Wordtrade LogoWordtrade.com
Social Science


Review Essays of Academic, Professional & Technical Books in the Humanities & Sciences


Migration Economics

The Economics of Migration edited by Klaus F. Zimmermann, Thomas Bauer (The International Library of Critical Writings in Economics Series: Edward Elgar) In the age of globalization, the importance of migration for the industrialized countries has increased. Inflows of migrants have steadily risen in the 1980s and the early 1990s. Yet while the public debate about policy responses to these developments continues unabated, research findings of economists are often ignored. In this four-volume collection the editors have selected some of the most significant contributions on the economics of migration, which provide an overview of the present state of empirical migration research. Topics covered include the migration decision, the integration of immigrants into the labor market and society, their economic behavior, empirical and theoretical contributions to migration policy, and the effects of immigrants on the native population.
The history of humankind is a history of migration. Labor mobility is an essential feature of the development of societies and also a prerequisite for economic success. During the 1980s and the early 1990s the inflow of migrants increased substantially in almost all OECD countries (see SOPEMI, 1999). A remarkable feature of these increasing migration flows is that- unlike in earlier decades - not only the traditional immigration countries (the US, Canada and Australia) are affected, but also many European countries. Between 1988 and 1997 the share of the foreign population increased from 4.5 percent to 9.1 percent in Austria, from 7.3 percent to 9 percent in Germany, and from 5 percent to 6 percent in Sweden (SOPEMI, 1999). Even countries that historically have been emigration countries, such as Italy, Spain and Ireland, recently became immigration countries.

The rapid increase in the size of the migration inflows over the past decade has led to rising tensions in various countries. The main issues in the policy debate were concerned with the causes of the increased immigration flows, the (mostly missing) integration of migrants into society and the labor market of the receiving countries, the effects of immigrants on the welfare of the native population, and possible policy responses to avoid mass immigration. A recent development in almost all major receiving countries has been to identify the high potential of migrants to foster rather than to hinder economic development. This can be linked to the increased competition for high quality labor on the global markets for experts. The information society has created a huge demand for additional high-skilled workers, and this will hardly be stopped by the current weakness in the US and European economies.

Following the public debate, economic research on migration topics has exploded in the last two decades. The articles collected here in four volumes attempt to reflect important developments in this research. An aim is to provide a useful point of departure for economists starting research in this area as well as a reading list for lecturers and graduate students. Those seeking an even broader overview on the topic are referred to some of the few textbooks and more popular books in the field such as Simon (1989), Jasso and Rosenzweig (1990), and Borjas (1990, 1999a). However, non-economists should also be attracted to our collection since the articles were selected for their relevance to the policy-oriented debate.

To choose articles from this large body of research for a project like this turned out to be a very demanding task. Our selection process has been guided by the following criteria: First, we have tried to identify influential papers or milestones for the development of the economics of migration. Due to the wide range of topics covered by research, and owing to lack of space, it was impossible to include all of these papers. Some of this literature was also omitted if it is nicely summarized elsewhere. For instance, we excluded papers associated with development economics, because this research has been covered already in the contributions of Lal and Coleman (1992) and Stark (1991).

Second, we have searched for an adequate balance of work related to North America, Europe and Australia. Many new influential developments originated in the US or Canada and were followed by large numbers of papers or books. This literature is also well documented and summarized. However, due to important differences in the structure of immigration and the labor market institutions, conclusions from this research are often not applicable especially in the European context. Hence, one attempt here is to support the research on Europe and Australia by collecting the somewhat more scattered research for this area.

Third, while there is a huge amount of work on the theoretical background of the migration issue, theory often gives no clear guidance to important questions in the policy debate. To provide focus, it was our decision to rely largely on empirical papers and literature that is, in our view, of a more immediate value to applied questions.

As usual, many colleagues will disagree with some of our choices for or against inclusion in this project. We are also aware that due to space constraints we would never be able to cover all important papers on the economics of migration. To fill the gap, we will refer in the sequel as much as possible to existing surveys that can be used to obtain an even more complete picture of the area. We will further provide a much more detailed bibliography on the economics of migration on the web page http://www.iza.org/migration. An attempt will be made to keep this web page updated and to provide interested researchers with the possibility of adding their work to this bibliography.

