Immigration within Europe has greatly increased in the last decades. Migrants may be very beneficial to the labor supply of the destination countries e.g. by increasing diversity of productive skills in the labor market, raising productivity levels and creating new employment opportunities that can be enjoyed by natives too. To value the contribution of migrants to the destination economy, it is important to take into account that the individual decision to supply labor is simultaneous to other important family choices e.g. regarding marriage, fertility, investments in children, and allocation of tasks within the family. These choices depend on values, beliefs and preferences partly rooted into individuals’ “culture”, i.e. within the broad set of knowledge, understanding, and practices shared by people from the same ancestry (this definition is by Fernández 2009). The relevance of cultural factors to economic decisions is nowadays undisputed. In particular, preferences for leisure vs. work or consumption, altruism, and beliefs regarding individual and gender roles in the family and the labor market are important cultural determinants of labor supply decisions. Nevertheless, the way we integrate cultural factors into models predicting the effects of immigrants on the long-term patterns of labor supply in Europe remains largely unsatisfactory.

 

The objective of the “Migration And Labor supplY wheN culturE matterS” (MALYNES) project is to propose an encompassing framework suited to predict the future effects of migration on labor supply in the European Union. MALYNES pursues this general objective following a three-fold strategy. First, it wants to create knowledge about future scenarios regarding labor mobility and migration in Europe. These scenarios depend on natives’ perceptions of the value of immigration, and political support towards an open migration policy. Second, it applies well-known empirical approaches from the cultural economics literature to the detailed information available from European time use and register data. In this way, it derives a cross-cultural “map” of the most important values and preferences (many of them related to family behaviors) that affect the individual labor supply decision. Third, it develops a quantitative theory for the impact of migration on labor supply. The model incorporates cultural differences as a key ingredient of behaviors like fertility, marriage and time use of both migrant and native families in a wide variety of European countries. We estimate the parameters of the model using structural estimation techniques. These objectives are declined in three tasks:

​

​

Task 1. Creating knowledge about future scenarios regarding migration and migration policy in Europe. 

 

This task is dedicated to the analysis of the possible migration policy scenarios in Europe. We analyze the impact of immigrants on political attitudes of European natives measured by the “salience” of migration in the political debate and the “revealed political preferences” of natives regarding migration policy in Europe. These factors will play a key role in determining the future migration policy outcomes. Possible policy scenarios range between (i) freedom of movement with no frontier controls, and (ii) closed borders with no migration. In between these two extremes, intermediate scenarios may arise e.g. (iii) the creation of a coordinated European migration policy determining and implementing a maximum stock of migrants per country; (iv) the implementation of preferential entry-mechanisms by countries of origin; or (v) heterogeneous policy outcomes across European countries, determined by the relative importance of anti-immigrant sentiment e.g. at the regional level within country as in the recent Brexit.

 

Task 2. Applying the scientific methods coming from the cultural economics literature to derive a cross-cultural map of values and preferences that are relevant to labor supply decisions.

 

The objective of this task is to investigate the cultural determinants of important family choices, which are simultaneous to the individual labor supply decision. Intuitively, the value of labor supply (e.g. in terms of hourly wages earned in the labor market) is the opportunity costs of time spent privately (e.g. to consume leisure time) or within the family (e.g. to raise kids, share housework with the partner, etc.). Individual preferences in these dimensions are partly related to culture, and determine the decision to marry, to have kids and the number of births.

 

We use migrant families as a laboratory to point out these mechanisms. This is a crucial task to disentangle the cultural contribution of migration to the labor supply of the destination country. Migrant individuals come from multiple cultural backgrounds. These backgrounds interact with those of native individuals in the destination, e.g. through inter-cultural marriages, and may be persistent in second and higher generations (Fernández and Fogli, 2009). We account for the fact that migration itself is a decision taken at the family level, thus correlated with our outcomes of interest. We thus analyze systematic differences between a country’s natives and emigrants along the cultural dimension (Docquier et al., 2017), and discuss the implications of such differences for the estimation of cultural effects. To these purposes, we compare cultural traits estimated among natives only, emigrants only, or both natives and emigrants from a country of origin (see e.g. Moriconi and Peri, 2015).

