Introduction

Marriage dissolution is the legal process in which a judge or other legal authority dissolves the bonds of matrimony existing between two people, restoring them to the status of being single. The issue has been extensively analyzed in the last few decades, looking for the determinants of this well-known phenomenon. Researchers have focused on the institutional environment (Goldin & Katz, 2002; González & Viitanen, 2009; González-Val & Marcén, 2012; Heim, 2003; Nixon, 1997; Wolfers, 2006), as well as on the economic conditions of married couples (Bremmer & Kesselring, 2004; Jensen & Smith, 1990; Nunley, 2009). Other studies have examined the social and cultural environment, that is to say, fertility behavior (Svarer & Verner, 2008), education (Isen & Stevenson, 2010; Peters, 1986), age of the couples (Lehrer, 2008), religiosity and church values (Lehrer & Chiswick, 1993; Vaaler, Ellison, & Powers, 2009), and culture (Chong & La Ferrara, 2009; Furtado, Marcén, & Sevilla, 2013). Finally, some investigations have been concerned with sexual relations between the spouses, analyzing marital satisfaction in terms of frequency and quality of sex within the marriage (Allen & Brinig, 1998; Byers, 2005; Lawrance & Byers, 1995), infidelity (Amato, 2010; Amato & Previti, 2003), and pornography (Doran & Price, 2014; Manning, 2006).

Although research on the topic is diverse and extensive, so far the relationship between female migration and marriage dissolution does not seem to have been investigated. In other words, no specific studies have concentrated on whether there exists a correlation between marital dissolution and migration, which might bias the sex ratio within society and, at the same time, introduce new sexual, social and cultural values. Indeed, so far the literature has been focused exclusively on marriage migration, i.e., on foreign females who migrate, in order to improve their socioeconomic status through marriage, particularly between Southeast Asia and East Asia (e.g., Belanger, Lee, & Wang, 2010; Hugo, 2005; Jones & Shen, 2007; Kim, 2012; Wang & Chang, 2002). Could the influx of new people—with different values and/or features, as well as sexual behavior—affect the rate of marital dissolution? This was the research question posed in this study.

Theoretical Background: Female Migration and Marital Dissolution

According to Winking, Kaplan, Gurven, and Rucas (2007), sexual infidelity is quite common (Allen et al., 2005; Collumbien, Das, & Campbell, 2001; Frederick & Fales, 2016; Lawoyin & Larsen, 2002) and men are more likely to commit adultery than women (e.g., Atkins, Jacobson, & Baucom, 2001; Buss & Schmitt, 1993; Frederick & Fales, 2016; Schmitt, 2003; Tafoya & Spitzberg, 2007; Wiederman, 1997), especially if opportunities are available (Atkins, Jacobson, & Baucom, 2001; Greeley, 1994), since men tend to desire promiscuous sex (Buss & Schmitt, 1993; Symons, 1979; Trivers, 1972). Variations in the current sex ratio might represent an opportunity for infidelity, especially if the ratio is biased toward a higher number of women (Geary, Vigil, & Byrd-Craven, 2004; Guttentag & Secord, 1983; Pedersen, 1991). As highlighted by Schmitt (2005), when there are far more women than men (i.e., the sex ratio is low), men become a scarce resource for which women must compete with even more intensity than normal, that is to say, we can expect a higher level of sociosexuality (i.e., more promiscuity). In other words, women have to be more competitive among themselves, fulfilling men’s greater desire for promiscuous sex, thus multiplying opportunities for infidelity. According to McNulty and Widman (2014), the main consequences of infidelity are decreased relationship satisfaction (Sanchez Sosa, Hernandez Guzman, & Romero, 1997; Spanier & Margolis, 1983), increased psychological distress (e.g., Allen et al., 2005; Cano & O’Leary, 2000) and, obviously, marriage dissolution (Amato & Rogers, 1997; Betzig, 1989).

