How To Calculate The Labor Market Effects Of Immigration And Emigration In Oecd Countries

Transcription

1 The Labor Market Effects of Immigration and Emigration in OECD countries Frédéric Docquier (FNRS-IRES, Université Catholique de Louvain) Çağlar Özden (The World Bank) Giovanni Peri (University of California, Davis) September 1st, 2011 Abstract This paper uses an improved database of the stocks of migrants by education level and country of origin for the years 1990 and 2000 to simulate the labor market effects of net immigration and emigration during the 1990 s in all OECD countries. Due to the much higher international mobility of college graduates relative to all other individuals, we find that net migration flowsare, ingeneral, verycollege-intensive relative to the population of non-migrants. Using a simple aggregate model of labor demand and supply in each country we simulate the long-run employment and wage effects of immigration and emigration. We use a range of parameter values that spans most of the estimates in the literature. In all cases we find that immigration in the 1990 s had a positive effect on the wage of less educated natives and it also increased or left unchanged the average wages in OECD countries. It also had positive or null effect on native employment. To the contrary, emigration had a negative effect on the wage of native less educated workers and it contributed to increase inequality in all OECD countries. These results hold true also when we correct for the estimates of undocumented immigrants, for the skill-downgrading of immigrants, when we focus on immigration from non-oecd countries and when we consider (still preliminary) measures of more recent immigration flows for the period JEL Codes: F22, J61, J31. Keywords: Immigration, Emigration, Complementarity, Schooling Externalities, Average Wage, Wage inequality. Frédéric Docquier, frederic.docquier@uclouvain.be; Çağlar Özden, cozden@worldbank.org; Giovanni Peri, gperi@ucdavis.edu. This article was funded by the project "Brain drain, return migration and South- South migration: impact on labor markets and human capital" supported by the Austrian, German, Korean, and Norwegian governments through the Multi-donor Trust Fund on Labor Markets, Job Creation, and Economic Growth administered by the World Bank s Social Protection and Labor unit. The first author also acknowledges financial support from the Belgian French-speaking Community (convention ARC 09/ on "Geographical Mobility of Factors"). We thank Francesco D Amuri, Francesc Ortega and participants to seminars at Bocconi University, the OECD, the World Bank, Universitat Autonoma de Barcelona, Copenhagen Business School and University of Helsinki for valuable comments and suggestions. The findings, conclusions and views expressed are entirely those of the authors and should not be attributed to the World Bank, its executive directors or the countries they represent. 1

2 1 Introduction Immigration rates in OECD countries are larger than in the rest of the world and they increased significantly in the last 20 years. As of year 2000 about 7.7% of the adult residents of OECD countries were born in another country (hence migrants), versus only 2.9% in the average world country (Freeman, 2006). Since then the number of foreign-born has further increased and they were estimated to be about 10% of the OECD resident population in Worries about immigration feature prominently in the rich countries media and are voiced by several European government officials. The common representation of recent migratory flows portrays a multitude of uneducated individuals, coming from countries at the margin of the developed world, trying to gain access to the labor markets and to the welfare systems of rich countries. Such phenomenon, the simplistic view continues, is a main contributor to depressed wages and job losses for the less educated natives, a group of workers that has under-performed in the labor markets during the last 20 years. The available data and recent analysis however (e.g. Docquier and Marfouk, 2006; Grogger and Hanson, 2011) are uncovering a different picture of international migrants. First, a large part of the mobility to OECD countries is from other OECD countries (about 4% of the 7.7% foreignborn residents of OECD countries in 2000 were from another OECD country). Second, the share of college educated among immigrants is usually larger than (or not different from) the share of college educated in the native population of the receiving country. In some OECD countries the percentage of college educated among immigrants is much larger than among natives (up to 4-5 times). Our paper uses a new database that combines original Census data for 76 receiving countries in year 2000 and 61 receiving countries in year 1990 to measure (or estimate in the case of the missing host countries) bilateral net immigration and emigration flows of all OECD countries in the 1990 s. The data distinguish between college educated migrants and non college educated ones and we can distinguish net flows for each bilateral relation between a OECD country and any other country in the world. These data are new, they are described in detail in the report by Docquier et al. (2010). They constitute a substantial improvement on Docquier and Marfouk (2006) especially in the construction of net emigration data from OECD countries because they include more than double the number of receiving countries. Using these data and simple models of the national economies and labor markets we simulate the employment and wage effect of immigration and emigration on nationals of each OECD country. The model used is a simple application of production-function based models that have become popular in the labor literature of the national analysis of effects of immigration (Borjas 2003; D Amuri et al., 2010; Ottaviano and Peri, forthcoming, Manacorda et al forthcoming). Macroeconomic studies analyzing growth, productivity and skill premium in the US and other countries have also used similar framework (e.g. Acemoglu and Zilibotti 2001, Caselli and Coleman 2006, Card and Lemieux 2001, Goldin and Katz 2008). To this framework that enables us to derive a labor demand by skill group we add a simple labor supply decision that generates an aggregate supply curve (also by skill group). The two facts about migration mentioned above have clear implications on the potential labor market effects of immigration and emigration. First, as OECD countries are important 1 These figures are from OECD (2011). 2

