Marriage, fertility, and the cultural integration of immigrants in Italy
Migration to Western countries has steadily increased over the past few decades, representing one of the most contentious political and socioeconomic phenomena these countries face. The very nature of migratory inflows has also been transformed, giving rise to the perception that current immigrants are culturally more distant than those of past eras. Anti-immigrant sentiments appear widespread, and voters’ attitudes reveal substantial support for restrictive immigration policies.
These tensions seem motivated in large part by the perception that immigration imposes cultural externalities on the native population and that immigrants integrate slowly if at all. In fact, sizeable adverse labour market and welfare effects are far from well-documented (Bisin and Zanella 2017). As cultural boundaries are increasingly salient, the successful integration of minorities represents a crucial issue for the future of Western societies (Harder et al. 2018, Abramitzky et al. 2019).
In recent work (Bisin and Tura 2019), we study the cultural integration of immigrant minorities. We interpret cultural integration as an equilibrium phenomenon – the result of a demand for integration on the part of immigrants and a supply, in the form of cultural acceptance, on the part of native populations. We focus specifically on the role of the family in the process of cultural integration of new generations. The family has a pivotal function in this process as the first and possibly the most significant place where values, attitudes, and beliefs are transmitted from parents to children.
This approach allows us to uncover the crucial mechanisms that may contribute to slowing down immigrants’ integration, to evaluate the dynamics of cultural integration for future generations, and to assess the integration response to counterfactual immigration policies.
More specifically, we estimate a structural model of marital matching along cultural lines and intra-household decisions, exploring in detail the roles of fertility and cultural socialisation in the integration process. Parents care about socialising their children and are endowed with technologies to transmit their own cultural-ethnic traits to children (Bisin and Verdier 2000, 2001). Importantly, socialisation incentives and technologies vary between homogamous and heterogamous marriages. Moreover, socialisation behaviour depends on the distribution of the population across ethnic groups at the local level. As a consequence, the model implies a systematic dependence of fertility, socialisation, and divorce patterns across both household ethnic characteristics and geographical regions.
We estimate the parameters of the model, exploiting variability across marital matches and regions, and we use restricted administrative Italian data on the universe of marriages, births, and separations over two decades (1995-2012). We measure cultural-ethnic transmission with language socialisation, proxied by the (self-reported) language spoken at home with the family.
The main parameters of interest in the model are the cultural intolerance parameters – that is, the (psychological) value a parent obtains from socialising a child to his/her own ethnic identity, relative to having a child with a distinctly other cultural-ethnic identity. Cultural intolerances represent a measure of the strength of the preferences for socialisation of a specific ethnic group and hence, a measure of the strength of its resistance to cultural integration.
Figure 1 graphs the estimated cultural intolerance of minorities with respect to the native population (Panel A). Strong preferences towards socialisation of children appear common across cultural-ethnic groups, but particularly so for parents from North Africa or the Middle East, whose estimated intolerance is about seven times as high as the one of the EU15, followed by sub-Saharan Africa and East Asia migrants’ minorities. As reported in Panel B, we also estimate the highest cultural intolerance of the Italian majority towards immigrants originating from sub-Saharan Africa, North Africa, and the Middle East (three times as high as that towards immigrants from the EU15).
Figure 1 Cultural intolerance parameters
a) Migrants towards natives
(b) Natives towards migrants
Notes: This figure reports parameter estimates for the cultural intolerance of migrants versus natives (Panel A) and natives versus migrants (Panel B) for all cultural-ethnic minorities from Europe-EU15, Other Europe, North Africa/Middle East, Sub-Saharan Africa, East Asia, Latin America.
The substantial amount of heterogeneity in cultural intolerances we estimate across cultural-ethnic groups imply potentially significantly different patterns of integration across minorities. In this respect, we investigate the long-run dynamics of cultural integration of minorities, simulating our model of marital matching and intra-household choices over successive generations. Results have to be interpreted in light of the strong assumption that parameters are invariant over time. It is also important to note that the notion of integration we must adopt, given our data, refers to the practice of speaking Italian at home (i.e. we say that a minority is ‘integrated’ when their descendants speak Italian at home).
Figure 2 Long-run dynamics of cultural traits
Notes: This figure shows the long-run dynamics of the distribution of cultural traits in the population for minority groups over successive generations. The share of each cultural-ethnic group over the total population is indexed to 1 in t=0.
Despite cultural intolerance estimates that highlight strong preferences among immigrants for maintaining their cultural identity, in the long run, all cultural-ethnic minorities are simulated to converge to the Italian majority along the language dimension, as depicted in Figure 2. Indeed, 75% of immigrants integrate into the native culture over the period of a generation – in other words, 75% of the second-generation immigrants are simulated to speak Italian at home with their children. However, the pace of convergence is heterogeneous across cultural-ethnic groups. We find that the EU15 and ‘Other Europe’ minorities converge almost completely to the native culture in a single generation (similar to the North Africa/Middle-East minority), while a significantly slower convergence rate is simulated for the Latin America minority. Their integration process does not start before the second generation, and after four generations, only 70% of immigrants from Latin America are culturally integrated. A slower convergence rate also characterises the East Asia and sub-Saharan Africa minorities.