Although the topic has been extensively investigated, no evidence has been provided about the main components of this ratio. In particular, there are no studies on the expected impact of a heavier presence of foreign women, characterized by new features, social values, as well as sexual attitudes. Indeed, a migration influx of females might bias the sex ratio (i.e., the ratio between men and women in society), driving husbands to sexual infidelity and, consequently, to marital dissolution. Furthermore, no light has been shed on the differences, in terms of marital dissolution, between an internal and an external influx of females. Is it admissible to hypothesize that external immigration might have a different impact than domestic influx?

Foreign women might be in a less economically stable position, and for this reason, they might be more willing than nationals to enter into a relation that offers this stability, even with a married man. On the one hand, we can observe the flow of young women with specific social/cultural values and sex attitudes, moving from a very poor country of origin with the aim to find a better future. On the other one, we can observe married men able to fulfill all of these wishes, i.e., provide protection and economic stability. Consistently with the evolutionary theory, the attractiveness of a young woman, combined with greater willingness, could drive these married men to dissolve their household. Obviously, this effect might be greater depending on the economic gap between immigrant females and the new host country.

Hence, the aim of this work was to look into this open issue, in particular, to try and confirm the proposed hypothesis, which was our main starting point:

H1

There is a positive relation between the level of marital dissolution and the sex ratio, i.e., given a number of males, the higher the number of females within an area, the higher the level of household dissolution; afterward, our second suggested hypothesis is as follows:

H2

There is a positive relation between the level of marital dissolution and female migration, which might vary according to the nature of said migration (i.e., foreign or national female influxes).

Due to a lack of micro-data, it was not possible to determine whether men decide to leave their household for a new life with a foreign partner (i.e., long-term mating) or whether this simply constitutes a hidden getaway from their current life (i.e., short-term mating), which is then discovered by their partner. Moreover, we were unable to ascertain whether the new social/cultural values, willingness or atypical sexual attitudes of female immigrants might positively affect this attraction. However, our investigation made it possible to detect the existence of differences, which might be consistent with the proposed thesis and the current literature on evolutionary theory (Geary, Vigil, & Byrd-Craven, 2004; Pedersen, 1991; Schmitt, 2005).

Method

Considering the Regions of Italy between 2002 and 2007, this work proposed an empirical analysis to test whether a relation between female migration and marital dissolution might exist. In detail, several multiple regression models (OLS) were implemented, applying the robust option. The dependent variable of our empirical investigation was a marital dissolution index, while the key explanatory variables were a sex ratio and several female rates. Finally, a dummy variable equal to 1, according to the year in which the observations were made (i.e., 7 variables, from 2002 to 2007), was introduced to control for the time factor.

Italian Judiciary and Marital Dissolution

Legal marital dissolution was introduced in Italy in 1970, with Law n. 898, designating the courts of first instance as competent for this matter. A special feature of the Italian legal procedure is that divorce cannot be obtained directly, since a long period of marital separation is necessary (i.e., 5 years, reduced to three in 1987 and reduced again, in 2015, to 1 year in the case of judicial separation and to 6 months in the case of separation by mutual agreement). Separations might be litigious or non-litigious, depending on whether an agreement can be reached between the parties concerning children, alimony, and/or family goods (e.g., houses and cars). This work considered only the first step of this legal procedure in order to analyze marital dissolution, which means that the empirical analysis focused exclusively on separations, both litigious and non-litigious.

The proposed measure of marital dissolution is calculated as follows:

$$ {\text{Marital dissolution}}_{jt} = \frac{{I_{jt} }}{{H_{jt} }} $$
(1)

where j represents the j-th administrative region considered at year t, I jt is the total number of incoming marital dissolutions (i.e., all applications for dissolution, both litigious and non-litigious, within a region), and H jt is the total number of households, calculated for 100,000 marriages (source: Italian Institute of Statistics). This approach was consistent with Friedberg (1998), González and Viitanen (2009), and Wolfers (2006), although these researchers chose to use the number of resident people (expressed in hundreds) instead of households. The measure proposed here may be considered a more realistic estimation of the dissolution rate, since it was normalized for the number of marriages.

