Artículo de revista: Esteve, Albert; Elwert, Annika; Batyra, Ewa (2023). “Gender Asymmetries in Cross-National Couples”. Population and Development Review, 49 (2): 379- 396 (ISSN: 0098-7921). https://doi.org/10.1111/padr.12565 Gender Asymmetries in Cross-National Couples ALBERT ESTEVE, ANNIKA ELWERT AND EWA BATYRA We use census microdata for Spain and the United States to examine gender dif- ferences in cross-national heterosexual couples (marriages and cohabitating unions involving natives and immigrant spouses). We examine whether native men and native women in cross-national couples exhibit similar patterns regarding the country of birth, age, and educational attainment of their international part- ners. Countries are ranked by their level of the human development index (HDI). In both Spain and the United States, native men are slightly overrepresented in cross-national couples. Native men are more likely to be married to individu- als from low HDI countries than native women. On the contrary, native women are predominantly married to men born in countries with similar or higher lev- els of HDI. Unions of native men are characterized by large spousal age dif- ferences, which increase with the HDI gap; this pattern reverses among native women. These gender asymmetries do not extend to education level, with cross- national couples, overall, involving highly educated individuals. These results point to the endurance of certain features of traditional marriage patterns in cross- national couples, but the patterns differ starkly between native men and native women. Introduction An increasing number of people in high-income countries marry or cohabit with partners who were born abroad (Niedomysl, Östh, and van Ham 2010; Lanzieri 2012; González-Ferrer et al. 2018; Ryabov and Zhang 2019). We refer to these unions as cross-national couples. According to data from the 2010 round of population censuses, in Spain and the United States, the two countries we examine in this research note, roughly 10 percent of all young couples (aged 30–39) involve a native and a foreign-born partner. Albert Esteve, Universitat Autònoma de Barcelona, Campus UAB, Bellaterra, 08193, Spain. Email: aesteve@ced.uab.es. Annika Elwert, Centre for Economic Demography, Lund University, Scheelevägen 15B, 223 63 Lund, Sweden. Ewa Batyra, Centre for Demographic Studies Edifici E2, Campus UAB, Bellaterra, 08193, Spain. P O P U L AT I O N A N D D E V E L O P M E N T R E V I E W 4 9 ( 2 ) : 3 7 9 – 3 9 6 ( J U N E 2 0 2 3 ) 379 © 2023 The Authors. Population and Development Review published by Wiley Periodicals LLC on behalf of Population Council. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 380 G E N D E R A S Y M M E T R I E S I N C R O S S -N A T I O N A L C O U P L E S However, the macro-level characteristics of these couples have been hardly researched. How many cross-national couples are there? Are native men more likely to be involved in cross-national couples than native women? Do native men and women form couples with migrants of the same origins? If not, is there any gender pattern to these differences? How do gender gaps in age and education in cross-national couples compare with native couples? Do these gaps vary across different dyads of cross-national couples? For the reasons we develop in the background section, we expect that the answers to these questions will disclose gender asymmetries that are unknown to social scientists. These asymmetries will emerge from the interactions be- tween native and foreign-born populations in the marriage markets. Even though available census data lack many key details that would be needed to formulate refined hypotheses on such interactions (e.g., the time and place of couple formation), censuses allow picturing macro-level patterns and associations with more detail than any standard survey could. To date, most research on cross-national couples in high-income West- ern countries has focused on the migrant’s perspective and has been mainly framed under the assimilation and segmented assimilation frameworks. Within these frameworks, marriage across ethnic boundaries or origins is considered solid evidence of assimilation (Gordon 1964; Qian, Glick, and Batson 2012; Dupont et al. 2017; Kulu and Hannemann 2018; Uggla and Wilson 2021). Assimilation theories first appeared in the United States, but, in modified forms, spread later throughout other Western societies (Portes and Zhou 1993). In this research note, we flip the usual perspective. We shift the attention from the migrant to the previously understudied native’s perspective. We examine gender differences among native populations in cross-national couples to stimulate further research on the gender dynamics that shape their interaction with foreign-born populations in increasingly global marriage markets. Our study highlights the need to move beyond the integration/assimilation framework in the study of cross-national unions. In order to fully understand those unions, we focus on the native majority population and their partner preferences. Background Demographers, economists, and sociologists have traditionally examined the characteristics of couples regarding age and education to learn about the social and gender dynamics that drive the selection of spouses in het- erosexual marriage markets (usually studied under the term assortative mat- ing; Oppenheimer 1988; Becker 1993; Mare 1991). Marriage markets are defined as the symbolic place in which males and females look for long- term partners (Akers 1967). The various gender gaps observed within cou- ples (e.g., in age and education) suggest that the marital search process and the context of opportunities might be different for men and women. For 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License ALBERT ESTEVE / ANNIKA ELWERT / EWA BATYRA 381 instance, across societies, women marry younger than men and, therefore, are on average younger than their spouses. Traditionally, women have had also lower levels of educational attainment and lower levels of labor market engagement than their partners. These gaps were