Press Release. Immigration and Naturalization Service, Office of Policy and Planning, Statistics Branch.

Massey, Douglas S., and Kristin E. Espinosa. 1997. "What's Driving Mexico-U.S. Migration? A Theoretical Empirical, and Policy Analysis." American Journal of Sociology.102: # 4 939-999.

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

Palloni, Alberto, Mike Spittel, and Miguel Ceballos. 1999. Using Kin Data to Falsify Social Networking Hypotheses in Migration. Paper presented at the annual meeting of the Population Association of America.

Razin, Eran, and Ivan Light. 1998. "Ethnic Entrepreneurs in America's Largest Cities." Urban Affairs Review Vol 33, No 3: 332-360.

Sanders, Jimy M. and Victor Nee. 1996. "Immigrant Self-Employment: Family as Social Capital and the Value of Human Capital" American Sociological Review. Vol. 61 April 231-249.

Shaughnessy, Rick. January 31, 1995. "Many Immigrants Demonstrate A Flair for Creating Jobs." San Diego Union-Tribune.

Simon, Julian L. 1989. The Economic Consequences of Immigration. Cambridge MA: Basil Blackwell, Inc.

Spencer, Gregory. 1996. "Population Projection of the United States..." Current Population Projections. Washington, DC: GPO. p. 25-1130.

Unz, Ron K. 1994. "Sinking our State; Immigration." Reason. Vol. 26, No. 6. p. 46.

Van Skambrouck, Paul. August 27, 1999. "Give Me Your Huddled Masses ... er Geniuses." Christian Science Monitor. USA section, p. 3.

Wattenberg, Ben J., and Karl Zinsmeister. December 1989. The Comparative Advantage of the First Universal Nation. Discussion Paper for American Enterprise Institute Annual Policy Conference.

Warren, Robert. 1999. Unauthorized Immigrants Residing in the United States: Estimating the Populations, Components of Change, and Trends, by Broad Area of Origin, 1987 to 1997. Paper presented at the annual meeting of the Population Association of America.

Yuengert, Andrew M. 1995 "Testing Hypotheses of Immigrant Self-Employment." Journal of Human Resources. Vol. 30, No1. p. 194-204.

Other Recent Publications

The Impact of New Americans: A Review and Analysis of the National Research Council's The New Americans: Economic, Demographic, and Fiscal Effects of Immigration

By Steven A. Camarota and Leon Bouvier, (November 1999) $12

 

Importing Poverty: Immigration's Impact on the Size and Growth of the Poor Population in the United States

By Steven A. Camarota

Shows that immigration is a major factor in the size and growth of the poor population. (September 1999) $12

 

Doctors and Nurses: A Demographic Profile

By Leon Bouvier and Rosemary Jenks

Examines the impact of foreign-born doctors and nurses on the United States and provides a detailed demographic profile of these professions. (February 1999) $12

 

"What, Then, Is the American, This New Man" Conference Report.

"Is There an American People," by Nathan Glazer; "Is America Too White?" by John Isbister; and "Do We Really Want Immigrants to Assimilate?" by Peter Skerry as well as responses by Orlando Patterson, Noah Pickus, Linda Chavez, Peter Brimelow, Lawrence Fuchs, and John Fonte. (October 1998) $12