 

Research in this task requires a higher level of detail of the information about migrants and their families, than it is available from cross-country survey data (e.g. ESS). We use individual time use, household survey and register data, available from national sources. The use of these type of data is new in the identification of cultural effects. These data have detailed information on family composition and country of origin of family members. This helps identifying the persistence of culture in three generations of migrants (from children to their grandparents). Moreover, these data present a detailed time use dimension (e.g. daily diaries), and information on ancestry of the family, which allow to investigate cross-cultural interactions due to inter-cultural marriages, and effects on offsprings.

 

Task 3. Developing a quantitative model able to predict the impact of migration on labor supply, enriched by cross-cultural differences between individuals.

 

As discussed above, most of the existing literature studying the effect of culture on labor supply applies the epidemiological approach by Fernández (2007). This empirical approach prompts a straightforward evaluation of the importance of culture for economic behaviors. Nevertheless, it has a main limitation of being “reduced form”, i.e. it does not allow modelling the multiple channels through which culture affects individual decision to supply labor. The argument against this type of approach can be casted in terms of the Lucas’ critique (Lucas, 1976): it is naive to predict the effects of e.g. the education, marital, or fertility decisions on labor supply, entirely based on reduced form relationships estimated through the epidemiological approach. In fact, these decisions are themselves shaped by a complex combination of mechanisms, which are object of studies in family, cultural and institutional economics.

​

This task wants to lift these scientific barriers and develop a quantitative theory able to predict the impact of migration on labor supply, enriched by cultural differences between individuals. We will use this theory for structural estimations of the deep parameters attached to decisions on time use, marriage and fertility. 

  

 

 

Originality and relevance in relation to the state of the art

 

Figure 1 provides a synthetic description of migration patterns in the European continent. Arrows indicate the stock of migrants from / to each European country. The color and size of the arrow signal respectively the direction of the migration phenomenon, and its relevance in quantitative terms.[1]The relatively large size and frequency of yellow and orange arrows (indicating European immigration) compared to blue arrows (indicating European emigration) provides a visual portrait of the fact that the stock of living European immigrants is large relative to the stock of European emigrants elsewhere. This suggests that European countries are a “melting pot” of immigrants from multiple cultural, ethnic, and linguistic backgrounds. If we look at the European Union (EU28), in 2015, the average stock of migrants touched the 11% of the total population, as opposed to the 6% in 1990 (World Bank, United Nations Population Division). This implies that European countries are facing a rising birthplace diversity, which may induce beneficial economic effect, particularly of skilled migration (Alesina et al.,2016). This diversity also increases the frequency of interactions between migrants from different origins, and between migrants and natives in each European destination. 

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

 

 

 

What is the overall impact of migration on the labor supply of EU destination countries? To answer this question, we should consider multiple channels. From a demographic perspective, migration reduces the age of workers: own calculations based on Eurostat (2017) suggest that the size of the youngest cohort of working age population (i.e. age 15-24) relative to the pre-retirement cohort (i.e. age 55-64) is twice as large among immigrants as among natives, on average. In addition, immigrants are more fertile than natives, with total fertility rates that, if shared by European natives too, would support a steady state level of the working age population (World Bank, 2015, 2017).[2]Even though immigrants tend to become more similar to natives over time in this dimension, their fertility behavior is partly rooted in their culture of origin (Fernández and Fogli, 2009). Spillover effects from migrants to natives may be also relevant, particularly in the presence of inter-cultural marriages (Pascale et al.,2015). 

From an economic perspective, migrants have an impact on the labor supply of natives in the destination country. A number of papers claim that an increasing presence of migrants is detrimental to native workers, as it increases the overall supply of labor, thus reducing natives’ employment prospects and wages (Borjas, 1995). Such concerns are very common among European natives, ending up into negative attitudes towards migration, and support to nationalist parties that pursue a restrictive agenda on migration policy (Scheve and Slaughter, 2000; Halla, et al., 2017). However, contributions that are more recent suggest that these fears are not fully justified: labor market competition between natives and migrants obtains in models where labor is a homogeneous factor of production. Once we consider a less stylized framework with heterogeneous workers (e.g. by their education level or specific skills), and account for natives’ economic responses to immigration, migrants have a positive impact on the labor supply of the destination country (Peri, 2016). In fact, immigrants increase diversity of productive skills in the labor market, favor innovation and raise productivity of the destination economy (Ortega and Peri, 2014). They fill up manual intensive and unskilled jobs, allowing companies to expand and create vacancies that also natives will take (Peri and Sparber, 2009). Through job creation, immigration has a positive effect also on wages, particularly of less educated native workers, with no negative consequences on native employment (Docquier et al., 2014). 