fully consistent with the prevalence of the male breadwinner model (Becker 1993). Changes in women’s roles have, however, eroded the male bread- winner model in several directions (Oppenheimer 1988). The expansion of women’s educational attainment and labor force participation has closed or even reversed, traditional gender gaps in age, education, and labor force participation (Esteve, Cortina, and Cabré 2009; Dupont et al. 2017; Van de Putte et al. 2009; Esteve, García-Román, and Permanyer 2012; Esteve et al. 2016). There is widespread agreement in interpreting these trends as signs of growing gender equality in the marriage market. But there is also consensus in seeing that the gender revolution has not been completed (England 2010; Esping-Andersen and Billari 2015). For example, women’s educational ad- vantage in couples does not necessarily translate into the same economic advantage as when men hold the educational advantage within the couple (van Bavel, Schwartz, and Esteve 2018). Such an asymmetry reflects signif- icant gender gaps in occupation and income. In short, changes in women’s roles in society might not have eroded at the same pace as that of gender ideologies. Marriage markets and couples are not immune to these conflict- ing tensions, as widely shown in the literature (e.g., Oppenheimer 1988; Goldscheider, Bernhardt, and Lappegård 2015). However, there is hardly any research that links these trends to the study of cross-national couples involving natives and foreign-born migrants. Recent studies on cross-national couples shed new light on gender asymmetries in the marriage market, which previous studies had not ad- dressed. Balistreri, Joyner, and Kao (2017) show that age gaps between partners are larger in unions where (female) noncitizens marry U.S. cit- izens or where (female) noncitizens marry U.S. citizens before migrating to the United States. The authors interpret this finding as evidence of sta- tus exchange, where migrants exchange their youth for citizenship. Similar patterns have been found in Sweden by Elwert (2020). Her findings indicate that immigrant women and men become more attractive marriage partners if they are considerably younger than their native spouse. The patterns are stronger for immigrants from certain non-Western regions of origin. Simi- larly, Guetto and Azzolini (2015) find for Italy that cross-national couples are more likely when less-educated older native men marry better-educated younger immigrant women, especially when the latter originate from non- Western countries. Serret Sanahuja, Esteve, and López Gay (2013) and González-Ferrer et al. (2018) also found for Spain notorious gender differ- ences among natives regarding the country of origin and the characteristics of the international partners with whom they marry. Gender differences in the characteristics of cross-national unions are likely related to ethnic 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 382 G E N D E R A S Y M M E T R I E S I N C R O S S -N A T I O N A L C O U P L E S and racial dating preferences. Such preferences have been found in Europe (Potârcă and Mills 2015) as in the United States. (Hwang 2013), and there are gendered patterns therein, for example, white men have preferences for Asian women and white women have preferences for black men (Feli- ciano, Robnett, and Komaie 2009; Robnett and Feliciano 2011). Braack and Milewski (2019) show that natives in cross-national unions are more open to sexual liberalization and gender equality. Spurred on by the evidence on gender asymmetries in the marriage market, we examine cross-national couples in two high-income countries that have sufficiently detailed micro- data to count on individual countries: the United States and Spain. They represent old—the US—and new—Spain—immigration countries (Cortina, Esteve, and Domingo 2008). In both countries, cross-national couples rep- resent more than 10 percent of the unions in which women and men of age 30–39 are involved. With this paper, we expect to unveil unknown gender asymmetries in marriage markets. We explore gender asymmetries in three main dimen- sions: (i) order of magnitude: are native men more likely to be involved in cross-national couples than native women?; (ii) country of origin: do native men and women form unions with foreign-born partners of the same ori- gin?; (iii) and gender differences in age and education. For the order magni- tude, we examine the distribution of couples in crude numbers. For the country of origin, we classify cross-national couples based on the coun- try of birth of the international partner and examine if native men are more involved than native women in each dyad. To systematize the gen- der patterns that might exist in relation to these unions, we classify coun- tries based on their level in the human development index (HDI). For each dyad, we compute the HDI gap between the native country (i.e., the United States or Spain) and the country of birth of the international part- ner. We test if native men and women are equally likely to be involved with international partners born in countries that represent all the HDI spec- trum, or, on the contrary, they show different patterns. HDI is a widely used indicator to rank countries in their health, human, and economic de- velopment. It can be used to rank countries on a traditional scale, as it is strongly, although not perfectly, correlated with other indicators such as the gender inequality index (GII), to which we will refer during this research note. Finally, we examine gender differences in characteristics of natives and their immigrant spouses according to individuals’ age and education. The studies mentioned above on cross-national couples support the idea that the development level of the immigrant partner’s country of birth is a factor that contributes to inequalities in the marriage market. Our guid- ing hypothesis is that, if persistent, gender inequalities will be manifested in very distinct marriage patterns between native men and native women, with native men partnering with immigrant women from less developed 