At this point we already wish to refer the reader to some excellent collective volumes which include many path-breaking contributions in the economics of migration and should be on the desk of every researcher interested in this area, such as Abowd and Freeman (1991), Borjas and Freeman (1992) and Chiswick (1992). We also recommend the collected volume by Hamermesh and Bean (1998), which includes several articles dealing with new questions of the effect of immigration on the native population, such as the impact of immigration on the quality of jobs and on the educational attainment of natives. Also worth reading is the volume by Smith and Edmonston (1998), which contains the background paper of the report of the Panel on the Demographic and Economic Impacts of Immigration to the US Commission on Immigrant Reform. An extensive survey of the immigration experience and the research on immigration in Europe is given in the volume edited by Zimmermann (2002).

The selected papers have been structured in four volumes, each of which deals. with one particular issue, namely Volume I, The Migration Decision and Immigration Policy, Volume II, Assimilation of Migrants, Volume III, Quality and Behavior of Migrants, and Volume IV, Migration and the Natives. This decision reflects the previously outlined focus on applied issues as well as the major strands of the migration literature research: Part I (The Migration Decision), Part II (Family Migration), Part III (Repeat and Return Migration), Part IV (Illegal Migration), and Part V (Immigration Policy).

Part I (The Migration Decision) deals with the application of the standard economic framework to the migration problem. Following the work of Becker (1962), the human capital model has been introduced to migration research by Sjaastad (Volume I, Chapter 1). This model, which treats migration as an investment problem, probably became the most influential and widely used approach to explain the migration decision. According to this approach, individuals calculate the present discounted value of expected returns in every potential region, including the home location. Migration occurs if the returns in a potential destination region net of the discounted costs of movement are larger than the returns achieved in the region of origin. The model has been extended in various ways. Carrington et al. (Volume I, Chapter 7) endogenized migration costs. Burda (1995) and Burda et al. (1998) (neither included in this collection) introduced uncertainty about the wage differential between the home and the receiving country and added the effect of the option value to wait for new information as an important determinant of the migration decision.

Many empirical investigations of the determinants of migration have been published for developed and developing countries. Surveys of this literature are provided by Krugman and Bhagwati (1976) and Bauer and Zimmermann (1997). Greenwood (1985) reviewed the empirical findings for developed countries and Todaro (1980) for developing countries. The latter deals largely with papers related to the Harris-Todaro model, which explains rural-urban migration in the context of a less developed economy. Empirical evidence based on aggregated data is contained in the work of Fields (Volume I, Chapter 4), who investigates migration within the US, and Greenwood and McDowell (Volume I, Chapter 5) who analyze migration to Canada and the US. Zimmermann (Volume I, Chapter 6) provides an analysis of labor migration to Germany, while Rotte et al. (Volume I, Chapter 8) investigate the causes of asylum migration from developing countries to Germany. Finally, the contribution of Hatton and Williamson (Volume I, Chapter 9) provides explanations for the mass emigration from Europe in the nineteenth century. An even more complete picture of the determinants and effects of mass migrations that took place between 1850 and 1914 is presented in the book by Hatton and Williamson (1998).

Empirical investigations using individual data to study the determinants of migration are rather scarce. This substantial deficit of applied research can be explained by the lack of adequate data, especially in the context of international migration. Hence, most of the existing studies using individual data tend to focus on internal migration. Bartel's (Volume I, Chapter 2) analysis of migration within the US serves as one example. Pissarides and Wadsworth (Volume I, Chapter 3) investigate the interaction between unemployment and mobility within the UK.

The work summarized in Part I of Volume I treats migration as a problem of individual decision-making. Another approach challenged many of the conclusions of this research by postulating that the migration decision is typically made by families or households rather than by individuals. This research is represented by papers in Part II (Family Migration). A first attempt to consider migration as the outcome of household decision-making has been made by Sandell (Volume I, Chapter 10) and Mincer (Volume I, (~h"pter 11). According to their approach a family will only migrate if the gains of one family member internalize the losses of other family members. Different to Sandell and Mincer, Borjas and Bronars (Volume I, Chapter 12) investigate the role of the family in the determination of the skill composition of immigrants to the US.