 

The analysis of the contribution of migrants to the labor supply of the destination economy becomes richer once we account for cultural factors. The cultural economics’ literature has very often identified cultural traits applying the “epidemiological approach” by Fernández, (2007). This links the individual preferences of children of immigrants (in Europe or the United States) to cultural attitudes measured in the country of origin of their parents. These cultural traits differ across countries, change slowly and may play an important role in labor supply decisions of families, and in the allocation of time between men and women. 

 

 

In Figure 2 we follow this approach to plot the “cultural distance” between immigrants in EU28 (including intra and extra European immigration), and natives. We focus on three preferences’ traits that affect individual labor supply decisions. The first is preferences for work vs. leisure among male individuals, which we measure by strong individual agreement to the question “I would enjoy having a paid job, even if I did not need money” (Moriconi and Peri, 2015). A second important dimension is women fertility (Fernández and Fogli, 2009), measured by the fact of women having at least one child. The third trait is women’s attitudes towards work vs. family (Fernández, 2007), measured as strong disagreement among women to the statement “women should be ready to cut off work for the sake of family”. We draw these traits from the 2010 wave of the European Social Survey (ESS). Following Fernández (2007), we retrieve from the sample of immigrants the percentage of respondents by country of origin that complies with each trait, controlling for individual and parental background. Finally, we construct a synthetic indicator of the “cultural distance” between immigrants and natives in EU28, by taking the difference between the percentage of immigrant respondents and the corresponding percentage of native respondents in each dimension. 

The black dotted line in Figure 2 plots the “zero distance” benchmark between immigrants and natives.[3]The highly volatile green, red and black lines suggest a large cross-cultural heterogeneity of immigrants around the culture of destination. Most immigrants come from high fertility cultural backgrounds, relative to the EU28 average. For example, by their own culture, the share of women from some African and Middle-Eastern countries with at least one child is between 20 and 40 percentage points higher than the corresponding share of European native women (green line in Figure 2). Interestingly enough, women from these cultures in most cases present a weaker attachment to work relative to women in the destination culture (with some exceptions e.g. Lybia, Egypt. See the black bold line in Figure 2). This recalls the importance of “family culture” in determining family sizes, and household arrangementsthat may discourage female labor supply (see e.g. Fernández, 2007; Fernández and Fogli, 2009; Algan and Cahuc, 2007; Alesina and Giuliano, 2010. See also Alesina and Giuliano, 2016 for a review). In other cultures however, the percentage of women that is not willing to cut work for the sake of family is remarkably higher than in their European counterpart (e.g. Brazil, Ecuador). Among these cultures, the strong attachment to work is not necessarily associated with low fertility norms. This is consistent with the cross-cultural heterogeneity in family work experiences discussed in demographic research (see e.g. Mc Donald, 2000). Finally, it is remarkable that the majority of cultures of origin seem to present higher preferences for work vs. leisure compared to European natives (red line in figure 2). Thisevidence is consistent with the view that immigrants generally “export” to the country of destination a preferences’ trait relatively favorable to labor (Moriconi and Peri, 2015). High preferences for work characterize both cultures with seemingly progressive attitudes towards women’s work (e.g. some Central-Eastern European cultures), and cultures that preserve a more conservative trait where the man is the family breadwinner (e.g. some Middle-Eastern and Eastern cultures). In fact, some cultures are more gender equal compared to others that present a clear distinction between labor-oriented males and family-oriented females (see e.g. Giavazzi et al., 2013; Lippman et al.2016). This is important, as in the presence of a cultural-mix, the culture of the mother is often found to be more relevant than the culture of the father for the labor supply of the male children and his partner (see e.g. Fernández et al., 2004). 

 

The MALYNES research project builds upon this knowledge, and is motivated by the following observations:

• The lack of a unified framework able to comprise the multiplicity of channels just discussed prevents the existing literature from reaching a straightforward conclusion regarding the overall effects of immigration on the labor supply of the destination country. In particular, very little has been said about the interplay between purely economic mechanisms (e.g. through skills’ composition, and market responses by economic agents) and cultural forces (e.g. through cultural interactions between migrants and natives in the destination).

