Skills or Culture? An Analysis of the Decision to Work by Immigrant Women in Italy

Activity and employment rates for immigrant women in many industrialized countries display a great variability across national groups. The aim of this paper is to assess whether this well-known fact is due to a voluntary decision (i.e. large reservation wages by the immigrants) or to an involuntary process in that the labor market evaluation of their skills is low. This is done by estimating the reservation wages for each individual in the dataset. Our results show that low activity and employment rates for certain national groups are not associated with high reservation wages. This implies that low participation should not be interpreted as a voluntary decision.


Introduction 1
The immigrant labour force makes an essential contribution to the growth and development of most of the industrialized countries. At the beginning of the 2000s, 85 million people residing in the OECD countries had been born abroad, almost three times more than in the 1960s. The share of immigrants in total population is quite high for economically advanced countries and it ranges from 6 per cent in Italy to almost 25 per cent in Canada.
In many European countries, characterized by a rapidly aging population, the economic integration of immigrants is a necessary condition for the achievement of the Lisbon targets of full employment and sustainable growth under the European Employment Strategy. However, although labour market integration for immigrant men is not generally an issue, employment rates for women are often low and characterized by great variability according to country of origin.
There are two possible explanations for this fact. The first is based on "cultural bias".
For certain nationalities, gender differences in the division of the family burden may play an important role in the labour supply. In other words, traditions may relegate women to the more traditional roles and tasks of housekeeping and child-raising. The second is a skill mismatch on the labour market. In this case, immigrants are willing to work outside the family but their human capital does not satisfy local employers' requirements.
This issue is particularly important for policy making. Whether the cultural bias explanation is true or not, migration policy should focus on the borders: visas should be issued mainly to the nationalities that are more willing to participate in the labour market. In the case of skill mismatch, the policy options are more complex and may range from selective migration for immigrants whose characteristics are more compatible with the host 1 We wish to thank Prof. G. Blangiardo, ISMU, for kindly providing us with the dataset. We are also indebted to Federico Cingano, Sauro Mocetti, Laura Pagani, Alfonso Rosolia, Alberto Zazzaro, two anonymous referees, the seminar participants at the Bank of Italy, University of Bari, Bocconi University and AIEL (Sassari, 2009) and Jennifer Parkinson for editorial assistance. The views expressed are the authors' own and do not necessarily reflect those of the Bank of Italy. Usual disclaimers apply. E-mail: antonio.accetturo@bancaditalia.it, luigi.infante@bancaditalia.it -6 -country's needs to labour market policies aimed at educating and training workers to help them become more suited to local labour market requirements.
The aim of this paper is to disentangle these two effects by assessing the impact of cultural background on women's decision to participate in the labour market. In particular, we investigate whether the observed high variability in the employment and activity rates across nationalities is attributable to a cultural or a skill effect by estimating individual reservation wages.
We make use of an extremely rich database compiled by the non-governmental organization Iniziative e Studi sulla Multietnicità -ISMU (Initiatives and Studies on Multiethnic Society) from surveys of immigrants from the least developed, emerging and transition countries now resident in the region of Lombardy in the North-West of Italy. The fact that the interviews are collected in the same (relatively small) area of Italy is particularly useful for our analysis. The women surveyed face the same labour market conditions; moreover, given Italy's recent tradition in immigration, they are most likely to preserve their cultural attitudes with respect to labour market participation.
There is a considerable body of international evidence on immigrant female participation. Many studies (Dustmann and Schmidt, 2000;Constant et al., 2006;Bevelander and Groeneveld, 2007;Dustmann and Fabbri, 2005;Adsera and Chiswick, 2007) find that the probability to work for an immigrant woman is significantly lower even after controlling for observable characteristics. This is generally attributed to an unbalanced family burden and a cultural bias. Moreover, as Dustmann and Schmidt (2000) point out, the same motives also prevent women from accumulating human capital. By using the British Time Use Survey, Zaiceva and Zimmermann (2007) obtained a similar result for the UK and showed that non-white females usually spent more time in "traditional" duties such as religious activities and food management than white females.