Between 2002 and 2007, there were more than 600,000 marital dissolutions in Italy. According to national statistics (Table 1), on average, the couples filing for marital dissolution had married at the age of 25–28 years old and, after about 15 years, appeared in front of a judge to put an end to their bond. Most of these men and women had a job (86 and 66%, respectively), a high level of education (half of them had, at least, a high school degree), as well as a family with children (70%). How did they come to the decision to dissolve their bond? According to the National Association of Lawyers specializing in this field, the main cause of marital dissolution in Italy is the male partner’s infidelity (ANM Report, 2013), confirming the current literature on male sexual attitudes (Frederick & Fales, 2016; Schmitt, 2003; Tafoya & Spitzberg, 2007). This result is highlighted in Table 1, which shows that three-quarters of applications for marital dissolution were submitted by women, who were clearly the more disappointed party.

Table 1 National descriptive statistics on marital dissolution (Italy, 2002–2007)
Full size table

Migration and Sex Ratio

A sex ratio and several female rates were estimated to better grasp the proposed evolutionary perspective, based on the hypothesis that female migration can explain marital dissolution. In detail, the sex ratio is estimated as follows:

$$ {\text{Sex ratio}}_{jtc} = \frac{{F_{jtc} }}{{F_{jtc} + \, M_{jtc} }} $$
(2)

where j represents the j-th region considered at year t, with respect to the c-th age class, F jt is the total number of resident females, and M jt is the total number of resident males (source: National Institute of Statistics). The proposed approach, consistent with the current literature (Schmitt, 2005), might support the hypothesis that marital dissolution is male driven, that is to say, the obtained coefficient confirmed that males are driven to infidelity if opportunities are available (e.g., a more favorable sex ratio). Note that the current literature also proposes a sex ratio equal to the total number of resident males over the total number of resident females (e.g., Becerra-Fernàndez et al., 2017; Blanchard & Lippa, 2008; Bogaert, 2005). The only difference between these two approaches is the reference value, which is equal to 0.5 in this case (i.e., when the total number of resident females is equal to the total number of resident males), as well as the interpretation of its change. In order to collect more robust results, several age classes were considered (i.e., 18–29, 18–34, and 18–39). This made it possible to determine whether differences might exist among age classes, as suggested by the National Association of Lawyers specializing in this field (ANM Report, 2013).

Subsequently, we introduced two female rates, estimated as follows:

$$ {\text{Foreign female rate }}_{jtc} = \frac{{{\text{EF}}_{jtc} }}{{F_{jtc} }} $$
(3)
$$ {\text{Married female rate }}_{jtc} = \frac{{{\text{MF}}_{jtc} }}{{F_{jtc} }} $$
(4)

where j represents the j-th region considered at year t, with respect to the c-th age class, while F jt is the total number of resident females (the same variable proposed in 2). EF jt is the total number of resident females from a foreign country (i.e., immigrants), while MF jt is the total number of resident females already married (source: National Institute of Statistics). Obviously, 1 minus the foreign female rate is equal to the national female rate, i.e., the weight of Italian females in each region. At the same time, 1 minus the married female rate is equal to the unmarried female rate, i.e., the weight of Italian females being legally free (i.e., divorced, widowed, and unmarried women). Therefore, these rates represented the composition of the sex ratio (2), considering the female side, thus providing a measure of female characteristics within the proposed sex ratio (2). Afterward, the empirical analysis helped determine which the differences might be, in terms of marital dissolutions, between an influx of foreign females and one of national females, given the sex ratio. Tables 2 and 3 present some descriptive statistics of the proposed sex ratio and its composition, looking at the female side.

Table 2 Sex ratio and female rates, considering the population aged between 18 and 49 (Italy, 2002–2007)
Full size table
Table 3 Foreign female rates, considering the population aged between 18 and 49 (Italy, 2002–2007)
Full size table