Several alternative concepts of modeling migration were developed. Many of them are included in the excellent survey by Stark (1991). Since this book is a collection of the important pieces of the literature that was called the "new economics of migration", single articles were not included here. One approach models migration as a risk-sharing behavior of families. In contrast to individuals, households are able to diversify their resources such as labor in order to minimize risks to the family income. This goal can be reached by sending some family members to work in foreign labor markets where wages and employment conditions are negatively or weakly correlated with those in the local region. Another approach in this literature develops from the assumption that families evaluate their income not only in absolute terms but also relative to other households. In the relative deprivation approach, migration occurs in order to improve the income of the household relative to a reference household.

Part III (Repeat and Return Migration) deals with the temporary nature of many migration flows. The theoretical literature treats return migration as part of life-cycle planning where return migration is an optimal decision-making phenomenon. In most existing models, return migration of individuals results from the assumption that the marginal utility of consumption is higher in the home country than in the receiving country (see Hill, Volume I, Chapter 14, Djajic and Milbourne, Volume I, Chapter 15, and Dustmann, Volume I, Chapter 17). Alternative motives for return migration, which are not included in this volume, are developed by Dustmann (1995) who assumes that relatively high returns to human capital investments made in the receiving country are responsible for return migration, and by Stark (1995), who models return migration as the result of employer learning about the skills of temporary migrants. Empirical evidence on the determinants and the structure of temporary migration is rather scarce. DaVanzo (Volume I, Chapter 13) investigates repeat and return migration within the US. An empirical analysis of the determinants and the structure of emigration of foreign­born in the US is provided by Borjas and Bratsberg (Volume I, Chapter 16).

Illegal migration (Part IV) has always been at the center of the policy debate in North America, but it also became increasingly important in Europe, especially for the South European countries (see also the special issue on illegal migration in the Journal of Population Economics edited by Katseli et al., 1999). Illegal migrants break laws and rules. They challenge the public transfer system since they pay no taxes and social security contributions, but they use public goods or publicly subsidized services, and they compete with natives and legal immigrants on the labor market. Apart from the heated public debate, illegal migration has only recently been captured by economic research.

A first rigorous treatment of the economic consequences of illegal migration is provided by Ethier (Volume I, Chapter 18), who investigates the implications of alternative policies towards illegal immigrants for the level of welfare and the income distribution in the host country. Djajic (Volume I, Chapter 20) gives a theoretical analysis of the effects of illegal migration on resource allocation, commodity prices, and wages of workers with different skills. He explicitly studies both the short-run effects when occupational mobility is restricted, and the long-run effects when the skill composition of the native workforce reacts to illegal migration. Building on the Harris-Todaro model (see Todaro, 1969, and Harris and Todaro, 1970), Todaro and Maruszko (Volume I, Chapter 19) provide an analytical framework of an individual's decision to migrate illegally when legal migration is possible.

Chiswick (Volume I, Chapter 21) provides a detailed economic discussion of the efficiency of the US Immigration Reform and Control Act (IRCA) of 1986, which granted legal status or amnesty to certain illegal immigrants and introduced penalties against employers who hire illegal immigrants. Hill and Pearce (Volume I, Chapter 22) use a general equilibrium incidence analysis to assess the long-run effects of the sanctions against employers hiring illegal immigrants on resource allocation and income distribution in the US. Another feature of the IRCA was an increased budget for apprehending undocumented migrants. Using Mexican data, Kossoudji (Volume I, Chapter 23) investigates whether border control and apprehension activity lead to changes in the migration behavior. Her results indicate that the main effects of this policy were that migrants stay in the US longer while they stay in Mexico for shorter spells between trips to compensate for the costs of increased apprehension. Cobb-Clark et al. (Volume I, Chapter 24) investigate the effects of IRCA on the wages of production workers in manufacturing. They find that sanctions against employers hiring undocumented migrants have a negative effect on wages whereas the amnesty of illegal workers affected wages positively. Both effects however, are very small.