The poor labour market outcomes by foreign women is usually attributed to culture or religion. For the US, Fernandez and Fogli (2009) find that intergenerational transmission of cultural attitudes helps to explain the low contribution to the labour forces by second generation immigrant women. Vella (1994) obtains a similar result for educational choices.
-7 - Heineck (2004) confirms the influence of religion, finding that the frequent attendance of places of worship reduces female participation in the labour market in Germany.
Compared with the previous literature, this paper disentangles the demand (skill) and the supply (cultural) components in the decision to participate in the labour market. We assess whether low participation rates for certain national groups are due to the fact that they attach a high economic value to the time spent at home or because there is a weak demand for their skills. This is done by estimating their reservation wages. Our results show that culture has a low impact on the decision to work by immigrant women in Lombardy. We show that low activity and employment rates for certain national groups (mainly those from North African, the Middle East and central Asian) are involuntary since their estimated reservation wages are no greater than those of nationalities characterized by higher employment levels (for example, Central and Eastern Europeans). 2 The paper is organized as follows: Section 2 describes the ISMU dataset and shows some of the descriptive statistics; Section 3 explains the most relevant econometric issues; Section 4 presents the results; and Section 5 concludes the paper. Appendix 1 shows how countries of origin are aggregated in homogenous groups while Appendix 2 presents a comparison between our estimated reservation wages and Labour Force Survey (LFS) data.

Data
Our main data source is a dataset collected by the ISMU in Lombardy (with the financial support of the Lombardy regional government and other private institutions) in the period [2001][2002][2003][2004][2005]. Since 2001, the ISMU has conducted a yearly survey of immigrants living in Lombardy. As for migration studies, the dataset has a strong comparative advantage. Its most important characteristic is that it is also able to collect information for undocumented aliens due to its data-collection process based on the method of aggregation centres developed by Blangiardo (1993). 2 The result has an antecedent in Europe in Niesing et al. (1994). By analysing the high unemployment rates of ethnic minorities in the Netherlands, they conclude many national groups suffer from a "discrimination in employment possibilities" due to employer choices.
-8 -Survey design -Surveys on aggregation centres are specifically designed to collect information on a representative sample of immigrants that include also irregular migrants.
The idea is that even undocumented individuals generally lead a social life by attending, for example, places of religious worship or cultural centres. Blangiardo's method hinges on those centres and is based on a three-stage design. In the first, the ISMU interviewers allocate the total number of questionnaires (roughly 8,000 each year) across the 11 provinces into which Lombardy is partitioned; this is aimed at obtaining significant estimates at provincial level by having roughly the same sample variability within each province. In the second stage, the ISMU selects a number of representative municipalities (slightly less than 350, almost 25 per cent of all the towns in Lombardy) within each province according to the social and economic characteristics of the area. In the third stage, the ISMU interviewers visit all possible aggregation centres for immigrants within each municipality and randomly meet the potential interviewees. The aggregation centres usually fall within 11 categories: help and counselling centres for immigrants, Italian language courses for foreigners, places of worship, healthcare centres, cultural centres, phone/money transfer centres, public offices (police stations, town halls, etc.), ethnic restaurants/bars, ethnic shopping centres, other (train stations, etc.). 3 Within each aggregation centre, the ISMU interviews usually have the support of the leading personalities of the centres (priests, shop owners etc.); this ensures that the interviewers are seen as "trusted" people and not public officials. As for the ability of the survey to truly detect irregular immigrants, the misreporting of legal status seems tiny.