Considering Italy between 2002 and 2007 and focusing on individuals aged between 18 and 49, on average, there were more men than women (i.e., sex ratio equal to 0.4952), with a high number of married females (i.e., 58%). Taking the foreign females into account, on average, their weight varied between 2% (e.g., in Calabria, Sardinia, and Sicily) and 9% (e.g., in Emilia-Romagna, Tuscany, and Veneto). The concept of female influx took on even greater significance when looking at its trend (Table 3). For example, the foreign female rates in Veneto and Tuscany were double in 2007 compared to 2002, representing a considerable portion of the proposed sex ratio (i.e., greater than 10%). Finally, Table 4 shows descriptive statistics according to geographical areas and age classes, indicating that young women made up most of the foreign female influx. Indeed, more than 80% of foreign females in Italy were aged between 18 and 49, while this percentage among Italian females of the same age decreased dramatically to 40%. Considering European female immigration, the Romanian ethnic group was the main community in Italy (i.e., 25% of European female immigrants in 2007). This Romanian flow was characterized by a sizeable presence of young people and, among them, a high percentage of women (Stan, 2005; Uccellini, 2010).

Table 4 Number of foreign females, according to age classes and geographical areas (Italy, 2007)
Full size table

Results

Table 5 shows the results of the suggested OLS models, considering a pooled sample and different age classes: 18–39 (Model A), 18–34 (Model B), and 18–29 (Model C). 2002 was the dummy variable against which the models were estimated (i.e., the dropped variable).

Table 5 Multiple regression model—robust option (Italy, 2002–2007)
Full size table

All the models were statistically significant, i.e., test (F) confirmed that all the coefficients in them were different from zero (p < .000), while the R-square was between 0.31 (Model C) and 0.37 (Model A). The normality of the residuals was graphically tested with optimal results, while the mean of the variance inflation factor (VIF) ranged between 1.66 and 1.73. Finally, as for the coefficients, several statistically significant results were observed. Both the sex ratio and the foreign female rate were positive and statistically significant (.01 < p < .1), although they obviously relied on the selected sample of population. Focusing on the age classes, the greatest impact of female migration was detected in the middle class, that is to say, in the 18–34 age bracket. Looking at the married female rate, the coefficient was again statistically significant but negative (.01 < p < .05).

Table 6 summarizes the impact of an increase in the presence of Italian and foreign females adopting the predicted variables (i.e., coefficients of Table 5), according to their age classes, simulating either internal (i.e., from another region of the same nation) or external immigration (i.e., from another country) generating a 1% rise in the sex ratio. According to the results proposed in Model B (18–34 age bracket), ceteris paribus, a 1% rise in the sex ratio led to an increase in marital dissolutions equal to 3748 for every 100,000 marriages. Assuming that this sex ratio variation was due to an influx of new females into the region, a different impact can be expected depending on the nature of these females. Indeed, according to the results proposed in Model B, ceteris paribus, a 1% rise in the rate of legally free foreign females (i.e., a 1% decrease in the married female rate and a 1% increase in the foreign female rate) led to an increase in marital dissolutions equal to 5381 for every 100,000 marriages. At the same time, a 1% rise in the rate of legally free national females (i.e., a 1% decrease in the married female rate and a 1% decrease in the foreign female rate) led to an increase in marital dissolutions equal to 3173 for every 100,000 marriages. Therefore, the expected impact of external migration was higher than of internal migration.

Table 6 Simulation according to the results of the multiple regression model—robust option (Italy, 2002–2007)
Full size table