Part V (Immigration Policy) collects papers that deal with direct policy concerns. The contribution of Simon (Volume I, Chapter 25) originated in the US context. It summarizes the major options for an explicit immigration country to choose economic migrants properly. The European context has been quite different, which has been well reflected by the slogan of a "Fortress Europe". Straubhaar and Zimmermann (Volume I, Chapter 26) argue that the stagnating and aging West European populations, the prosperous economic development of the European Union member countries, and the drastic political changes in Eastern Europe call for a substantial reform of immigration policies to open up the European labor markets. The paper proposes an active European migration policy. It is typically conjectured that immigrants would be harmful to the European economies in the face of high unemployment rates in the member states. The paper by Zimmermann (Volume I, Chapter 27) argues that a selective immigration policy is called for to cope with this issue. European unemployment is to a large extent a problem of unskilled labor. However, high-skilled labor migrants are needed and can contribute to creating jobs for low-skilled natives. So far, the literature on the economics of migration policy has been rather scarce. An exception is the paper by Benhabib (Volume I, Chapter 28). It studies how immigration policies that impose skill and capital requirements would be determined under majority voting when natives vote to maximize their income.

 Volume II: Assimilation of Migrants

 This volume investigates the way in which migrants assimilate to the destination country. Part I (Immigrants' Labor Market Assimilation: Evidence from North America) collects evidence from North America, Part II (Immigrants' Labor Market Assimilation: Evidence from Europe and Australasia) deals with Europe and Australasia. Part III is Migration and Self-Selection, Part IV The Role of Language. The book by Zimmermann (2002) contains a recent summary of the European experience.

Part I (Immigrants' Labor Market Assimilation: Evidence from North America) documents the origin and major steps of this debate. Following the pioneering work of Chiswick (Volume II, Chapter 1), the analysis of the earnings assimilation of immigrants towards those of natives has become one of the predominant topics in the economics of migration. Borjas (1994, 1999b) gives a survey of this literature. In his seminal article, Chiswick estimated a standard wage equation with a dummy variable indicating whether an individual is an immigrant and the number of years of residence of an immigrant worker in the US as additional explanatory variables using the 1970 cross-section of the US census. The estimation results by Chiswick indicate that at the time of entry to the US labor market an immigrant has an earnings disadvantage compared to a similar native worker. With time of residence the earnings of immigrants converge to those of natives and after about 15 years of residence the earnings of immigrants even exceed those of natives. Chiswick explains the initial earnings disadvantage with imperfect transferability of human capital across countries. As immigrants acquire the skills that are important in the US labor market, the human capital stock of immigrants grows faster than that of natives, which explains the convergence of earnings. Duleep and Regets (Volume II, Chapter 6) provide a theoretical model explaining why immigrants invest more in human capital than natives, and how human capital investments vary by type of immigrant. Chiswick's results, which showed that immigrants earn more than similar natives after about 15 years in the US, are explained with a favorable self-selection of immigrants in terms of unobservable characteristics such as ability and motivation.

The paper by Borjas (Volume II, Chapter 2) questions Chiswick's findings, arguing that his results are spurious because of cohort effects. When using cross-section data to estimate earnings assimilation of immigrants, a positive correlation between years since migration and earnings might evolve if the quality of immigrants decreases over time. Using the 1970 and 1980 US census, Borjas shows that for most immigrant groups within-cohort earnings growth is significantly smaller than the earnings growth predicted by cross-section regressions, indicating that there has been a decrease in the quality of immigrants to the US.

The contribution of Borjas initiated an intense debate on the measurement of earnings assimilation of immigrants. In addition to the problem of biased results on the estimated coefficient on years since immigration in a wage regression, researchers investigated the sensitivity of empirical results of immigrants' earnings assimilation with respect to the choice of the comparison group (see Yuengert, Volume II, Chapter 3) as well as the importance of period effects, which in most empirical studies are assumed to affect natives and immigrants in a similar way (see for example the article by LaLonde and Topel (1992), which is not included in this collection).