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License ALBERT ESTEVE / ANNIKA ELWERT / EWA BATYRA 383 backgrounds and their unions exhibiting more traditional patterns in terms of spousal age than cross-national couples of native women. Data and methods We use recent samples from the International Integrated Public Use Mi- crodata Series (IPUMS International; Minnesota Population Center 2020) for the United States (2010) and Spain (2011). The choice of restricting our sample to these two countries is based on the availability of detailed country classifications for the foreign-born spouse (i.e., no continental groupings). Sample sizes are 1 percent for the United States and 10 percent for Spain, which are large enough to explore dyads of natives and international mi- grants. Censuses provide information about key characteristics of the cou- ples, such as partner’s age and education, which allow us to capture the within-couple dynamics. To minimize the potential biases that union dis- solution or differences in age at union formation may cause, we focus on young couples. We select all heterosexual couples in which at least one partner is of age 30–39. Alternative, wider range age classifications (e.g., 25–49) yielded similar results. We combine married and cohabiting cou- ples. Given the large racial disparities in union formation between whites and blacks in the United States and the large presence of migrant descen- dants, we restrict the native population to the non-Hispanic whites born in the United States. Similarly for Spain, we restrict the native population to people born in Spain whose parents are also born in Spain. These re- strictions are intended to minimize the impact that cross-national couples between international migrants and descendants of migrants of the same origins might have in our analyses (ex. Mexicans in the United States or Moroccans in Spain). Table 1 shows the sample sizes and key variables se- lected for the analysis, for couples by partners’ nativity status. Results have been weighted using analytic weights. For every partner, we select their age and educational attainment. Age is used to compute the age gap between spouses (the age of a man minus the age of a woman). Regarding education, we dichotomize educational attain- ment according to whether a person has tertiary education or not. Finally, we assign the HDI value (World Bank 2022) for each partner’s country of birth. All partners from the same country of birth share the same code re- gardless of any other individual characteristics. Native partners are given their own HDI values. We compute the gap in HDI between partners. This gap is positive if a spouse of a native comes from a less developed country (HDI native − HDI foreign > zero) and negative if they come from a more developed country (HDI native − HDI foreign < zero). The analysis is performed at the country- and macro-level; therefore, gaps reflect the countries’ averages. For graphical purposes only, we use color to identify the continent to which each country belongs (combining 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 384 G E N D E R A S Y M M E T R I E S I N C R O S S -N A T I O N A L C O U P L E S TABLE 1 Descriptive statistics of the sample and key variables selected for the analysis, for couples by partners’ nativity status, Spain and the United States Native Native Native man–native man–foreign woman–foreign woman woman man United States Frequency (n) 87,272 4,734 3,788 Mean age-gap 2.33 3.47 2.45 Proportion with higher education, men 0.35 0.50 0.42 Proportion with higher education, women 0.42 0.52 0.49 Spain Frequency (n) 224,782 16,107 12,198 Mean age-gap 2.28 4.94 0.99 Proportion with higher education, men 0.15 0.19 0.22 Proportion with higher education, women 0.21 0.25 0.26 Europe and North America). Our analysis proceeds as follows. First, sepa- rately for each of the two countries, we calculate a measure of gender asym- metry in cross-national couples between the natives and their foreign spouses. We compute this measure as a proportion of all unions involving a native individual and a foreign spouse from a given country, in which a native is a man. The higher this proportion, the more frequent the unions of na- tives with foreigners from a particular origin are among native men than among native women. When the gender gap is 0.5, native men and native women are equally likely to partner with immigrants from a given coun- try, pointing to gender equality in cross-national couples. We calculate such measures for all native-foreigner country combinations available, excluding those for which there were less than 50 unions in the sample. Using the gen- der asymmetry indicator, we explore whether gender gaps in cross-national couples vary depending on the HDI gap between the spouses’ countries of origin. This approach allows us to explore whether gender gaps in cross- national couples follow any high versus low development patterns, which ultimately permits capturing whether there is any evidence of asymmetry in native women’s and men’s preferences regarding the characteristics of their partners. Second, we explore whether gender asymmetries in cross-national couples across developmental gaps follow any specific patterns in terms of spousal age differences and educational level. In order to explore whether and how within-union dynamics differ between native women and native men, we conduct analyses by gender. 