3 sources as well as destinations for many of the recent world migrant it makes sense to analyze the potential labor market effects of both flows, immigration and emigration, and see whether OECD countries should be concerned with the domestic labor market effects of either of them. Second, if migrants have a large share of college graduates, immigration can actually help wages and employment of less educated (through complementarity) and possibly average wages (through human capital externalities). To the contrary emigration may hurt them. Immigrants in Barcelona or Athens are more visible to the European public eye than the Spanish or Greek engineers in Silicon Valley, but are they more harmful to the labor market options of Spanish and Greek workers with low education? While there are scores of studies on the labor market effects of immigration in OECD countries there is only a handful of studies on the labor market effects of emigration. This asymmetric view may lead to misconceptions of the economic effects of migration. The goal of this paper is to assess the impact of global labor movements in the 1990 s on the wages of those who did not migrate. We consider all OECD countries as well as the remaining non-oecd countries of Europe, some of which were affected by significant flows in the 1990 s. The existing estimates of the labor market effectsofimmigrants areoftenseenasdifferen- tiated or even as conflicting with each other 2. Most of the disagreement, however, is relative to the US labor markets and it is limited to moderate differences about the wage impact of immigrants on less educated workers. We take a differentapproachhere. Wecapturethe range of disagreement across economists with different estimates of the fundamental parameters of our labor market model. In particular, different choices of the elasticity of relative demand between more and less educated and between native and immigrants, of the elasticity of supply of more and less educated and of the strength of human capital externalities can be varied to produce different scenarios. Those can be associated with the most pessimistic, an intermediate and a most optimistic view of the labor market effects of immigrants. This is what we will do, without taking any stand on the current debate but showing what range of effects one obtains by allowing the whole reasonable spectrum of parameter estimates. Some general results emerge from these simulations. First, in general, over the period immigration had zero to small positive long-run effect on the average wages of non-migrant natives in all the OECD countries. These effects were, usually, positively correlated with the immigration rate of the country (the size of immigrant flow relative to the population). They ranged from 0 to +4%. Canada, Australia and New Zealand, which adopted immigration policies explicitly selecting in favor of highly educated had significant positive wage gains of natives from immigration. However also countries as Luxembourg, Malta, Cyprus, the United Kingdom and Switzerland (which did not explicitly select their immigrants based on education) experienced positive average wage effects in the order of 1 to 3%. Second, immigration had even more beneficial effects on wages of less educated (non-college) workers in OECD countries. These effects ranged between 0 and +6%. For some countries such as Ireland, Canada, Australia, United Kingdom and Switzerland the effects are likely in the 2-4% range. Only Austria, Denmark, Italy, Japan and Greece show estimated effects on the wages of less educated (in the most pessimistic scenario) that are close to 0. A corollary of this result is that immigration reduced the wage differential be- 2 The estimates in Borjas (2003) and Card (2001, 2009) are considered as spanning the range between the more pessimistic and more optimistic view of the labor market effects of immigrants. 3

4 tween more and less educated natives. Third,emigration, to the contrary, had a negative and significant effect on wages of less educated natives ranging between 0 and -7%. In countries as Cyprus, Ireland and New Zealand, less educated workers suffered wage decrease between 3 and 6% due to emigration of the highly skilled. Even in Portugal, United Kingdom, South Korea, Latvia and Slovenia less educated suffered losses between -1 and -2% because of emigration. All three results logically proceed from the nature of measured migrant flows from 1990 to OECD countries have experienced both immigration and emigration that were usually more "college-intensive" than their domestic labor forces. Under these conditions, in the long run, immigration is associated with average wage gains of less educated workers and emigration with average wage losses for non-migrant less educated workers. As the skill composition of migrants is crucial in determining our relative and average wage results, we attempt to correct for the "effective" skill content of immigrants in a series of checks. First, we use estimates of the extent of illegal immigration (from the recent "Clandestino" study, performed in several European countries) to correct for the inflows of undocumented migrants. Second, we account for the potential "downgrading" of immigrants skills in the host countries labor market, by using their occupational choice as of Third, we focus on immigration from non-oecd countries which is often portrayed as particularly low skilled. All corrections reduce the percentage of "effective" highly educated immigrants in OECD countries. However none of them eliminates or reverses the general picture described above: when immigration changed the educational composition of residents, it changed it by increasing the share of college educated (highly skilled). Finally we consider two further extensions of our data and simulations. First, we redo the exercises for a subset of countries for which we have provisional net immigration data for the 2000 s ( ). We choose some European countries that received large immigration flows (including Luxembourg, Spain and Greece which were very large receivers of immigrants) and the US. For these countries data from the EU Labor Force Survey or from the American Community Survey are available. They are based on smaller samples relative to the censuses and hence subject to larger measurement error. Even in this case, however, we find that the long-run effects of the more recent immigration flows are between zero and positive for all countries. For Luxembourg, the biggest recipient of immigrants in this period, they can be as large as +6% for less educated. For Spain, usually considered as the country most affected by immigrants in the 2000 s the wage effect on less educated natives range between 0 and +2%. The second extension is to account for sluggish capital adjustment in the short-run (rather than assume perfectly elastic capital supply), using estimates of the speed of adjustment from the literature. In this case, Spain and Cyprus exhibit a negative short-run impact on wage of less educated, when accounting for downgrading, undocumented and slow capital adjustment. The effects are between -1 and -2%. For all other countries experiencing net immigration, the short-run effect on wages of less educated are very small (between -0.5 and +0.5%) and for Luxembourg they are still positive at +2.2%. The rest of the paper is organized as follows. Section 2 presents the simple aggregate production and labor supply framework from which we derive wages and employment effect of exogenous immigration and emigration shocks. Section 3 describes the data, their construction the main sources and it shows some simple summary statistics about the educational structure of labor force data and migrant data. Section 4 presents the basic results 4

5 of the simulated wage effects of immigrants using our model and the range of parameters available from the literature. Section 5 considers the wage effect when accounting for undocumented workers, for downgrading of skills, looking at non-oecd immigrants and using the preliminary data on net immigration in the 2000 s. Section 6 concludes the paper. 2 Model We present here a simple aggregate model of each country for which we derive the labor demand and labor supply for different types of workers: native and foreign-born. This allows us to examine the long-run wage and employment effects of exogenous net immigration and emigration flows, which constitute change in the supply of workers of certain skill types. The analysis builds on the literature that aims at identifying the impact of immigration on national labor markets. Immigration has the main effect of changing the marginal productivity (demand) of native workers. Emigration changes the exogenous supply of potential workers. The long-run consequence of these changes on employment and wages of natives depend crucially on the elasticity of substitution across workers and on the elasticity of labor supply. The size and the educational composition of immigrants and emigrants relative to the non-migrants are the crucial factors in determining the long-run domestic wage and employment effects of international migrations. 2.1 Aggregate production function The prevalent model adopted in this literature (Borjas 2003, Card 2001, Ottaviano and Peri forthcoming) is based on a production function where the labor aggregate is represented as a nested constant elasticity of substitution (CES) aggregation of different types of workers. In the production function (we omit country subscripts for simplicity), we assume that output in year ( ) is produced according to a constant-returns-to-scale Cobb-Douglas production function with two factors, physical capital ( ), and a composite labor input ( ): = e 1 (1) The term e represents the total factor productivity (TFP), and is the income share of labor. Assuming that physical capital is internationally mobile in the long-run and that each single country is too small to affect the global capital markets, the returns to physical capital are equalized across countries. If denotes the international net rate of return to capital, the following arbitrage condition implicitly defines the equilibrium capital-to-labor ratio in the economy: =(1 ) e (2) Theaboveconditionholdsintheshortandinthelongruninasmallopeneconomy. In a closed economy as in Ramsey (1926) or Solow (1951) condition (2) holds in the longrun balanced growth path, with being a function of the inter-temporal discount rate of individuals (or of the savings rate) 3. Hence in the long-run we can substitute this arbitrage 3 As long as immigration does not change the saving rate of an economy the pre- and post- migration are identical. 5