The debate on earnings assimilation of immigrants in the US also started a rich literature on assimilation of immigrants in other important immigration countries (Part II: Immigrants' Labor Market Assimilation: Evidence from Europe and Australasia), such as Canada (see Baker and Benjamin, Volume II, Chapter 4, and Bloom, Grenier and Gunderson, Volume II, Chapter 5), Australia (Chiswick and Miller, Volume II, Chapter 9), New Zealand (Winkelmann and Winkelmann, Volume II, Chapter 14), Germany (Dustmann, Volume II, Chapter 11, and Schmidt, Volume II, Chapter 13), Sweden (Aguilar and Gustafsson, Volume II, Chapter 10), France (Granier and Marciano, Volume II, Chapter 7), and the UK (Chiswick, Volume II, Chapter 8, and Bell, Volume II, Chapter 12).

The conclusion by Borjas (Volume II, Chapter 2) that the quality of migrants to the US declined over time created another discussion within the migration literature, which is concerned with the development and the determinants of the quality of migrants in terms of the stock of human capital they bring with them and the transferability of this human capital stock to the labor market of the host country. A sample of papers within this strand of literature is provided in Volume III of this collection.

One explanation of the decreasing quality, however, rests on the presumption that migrants are a self-selected group. Hence, several studies have investigated this hypothesis. Part III (Migration and Self-Selection) documents this literature. Borjas (Volume II, Chapter 17) uses a version of the well-known Roy model to show that both the mean and the variance of labor market opportunities in the home and the host country determine the incentives for individuals to migrate. Cobb-Clark (Volume II, Chapter 18) extends this analysis of migrant selectivity to female immigrants. Changes in the labor market opportunities in the sending and receiving countries lead to a change in the structure and hence the quality of immigrants. Even before Borjas used the Roy model to explain self-selection of international migrants, this approach had been applied to explain internal migration. Nakosteen and Zimmer (Volume II, Chapter 15) made the first attempt to incorporate endogenous selectivity into a model of migration and income using data for the US, and Robinson and Tomes (Volume II, Chapter 16) used the Roy model to compare the wages of migrants within Canada to the wages of stayers.

The assimilation pattern of immigrants' earnings to those of natives has been explained by the differential accumulation of human capital by immigrants and natives. An important factor identified by the literature is the ability of migrants to speak and write the dominant language of the host country. Papers investigating this issue are collected in Part IV (The Role of Language). In an early investigation for Canada, Carliner (Volume II, Chapter 19) shows that individuals whose native language was neither French nor English earned significantly lower wages than monolingual or bilingual English or French workers. McManus, Gould and Welch (Volume II, Chapter 20) and Grenier (Volume II, Chapter 21) conclude that immigrants in the US who are proficient in English earn significantly more than non-proficient immigrants. Chiswick (Volume II, Chapter 22) investigates the determinants of language proficiency among immigrants. His results indicate that English speaking and reading fluency increases with years of residence in the US. This increase in language competence is higher for those with more schooling. Estimating earnings regressions, he further shows that English reading fluency is more important for the labor market success of immigrants than speaking fluency. Chiswick's results are supported by the findings of Dustmann (Volume II, Chapter 23), who shows that writing proficiency is more important for the earnings position of immigrants in Germany than speaking fluency.

Empirical investigations of the effect of language ability on earnings suffer from a potential endogeneity bias since immigrants who expect higher earnings when they are fluent in the host country's language are also more likely to become fluent. Chiswick and Miller (Volume II, Chapter 24) have dealt with this criticism. They confirm for Australia, Canada, Israel and the US the presumption that language proficiency is indeed endogenous. Most studies on the determinants of language proficiency among immigrants find duration of residence in the host country to be one of the most important determinants. In addition, a common result of these studies is that immigrants who live in regions that have a relatively high representation of individuals speaking the language of their home country are less likely to become fluent in the dominant language. Chiswick and Miller (Volume II, Chapter 25) show that this effect is mainly due to formal ethnic networks, i.e. that measures of minority language concentration mainly capture interactions in the marriage market, with family and with ethnic media.

