VoxEU Column Europe's nations and regions Migration

Break-ups of inter-ethnic marriages in Italy

Modern Italy has more inter-ethnic marriages – a consequence of recent immigration. This column uses recent census data to show that  inter-ethnic marriages in Italy have a significantly higher risk of separation, which persists even when accounting for spousal traits and self-selection. The difference is smaller for recent marriages, reflecting a more secular society.

On 19 March 2017, a popular Saturday talk-show called "Let’s talk about it" on Rai1, an Italian public television channel, caused outrage on social media and in the press when it listed a number of reasons why Italian men, according to a “poll,” prefer “Eastern European fiancées” to Italian ones. The stream of sexist remarks included “these mums quickly get back into shape after child-bearing,” “they are always sexy and do not wear baggy gym suits at home,” “they forgive their husband’s cheating,” “they obey their husbands,” “they are perfect housewives,” and so on.

As a result, the programme was cancelled, its popular host was fired, and Rai’s chief editor had to face to the public. The episode reflects prejudices on a relatively novel social phenomenon: inter-ethnic marriages.

Marriage stability in theory

Economic theory (Becker 1974, Becker et al. 1977), suggests that differences among spousal traits may worsen or improve the couple’s welfare, resulting in more or less stable unions, depending on whether these characteristics are complements or substitutes in the household’s production function. Shared values are complements for educating children (positive assortative mating). Different abilities among spouses, for example in work or in raising children, may be substitutes (negative assortative mating) and allow welfare-improving task specialisation within the couple.

Our recent research looks at whether inter-ethnic marriages in Italy are as stable as co-ethnic marriages (Bottazzi et al. 2017). We define ‘inter-ethic’ as marriages for which spouses were born in different countries. Our hypothesis is that differences in preferences and culture, such as the role of women in society, may explain any differences we found.

This question is very important because inter-ethnic marriages can be considered as the ultimate  successful integration, at a time when immigration issues in Europe and the US have become  political priorities.

Marriage stability in census data

Italy has traditionally been a country of emigration that has no relevant colonial past. It has only recently experienced a surge in immigration. The share of foreign (legal) residents in the population was only 2.2% in 2002, and rose to 8.1% in 2014. This created an increase in inter-ethnic marriages by our definition. The share rose from 4.4% in 1995 to 14.8% 2012 (ISTAT 2014). Mixed marriages have been analysed for many countries with a long tradition of immigration (Bratter and King 2008, Dribe and Lundh 2011, Feng et al. 2012, Furtado et al. 2013, Frimmel et al. 2012), there is almost no research on Italy (De Rose 1992 and Manasse and See 2013 are exceptions).

We analyse census data that recorded all marriage break-ups in Italy between 2008 at 2010. This represents about 250,000 marriages, lasting between 65 years and a few days. The data show that, contrary to Let’s talk about it, inter-ethnic marriages are likely to break up earlier than co-ethnic ones. Figure 1 shows the average duration of five types of marriages: those where both spouses were born in Italy, inter-ethnic marriages with Italian husband, inter-ethnic marriages with Italian wife, co-ethnic foreign spouses, and inter-ethnic foreign spouses (foreigners who were born in different countries).

The difference between the durations is striking. Marriages between Italians in the sample survived about 15.6 years on average, compared to 9.9 years and 10.5 years for inter-ethnic marriages with Italian men and women, respectively. The shortest-lived marriages on average were inter-ethnic marriages with two foreign spouses.

Figure 1 Average duration of marriages, by couple type (2008-2010)

Source: Italian census data.

Spousal characteristics

Does this mean that inter-ethnic marriages per se are much more likely to fail? Not necessarily. In our sample, spouses in inter-ethnic couples had different characteristics relative to the average Italian co-ethnic spouses, and these may have accounted for shorter duration of marriages. First, inter-ethnic marriages were twice as frequent for Italian men than for Italian women. Italian husbands in these marriages were also much older than the average husband and older than their foreign-born wives. They were also significantly less educated. Spouses in inter-ethnic couples were more frequently on their second marriage, and were much less likely to have children than Italian co-ethnic couples. They also had worse employment situations. Interestingly, inter-ethnic couples exhibited signs of lower ex-ante commitment to marriage. Civil, as opposed to religious, ceremonies were significantly more frequent, despite a large majority of these marriages involving spouses from Catholic countries. Moreover, inter-ethnic couples more often chose not to merge their wealth into a shared property arrangement. This arrangement can be seen as a commitment device to the marriage, given than it would create higher litigation costs in case of separation. Yet, even when these differences, region, and time effects are considered, inter-ethnic marriages are found to last less than co-ethnic ones: about 6.5 years.