6 condition into (1) to obtain an expression of aggregate output as linear function of the aggregate labor : = (3) where e1 [(1 ) ] (1 ) is an increasing function of TFP and is referred to as modified TFP. Production function 3 can be considered as an aggregate production function in which inputs (workers) can be adjusted immediately and the long-run and short run effects of a change in workers are the same as there is no slowly "accumulated" factor. Alternatively it can be considered as the long-run reduced version of a function with capital accumulation. In long-run (balanced growth path) analysis the two interpretations are equivalent. Following the labor (Katz and Murphy 1992, Card and Lemieux 2001, Acemoglu and Zilibotti 2001) and growth (Caselli and Coleman 2006) literature, we assume that labor in efficiency unit ( ) is a nested CES function of highly educated ( ) and less educated workers ( ): " 1 = +(1 ) 1 # 1 where and 1 are the productivity levels of highly educated workers (college educated) and less educated workers (non college educated). The parameter, is the elasticity of substitution between the two groups of workers. We also distinguish between natives and immigrants within each labor aggregate and. If native and immigrant workers of education level were perfectly substitutable, the aggregate would simply be equal to the sum of natives and immigrants labor supplies. However native and immigrant workers of similar education may differ in several respects. First, immigrants have skills, motivations and tastes that may set them apart from natives. Second, in manual and intellectual work, they may have culture-specific skills and limitations (e.g., limited knowledge of the language or culture of the host country), which create comparative advantages in some tasks and disadvantages in others. Third, even in the absence of comparative advantage, immigrants tend to concentrate in occupations different from those mostly chosen by natives because of migration networks, information constraints and historical accidents. In particular, new immigrants tend to cluster disproportionately in those sectors or occupations where previous migrant cohorts are already over-represented. Several studies (such as Card, 2009; D Amuri et al 2010; Ottaviano and Peri, forthcoming; Manacorda et al., forthcoming) find imperfect degrees of substitution between natives and immigrants. Hence, we assume that the quantities of high-educated ( ) and low-educated labor ( ) are both nested CES functions of native and immigrant labor stocks with the respective education levels. This is represented as: = 1 +(1 ) 1 1 (4) where = (5) where is the number of type- native workers and is the number of type- foreignborn workers who are present in the country. is the elasticity of substitution between natives and foreign-born in group. 6

7 2.2 Schooling externalities We also consider the possibility of a positive externality from highly educated workers, in the spirit of the recent literature (Acemoglu and Angrist 2000, Ciccone and Peri 2006, Moretti 2004a, 2004b and Iranzo and Peri 2009). There is a large body of growth literature (beginning with Lucas 1988, and extending to Azariadis and Drazen 1990, Benhabib and Spiegel 2005, Cohen and Soto 2007 and Vandennbussche et al 2009) that emphasizes the role of human capital (schooling) on technological progress, innovation and growth of GDP per capita. More recently, however, the empirical literature has pointed out that while it is sometimes hard to find an effect of human capital on growth of income per capita (Benhabib and Spiegel 2005), there seems to be evidence that human capital contributes to the level of income per person beyond its private returns. This implies that TFP is an increasing function of the schooling intensity in the labor force. Such formulation is particularly appropriate to be included in our model and, based on the expressions used in Moretti (2004a, 2004b), the TFP can be expressed as follows, µ + = 0 µexp (6) where 0 captures the part of TFP independent of the human capital externality, and is the semi-elasticity of the modified TFP to the fraction of highly educated individuals in the labor force (in parenthesis) that we call. is the total number of individual of education and nativity-status in working age 4. The upper bar distinguishes total population in working age from the actual employed individuals of the same group denoted as Acemoglu and Angrist (2000) and Iranzo and Peri (2009) also use a similar formulation to express schooling externalities and we use their estimates of the parameter. 2.3 Labor Demand We consider each country as a single labor market. Then we derive the marginal productivity for native workers of both education levels ( and ) by substituting (4) and (5) into (3) and taking the derivative with respect to the quantity of labor and respectively. This yields the following expression which defines the labor demand for each type of national workers: µ 1 µ 1 = (7) = (1 ) µ 1 µ 1 (8) By taking the logarithm of the demand functions above and calculating the total differentials of each one of them with respect to infinitesimal variations ( ) of the employment of 4 We assume that the schooling externality depends on the college intensity of the population in working age. In practice this is very similar to the college intensity of workers and simplify the analysis much. Noninstitutionalized individuals in working age represent the total pool of potential workers in the long-run. It is, therefore, reasonable to consider them as the relevant group in generating productive externalities. 7