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License ALBERT ESTEVE / ANNIKA ELWERT / EWA BATYRA 385 Results Figure 1 shows the relationship between gender asymmetry in cross- national couples (x-axis) and the HDI gap (y-axis), separately for Spain and the United States. Each plot includes a fitted line (second-degree polyno- mial, weighted by the size of specific country-origin combinations) with 95 percent confidence intervals. The size of each of the points reflects the num- ber of unions between natives and foreigners from a given country. Colors distinguish four geographical regions of foreign spouses: Africa, Asia and Oceania, Latin America and the Caribbean, and Europe and North Amer- ica. Although international partners come from different origins depending on the country, the results for the two nations consistently suggest a similar gendered pattern in cross-national couples. First, cross-national couples overall are more common among native men than native women. Around 57 and 56 percent of all unions between natives and foreigners in Spain and the United States, respectively, are among native men. In the figures, it can be seen that, for most countries of origin, unions of native individuals and foreigners are primarily formed by native men (gender asymmetry above 0.5 in the case of most countries). For example, in Spain, around 80 percent of cross-national couples involv- ing Colombian or Brazilian individuals are among Spanish men, and only around 20 percent among Spanish women (gender asymmetry of ∼0.80). Second, cross-national couples are highly gendered, with native males and native females partnering with immigrants from countries at a very different level of development. When focusing on countries from which a large num- ber of immigrants are in unions with natives (large points concentrated in the middle and right part of the plot), it can be seen that, in both countries, gender asymmetry in cross-national couples increases with the growing HDI gap. This means that unions involving a native and a foreigner who comes from a less developed country (HDI gap above zero) are formed predomi- nantly by native men, as opposed to native women. For example, see the case of the Philippines: in both countries, more than 80 percent of unions with individuals from that nation are among native men and only 20 per- cent among native women. A similar pattern is visible for other countries of origin such as, for example, Bolivia in Spain or Colombia in the United States. Thus, the larger the differences in the development level between the origin countries of partners, the more the unions tend to involve native men, as compared to native women. Conversely, women tend to be overrepresented in unions where a spouse is from a country with a similar or a higher level of development (HDI gap around or below zero). This pattern is clearly visible for unions of natives with immigrants from Western Europe such as Belgium, Germany, or the United Kingdom. For example, 60 percent of unions with Belgians in Spain involve Spanish women and only 40% involve Spanish men. Overall, Figure 1 shows that for the most frequent types of cross-national couples, FIGURE 1 Relationship between gender asymmetry and HDI gap, Spain and the United States NOTE: The development (HDI) gap is positive if a spouse of a native person comes from a less developed country (HDI native − HDI foreign > 0) and negative if they come from a more developed country (HDI native − HDI foreign < 0). The vertical line at 0.5 indicates gender symmetry in cross-national marriages. Second-degree polynomial fit, weighted by the size of country-origin combinations. AD Andorra; AR Argentina; AT Austria; AU Australia; BE Belgium; BG Bulgaria; BN Taiwan; BO Bolivia; BR Brazil; CA Canada; CH Switzerland; CL Chile; CN China; CO Colombia; CU Cuba; CZ Czech Republic; DE Germany; DK Denmark; DO Dominican Republic; DZ Algeria; EC Ecuador; ES Spain; FR France; GB United Kingdom; GQ Equatorial Guinea; GR Greece; GT Guatemala; GU Guam; HN Honduras; IE Ireland; IL Israel; IN India; IR Iran; IT Italy; JM Jamaica; JP Japan; KR Korea; LA Laos; LB Lebanon; LT Lithuania; MA Morocco; MD Moldova; MX Mexico; NG Nigeria; NI Nicaragua; NL Netherlands; PA Panama; PE Peru; PH Philippines; PK Pakistan; PL Poland; PR Puerto Rico; PT Portugal; PY Paraguay; RO Romania; RU Russia; SE Sweden; SN Senegal; SV El Salvador; TH Thailand; TR Turkey; UA Ukraine; US United States; UY Uruguay; VE Venezuela; VN Vietnam; ZA South Africa. 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License ALBERT ESTEVE / ANNIKA ELWERT / EWA BATYRA 387 the foreign partners of men are more likely to come from less developed nations than foreign partners of native women. On the other hand, native women are more likely to form unions with foreigners from countries with a similar or higher level of development. There are some notable exceptions to these patterns. First, the de- scribed relationships hold for the most frequent dyads, and the pattern reverses when an immigrant partner is from a less represented country (e.g., see the case of African countries such as Senegal or Nigeria in Spain). We hypothesize that these distinct union formation patterns for these small migrant populations are related to white women’s racial preferences (see, e.g., Feliciano, Robnett, and Komaie 2009), as well as to the selectivity of migrants from these underrepresented destinations. Second, the relation- ship also reverses for unions with immigrants from Morocco in Spain and from Mexico in the United States. Despite the relatively large HDI gaps, unions with individuals from these countries are more common among native women than native men. Morocco and Mexico are neighboring countries of Spain and the United States, respectively, and historically have had strong migration ties, which could be behind the distinct union formation pattern compared to the one observed for origin countries from which migration is a more recent phenomenon. Third, despite a relatively small HDI gap, unions in Spain with immigrants from South Korea, as well as in Spain and the United States with immigrants from Japan, are gender imbalanced (gender gap > 0.5) and involve mainly native men. Despite being high-HDI countries, the union formation pattern with individuals from this nation is more similar to that of countries with lower HDI levels. The similarity in the described patterns between Spain and the United States is striking, given that they differ regarding the origins of the mi- grant groups who form unions frequently. For example, contrary to Spain, where cross-national unions with Latin Americans are particularly com- mon, in the United States, unions with immigrants from Asia are frequently observed. Despite these migrant origin