In order to provide a comparison between the ISMU estimates and actual data, we compare  -9 -12 thousand individuals. The ISMU datasets are quite rich and contain some information on social and economic characteristics. Table 1 presents the summary statistics for the variables used in the empirical analysis. Average schooling is quite high (10.7 years) 4 and comparable with the figures for the Italian population. Almost 60 per cent of the interviewed women is married; they have been residing in Italy for a relatively short period of time (5.7 years). Irregular women make up 11.7 per cent of the total. As for religion, more than two thirds of the dataset is Catholic or Muslim. Other Christians (mainly Greek-Orthodox) are slightly more than one fifth of the sample. Table 1 Table 2, 5 which shows, as is usual in many industrialized countries, that there is considerable heterogeneity in labour market outcomes according to the country of origin. The employment rate averages 60 per cent for the entire sample, while the activity rate is much larger (73.4 per cent). The employment rate of the women coming from the Central and Eastern European Countries (CEEC) and Central and South America is roughly two-thirds higher than that for the Near East and North African countries (NENAC). 6 The figure for the Central Asia group is even smaller. Activity rates display the same great variability across national groups. 7 4 As usual in this literature, we assigned zero years of schooling when the individual does not have any formal education, 8 years for a compulsory school leaving certificate, 13 years for high school and 17 for at least a university degree. 5 To be clear, we split the the immigrants in three groups as regards employment status. The first comprises the inactive population, i.e. those who stated they were housewives or students. The second group is made up of the self-reported unemployed. The third is composed of all employed persons. The "active" group includes both unemployed and employed individuals. 6 A similar finding is reported by Amuedo and de la Rica (2006) for Spain, according to which African women suffer the lowest occupational attainment, with respect to other immigration groups (in particular if taken into account the years since migration), while Central and South Americans, along with immigrants within EU15, show high level of assimilation. 7 The characteristics of migration in Lombardy are quite similar to those of the rest of the country. According to the Italian LFS (2006-2008 averages), the average schooling level of immigrant women is 10.2 years. The activity rate is slightly lower (65 per cent) probably due to the different definitions of working status between LFS and ISMU. Cross-country differences in the working status for LFS confirm the ISMU data: activity rate for NENAC is 20 percentage points lower than the one of CEEC. The only relevant difference -10 -

How do economists obtain reservation wages?
The aim of this paper is to report an econometric estimate of the immigrants' reservation wages. In job search theory, the reservation wage is the lowest offered wage that an unemployed individual looking for work is prepared to accept (see Blundell and MaCurdy, 1999). 8 Although this is a crucial variable in the neoclassical theory of labour supply, there is still an open discussion on the best way to estimate it. Two prevailing methods are usually available in literature.
The first is based upon surveys in which unemployed respondents are directly asked what their reservation wage is. 9 This information is widely used in many studies: see, among others, Addison et al. (2004 and and, for the Italian case, Sestito and Viviano (2011).
The reliability of this information, however, is widely debated. As shown by Burdett and Vishwanath (1988) and Hofler and Murphy (1994), self-reported reservation wages are often biased and they are usually inconsistent with the actual behaviour of a worker. As Addison et al. (2010) point out, this is mainly due to the fact that respondents usually answer by indicating the prevailing wage on the labour market, rather than their true reservation value.
The second method treats reservation wages as unobservables, that must be inferred econometrically by the actual behaviour of a worker. This was pioneered by Heckman (1974) in his contribution on women's shadow prices in the labour market and was subsequently developed in further studies (see, Kiefer and Neuman, 1979;Fishe, 1982;Ferber and Green, 1985;Duncan, 1992;Sharpe and Abdel-Ghany, 1997). The idea is that observed wages for employed individuals are those that succeed in exceeding the individual between ISMU and LFS is in the country of origin composition: in LFS the share of women coming from CEEC is 54 percent, more than 20 percentage points greater than the ISMU figure. 8 In a static framework, with quasi-convex utility functions, the reservation wage is equal to the marginal rate of substitution between leisure and consumption. 9 For example, the Italian Labour Force Survey asks unemployed workers "What is the lowest net monthly wage you would be willing to accept?".
-11 -reservation values. This implies that, by controlling for the selectivity bias, actual market wages contain enough information to infer workers' reservation wages. 10 In this paper we adopt the second methodology for two main reasons. The first relies upon the above-mentioned reliability problems of self-reported reservation wagesproblems that can be greatly aggravated for migrants, whose understanding of the crucial features of the local labour market is more limited. The second relates to whether there is sufficient information in the LFS. In Appendix 2 we provide the results for self-reported reservation wages in the Italian LFS. By concentrating on immigrants living in the North (where Lombardy is located and where most of the immigrants in Italy live), we end up with just under 1,000 individuals, which is one-tenth of our baseline estimates database.