8 each type of workers, we obtain the percentage change in marginal productivity in response to employment changes. In particular, the percentage change of marginal productivity for native workers of schooling level (= ) in response to a percentage change in employment of foreign-born ( and ) and native ( and ) workerscanbewrittenasfollows 5 : = 1 µ µ 1 1 µ (9) 1 + for = In expression 9 the terms represent the share of wage bill going to workers of schooling level (= ) and place of origin (= ) The firstterminthesummationcapturesthe effect of changes in the employment of each group on the marginal productivity of natives of type through the term in the wage equation. The second term in the summation, which depends only on the change in supply of workers of the same schooling type (natives or immigrants), is the impact on marginal productivity of natives of type through the terms inthewageequation.theterm 1 captures the impact through the term 1. Finally the terms is the effect of a potential change in college intensity, measured as a change in the share of the college educated in the working-age population, through, the Total factor Productivity. 2.4 Labor Supply We assume that a native household of type (= ) chooses how to split one unit of labor endowment between work and leisure 1 to maximize the following instant utility function 6 which depends positively on consumption and negatively on the amount of labor supplied : = (10) The constant parameters, and (= ) can be specific to the schooling group. Considering production without capital as in expression 3, in each period individuals consume all their labor income and hence the budget constraint implies = for each period 7. The term isthewagerateofnativeworkerofschoolinglevel. Substituting the 5 The details of the derivation are in the Appendix B. 6 The more common form of the utility function when analyzing the labor-leisure choice in a macro model is: = 1 exp( ) 1 1 This function would generate an individual labor supply that is independent from the wage paid. Hence it would produce a special case of expression 11, with perfectly rigid (vertical) supply curve. 7 The model with savings and capital accumulation could be solved with the alternative utility function = 1 exp( ) 1 1 as an intertemporal optimization model. In that case, which is illustrated at page of Barro and Sala i Martin (2004) the labor supply in balanced growth path does not depend on wages. Consumption would be a constant fraction of income and, in balanced growth path wages would be growing at the rate of e Hence it would be a special case with perfectly inelastic individual labor supply. 8

9 constraint into the utility function and maximizing with respect to we obtain the labor supply for the individual household. where = = ³ is a constant and = = 0 is a positive parameter, that captures the elasticity of household labor supply. As there are individuals in working age among households of type then the aggregate labor supply of that type is given by: = for = (11) As described above isthewagepaidtoanativeworkerofschooling, (as defined in section 2.2) is the working-age population in group, > 0 is the elasticity of labor supply and is an unimportant constant. Forimmigrantswemakeafurthersimplification assuming that all immigrants in working age supply a constant amount of labor (call it ) so that total employment of immigrants is: = for = This is equivalent to the assumption that immigration employment is rigid to changes in wages. It is also equivalent to considering immigration as an exogenous shock and tracking its labor market impact on natives which is customarily done in the labor literature on the impact of immigrants (e.g. in Borjas 2003, Borjas and Katz 2007, Ottaviano and Peri, forthcoming). 2.5 Equilibrium effect of exogenous immigration and emigration An exogenous change in the foreign-born population in working age ( and ) and in the native population in working age ( and )arewhatwecallnet immigration and net emigration. They generate changes in the marginal productivity of native workers as well as changes in the supply of native workers. In the new equilibrium each of the two labor markets for more and less educated native workers will respond these flows reaching a new wage and employment combination. In particular considering an immigration "shock" represented by and and an emigration "shock" represented by and the following two conditions (for each market) represent the percentage change in demand and supply of native labor: = 1 µ µ 1 1 µ (12) 1 + for = = 1 µ for = (13) The new equilibrium is obtained by equating the percentage wage changes in the demand and supply of native labor in each market and solving for and For compactness 9

10 we define as b = thepercentagechangeofanyvariable. We also denote the change of marginal productivity of native workers of schooling due to exogenous immigration as: \ 1 ³ µ = [ + [ µ d + for = (14) Solving the system of (12) and (13) and simplifying we obtain the following two linear equations in two unknown that summarize the equilibrium in each labor market: \ + [ + 1 µ [ 1 + [ =0 (15) \ + [ + 1 µ [ 1 + [ =0 (16) In expression (15) the first two terms \ + [ represent the shift (in percentage terms) of the intercept of the logarithmic demand function for native workers of schooling while 1 [ is the (negative of) the change in the intercept of the logarithmic ³ supply of native workers of schooling 8 The coefficient = is the (absolute value) of the slope of the logarithmic demand function for native workers of type while 1 is the slope of the logarithmic supply for those workers 9. The interactions between the two markets ( and ) isduetothefactthatachangein employment of workers with schooling affects the marginal productivity of, and hence the demand for, workers of schooling level through the term [ In turn, employment in the market affects productivity (and demand) for workers of type (through [ ). Solving the system (15)-(16) with respect to [ and [ we obtain the following equilibrium changes in employment and wages of natives in response to immigration and emigration that we denote with a star: [ = ³ [ = ³ 1 + ³ \ + 1 [ + ³ \ ³ ³ ³ \ + 1 [ + ³ \ ³ ³ [ (17) + 1 [ (18) 8 Correspondingly, in equation (16) the term \ + \ is the shift in labor demand for native workers of schooling and 1 \ is the negative of the change in labor supply for those workers. 9 Correspondingly = 1 + ³ 1 1 demand for native workers of schooling while 1 is the (absolute value of) the slope of the logarithmic is the slope of their logarithmic supply. 10

11 and = 1 ³ d d for = (19) 2.6 Special cases Two extreme cases are of interest to define the boundaries of potential wage and employment effects. If wages are perfectly flexible and supply of workers perfectly inelastic ( = and = ) the whole effect of exogenous immigration and emigration shocks accrues to wages. This is a reasonable scenario, especially in the long-run, and it is the one considered by most national studies (such as Borjas 2003). In such a case there is no endogenous response of native employment to the exogenous immigration and emigration shocks and the change in equilibrium wages is: d = 1 ³ [ + [ + [ + [ + (20) µ 1 1 µ d + d 1 d + for = At the opposite extreme of the spectrum the wage would be completely rigid so that any change in exogenous immigration is absorbed by changes in employment. In this case there is no wage impact of immigration and the employment response of natives to the exogenous inflow of immigrants is given by: ³ ( ) \ [ = ³ ( ) \ [ = + + ³ \ (21) ³ \ (22) Such case is well outside the reasonable range of long-run effects, as assuming full wage rigidity in the long-run is unreasonable. Most of the estimates of the labor supply imply that it is very rigid (the elasticity of labor supply to wages is estimated much closer to 0 than to infinity) and hence this perfectly elastic case is simply a theoretical case. 2.7 Imperfect physical capital adjustment and the short-run effects With full capital adjustment to its balanced growth path equilibrium the effects of migration on wages are as described above. However, immigration in the short run may affect the capital labor ratio 10 = pushing the economy outside its balanced growth path. 10 We use the slightly modified expression for the capital-labor ratio that includes the labor composite (rather than employment) at the denominator. 11