• The main challenge is going beyond the causal relationships on each specific determinant of labor supplies that can be estimated by means of a reduced-form approach. A structural approach is better suited to define how labor supply outcomes relate to preferences, the relevant economic and policy determinants (e.g. migration policy and patterns), and the family level choices that determine labor supply outcomes.

 

MALYNES aims at lifting these scientific and technical barriers to offer general predictions regarding the effects of immigration on labor supply in Europe. We propose a fully specified structural model in which marriage, fertility and children investment decisions, shape labor supplies. We will start by producing a theoretical model, which uses collective cooperative household decision theory à la Bourguignon and Chiappori (1992). This model will extend the one of Baudin et al.(2015a, 2015b) in order to reproduce fertility, marriage decisions as well as households’ decisions regarding the time that partners allocate to leisure, work and childcare in European countries. The model will encompass both gender differences in preferences, and cultural heterogeneity between natives and migrants. 

 

To estimate the fully specified structural model, we aim at using the Simulated Method of Moments (SMM hereafter) introduced by Mc Fadden (1989). As anticipated by Sickles and Taubman (1997) and demonstrated by Jeremy Greenwood and his co-authors, this technique is very relevant for research in household and family economics. For instance, Greenwood et al.(2016) estimate the deep parameters of a unified model where decisions on marriage, divorce, female labor force participation and education are endogenous. They show how their model is suitable to reproduce postwar U.S. data. Using this technique, they have been able to disentangle the respective roles of education, female labor force participation and assortative mating in explaining income inequalities in the US (see also the pioneering contribution by Aiyagari et al., 2000). Moreover, Baudin et al.(2015a,b), and Baudin and Stelter (2016) have applied this technique on the dynamics of marriage, fertility, childlessness and internal migration in both developed and developing countries.

 

 

Selected References

 

- Alesina, A. and P. Giuliano (2010), “The power of the family,” Journal of Economic Growth, Springer, vol. 15(2), pages 93-125, June.

- Alesina, A. and P. Giuliano (2016), “Culture and Institutions”, Journal of Economic Literature, 53(4): 898-944.

- Alesina, A., Harnoss, A. and H. Rapoport (2016) “Birthplace diversity and economic prosperity”, Journal of Economic Growth, 21(2): 101-138.

- Algan, Y. and P. Cahuc (2007), “The Roots of Low European Employment: Family Culture?” In NBER International Seminar on Macroeconomics 2005, edited by C. Pissarides and J. Frenkel. MIT Press, pp. 65-109.

- Baudin, T., de la Croix, D. and P.E. Gobbi (2015a), “Fertility and Childlessness in the US”. American Economic Review 105(6):1852-1882. 

- Baudin, T., de la Croix, D. and P.E. Gobbi (2015b), “Endogenous childlessness and the stages of development”, IRES Discussion Paper 2015-03.

- Baudin T. and R. Stelter (2016), “Rural exodus and fertility at the time of industrialization”, Discussion Paper 2016-20.

- Borjas, George J. (1995), “The Economic Benefits from Immigration.” Journal of Economic Perspectives 9(2): 3–22.

- Bourguignon F., Chiappori P.A. (1992), “Collective models of household behaviour”, European Economic Review 36, pp 355 – 364.

- Docquier, F. & Çağlar Ozden & Giovanni Peri (2014), "The Labour Market Effects of Immigration and Emigration in OECD Countries," Economic Journal, Royal Economic Society, vol. 124(579), pages 1106-1145, 09.

- Docquier, F., Tanselb, A. and Turati, R. (2017), “Do emigrants self-select along cultural traits? Evidence from the MENA countries”, GLO Discussion Paper Series 146, Global Labor Organization (GLO).

- Eurostat (2017), http://ec.europa.eu/eurostat/statistics-explained/index.php/Fertility_statistics

- Fernández, R. (2007), “Women, work, and culture.” Journal of the European Economic Association 5(2-3): 305-332.

- Fernández, R., A. Fogli and C. Olivetti (2004), “Mothers and sons: preference formation and female labor force dynamics.” Quarterly Journal of Economics November: 1249-1299.

- Fernández, R. and A. Fogli (2009), “Culture, an Empirical Investigation of Beliefs, Work, and Fertility”. American Economic Journal: Macroeconomics, 1:1, 146-177.