The methodology we use in this paper is based on the Mohanty's (2005) extension of the Heckman model with frictional unemployment and feedbacks between labour demand and supply. As will soon be clearer, from a technical point of view, the only difference between this methodology and Heckman's lies in the first stage, which is bivariate-probit rather than probit estimated. This allows us to take into account a double selectivity bias due to involuntary unemployment or feedback effects between demand and supply. 11

3.2
The estimate of reservation wages: an econometric approach A woman i decides to participate in the labour market (i.e. to be active) whenever the wage offers she expects to receive are greater than her own reservation wage. In formulas, this implies that she is active whenever w o i -w r i = y 1i 0, where w o i is the expected wage offer, w r i is her reservation wage and y 1i represents the (normalized) individual preference for labour market participation. Whenever y 1i is greater than zero, individual i participates in the labour market, whenever it is negative she prefers to stay at home. It immediately follows that reservation wages can be obtained as w r i = w o i -y 1i and can be computed by estimating preferences (y 1i ) and wage offers (w o i ). 10 An important advance on this topic is the estimation of structural search models by using the panel structure of many LFSs. As will be clearer later, this method cannot be used in this paper because the ISMU dataset is compiled on the basis of pooled cross-sections. 11 Mohanty (2005), in turn, generalizes the procedure to cases with multiple selection bias following the approach in Meng and Schmidt (1985). Recently Baffoe-Bonnie (2009) uses a similar framework to study wage differentials.
-12 -Estimates for y 1i and w o i are obtained in two steps.
In the first step we consider both the participation decision and the hiring process in the labour market. Woman i is employed only if she decides to participate in the labour market (Participate i =1) and is hired by an employer (Employed i = 1). Formally, where y 2i represents the (normalized) employers' preferences over individual i.
The aim of the first step is to compute the latent variables y 1 and y 2 by estimating the following two equations: (1) y 2 = x 2 b 2 + e 2 (2) by a bivariate probit with partial observability. The choice of the bivariate probit is particularly useful since it allows us to treat demand and supply components simultaneously.
As mentioned above, this allows us to take into account the existence of feedbacks between the decision to participate and the expected labour market outcome. Operationally, x 1 contains a set of variables aimed at capturing the economic and cultural determinants for the labour supply, while x 2 includes all the possible personal characteristics which are likely to influence the employer's willingness to hire an individual.
As for the supply components, x 1 includes schooling, potential experience, religion dummies and their interaction with marital status, number of children below and above 18 years of age and a set of time dummies. The number of under age children may indicate a greater interest for childcare and housekeeping while the number of children over 18 should have a positive effect on the labour supply since offspring could need financial support from the family of origin. Religion dummies indicate a cultural attitude toward labour, especially -13 -when a woman is married. As will be clear later, religion dummies play a fundamental role as identification variables in the empirical strategy.
On the demand side, x 2 contains schooling; years since migration; dummies for country of origin; time; and space. 12 Years since migration are expected to enhance the probability for a worker to be employed since during these years the worker is likely to increase his ability to understand the crucial features of the host country's labour market and local language; country dummies capture the workers' heterogeneity in terms of the quality of the institutions in their area of origin (for example, educational system, sectoral specialization) while spatial dummies capture time invariant local characteristics that are likely to influence employment levels.
x 1 and x 2 share the schooling variable and time dummies, since education is likely to have an effect on both the demand and supply components and year dummies take into consideration business cycle fluctuations.
The latent variable of interest (ŷ 1 ) is calculated by taking the predicted values (linear prediction) of equation (1).
In the second step, we compute expected wage offers. We estimate the following wage equation using a correction for a double selectivity bias (Tunali, 1986): .  and  represent, respectively, the density and the cumulative function of a standard normal distribution, while  (rho) is the correlation of the 12 Spatial dummies include one dummy for each of local labour markets in the partitions of Lombardy.
-14 -error terms in the bivariate probit. x 3 includes standard variables in migration-augmented mincerian equations: schooling, potential experience and years since migration. The regression includes spatial dummies ( s D ) to take into account spatial differences in wage levels; time dummies ( t D ) to control for business cycle effects; and country of origin dummies ( c D ) to control for institutional factors such as sectoral specialization or educational quality.