Volume III: Quality and Behavior of Migrants

 There have been substantial concerns about the development of immigrant quality upon arrival and the economic behavior of immigrants. Volume III provides a structure to this debate by collecting articles in the various areas: Part I (Immigration Policy and Immigrant Quality), Part II (Labor Supply), Part III (Self-Employment), Part IV (Welfare Benefits) and Part V (Intergenerational Issues).

According to the literature on the economic assimilation of migrants, an important determinant of economic success of migrants - and hence the costs associated with immigration - is the skill level of the migrant as well as the transferability of these skills to the new environment. Furthermore, changes in the skill levels of immigrants may alter the effect of immigration on the income level and the income distribution among natives. Part I (Immigration Policy and Immigrant Quality) addresses this issue. The article by Douglas (Volume III, Chapter 1) shows that the quality of immigrants was a main issue of the migration debate even early in the last century. A crucial issue of the debate on the quality of immigrants is the effectiveness of different immigration policies in selecting high-quality immigrants. As part of the debate on the earnings assimilation of immigrants (see Volume II), the debate on the quality of immigrants was revived. Using the 1980 US Census, Chiswick (Volume III, Chapter 2) analyzes the changes in the quality of migrants who entered the US between 1950 and 1980. His work leads to the conclusion that the rationing of immigration visas by skills tended to increase the average skill level of immigrants, whereas the growing number of immigrants admitted through kinship and the increasing number of illegal immigrants and immigrants from less developed countries lead to a lower average skill level.

Borjas (Volume III, Chapter 3) shows that changes in the US immigration policy in 1965 led to major changes in the country-of-origin mix of the immigration flows. The change in the origin of immigrants can explain the recent decline in skills and relative wages of successive immigration cohorts. Borjas (Volume III, Chapter 3) concludes that over 90 percent of this decline can be explained with changes in the source-country mix. Based on CPS data, Funkhauser and Trejo (Volume III, Chapter 6) conclude that immigrants who entered the US in the late 1980s are more skilled than those who entered earlier, which means that there is a break in the steady decline of immigrant skills between 1940 and 1980.

Jasso and Rosenzweig (Volume III, Chapter 5), Barrett (Volume III, Chapter 8) and Duleep and Regets (Volume III, Chapter 9) compared immigrants in the US who migrated under different admission criteria. Using longitudinal data on the 1977 US arriving cohort, Jasso and Rosenzweig (Volume III, Chapter 5) show that at the time of immigration those migrants admitted on skill-based reasons are more skilled than those admitted under family reunification. Evidence is found that those admitted on skill-based reasons experience downward mobility and those admitted because of family reasons face upward mobility. Hence, the difference between the two groups of migrants becomes significantly smaller with the time of residence in the US. Duleep and Regets (Volume III, Chapter 9) confirm the results of Jasso and Rosenzweig (Volume III, Chapter 5). Using a matched Immigration and Naturalization Service (INS)-Social Security Administration (SSA) data set, Duleep and Regets (Volume III, Chapter 9) find that immigrants admitted on the basis of kinship into the US earn substantially less at the time of immigration than immigrants admitted primarily on the basis of their occupational skills. The former, however, experience a faster wage growth with time in the US than migrants

finds that more than 50 percent of the native/immigrant self-employment rate gap is due to the fact that immigrant groups generally have higher home country self-employment rates than the US. Fairlie and Meyer (Volume III, Chapter 15) find that high relative returns to self­employment, the level of education, and years of residence are important determinants of self­employment. They cannot support the hypothesis that immigrants from countries with high self-employment rates also have a higher probability of becoming self-employed in the US.

Another important question in the immigration debate is whether immigrants pay their way in the welfare state. Part IV (Welfare Benefits) studies this issue. Existing research on the welfare dependence of immigrants focuses again on the assimilation of migrants in the host country and the development of immigrants' welfare participation behavior with time of residence (see Blau, Volume III, Chapter 16, Simon, Volume III, Chapter 17, Borjas and Hilton, Volume III, Chapter 20). The empirical results on welfare dependency further suggest that immigration policy plays a role: whereas immigrants are initially less likely to depend on welfare in Canada (see Baker and Benjamin, Volume III, Chapter 19), the dependency rates of immigrants in the US exceed those of natives (see Borjas and Trejo, Volume III, Chapter 18).