- Giavazzi, F., Petkov, I., and F. Schiantarelli (2014), "Culture: Persistence and Evolution," NBER Working Papers 20174, National Bureau of Economic Research, Inc.

- Giavazzi, F., Schiantarelli, F. and M. Serafinelli (2013), “Attitudes, Policies and Work”, Journal of the European Economic Association, 11(6):1256-1289.

- Greenwood, J., Guner, N., Kocharkov G., and C. Santos (2016), “Technology and the changing family: A unified model of marriage, divorce, educational attainment and married female labor-force participation”. AEJ: Macroecroeconomics 8 (1): 1 41. 

- Aiyagari S.R., Greenwood J. and N. Guner (2000), “On the state of the Union”, Journal of Political Economy, 108(2), 213-244.

- Halla, M., Wagner, A.F., and J. Zweimüller (2017), “Immigration and Voting for the Far Right”, Journal of the European Economic Association, Volume 15, Issue 6, 1 December 2017, Pages 1341–1385. 

- Lippman, Q., Georgieff, A. and C. Senik (2016), “Undoing Gender With Institutions. Lessons from the German Division and Reunification”, PSE Working Papers n°2016-06.

- Lucas, R. (1976), "Econometric Policy Evaluation: A Critique". In Brunner, K.; Meltzer, A. The Phillips Curve and Labor Markets. Carnegie-Rochester Conference Series on Public Policy. 1. New York: American Elsevier. pp. 19–46. ISBN 0-444-11007-0.

-McFadden, D. (1989), "A Method of Simulated Moments for Estimation of Discrete Response Models Without Numerical Integration". Econometrica, 57(5):995–1026.

- Moriconi, S. and G. Peri (2015), “Country-Specific Preferences and Employment Rates in Europe”. NBER Working Paper w21561.

- Ortega, F. and G. Peri (2014), "Openness and income: The roles of trade and migration," Journal of International Economics, Elsevier, vol. 92(2), pages 231-251.

- Pascale I. van Zantvliet, M.K. and E. Verbakel (2015), “Early Partner Choices of Immigrants: The Effect of Preferences, Opportunities and Parents on Dating a Native”, Journal of Ethnic and Migration Studies Vol. 41, Iss. 5, 2015.

- Peri, G. and C. Sparber (2009), “Task Specialization, Immigration, and Wages”.  American Economic Journal: Applied Economics · vol. 1, no. 3, July 2009. (pp. 135-69).

- Peri, G. (2016), "Immigrants, Productivity, and Labor Markets." Journal of Economic Perspectives, 30(4): 3-30.

- Scheve, K. F. and M. J. Slaughter (2001), “Labor Market Competition and Individual Preferences over Immigration Policy?”, Review of Economics and Statistics, 83, 1, 133-45.

- Sickles, R.C., and P. Taubman (1997), "Mortality and morbidity among adults and the elderly." Handbook of population and family economics 1 (1997): 559-643.

- World Bank (2015), “Golden Aging: Prospects for Healthy, Active, and Prosperous Aging in Europe and Central Asia”. Washington, DC: World Bank. 

- World Bank (2017), “World Development Indicators”, The World : http://data.worldbank.org/products/wdi.

 

​

Notes

​

[1]Stocks include both economic migrants and asylum seekers. For expositional purposes migration is signaled by arrows only when the total stock exceeds 0.25 million of migrants from one origin to a destination. Yellow and orange arrows signal intra-European and extra-European migration, respectively. Blue arrows signal emigration from Europe to extra-European destinations. The figure assigns two colors to arrows from/to Turkey to signal that part of Turkey is located in Europe. 

[2]In 2017, the total fertility rates from sender regions of Europe and Central Asia, Middle-East and North Africa, and Sub-Saharan Africa touch upon 2, 3, and 5 children per women, respectively. The average birth rate in EU28 is 1.58 children per woman, well below the replacement rate of 2.1, which would be required to maintain population constant without immigration (World Development Indicators, 2017).

[3]In this exercise, EU countries are recorded both as cultures of destination (measured using natives’ preferences) and cultures of origins (measured using migrants’ preferences). Preferences of European migrants are generally close to the black dotted line, net of selection effects along the cultural dimension, which are discussed in the literature (see Moriconi and Peri, 2015).