Expected wage offers are computed as follows: Where l n m w is the predicted value of equation (3). By comparing all the variables in x 1 , x 2 and x 3 , the crucial role of religion dummies and their interaction with marital status as identification variables is now apparent. The idea is that religion is a private matter that affects the individual working decisions but should not affect the labour market evaluation (wages) and the hiring decision by a non-discriminating employer. In other words, the worker's productivity should be influenced by the country of origin's institutional setting (school quality, sectoral specialization etc.) but not by the migrant's private attitudes toward religion (which does, however, influence her decision to work). This implies that by inserting country dummies in equations (2) and (3)  in all the analyses we exclude all countries with a HI equal to one.

Cross-country differences and robustness
After computing the reservation wages, we test whether they systematically differ across groups of nationalities. We calculate the percentage differences between each group and our reference cluster (Central and Eastern European Countries, CEEC). The choice of CEEC as a benchmark 15 relies on the fact that, Central and Eastern European women share similar institutions with the host country. We focus, in particular, on the differences between the CEEC group and two nationalities that display the lowest activity and employment rates: NENAC and Central Asia. If the reservation wages for those groups were higher, the observed low labour market participation would be interpreted as voluntary: the value NENAC and Central Asian women attach to their time spent at home is so high that they are not attracted by the local labour market. Conversely, if their reservation wages were comparable or lower to that of our reference group, their inactive status would be interpreted as involuntary: their reservation values are not particularly high but they remain unemployed because the arrival rate of job offers is quite low.
We further check the robustness of these estimates along four lines. 14 HI for country c is calculated as follows: where R is the set of religions, rc P is the number of immigrants of religion r coming from country c and c P is the number of people coming from country c. When in country c there is only one religion, HI is equal to one. 15 As in most papers, our first choice would have been native women; but since the ISMU dataset only concentrates on immigrants, we chose the CEEC group as the closest to the native one.
-16 - The first check is based on Italian migration law according to which it is necessary to have a job in order to obtain a visa. As an exception to this rule, immigrants can enter Italy to join their family and obtain a visa based on family reunification. For those cases, migrants' true shadow values should be revealed since they do not need to work to obtain a residence permit. We check this issue by restricting the analysis to those women who migrated to Italy with a family reunification visa. This information is available for all years except for 2004: migrants entering the country for with a family reunification visa amount to 2,408.
The previous scheme obviously implies that the immigrants have a good knowledge of the Italian migration laws. We can generalize this scheme by focussing on migrants with a blood relative already residing in the country. Again, for those individuals their shadow values could be higher, since they can rely on the family's financial support while looking for work. We check this issue by restricting the analysis to women who entered the country when a next of kin was already residing here. This leaves us with 1,753 individuals.
The third check is based on the analysis of the irregular migrants. Undocumented aliens have very weak bargaining power with respect to their employers as they cannot join a union and must work off the books. This implies that wage offers are usually quite low (see Accetturo and Infante, 2010, on this topic) and, therefore, they may be quite close to the reservation wages. Moreover, illegal aliens' incentives to work are particularly strong since their only chance of being regularized under one of the recurrent amnesties is strictly linked to evidence that they are employed on Italian soil. In other words, they are more likely to accept the jobs they are offered. In the ISMU dataset, irregular women are surveyed each year and they amount to 1,303 individuals.
The fourth check is based on the relationship between religions and countries. As we said before, we already discard all the observations coming from one-religion countries.
However, in some countries a religion could prevail but not be the only one (for example, Islam in Arab countries or Roman Catholicism in Central and South America); this implies that from the employer's point of view religions and countries of origin are quite indistinguishable thus invalidating our identification structure. We cope with this problem by restricting our analysis to a group of countries that can be considered truly multi-religious.
-17 -This is done by discarding all observations coming from countries whose HI based on religions exceeds 0.75: this leaves us with 8,305 individuals.