differences between Spain and the United States, with the exceptions described above, the cross-national cou- ples follow a similar gendered pattern such that native men are more likely to form unions with individuals from countries with a lower level of de- velopment than native women. Finally, although the analysis encompasses fewer countries, similar conclusions are reached when, instead of HDI, a measure of gender inequality from the World Value Survey is used. This suggests that the HDI gaps are likely to reasonably well proxy differences in gender roles between countries (see Figure A1 in the Supporting Infor- mation). The subsequent figures—Figure 2 for Spain and Figure 3 for the United States—cast light on whether the cross-national couples are more or less symmetric when it comes to spousal age and education level. Figures 2 and 3 provide further evidence about gender (a)symmetries in cross-national couples, both within cross-national couples (according to 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 388 G E N D E R A S Y M M E T R I E S I N C R O S S -N A T I O N A L C O U P L E S FIGURE 2 Relationship between HDI gap and spousal age gap (left), proportion in higher education (right), Spain NOTE: The development (HDI) gap is positive if a spouse of a native person comes from a less developed country (HDI native − HDI foreign > 0) and negative if they come from a more developed country (HDI native − HDI foreign < 0). First-degree polynomial fit, weighted by the size of country-origin combinations. measures of age and education, defined on the x-axis) and between the pro- files of native men (top panels) and native women (bottom panels) (first- degree polynomial, weighted by the size of specific country-origin combi- nations). When it comes to the spousal age gap, there is a clear relationship with the HDI gap in both countries, which nonetheless differs between na- tive women and native men (left panels). In both Spain and the United States, native men in cross-national couples (top, left panels) tend to part- ner with younger women (spousal age gaps > zero) and these age gaps tend to be larger than the average spousal age gap (native–native age gap in a dotted, vertical line). Moreover, the difference in age between partners in unions between native men and women from countries at a similar stage of 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License ALBERT ESTEVE / ANNIKA ELWERT / EWA BATYRA 389 FIGURE 3 Relationship between HDI gap and spousal age gap (left), proportion in higher education (right), United States NOTE: The development gap is positive if a spouse of a native person comes from a less developed country (HDI native − HDI foreign > 0) and negative if they come from a more developed country (HDI native − HDI foreign < 0). First-degree polynomial fit, weighted by the size of country-origin combinations. development (e.g., France and Germany) is relatively small and close to the average age gap; this difference increases sharply with the HDI gap. These results highlight a pattern of gender asymmetry when it comes to age in cross-national couples across development gaps in both countries. Among native women in cross-national unions (bottom, left pan- els), for the largest origin groups (e.g., Germany, France, or the United Kingdom), the pattern of spousal age differences is similar to the pattern observed among native couples. Interestingly, the positive relationships be- tween spousal age gaps and the HDI gap observed among native men reverse in the case of native women. In both Spain and the United States, native women tend to partner with men of similar age (e.g., from Venezuela) and, in Spain, frequently with men younger than themselves (e.g., from 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 390 G E N D E R A S Y M M E T R I E S I N C R O S S -N A T I O N A L C O U P L E S Romania or Ecuador). Moreover, spousal age gaps tend to decrease and be- come negative with an increasing HDI gap, which means that many native women partner with men from less developed countries who are younger than themselves. In the United States, the pattern is less clear with unions with some nationalities exhibiting larger (e.g., Philippines) or smaller (e.g., Indonesia) than the average age gaps. Finally, there are some notable regional patterns common to both countries. For example, age gaps smaller than the average are clearly visible for the majority of Latin American (e.g., Mexico or Venezuela) and African countries (e.g., Morocco in Spain and South Africa in the United States); in Spain, there is a clear division between unions of native women with individuals from Western Europe (similar to the age gap among natives) and Eastern Europe (mainly negative age gap). Despite stark asymmetries in age between spouses in cross-national couples and between unions of native women and native men, Figures 2 and 3 show that the educational profiles are rather similar across genders (right panels). In Spain, both native men and native women who part- ner with individuals from countries at a similar or higher development level tend to be more educated than average (e.g., from Western European countries like France or Germany or Latin American countries like Ar- gentina). In a number of cases, this pattern reverses when a partner is from a less developed country. For example, natives who form unions with mi- grants from Romania or Morocco are average or less than average educated. While the patterns are quite mixed in Spain, in the United States a pattern is very consistent in that both native women and native men who form cross- national unions are similarly or more educated than an average native who partners with a native. This is except for native women in unions with Mex- icans and individuals from a few, mainly “small” Latin American countries, who tend to be less educated than average. Overall, although native women and men tend to form unions with individuals from very different origins, as shown in Figure 1, except for a number of cases explained above, there is a high degree of symmetry in their educational profiles in both countries and there is evidence that, overall, cross-national couples largely involve well-educated individuals. Overall, while the results pertaining to the country of origin of in- ternational partners and spousal age differences across the developmental gaps are similar between Spain and the United States, the pattern pertaining to education exhibits more heterogeneity between the two countries. The difference can also be seen when comparing unions of natives with individ- uals from particular origins that are present in both Spain and the United States. For example, among many, the case of Colombia shows how unions with individuals from that country exhibit similar patterns with respect to our measure of gender asymmetry and age gaps, but this is less so the case when it comes to the measures of education. 