In contrast to the extensive literature on assimilation of first-generation immigrants, relatively little is known about the labor market performance of their children and grandchildren born in the host country. A selection of relevant articles is contained in Part V (Intergenerational Issues). Chiswick's (Volume III, Chapter 21) results indicate that, if anything, second generation immigrants have a slight earnings advantage over individuals with native-born parents. A positive relationship between the earnings of first and second-generation immigrants is found by Borjas (Volume III, Chapter 22). He further finds that the earnings of second-generation immigrants in the US are significantly affected by the economic conditions in the source countries of their parents. Gang and Zimmermann (Volume III, Chapter 23) show for second­generation immigrants in Germany that ethnic differences in educational attainment persist even after controlling for household and individual characteristics, and that parental education has no independent effect on the educational outcome for children in immigrant households but does have an effect for those in native households.

Volume IV: Migration and the Natives 

One of the most important and most sensitive debates in the migration literature deals with the effects of immigrants on the native population (see Greenwood and McDowell, 1986, 1994, Borjas, 1994, and Friedberg and Hunt, 1995, for surveys of the literature). The book by Zimmermann (2002) contains a recent summary of the European experience. We have allocated the relevant literature in the following four parts: Part I (The Effect of Immigrants on Natives: Theory), Part II (Labor Market Impact of Immigration: Evidence from North America), Part III (Labor Market Impact of Immigration: Evidence from Europe and Australasia), and Part IV (Migrants and Mobility of the Natives).

Different to the question of immigrant assimilation, which is to a large extent empirical, a growing theoretical literature analyzes the economic effects of migration on natives. Theoretical predictions of the economic effects of immigration depend on the specific assumptions of the model used. The most critical assumptions in this setting are the openness of sending and receiving countries to international trade, the technological relationship between immigrants

and natives, and whether one considers a perfect or an imperfect labor market in the host country. Part I (The Effect of Immigrants on Natives: Theory) presents a selection of relevant articles.

An early contribution is Reder (Volume IV, Chapter 1), who discusses the consequences of immigration on the economy of the host country in a standard neoclassical framework, where immigrants and natives are perfect substitutes. Berry and Soligo (Volume IV, Chapter 2) study the effect of emigration on the remaining population using a one-sector neoclassical growth model with two types of labor: unskilled and skilled workers. They show that if the two types of labor have a different savings behavior, and if migrants take their capital with them, then emigration leads to a decreased utility of the remaining population. Rodriguez (Volume IV, Chapter 3) finds that some of the conclusions of Berry and Soligo change when the savings behavior is modeled similar to the Samuelson-Diamond life cycle model of savings instead of being treated as exogenous. Usher (Volume IV, Chapter 4) extends the standard analysis of the economic effects of migration on the sending and receiving countries by taking into account the use of public goods by migrants. Johnson (Volume IV, Chapter 5) considers the impact of unskilled immigration on wages and employment among the native population of the host country explicitly allowing for the possibility of rigid wages for low-skilled workers. Ethier (Volume IV, Chapter 6) introduces international labor mobility of two types of labor into a standard factor-endowment model of international trade. Schmidt, Stilz and Zimmermann (Volume IV, Chapter 7) show that unskilled immigration may be beneficial even if there is an imperfectly competitive labor market with persistent unemployment among native unskilled workers. The contributions by Borjas (Volume IV, Chapter 8) and Bauer and Zimmermann (Volume IV, Chapter 9) provide simulation studies to quantify the potential labor market effects of immigration.