12 Due to slow adjustment of capital there will be an extra effect of immigrants on marginal productivity of each type of worker, through. We can re-write the expressions of marginal productivity in (??) and(??) noting explicitly their dependence on the capital-labor ratio: µ 1 = e ( ) 1 = e ( ) 1 (1 ) µ µ 1 1 (23) µ 1 (24) In balanced growth path (or in an open economy) the capital-labor ratio is given by the following expression = 1 1 e 1 and it does not depend on the inflows and outflows of migrants. In that case expressions (23) and (24) reduce to (7) and (8) and the total percentage effect of immigrants and emigrants on marginal productivity are as in (15) and (16). In the short run, however, may deviate from its long-run equilibrium due to exogenous migration shocks. In this case the percentage effect on marginal productivity, included in expressions 17 and 18 of each group, includes an extra term, namely: \ = \ short run +(1 )b short run = (25) where b short run is the short-run percentage deviation of the capital-labor ratio from due to exogenous immigration or emigration. Considering, immigration 11,withimmediateandfull capital adjustment, b short run equals 0. At the opposite extreme, with fixed total capital, = then b short run is approximately equal to the negative percentage change of labor supply due to exogenous immigration and emigration which is: +, where the numerator is the net change in workers due to the total net immigration and is, approximately, the employment at the beginning of the period. While for short periods or for sudden shocks the short run effect may be evaluated considering capital as fixed in the case of ten years of immigration, which is a slow and sustained phenomenon, we need to account explicitly for the dynamics of the capital-labor ratio to recover a short-run effect. A popular way to analyze the deviation of ln( ) from its balanced growth path trend, used in the growth and business cycle literature, is to represent its time-dynamics in the following way: ln( )= ln( 1 )+ 2 ( )+ + + (26) where ³ the term 2 ( ) captures the balanced growth path trajectory of ln( ) equal to 1 ln 1 e and the term 1 ln( 1 ) captures the sluggishness of yearly capital adjustment. The parameter (1 1 ) is commonly called the speed of adjustment since it is the 11 The short run effect of exogenous emigration could be accounted similarly. The difference is that in that case we should consider actual, rather than exogenous, employment changes. However in our case due to the small values of net emigrants (relative to the total lab or force) we neglect the endogenous employment response on in the short-run and we consider only the effect of exogenous emigration on it. 12

13 share of the deviation from the balanced growth path (trend) eliminated each year. Finally, + represents the yearly net exogenous immigration flows as share of the labor force and are other zero-mean uncorrelated shocks. Once we know 1, and the sequence of yearly net immigration flows, + we can use (26) to obtain an impulse response of ln( ) as of 2000 in deviation from its trend (short run). In order to obtain the short-run effect on native employment and wages of immigration we need to evaluate the yearly short-run deviation of ln( ) from its trend due to the net inflows of immigrants + and correct the marginal productivity effects in by (1 )b short run When analyzing the short-run effects of immigration it is reasonable to consider the labor supply as rigid, to be consistent with the optimizing model (which has a rigid labor supply) and to isolate only this channel. 3 Description of the New Data-Set 3.1 Net migration data: sources and definitions The relevant migration flows to be used in our analysis are net immigration and emigration flows, namely gross flows of immigrants and emigrants net of returnees. The figures for immigrants are calculated using the stock of foreign-born (or foreign nationals) individuals in each countries in two different years and taking the difference between the two. The figures for emigrants are calculated for each country of origin by aggregating the stock of people living in other countries in two different years and taking the difference between the two. There are several sources documenting yearly migration flows by receiving country (e.g. OECD International Migration Database, UN migration statistics) but those only include gross inflow of people in a country and they do not correct for migrants who leave or go back to their country of origin. Moreover they never record undocumented migrants and they often record immigrants when they achieve their resident status rather than when they first enter the country. Most importantly for our purposes, those data do not have information on the education level of migrants. The flows of immigrants to a country can only be recovered by measuring the stock of foreign born people in a destination country (from a certain origin country) at different points in time and then taking the difference. The other advantage of using data on stocks of migrants is that they are from national censuses which tend to be more representative, more accurate and more complete than other data sources. Censuses (i) often account for undocumented immigrants at least in some countries like the US, (ii) they categorize immigrants by place of birth, rather than nationality which can change over time and across countries due to naturalization laws and (iii) report their education levels. Our database was constructed by two of the authors and is described in greater detail in Docquier et al. (2010) 12. It consists of measures of bilateral immigrant and emigrant stocks for 195 countries in 1990 and The starting point for the new data is the database assembled by Docquier and Marfouk (2006) which collected the stock of foreign-born in all OECD destination countries in 1990 and 2000, by country of origin and level of schooling 12 Further detsails of the construction and specific references are also in the Appendix A. 13