Baseline sample
The first part of this section is devoted to the results of the baseline sample. This group is constituted by all the regular women of working age interviewed in the 2001-2005 waves. Before starting with the comparison across national groups it is worth assessing whether estimated reservation wages are consistent with the predictions of the standard theory of labour supply (Table 4). As Blundell and MaCurdy (1999) point out, individuals with a higher inter-temporal elasticity of substitution should have lower reservation wages: this is confirmed in our estimates since younger individuals attach a lower economic value to 16 This is due to the fact that the dependent variables in the biprobit are highly correlated since they only differ in the values they assume for unemployed individuals. This may generate a collinearity problem between  1 and  2 in the estimate of the wage equation. A possible way to circumvent this problem is to treat equations (1) and (2) in a sequential way by estimating an Heckprob model. The results, not shown here to save space but available upon request, are in line with those presented in the text.
-18 -time spent at home compared with older cohorts. Even education plays an important role in setting the individual's reservation value. Consistently with the standard theory, we find that more educated individuals usually set higher reservation wages. Finally we check whether individuals residing in a city 17 have a lower reservation value because of lower job-search costs thanks to better access to the transportation network. This is confirmed by the bottom panel, in which we show reservation wages for workers in a city by educational group.
We subsequently deal with the core estimates of the paper. In Table 5  Estimates show that while the share of active and working population is substantially low for the NENAC and Central Asia group, their reservation wages are statistically smaller than those computed for the CEEC group. This suggests that low participation should not be attributed to a cultural attitude that raises the economic value of the time spent at home, but, rather, to a skill mismatch between local labour market requirements and immigrants' abilities.
We further split the computed reservation wages across educational levels and years since migration. Table 6 shows the differentiated effects across four classes of education: no education, primary, secondary and tertiary schooling. Baseline results are confirmed as reservation wages for NENAC and Central Asian are always statistically lower than the estimate for the CEEC. In columns [1] to [4] of table 7 we report the percentage differences in reservation wages for subgroups belonging to (respectively) the first to the fourth quartile of the distribution of years since migration. Even in this case the results remain undisputed. 17 We define "city" the local labour market of the three most important towns of Lombardy: Milan, Brescia and Bergamo. 18 Reported figures are quite similar to those in Table 2 since the regular aliens constitute almost 90 per cent of the ISMU sample.

Robustness
So far we have obtained the interesting result that the low employment/activity rates registered for certain groups are not associated with higher reservation wages.
In this section we present four robustness checks by restricting the analysis to: (i) a women who benefited from a visa based on family reunification, (ii) immigrants with a blood relative at the time of migration to the host country, (iii) undocumented immigrants, and (iv) immigrants coming from truly multi-religious countries.
Results for the first check are shown in Table 8. Columns [1] and [2] report, respectively, the shares of active and working population among those who entered with a family reunification visa: consistently with the standard theory of labour supply, activity and employment rates are quite small now, thus reflecting their possibility to sustain larger spells of unemployment; 19 however, there are still differences across national groups. In column [3] we report the estimated reservation wage. Baseline results are still confirmed: NENAC and Central Asia are among the lowest participant group in the labour market, while the reservation wage in this subsample is also the lowest for the Central Asian women, and statistically equivalent to the benchmark group for the NENAC women. A very similar picture is depicted in Table 9, which shows the estimates for immigrants with a blood relative already residing in Italy.
Estimates for the irregulars are reported in Table 10. Columns [1] and [2] (activity and employment rates) show that incentives to work are particularly strong for the irregulars.
Activity rate averages around 90 per cent, while the employment rate is much lower (64 per cent). Cross-national differences are still in place, especially in the employment rates, although they narrow with respect to the baseline sample. However, our baseline result on reservation wages remains undisputed: the difference in reservation wages is negative and statistically significant for the NENAC group, while negative but not significant for Central Asian migrants.
Finally, Table 11 tackles the issue of the multi-religious countries. As explained above, even in nations with an HI based on religions smaller than one, we can observe the 19 This result is confirmed by Constant and Zimmermann (2005).
-20 -prevalence of one religion that can invalidate our identification strategy. To address this concern, we only consider countries with an HI lower than 0.75. Results remain quite similar to the baseline ones, thus confirming the correctness of the identification strategy.