Discussion

In line with the proposed theory, this research aimed to determine whether the external shock of an influx of foreign females might disrupt household balance, by testing the existence of a statistically significant correlation. Due to lack of micro-data, we were unable to understand whether men decide to leave their household for a new life with a foreign partner (i.e., long-term mating, driven by new social and cultural values or atypical sexual attitudes) or whether this simply constitutes a hidden getaway from their current life (i.e., short-term mating), which is then discovered by their partner. However, the results showed that an effect does actually seem to exist, supporting the hypothesis that a more favorable sex ratio might drive couples to marital dissolution. Indeed, according to the results, regions with a larger stock of females were characterized by a higher number of marital dissolutions (i.e., positive coefficient), that is to say, hypothesis H1 were supported. These results were consistent with the current literature on male sexual infidelity (Frederick & Fales, 2016; Schmitt, 2005) and sex ratio (Geary, Vigil, & Byrd-Craven, 2004; Pedersen, 1991; Schmitt, 2005), confirming that marital dissolution is male driven. As suggested by our results (i.e., sex ratio coefficient), an increase in the number of males means that they have to be more competitive, reducing their promiscuity to try and satisfy women’s expectations; conversely, an increase in the number of females provides males with opportunities for infidelity, driving women to be more competitive, i.e., trying to satisfy their promiscuity (see Pedersen, 1991). At the same time, considering female migration influxes, we support hypothesis H2, referring to a positive relation between marital dissolution and female migration, which may vary according to the nature of said migration. Indeed, by comparing the two types of influxes, it clearly emerged that the impact of external migration is higher than the impact of internal female influxes. Finally, focusing on the age classes and their impacts, our results can shed new light on the proposed evolutionary theory. The key age to attract a married man is between 29 and 34 years, which can be considered as the best years for a woman. According to results, it looks like married men wish young but mature women. Indeed, in terms of marital dissolutions, the expected impact of this age class (18–34) was higher than the others, that is to say, the younger (18–29) and the older one (18–39). Figure 1 supports the proposed interpretation, showing the relationship between foreign female rate (estimated in 2002), suggested as a key determinant of household dissolution, and percentage of mixed marriages (i.e., between divorced men and foreign women) over the total number of second marriages (estimated in 2008). Note that the proposed time gap (6 years) is the time needed to obtain marital dissolution in Italy—based on national judicial procedures and the efficiency of courts—i.e., the time it takes a husband to be legally free. These results appeared to be even more significant when considering Table 7, which highlights the differences in mixed marriages (i.e., between Italian husband and foreign wife) in the case of first marriage or second marriage (i.e., when the husband is divorced). Our evidence suggested that, on average, the percentage of mixed marriages with a divorced man is three times higher. In some regions, such as Umbria and Marche, the percentage of marriages between divorced Italian men and foreign women is as high as 25%, four times higher than the percentage of mixed first marriages.

Fig. 1
figure 1

Two-way scatter plot with the relation between foreign females (2002) and mixed marriages of divorced men (2008)

Full size image
Table 7 Rate of mixed marriages between Italian husband and foreign wife, considering first marriages and second marriages (i.e., divorced men) (Italy, 2008)
Full size table

Therefore, a statistically significant positive relation between the influx of female immigrants and household dissolution was detected, based on the idea that opportunities might drive males to infidelity, confirming our main hypothesis. At the same time, the suggested hypothesis was further supported by evidence of a positive relation between foreign female influxes and the percentage of mixed marriages with divorced men, after the conclusion of legal procedures leading to household dissolution.

Limitations

Naturally, our empirical investigation had clear limits, since results show the existence of a correlation between immigration and marital dissolution rather a clear effect. Indeed, this was a macro-analysis, and although the results suggested the existence of a statistically significant relation, it proved impossible to determine whether an effect might exist and which characteristics of the new female immigrants might positively affect the decision by married men to look for alternatives to their wives. Alternative explanations of these correlations might concern the socioeconomic values of the host regions, which might be favorable both to marital dissolutions and to female immigrations. Due to scarce data availability, the analysis could not be pursued further, yet our study may stimulate debate among scholars. Further research could be carried out, perhaps based on specific surveys, to shed new light on this key topic. The die is cast!

Conclusions

Men look for women with clear signs of fertility, genetic quality, and fecundity (Buss & Schmitt, 1993; Furnham & Swami, 2007; Kozee, Tylka, Augustus-Horvath, & Denchik, 2007; Meltzer & McNulty, 2014; Quinn, 2002; Zelazniewicz & Pawlowski, 2011), tending to prioritize the pursuit of short-term mating partners (Wlodarski & Dunbar, 2014). Obviously, a higher sex ratio in society increases the opportunities for men to meet women able to satisfy their need for short-term mating. However, not all women are the same and it would be wrong to simply postulate that a more favorable sex ratio boosts the probabilities of realizing this sexuality. Females have specific social and cultural characteristics or sexual attitudes, which might be able to increase (or decrease) said probabilities in different ways, as highlighted by this work.