Causality and self-selection

Does this mean that an inter-ethnic marriage causes early separation? To establish this, we should observe the counterfactual: how long would spouses in inter-ethnic marriages have stayed together had they married a co-native instead? In our case, the treatment (inter-ethnic marriage) is not necessarily randomly assigned, because the decision to marry a foreigner may depend on unobserved characteristics of the spouse that might also affect the marriage duration. Consider the possibility that people who have liberal or secular attitudes are more open to cultural diversity (for example, they studied or lived abroad) and, as consequence, are more likely to choose an inter-ethnic marriage. Their attitude means they may also believe less in the permanence of marriage, and may be more likely to separate. In other words, more 'secular' people are keener to marry a foreigner and less averse to a divorce.

The bias could go either way, however. Consider a native spouse who is less 'marketable' in the Italian marriage market due to some unmeasured trait (a health problem, for example). As a result, he or she may be averse to separation because he is less likely to find a native partner. If we neglect the issue of self-selection into mixed marriages, we may respectively over-estimate (in the case of liberal attitudes) or under-estimate (in the case of a non-marketable spouse) the impact of inter-ethnic marriages on the marriage duration. We found that self-selection due to a non-marketable spouse, is prevalent when we consider older marriages, while liberal spouse self-selection prevails in more recent ones. Over all, the two effects largely cancelled out in the sample. We estimated that inter-ethnic marriages caused a reduction of about five years in the duration of a marriage.

O Tempora! O Mores!

This is not the end of the story. As Cicero wrote, moral values change. Attitudes towards the family, marriage, and divorce have evolved in Italy since divorce was legalised in 1974.  

The fraction of marriages ending in divorce has more than doubled, from 2% in 1980 to 9% in 2014 (Eurostat 2016). Single-parent families have become more common, same-sex unions have been legalised and the social stigma associated to separations has disappeared. Even the Church is now revising its exclusion policy towards divorcees. As a result, the shorter duration of inter-ethnic marriages may partially reflect the fact that they were celebrated much more recently (a cohort effect), at a time when separations had become more common, and faster, for all marriages. Indeed, we found that co-ethnic marriages lasted between five and six years longer (43%) than inter-ethnic ones when we considered all marriages in the sample, but only 1.2 years longer (16%) when we considered marriages celebrated since 1994.

Our results indicate that inter-ethnic marriages in Italy had a significantly higher risk of separation which persisted even when we accounted for spousal characteristics and allowed for self-selection into mixed marriages. More recent marriages, however, displayed convergence in longevity, reflecting a society that is becoming more secular, as well as increasingly open to cultural and ethnic diversity.


Becker, G (1974), “A Theory of Marriage”, Journal of Political Economy, 81(4): 813–846.

Becker, G, E M Landes and R T Michael (1977), “An Economic Analysis of Marital Instability”, Journal of Political Economy, 85(6): 1141-1187.

Bottazzi, L, P Manasse and S G See (2017), “Better Wed Over the Mixen than Over the Moor? Break-Ups of Inter-Ethnic Marriages in Italy”, Quaderni Working Paper DSE, No. 1098.

ISTAT (2014). Matrimoni, Separazioni e Divorzi.

Bratter, J L and R B King (2008), “But Will It Last?: Marital Instability Among Interracial and Same Race Couples”, Family Relations, 57(2): 160–171.

De Rose, A (1992), “Socio-Economic Factors and Family Size as Determinants of Marital Dissolution in Italy”, European Sociological Review, 8(1): 71–91.

Dribe, M and C Lundh (2011), “Cultural Dissimilarity and Intermarriage. A Longitudinal Study of Immigrants in Sweden 1990-2005” International Migration Review, 45(2): 297–324.

Eurostat. (2016). Marriages and divorces statistics.

Feng, Z, P Boyle, M Ham and G Raab (2012), “Are Inter-Ethnic Unions More Likely to Dissolve Than Co-Ethnic Unions? New Evidence from Britain”, European Journal of Population, 28(2): 159–176.

Furtado, D, M Marcen and A Sevilla-Sanz (2013), “Does Culture Affect Divorce Decisions? Evidence from European Immigrants in the US”, Demography, 50(3): 1013–1038.

Frimmel, W, M Halla and R Winter-Ebmer (2012), “Assortative mating and divorce: evidence from Austrian register data”, Journal of the Royal Statistical Society: Series A (Statistics in Society), 176: 1–23.

Manasse, P and S G See (2013), “Marital Dissolution in Italy. Are Inter-Ethnic Couples More Likely to Separate?”, mimeo, University of Bologna, presented at the DSE Economics Seminar.

4,619 Reads