Concluding remarks
As in many advanced countries, activity and employment rates for immigrant women in Italy display great variability across nationals groups. Some nationalities (NENAC, Central Asia), in particular, are characterized by extremely low participation in the labour market. This fact can be due to either a supply "cultural" or a demand "skill" effect. The aim of this paper is to disentangle these two components by estimating the economic value of the time spent at home (reservation wages).
From a technical point of view, we use religion as an identification variable for the effects of culture on labour supply in multi-religious countries. Our results show that low activity and employment rates for NENAC and Central Asia are not associated with higher reservation wages; this implies that low migrant participation is attributable to a weak demand for their skills rather than to a high value ascribed to time spent at home. This result is quite robust to a number of sample selections and specification tests. In particular, the reservation wage differentials do not change according to the mode of entry into the host country (visa based on family reunification or the presence of blood relatives) or to legal status within the host country.
This result can hardly be overstated and it may have some important policy implications. As low participation is mainly involuntary, policymakers have a set of policies that can be used to raise participation and employment rates. The first option is related to selective migration. As far as the policymakers are concerned, with a quantitative target (e.g. Lisbon criteria), admissions could be aimed at those nationalities whose capabilities to integrate are particularly high. The policy, in this case, could also be designed to require different educational levels for different countries of origin. Alternatively, for the nationalities whose skills are particularly mismatched, the policymaker can think of stricter educational requirements. The second set of options relates to policies for labour market inclusions. The idea is to create a set of training programmes with the aim of helping immigrants to become more suited to local labour market requirements.
-23 -   Source: Authors' calculations on ISMU dataset. Robust standard errors in parenthesis. Stars show significance levels, *** up to 1 per cent, ** between 1 per cent and 5 per cent, * between 5 per cent and 10 per cent. All regressions are weighted according to the sample design.
-26 -  (1) Percentage differences in computed reservation wages. Stars show significance levels, *** up to 1 per cent, ** between 1 per cent and 5 per cent, * between 5 per cent and 10 per cent. Mean differences are weighted according to the sample design.
-28 - Source: Authors' calculations on ISMU dataset. Percentage differences in computed reservation wages. Stars show significance levels, *** up to 1 per cent, ** between 1 per cent and 5 per cent, * between 5 per cent and 10 per cent. Mean differences are weighted according to the sample design.
-29 - Source: Authors' calculations on ISMU dataset. Percentage differences in computed reservation wages. Stars show significance levels, *** up to 1 per cent, ** between 1 per cent and 5 per cent, * between 5 per cent and 10 per cent. Mean differences are weighted according to the sample design.
-30 - (1) Percentage differences in computed reservation wages. Stars show significance levels, *** up to 1 per cent, ** between 1 per cent and 5 per cent, * between 5 per cent and 10 per cent. Mean differences are weighted according to the sample design.
-31 - (1) Percentage differences in computed reservation wages. Stars show significance levels, *** up to 1 per cent, ** between 1 per cent and 5 per cent, * between 5 per cent and 10 per cent. Mean differences are weighted according to the sample design.
-32 - (1) Percentage differences in computed reservation wages. Stars show significance levels, *** up to 1 per cent, ** between 1 per cent and 5 per cent, * between 5 per cent and 10 per cent. Mean differences are weighted according to the sample design.
-33 - Appendix 2. A comparison with the self-reported reservation wages in the Italian Labour Force Survey.
In this Appendix, we compare our estimates for reservation wages on ISMU data with those provided in the Labour Force Survey (LFS). We use on the 2006-08 waves of LFS, as previous ones do not provide information on the citizenship of foreign workers. Moreover, in order to provide a comparable result we concentrate on unemployed females in the 15-64 age bracket residing in the North of the country, 20 where Lombardy is located. We provide here the answer to the question "What is the lowest net monthly wage you would be willing to accept?" Results (Table A1) show that the magnitude for reservation wages is quite similar across national groups. Moreover, cross-national differences in LFS basically mirror those in our estimates. 20 The extension to other regions provides very similar results since immigrants in Italy mostly settle in the North.