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License ALBERT ESTEVE / ANNIKA ELWERT / EWA BATYRA 391 Discussion In this paper, we have shown that men are more frequently present in cross- national couples and such involvement differs by the country of birth of the foreign spouse. More specifically, by plotting the country differences in HDI, we have found that the presence of men, compared to women, in cross-national couples increases with the HDI gap, showing that gender im- balances in the characteristics of cross-national unions are higher when the immigrant partner comes from a less-developed country. Given that indi- viduals from many non-EU and low-development countries face barriers to migrating to Western countries, marriage can be an opportunity for migra- tion (Charsley et al. 2020) not only in transnational communities but also in cross-national couples (Elwert 2020). If, for explanatory purposes, we use development as a proxy for tra- dition and level of gender inequality, our results suggest that the involve- ment of men in cross-national couples increases when the foreign partner comes from more “traditional” countries. For the majority of unions where the foreign-born spouse comes from a country with a level of development close to those of Spain and the United States, the distribution of men and women is close to equal; when the development gap reverses, the distribu- tion is in favor of women. Cross-national couples involving Japanese and South Koreans are the exception to this rule, as despite having similar lev- els of HDI, these couples involve mainly native men. Precisely, these are two of the countries in which HDI might not proxy well the level of gender (in)equality. Interestingly, in the analysis where the gender roles gap based on the World Values Survey measure is used instead of the HDI gap, Japan and South Korea fit well into the average pattern (see Figure A1 in the Sup- porting Information). Asian countries tend to be gender asymmetric and exhibit a more “traditional” pattern, which might also explain why native women are not involved with international partners from these countries. A second finding is that the gap in development is also correlated with the spousal age gap. Cross-national couples of native men with women from countries with a lower HDI typically involve women younger than aver- age and younger than their native spouse. Age differences widen as the gap in development increases. Our results contest the idea that being in a cross-national union is associated with more gender-equal attitudes (Braack and Milewski 2019). Instead, our findings suggest that whether such atti- tudes translate into gender equality in spousal characteristics may highly depend on the origin country of the immigrant spouse and its level of de- velopment. The age difference patterns observed in native men–immigrant women cross-national couples replicate traditional marriage patterns with the woman being considerably younger than the man, which is not only a representation of male superiority (Bourdieu 2002) but potentially has consequences for women’s careers. 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 392 G E N D E R A S Y M M E T R I E S I N C R O S S -N A T I O N A L C O U P L E S The association between the spousal age difference and the HDI gap, however, differs by gender. Among native women, the age gap decreases with a growing developmental gap. The case of Spanish women is very clear in this respect. Women who partner with men from less developed countries are either of the same age or even older than them, thus revers- ing the traditional age gap. Results suggest that these women (albeit smaller in number than men) might be showing reversed age-hypergamic patterns. A similar pattern is observed in the United States, although it is much less clear. Women from the United States who are in unions with men from more traditional countries are closer in age than when they are in unions with natives, but this is not the case for all origins. Native women’s cross- national union patterns thus challenge traditional marriage patterns with regard to the male–female age gap. Regarding education, men and women involved in cross-national couples are in general more educated than their counterparts partnering with natives. This is clearly the case for almost all types of unions in the United States, but it is somewhat less clear in Spain. These findings suggest that there is no gender difference among the native population regarding selectivity into these unions with respect to education. Overall, it is this im- portant to highlight that, for some origins, despite being from countries with lower development levels, immigrant spouses are highly educated, which suggests that these individuals come from higher socioeconomic strata of the origin country (e.g., Mexico, Venezuela, or Argentina). This means that foreign partners of natives might be positively selected when it comes to their educational profiles. Overall, the study of cross-national unions, which now represent around or more than 10 percent of the cases for individuals aged 30–39 and are likely to raise in the future, is opening a new opportunity to study the endurance of traditional arrangements and hypergamic patterns. Our results on