Several approaches have been used to estimate the impact of immigration on the labor market outcomes of natives. Part II (Labor Market Impact of Immigration: Evidence from North America) and Part III (Labor Market Impact of Immigration: Evidence from Europe and Australasia) contain research findings on the labor market impacts of immigration on natives. Various studies, such as Grossmann (Volume IV, Chapter 10) and Gang and Rivera-Batiz (Volume IV, Chapter 16), used a production function approach to calculate the elasticity of substitution between immigrants and natives. Most existing studies, however, look at the labor market effects of immigration on natives by estimating a reduced form wage equation, where the share of immigrants in a region or an industry is the explanatory variable of main interest. An analytical derivation of such a reduced form equation is given by Altonji and Card (Volume IV, Chapter 12). Using the reduced form estimation, Altonji and Card provide evidence on the wage effects of immigration to the US, De New and Zimmermann (Volume IV, Chapter 17) focus on Germany, and Winter-Ebmer and Zweimuller (Volume IV, Chapter 19) study Austria. Some studies analyze the employment effects of immigration by using the employment growth or the unemployment rate of natives as a dependent variable; see Winkelmann and Zimmermann (Volume IV, Chapter 15), Pischke and Veiling (Volume IV, Chapter 20), Winter-Ebmer and Zweimuller (Volume IV, Chapter 21) and Shan, Morris and Sun (Volume IV, Chapter 22). Finally, Borjas, Freeman and Katz (Volume IV, Chapter 13) consider the effects of immigration and trade on wage inequality.

Several problems are inherent in studies on the effects of immigration on wages and employment of natives. First, the share of immigrants in a region or industry is likely to be

endogenous, since immigrants may choose the location or industry in the host country where high wages can be achieved. To circumvent this problem, most authors rely on instrumental variable estimations. Another approach to avoid biases in the analysis of the wage and employment effects of immigration is to analyze "natural experiments" in immigration, where the timing and the location of immigration is not economically motivated. Examples of such studies are Card (Volume IV, Chapter 11), who investigates the migration of Cubans to Miami around the May 1980 Mariel boatlift, Hunt (Volume IV, Chapter 14), who looks at the repatriation of French residents of Algeria to France resulting from Algerian independence in 1962, and Carrington and de Lima (Volume IV, Chapter 18) who investigate the return of colonialists from Africa to Portugal after a revolution in 1974.

Overall, the empirical results of all these studies suggest that there is, if anything, only a negligible effect of immigration on the wages and employment of natives. However, it might be the case that the measured effects are biased downwards if increased immigration into an area results in an increased outflow of natives from this area. The issue is studied in Part IV (Migrants and Mobility of the Natives). The empirical evidence on the migration response of natives to an increased inflow of migrants is mixed. Whereas Frey (Volume IV, Chapter 24) and Filer (1992 - not included in this volume) find a positive link between an increased inflow of migrants and the outflow of natives, this conclusion was not confirmed by Butcher and Card (Volume IV, Chapter 23), Wright, Ellis and Reibel (Volume IV, Chapter 25) and White and Liang (Volume IV, Chapter 26).



Some Final Remarks


Much progress has been made in the study of human migration. Economists have contributed their experience in modeling human behavior in a straightforward and consistent way. Highly sophisticated econometric techniques have been applied to test theories and to provide empirical evidence on current effects and conditions. However, there is still a long way to go before a sufficiently acceptable stage is reached. What are, in our limited judgment, the most important and promising ways to go? First, due to the bias of the literature towards the US there is still a considerable lack of international comparable empirical evidence for many important questions. Topics where there is an urgent need for empirical evidence in other immigration countries than the US include the effects of different migration policies on the quality of migrants, the labor supply of migrants in a family context, self-employment of migrants, the use of social benefits and the effects of immigration on the location choice of natives. Second, migration is a multidisciplinary and strongly policy-oriented research topic. It seems important to incorporate institutional constraints and to evaluate policy measures. Third, international migration flows are not sufficiently monitored at the individual level. Therefore, we need to create international panel data sets that should be made widely available to the research community. Fourth, migration is an issue of dynamic economies. The movement of people is caused by the need for adjustments in economies, and the major economic virtue stems from the migrants' contribution to a faster adjustment process. Our theories and measurement methods do not yet provide sufficient tools to deal with this problem.

Headline 3

insert content here