Intolerance parameters are not the only determinants of the dynamics of integration of different cultural-ethnic groups. Indeed, higher fertility and homogamy rates prove to be fundamental socialisation mechanisms that assist in slowing down the cultural integration of some immigrant minorities. A strong estimated selection into homogamous marriages for sub-Saharan Africa migrants, for instance, allows them to sustain their cultural heterogeneity by accessing superior direct socialisation technologies. On the other hand, estimated fertility rates are particularly high for East Asia minorities, a fundamental factor behind their integration patterns. Finally, the relative success of the Latin America migrants in securing their cultural distinctiveness over time is due, in large part, to their unique ability to socialise children in heterogamous marriages with the native-born population.
Figure 3 Long-run dynamics of cultural traits with Italians fully tolerant
Notes: This figure shows the long-run dynamics of the distribution of cultural traits in the population for minority groups over successive generations assuming complete tolerance of the Italian majority towards minorities. The share of each cultural-ethnic group over the total population is indexed to 1 in t=0.
The relative speed of the cultural integration of immigrants is also relevant given the cultural intolerance of Italians and the strength of their own cultural transmission preferences. This effect is somewhat surprising in its sign. In principle, a native population that is more accepting of the cultural traits and beliefs of immigrants might make their cultural integration easier and faster; for instance, by fostering heterogamous marriages. But the opposite holds true: when we simulate our model by counterfactually assuming a higher willingness of the majority to welcome cultural dissimilarities, we obtain, on average, a reduction in the dynamics of integration of minorities towards Italian culture on the order of 15% over the period of a generation. Results are reported in Figure 3. The ability of minorities to maintain their cultural identity in this counterfactual scenario is due to three mechanisms:
- a large increase in demand for intermarriages with the native population and a parallel lower demand for homogamous marriages previously motivated by a parental socialisation premium;
- a sizable increase in fertility rates in intermarriages with the native population, as the socialisation quality of children is higher;
- higher socialisation rates for marriages with native-born Italians due to the higher acceptance of cultural differences among the native-born in these cases.
Finally, we investigate the integration response to potential open-border immigration policies by simulating an exogenous rise in migration inflows. Specifically, we double the number of second-generation minorities, overweighting minorities from North Africa, the Middle East, sub-Saharan Africa, and East Asia. As can be seen from Figure 4, the effects are varied, with some minorities accentuating their successful transmission of cultural values dramatically. While North Africa/Middle East immigrants reduce integration by only 4 percentage points by the third generation, the response of East Asia and Sub-Saharan Africa minorities ranges from a 20-point to a 60-point reduction in integration, significantly slowing down the process of cultural integration.
Figure 4 Long-run dynamics with compositional increase in migration inflows
Notes: This figure shows the long-run dynamics of the distribution of cultural traits in the population for minority groups over successive generations. The solid line represents the dynamics at the baseline, while the dashed line represents the dynamics after raising the share of second-generation North Africa/Middle East, Sub-Saharan Africa, and East Asia minorities. The share of each cultural-ethnic group over the total population is indexed to 1 in t=0.
These findings have implications for the design of immigration policies capable of moving the conversation beyond well-meaning, across-the-board integration policies on the one hand and restrictive, closed-borders policies on the other. In this respect, our findings – that higher socialisation rates in marriages between immigrants and the native population encourages a higher acceptance of minority cultures on the part of the native population, allowing immigrants to better maintain their cultural traits – deserve careful attention.
Abramitzky, R, L Boustan and K Eriksson (2019), “Do Immigrants Assimilate More Slowly Today Than in the Past?”, American Economic Review: Insights (forthcoming).
Bisin, A and G Tura (2019), “Marriage, Fertility, and Cultural Integration in Italy”, NBER Working Paper 26303.
Bisin, A and T Verdier (2000), “Beyond The Melting Pot: Cultural Transmission, Marriage, and The Evolution of Ethnic and Religious Traits”, Quarterly Journal of Economics 115(3): 955–988.
Bisin, A and T Verdier (2001), “The Economics of Cultural Transmission and the Dynamics of Preferences”, Journal of Economic Theory 97(2): 298–319.
Bisin, A and G Zanella (2017), “Time-consistent Immigration Policy under Economic and Cultural Externalities”, Economic Policy 32(91): 415–446.
Harder, N, L Figueroa, R M Gillum, D Hangartner, D D Laitin, and J Hainmueller (2018), “Multidimensional Measure of Immigrant Integration”, Proceedings of the National Academy of Sciences 115(45): 11483–11488.