the differences between native men and native women when it comes to the HDI of the country of origin of the foreign spouse, as well as spousal age gaps, are consistent with the idea that traditional gender norms in marriage formation might have not completely disappeared in high-income countries. At the same time, some native women might be challenging this pattern by forming unions with younger men from coun- tries with a lower development level. We thus posit that the study of cross- national couples has the potential to capture and reflect the prevailing gen- der ideologies equally well, or possibly even better, than the study of gender differences in unions among natives. The results of this study invite us to conceptualize the marriage market from a global perspective and test in fu- ture research if the partner exchange across societies at different levels of development is gendered. Our results support this idea. Forming unions across borders could be seen as a strategy to bridge the gaps produced by internal mismatches due to changes in gender roles and expectations. 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License ALBERT ESTEVE / ANNIKA ELWERT / EWA BATYRA 393 This study does not come without limitations. First and foremost, our study is only on two immigration countries, Spain and the United States. While our study is limited by data availability, we encourage fur- ther research on other contexts to study gendered patterns of cross-national unions. We have focused on prevailing couples with different durations. Some of the cross-national couples formed around the same time as the ones observed in this research might have dissolved by the census date. We have not distinguished between married and cohabiting couples, which might display different patterns. International migrants from certain origins might be more prone to marry than cohabit if acquiring citizenship of legal residence is at stake. However, this should, in principle, not imply any gen- der difference. We also do not know when and where these couples were formed. As the world becomes increasingly globalized and web dating is widening the partner search, some couples might have met online. Because of partners increasingly meeting online, it should also not be assumed that the gender differences we find are purely reflecting the gender composi- tion of migrants, as some of the unions could have been formed abroad. Despite these limitations, our evidence suggests gendered interactions between native populations and international partners, which were rela- tively unknown to social scientists. Acknowledgments Albert Esteve acknowledges funding from PID2021-124267OB-I00 and HE-ERC-2021-AdG-GA No 101052787-CORESIDENCE. Annika Elwert ac- knowledges funding from the European Union’s Horizon 2020 research and innovation programme (grant agreement No 730998, InGRID-2 – Integrat- ing Research Infrastructure for European expertise on Inclusive Growth from data to policy) and the Jan Wallander and Tom Hedelius Foun- dation (Jan Wallander och Tom Hedelius Stiftelse samt Tore Browaldhs Stiftelse W18-0009). Ewa Batyra acknowledges funding from the Marie Sklodowska-Curie Action Individual Fellowship from the European Com- mission (101060237-DISPARFAM). We are grateful to Anna Turu for her help in producing the figures. Data Availability Statement The IPUMS International data (International Integrated Public Use Micro- data Series) is available from the IPUMS International homepage main- tained by the Minnesota Population Center (Minnesota Population Cen- ter 2020). The original producers of the data are the Bureau of the Census (United States) and the National Institute of Statistics (Spain). 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 394 G E N D E R A S Y M M E T R I E S I N C R O S S -N A T I O N A L C O U P L E S Note 1 We have considered alternative mea- based on the question “Men make better po- sures that could potentially be better suited litical leaders than women” is considered a to capture the degree of gender inequalities good proxy of gender roles but its disadvan- between countries. We decided not to use tage is that it is available for an even lower the United Nations GII as it is strongly corre- number of countries. Nonetheless, we ex- lated with HDI with the disadvantage of be- ploited that alternative measure and it points ing available for fewer countries than HDI. to similar conclusions as those when HDI is The indicator from the World Value Survey used (see the Supporting Information). References Akers, Donald S. 1967. “On Measuring the Marriage Squeeze.” Demography 4(2): 907–924. https: //doi.org/10.2307/2060328. Balistreri, Kelly Stamper, Kara Joyner, and Grace Kao. 2017. “Trading Youth for Citizenship? The Spousal Age Gap in Cross-Border Marriages.” Population and Development Review 43(3): 443– 466. https://doi.org/10.1111/padr.12072. Becker, Gary S. 1993. A Treatise on the Family. Enlarged ed. London: Harvard University Press. Bourdieu, Pierre. 2002. Masculine Domination. Stanford, CA: Stanford University Press. Braack, Mirko K., and Nadja Milewski. 2019. “A Different Perspective on Exogamy: Are Non- Migrant Partners in Mixed Unions More Liberal in Their Attitudes toward Gender, Family, and Religion than Other Natives?” Journal of Family Research 31(3): 361–386. https://doi.org/ 10.3224/zff.v31i3.06. Charsley, Katharine, Marta Bolognani, Evelyn Ersanilli, and Sarah Spencer. 2020. “British South Asian Transnational Marriage.” In Marriage Migration and Integration, edited by Katharine Charsley, Marta Bolognani, Evelyn Ersanilli, and Sarah Spencer, 83–121. Cham: Springer International Publishing. Cortina, Clara, Albert Esteve, and Andreu Domingo. 2008. “Marriage Patterns of the Foreign-Born Population in a New Country of Immigration: The Case of Spain.” International Migration Re- view 42(4): 877–902. http://www.jstor.org/stable/27645283 Dupont, Emilien, Amelie van Pottelberge, Bart Van de Putte, John Lievens, and Frank Caestecker. 2017. “Partner Choices in Long Established Migrant Communities in Belgium.” Historical Life Course Studies 4: 20–40. Elwert, Annika. 