14 (primary, secondary and tertiary), using Censuses as primary data sources. These data were supplemented with original data from the censuses of a large number of non-oecd countries. Finally, for many destination countries with no data, bilateral migrant stocks were predicted using a gravity framework as described in greater detail in Docquier et al. (2010). Table A1 in the appendix shows the estimated total stock of migrants in OECD countries, in non- OECD countries with observed data and in non-oecd countries with imputed data. As we can see about 70-77% of world migrants are in countries with census data and only 23-30% of them are in countries with imputed data. Measuring emigration from OECD countries requires data from all the possible destination countries, at least the most relevant ones. Emigrant stocks from a certain country of origin can only be measured by aggregating all migrants recorded in the censuses of all destination countries. As some important destination countries (such as Russia, South Africa, Brazil, Argentina, and Singapore) are outside the OECD, this new database, ensures much better coverage of emigrants from OECD countries relative to Docquier and Marfouk (2006). Table A2 in the appendix show that the majority of emigrants from the countries considered in this study are in destination countries for which we have actual census data. 13 Less than 13% of emigrants from most countries of origin are in countries with imputed (rather than actual) migration data. The only countries relying on imputed data for a larger fraction of their emigrants (up to 40%) are Israel, the Baltic States and France 14. We distinguish two schooling types indexed by. = denotes college graduates (also referred to as highly educated) and = denotes individuals with secondary completed or lower education (referred to as less educated). The database covers the years 1990 and 2000 and the differences in stocks by country of origin and destination provides the measures of the bilateral net flows in the 1990 s. The data are relative to individuals aged 25 and over as a proxy of the working-age population. This choice maximizes comparability between data on migration and on labor force per educational attainment. Furthermore, it excludes a large number of students who emigrate temporarily to complete their education or children who migrate with their families and are not active in the labor market Labor force data per education level It is relatively easier to identify the number and average education level of adult individuals resident of each country of the world. Several data sources can be used to assess the size and skill structure of the labor force of each country. The size of the adult population (i.e. population aged 25 and over) is provided by the United Nations. Data is missing for a few countries but can be estimated using the CIA world factbook. 16 Adult population data is then split across skill groups using international indicators of educational attainment. We follow Docquier and Marfouk (2006) and Docquier, Lowell 13 This pattern is also confirmed in Ozden et.al. (2010) which presents global bilateral migration stocks but does not disaggregate by education levels. 14 Also for this reason we will consider Israel and the Baltic States as outliers in most of our simulations. 15 The dataset contains 195 source countries: 190 UN member states (after excluding North Korea), the Holy See, Taiwan, Hong Kong, Macao, and the Palestinian Territories. We consider the same set of countries in 1990 and 2000, although some of them had no legal existence in See 14

15 and Marfouk (2009) in combining different data sets documenting the proportion of postsecondary educated workers in the population aged 25 and over. Those studies use De La Fuente and Domenech (2006) for OECD countries and Barro and Lee (2010) for non-oecd countries. For countries where Barro and Lee s measures are missing, they estimate the proportion of educated using Cohen and Soto s measures (see Cohen and Soto, 2007). In the remaining countries where both Barro Lee and Cohen Soto data are missing (about 70 countries in 2000), they apply the educational proportion of the neighboring country having the closest enrollment rate in secondary/tertiary education, or the closest GDP per capita. 3.3 Description and General Trends Table 1 shows the immigration rates during the period for all the countries considered in this studies, namely all OECD countries plus all the remaining countries of Europe. The first column of Table 1 shows total immigration rates calculated as net inflow of immigrants (age 25 and older) over the period ( + in the notation of the model in section 2) divided by the initial population in working age as of 1990, 1990 For instance the figure of 14.35% relative to Israel means that the net inflow of foreign-born during the decade was equal to 14.35% of its population in This is a huge number and it is the consequence of the very well known migration of Russian Jews between 1990 and 1994 when migration restriction were lifted in a very unstable Soviet Union 17. Less well known is that also Luxembourg, Austria and Ireland received during this period very large inflow of immigrants relative to their population with total rates between 7.6 and 12.5%. Three countries at the bottom of the table are also worth mentioning. The three Baltic Republic (Estonia, Latvia and Lithuania), born after the collapse of the Soviet Union, experienced a massive negative net immigration. This means that the stock of foreign-born individuals decreased massively during this period. This is due to the return migration of many people of Russian ethnicity back to Russia, after their immigration into Latvia and Estonia (and to a less extent into Lithuania) during the Soviet period. Hence the negative immigration rates are really emigration of foreign nationals. Several other eastern European countries (e.g. Romania, Slovenia, Hungary and Poland) also experienced during this period net emigration of foreign nationals. The second column of Table 1 shows the net immigration rages for College educated, refereed to as "highly educated" in this study. They are calculated as the net change (between 1990 and 2000) in the stock of college educated foreign-born ( )relativetothe resident population aged 25 and older with a college degree in We notice two interesting things. First, in all countries with positive net immigration rates, except for one (Austria), the immigration rate of college educated was larger than the rate for the total population. In some cases (such as Israel, Ireland, Iceland, Canada, Malta, Australia and the United Kingdom) the immigration rate for college educated was more than double the overall immigration rate. Immigration, therefore, contributed to increase the share of college educated in the resident population in all but one of the countries considered. Second Latvia and Estonia, had negative immigration rates, implying large returns of immigrants 17 There are several studies analyzing the economic impact of this episode on the Israel economy. Friedberg (2001), Cohen-Goldberg and Pasermann (forthcoming) and are among those. 15

16 and even larger return rates for college educated. Third the immigration rate for college educated was remarkably high in many countries. Even omitting the whopping 60% experienced by Israel, ten more countries had immigration rates for highly educated larger than 10%. Among the countries with immigration rates most skewed in favor of highly educated were Ireland, Malta, United Kingdom and Switzerland. None of those is among the countries with immigration policies explicitly favouring highly educated immigrants. The third and fourth column of Table 1 show the total and college-educated immigration rates considering only foreign-born from non-oecd countries. In calculating those figures, while we left as denominator the stock of population or college educated resident in the country we only included in the numerator (net immigrants) those born outside the OECD countries. For some European countries (e.g. Luxembourg, Austria, Ireland and Iceland) the immigration-rates from non-oecd countries are less than half the total rates indicating that large part of the immigration is probably within-european mobility. However, in many cases, including several important receiving countries, the immigration rates from non-oecd countries were significantly more than half of the total. This was the case, for instance, in the US, Canada, Australia, New Zealand and Sweden. Moreover even focussing on non-oecd immigrants the immigration rates for college educated were larger than the average immigration rates for all but three countries (Luxembourg, Austria and Italy). Even non-oecd immigrants contributed to increase the college share of most OECD receiving countries. Table 2 shows the emigration rates for the countries in our sample. Column 1 displays the total emigration rate, calculated as the net outflow of natives (25 year and older) during the period ( + ) relative to the total resident population (25+) in Column 2 contains the net emigration rate of native college-educated. A negative emigration rate of natives implies that during the period the flow of returnees (native resident abroad) was larger than the flow of emigrants. Countries are ranked in the table in decreasing order of their college emigration rates. Few comments are in order. First, as in the case of immigration, emigration rates are also larger for college educated than on average for all but one country (Israel). For some small countries (Cyprus, Malta and Ireland) a very large emigration rate of college educated, which we may call "brain drain" is associated with negative overall emigration rates, implying large rate of return for non-college educated natives from abroad. Some of those small countries had, however, large immigration rates among college educated that compensated in part for the brain drain. Second there are some countries with large immigration and emigration rates. Typical is New Zealand with large immigration from Asian and Pacific countries and large emigration to Australia. Third several Eastern European countries (such as Poland, Romania, Slovenia, Slovakia) and some western European countries (Portugal, Greece) had significant emigration rates, but mainly very large rate of college emigration not compensated by similar immigration flows. For those countries emigration was a significant source of change in the supply of highly educated labor. Other European countries such as United Kingdom, Luxembourg, Switzerland and the Netherlands had significant rates of college-educated emigration but they made up with significant immigration rates in that group. Finally the United States, Canada and Australia were, as expected, mainly countries of immigration as the immigration rates (total and for highly educated) were much larger than the corresponding emigration rates. In summary, considering OECD countries during the 1990 s immigration and emigration were very college- 16

17 intensive, that many countries experienced emigration rates as large as immigration rates and that even immigrant from non-oecd countries were over-represented among college educated relative to the total. 4 Simulated Labor-Market effects 4.1 Parameterization and Measurement Our model allows us to calculate the percentage wage and employment effects of migration on natives. As one can see from the formulas in 17-19, in order to evaluate these effects we need to know three sets of variables, for each country. The firstistheshareofwage income to each of four groups (native and foreign-born, more and less educated) as of 1990 denoted as ( ) 1990 ( = and = ). The second is the percentage change in employment in each of those four groups due to immigrants and emigrants in the decade , d ( = and = ). The last is the change in college intensity due to immigration and emigration Then we also need to know the value of four fundamental parameters:, the elasticity of substitution between more and less educated; the elasticity of substitution between native and immigrants with same schooling, the intensity of college-externalities and the labor supply elasticity of more and less educated natives. I describe briefly in this section how we measure these country-specific variables and how we choose the range of the parameters. The share of wage income to more and less educated and native-immigrants depend on their employment and their wages. We use their number in the population in working age as measure of employment (from the Docquier et al. 2010, Database). As there is no international database on wages of college educated and less educated we proceed as follows. From the Hendricks (2004) database we take the estimated returns to one year of schooling in each of the considered countries estimated in a year as close as possible to From the Barro and Lee (2010) database we calculate the average years of schooling in each of the two schooling groups (college graduates and those with no college degree) for each country. We multiply the yearly return by the year of schooling difference between the two groups to identify the college wage premium in the country. Then from several estimates, most of which reviewed in Kerr and Kerr (2009) we use the country-specific estimate of the nativeforeign wage premium to correct the wages of immigrants at each level of wage. Even in this case we use the estimate for the closest country with most similar income per person if the data is not available for a specific country. Multiplying the group-specific employment by the group-specific wage (standardized for the wage of less educated natives) we obtain the wage-bill for the group, which divided by the total wage bill provides the share. The shares of wage income for each of the four groups in each country, calculated using this method, are reported in Table A3 of the appendix. The percentage change in the employment of each group during the period due to immigration and emigration, as well as the change in share of college-educated, are calculated from the dataset on stocks of migrants in If the estimate was not available for a country we choose the estimate for country sharing a border with theclosestlevelofincomepercapite. 17

18 Table 3 summarizes the values of the parameters chosen in each of three scenarios considered in the numerical simulations. The estimates span the range found in the literature. For the parameter the elasticity of substitution between more and less educated, there are several estimates in the literature. A group of influential papers propose specific estimated values for low, intermediate and high levels of substitution. For instance Johnson (1970) and Murphy et al. (1998) estimate values for around 1.30 (respectively 1.34 and 1.36); Angrist (1995); Katz and Murphy (1992), Ciccone and Peri (2005) and Krusell et al. (2000) estimate values around (respectively 1.47, 1.50, 1.66) and Ottaviano and Peri (forthcoming) estimate a value close to 2. Hence we use the values 1.5, 1.75 and 2 for the three scenarios. The parameter capturing the elasticity of substitution between natives and immigrants has been the subject of several recent papers and has generated a certain level of debate. This parameter is particularly relevant to determine the effect of immigrants on wages of natives and, as we will see, the choice of this parameter affects the significantly the estimated wage-effects of migration in some countries. Borjas et al. (2008), Peri (2010) and Ottaviano and Peri (forthcoming) use US data and Manacorda et al. (forthcoming) use UK data in their estimation of.thefirst study finds a value of infinity, the second and third estimate an elasticity between 10 and 20 and the paper on UK data findsavalueof6. Weuseinfinity, 20 and 6 as the three parameters spanning the range. The parameter, capturing the externality of college educated, whose magnitude has been estimated using data from US cities (Moretti 2004a, 2004b) or US states (Acemoglu and Angrist 2000 and Iranzo and Peri 2009) is also subject to a certain level of disagreement. Some studies find substantial schooling externalities ( =0.75 in Moretti 2004b) while others do not ( =0in Acemoglu and Angrist 2000). Finally, the estimates of the elasticity of labor supply as summarized in the review by Evers et al (2008), which covers several European countries and the US, range from 0 (with actually a few small negative estimates) to 0.17 (in a study relative to the US of Flood and MaCurdy 1992 and in one relative to the Netherlands by van Soest et al 1990). Hence we choose 0, 0.10 and 0.20 as representative values in this range. 4.2 Simulation: Wage Effects of Immigration The effects of immigration on native wages, in percentage points are shown in Figures 1 and 2. Each figure reports the values obtained using the formulas 19 and the three configurations of the parameters shown in Table 3 (pessimistic, intermediate and optimistic), for each of the considered countries. The thick solid line connects the estimates in the optimistic scenario, the thin solid line connects estimates in the intermediate scenario and the dashed line connects the pessimistic estimates. The numeric values corresponding to the figures are reported in Table A4 of the Appendix. We show the long-run percentage effect of immigration onthewageoflesseducatednativeworkers infigure1andthepercentageaveragewage effects on natives in Figure 2. We focus our attention on less educated workers, first, as they are those usually considered as suffering the most disruptive labor market effects from immigration (e.g. Borjas 2003). The average wage effect, on the other hand, provides us with an idea of the overall effect of immigrants in the long-run on income accruing to national factors. While in Table A4 we report the effect on natives for all the 39 countries considered (all OECD plus the remaining European Countries) in the Figures we omit Israel and the 18