2020. “Opposites Attract: Assortative Mating and Immigrant–Native Intermarriage in Contemporary Sweden.” European Journal of Population 36(4): 675–709. https://doi.org/10. 1007/s10680-019-09546-9 England, Paula. 2010. “The Gender Revolution: Uneven and Stalled.” Gender and Society 24(2): 149– 166. http://www.jstor.org/stable/27809263 Esping-Andersen, Gøsta, and Francesco C. Billari. 2015. “Re-Theorizing Family Demographics.” Population and Development Review 41(1): 1–31. https://doi.org/10.1111/j.1728-4457.2015. 00024.x Esteve, Albert, Clara Cortina, and Anna Cabré. 2009. “Long Term Trends in Marital Age Homogamy Patterns: Spain, 1922–2006.” Population 64(1): 173–202. https://doi.org/10.3917/popu.901. 0183 Esteve, Albert, Joan García-Román, and Iñaki Permanyer. 2012. “The Gender-Gap Reversal in Edu- cation and Its Effect on Union Formation: The End of Hypergamy?” Population and Development Review 38(3): 535–546. http://www.jstor.org/stable/41857404 Esteve, Albert, Christine R. Schwartz, Jan van Bavel, Iñaki Permanyer, Martin Klesment, and Joan García-Román. 2016. “The End of Hypergamy: Global Trends and Implications.” Population and Development Review 42(4): 615–625. https://doi.org/10.1111/padr.12012 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License ALBERT ESTEVE / ANNIKA ELWERT / EWA BATYRA 395 Feliciano, Cynthia, Belinda Robnett, and Golnaz Komaie. 2009. “Gendered Racial Exclusion Among White Internet Daters.” Social Science Research 38(1): 39–54. https://doi.org/10.1016/ j.ssresearch.2008.09.004 Goldscheider, Frances, Eva Bernhardt, and Trude Lappegård. 2015. “The Gender Revolution: A Framework for Understanding Changing Family and Demographic Behavior.” Popula- tion and Development Review 41(2): 207–239. https://doi.org/10.1111/j.1728-4457.2015.000 45.x González-Ferrer, Amparo, Ognjen Obućina, Clara Cortina, and Teresa Castro-Martín. 2018. “Mixed Marriages between Immigrants and Natives in Spain: The Gendered Effect of Marriage Market Constraints.” Demographic Research 39: 1–32. https://doi.org/10.4054/DemRes.2018.39.1 Gordon, Milton Myron. 1964. Assimilation in American Life: The Role of Race, Religion, and National Origins.New York: Oxford University Press. Guetto, Raffaele, and Davide Azzolini. 2015. “An Empirical Study of Status Exchange Through Migrant/Native Marriages in Italy.” Journal of Ethnic and Migration Studies 41(13): 2149–2172. https://doi.org/10.1080/1369183X.2015.1037725 Hwang, Wei-Chin. 2013. “Who Are People Willing to Date? Ethnic and Gender Patterns in Online Dating.” Race and Social Problems 5(1): 28–40. https://doi.org/10.1007/s12552-012-9082-6 Kulu, Hill, and Tina Hannemann. 2018. “Mixed Marriage among Immigrants and Their Descendants in the United Kingdom: Analysis of Longitudinal Data with Missing Information.” Population Studies, 1–18. https://doi.org/10.1080/00324728.2018.1493136 Lanzieri, Giampaolo. 2012. “Mixed Marriages in Europe, 1990–2010.” In Cross-Border Marriage: Global Trends and Diversity, edited by Doo-Sub Kim, 81–121. Seoul. Accessed March 14, 2014. http://www.academia.edu/2565558/Mixed_Marriages_in_Europe_1990-2010. Korea: Insti- tute for Health and Social Affairs. Mare, Robert D. 1991. “Five Decades of Educational Assortative Mating.” American Sociological Review 56(1): 15–32. https://doi.org/10.2307/2095670 Minnesota Population Center. 2020. “Integrated Public Use Microdata Series.” IPUMS, no. Inter- national: Version 7.3. https://doi.org/10.18128/D020.V7.3 Niedomysl, Thomas, John Östh, and Maarten van Ham. 2010. “The Globalisation of Marriage Fields: The Swedish Case.” Journal of Ethnic and Migration Studies 36(7): 1119–1138. https://doi.org/ 10.1080/13691830903488184 Oppenheimer, Valerie Kincade. 1988. “A Theory of Marriage Timing.” American Journal of Sociology 94(3): 563–591. https://doi.org/10.1086/229030 Portes, Alejandro, and Min Zhou. 1993. “The New Second Generation: Segmented Assimilation and Its Variants.” Annals of the American Academy of Political and Social Science 530: 74–96. Potârcă, Gina, and Melinda Mills. 2015. “Racial Preferences in Online Dating across European Countries.” European Sociological Review 31(3): 326–341. https://doi.org/10.1093/esr/jcu093 Qian, Zhenchao, Jennifer E. Glick, and Christie D. Batson. 2012. “Crossing Boundaries: Nativity, Ethnicity, and Mate Selection.” Demography 49(2): 651–675. https://doi.org/10.1007/s13524- 012-0090-3 Robnett, Belinda, and Cynthia Feliciano. 2011. “Patterns of Racial-Ethnic Exclusion by Internet Daters.” Social Forces 89(3): 807–828. https://doi.org/10.1093/sf/89.3.807 Ryabov, Igor, and Yuanting Zhang. 2019. “Entry and Stability of Cross-National Marriages in the United States.” Journal of Family Issues 40(18): 2687–2706. https://doi.org/10.1177/ 0192513x19860186 Serret Sanahuja, Joana, Albert Esteve, and Antonio López Gay. 2013. “Revelaciones De Una Década De Matrimonios Entre Españoles Y Extranjeros, 1998-08.” Estadística española 55(181): 213– 230. https://dialnet.unirioja.es/servlet/articulo?codigo=4534812 Uggla, Caroline, and Ben Wilson. 2021. “Parental Age Gaps Among Immigrants and Their De- scendants: Adaptation Across Time and Generations?” Population Studies. In press. https: //doi.org/10.1080/00324728.2021.1998583 van Bavel, Jan, Christine R. Schwartz, and Albert Esteve. 2018. “The Reversal of the Gender Gap in Education and Its Consequences for Family Life.” Annual Review of Sociology 44(1): 341–360. https://doi.org/10.1146/annurev-soc-073117-041215 17284457, 2023, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/padr.12565 by Readcube (Labtiva Inc.), Wiley Online Library on [20/09/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 396 G E N D E R A S Y M M E T R I E S I N C R O S S -N A T I O N A L C O U P L E S Van de Putte, Bart, Frans vanPoppel, Sofie Vanassche, Maria Sanchez, Svetlana Jidkova, Mieke Eeckhaut, Michel Oris, and Koen Matthijs. 2009. “The Rise of Age Homogamy in 19th Cen- tury Western Europe.” Journal of Marriage and Family 71(5): 1234–1253. https://doi.org/10. 1111/j.1741-3737.2009.00666.x World Bank. 2022. The World Bank Data. Washington, DC: World Bank.