19 Baltic Republics (Lithuania, Latvia and Estonia). The first experienced huge immigration rate in the 1990 s, unmatched by other countries. It is a clear outlier. Reporting its figures would overshadow the details for all other countries. As for Lithuania, Latvia and Estonia, they experienced large negative net immigration rates (of Russians returning after the forced migration of the Soviet period). This was really an emigration phenomenon with features and consequences more akin to those of emigration. The Figures show the countries arrayed from left to right in decreasing order of their total immigration rate (as defined in Table 1). A number of interesting features should be emphasized. First, all the simulated values, capturing the impact on the wage of less educated (excluding only the most pessimistic estimate for Austria) are positive. Less educated native workers gained from immigration. For some countries with large immigration rates, such as Luxembourg, Ireland, Iceland, Canada, Australia the effects are positive and non negligible ranging from 1% up to 6% gains. Also, for some countries with intermediate immigration rates, such as Belgium, United Kingdom and Switzerland the positive wage effects on less educated are non-negligible, between 1 and 2%. For most of the other countries the estimates range between 0 and positive 1%. Second, the distance between the pessimistic and the optimistic scenario illustrate how relevant are native-immigrant imperfect substitutability and college externalities in determining the wage effect on less educated. The most pessimistic effects, on the other hand, are purely driven by complementarity between more and less educated and are small but nearly all positive. Third, notice that the US, among countries with high immigration rates, is one whose less educated workers experience the lowest wage effects (only Austria experienced smaller effects). Still the estimated longrun effect of US less educated is positive and it is between 0.33 and 1.2%. The magnitude of the wage effect on less educated natives depends on two main features of the immigration rate: how large and how strongly biased in favor college educated immigration was. Notice that some countries whose immigration laws explicitly favoured the immigration of more educated (such as Australia and Canada) did experience very large benefits on the wages of less educated native workers. However, several countries with no such explicit laws (such as Luxembourg, Ireland, Malta, UK and Switzerland) also received significant positive effect to their less educated. This is because they still attracted many highly educated, possibly for the reward that they provide to skills, the dynamic environment to college educated or other virtues and those created productive opportunities for less skilled and for all workers. The average wage effects shown in Figure 2 mirror, qualitatively, the effects on wages of less educated reported in Figure 1. They are smaller quantitatively and they differ more, proportionally, between scenarios. In the most pessimistic case (with no externalities, perfect substitution between native and immigrants, low complementarity between more and less educated and rigid labor supply) they are very close to 0 in most countries. Countries to the right of Turkey, which include mostly Eastern European and Asian economies, received quite low immigration flows (less than 1% of their labor force during the whole decade). As expected, therefore, they also experienced small wage effects from those. Still, except for Poland (which experienced negative net immigration), in all countries the effects were positive both for less educated and on average. Countries between Turkey and Spain, experienced intermediate immigration rates (between 1 and 3.5%) and they all had positive wage effects, ranging from 0 to 2% for less educated and between 0 and 1.5% for the average. Finally countries to the left of Spain in Figure 1 and 2, representing those with 19

20 very large immigration rates, experienced quite variable, but always positive, wage effects. Immigration produced up to 6% gains, for less educated and up to 4% for the average wages. To give an idea of the aggregate magnitude of gains to the native labor force, the intermediate estimate for the US average wage (+0.37%) would imply that in year 2000, as a consequence of immigration over the previous decade, the total wage bill of natives was 3.1 Billion larger than it would have been otherwise. 4.3 Simulation: Employment Effect of Immigration Our model allows also to evaluate long-run effects of immigrants on native employment, by using expressions (18) and (17).This is shown in Figures 3 and 4. Usually these effects are rather small as the elasticity of labor supply is rather low in the aggregate (at most we useavalueof0.2)andtheeffects have the same sign of the wage effects, as they are also driven by increases (or decreases) in the marginal productivity of native workers due to the inflow of immigrants. Allowing employment of natives to respond to immigration has also an interesting multiplying effect, driven by complementarity. In fact as immigration increases the relative supply of college educated, less educated natives benefit from its complementarity and their employment increases. This increase, in turn, benefits the productivity of college educated who will increase employment and so on. This mechanism, that can be strong for labor supply elasticity in the order of one, is quite limited, however, in the range of our simulations. Figure 3 shows the employment effects for the less educated natives and Figure 4 for aggregate native employment. The Figures replicate the structure and the notation of Figures 1 and 2. In what we call the "optimistic" scenario we consider the labor supply as inelastic, so that we have the largest impact on wages. Hence by assumption the employment effect of immigration is 0. In the intermediate scenario we use an intermediate value of the labor elasticity (0.1) and in the pessimistic scenario we consider the largest labor supply elasticity (0.2). In this last case the transmission from marginal productivity to employment is maximized. However the configuration of the other parameters is such that the marginal effect of immigration in the most pessimistic scenario is the closest to zero. While the qualitative features of employment effects (sign and relative magnitude across countries) are similar to the wage effects their magnitude is much smaller. In no case the effect on less educated is larger than +0.8% and the aggregate effects are mostly between 0 and 0.2%. Ireland, Canada and Australia experience the largest positive effect for less educated workers. Austria is the only country to have a negative effect (although equal to a very small -0.05%) on employment of less educated workers in the pessimistic scenario. 4.4 Simulation: Wage effects of Emigration As we emphasized in Section 3 our data show that net emigration rates for some OECD countries were at least as large as net immigration rates during the considered period, in particular for college-educated adults. We analyze here the wage impact of these emigration flows considering them as a decrease in the supply of native workers. Figure 5 shows the effects of emigration on wages of less educated native adults in the country of origin. Figure 6 shows the effect on average native wages of those remaining. The actual figures for each 20