Geist Three worlds of marriage effects ? Gendered marriage earning differences in the United States , Germany , and Sweden

Being married is associated with many advantages. However, we do not know enough about the actual impact of entering marriage on individuals’ earnings, especially for women. In this paper, I examine the immediate and the short-term impact of marriage on men’s and women’s earnings in the United States, Germany, and Sweden. Studying the impact of marriage on earnings in three distinct socio-political settings provides insights into the context dependency of the link between marriage and earnings. Fixed effects models show that marriage transitions are not associated with women’s earnings in the United States and Sweden. For German women, I find an earnings penalty for marriage. Once I adjust for selection into employment, I find that employed German women with low employment propensities may experience instantaneous earnings boosts when they enter marriage, but that among women who are more firmly attached to the labor market, there is a short-term marriage penalty. For men in all three countries, I find no effect of marital transitions once employment likelihood is taken into account.

Schlagwörter: Ehe, Einkommen, Deutschland, Schweden, Vereinigte Staaten von Amerika, Selektion, Frauen, Männer Marriage is associated with many economic advantages as well as better physical and mental health and improved social networks for both men and women (Ross/Mirowsky/ Goldsteen 1990; Sigle-Rushton/McLanahan 2002; Simon 2002; L. J. Waite 1995; L. J. Waite/Gallagher 2000).For men, being married is associated with higher wages in a broad range of countries, but the evidence for women is less clear (Geist, unpublished document).
These results provide an important snapshot of existing marriage earnings differences, yet they do not tell us about the impact of entering marriage on individuals' earnings.In this paper, I examine the immediate and short-term impact of marriage on men's and women's earnings in the United States, Germany, and Sweden.A longitudinal and comparative approach has two advantages.First, the longitudinal nature of the data allows me to account for unmeasured individual characteristics that may be associated with both the chances of entering marriage and earnings.Second, examining the impact of marriage on earnings for men and women in comparative perspective provides me with three distinct institutional settings in which marriage earnings differences can manifest themselves, and as a result, this study provides important insights into the possible context variability of the short-term impact of marriage on individuals' earnings.
In this paper, I focus on the immediate and short-term impact of marriage entry on men's and women's individual earnings in the United States, Germany, and Sweden.My study addresses both selection into employment and, to a lesser extent, into marriage, and is therefore uniquely suited to illustrate the net effect of marriage for women and men in the three different contexts.Although even longitudinal data cannot fully assess the causal nature of the relationship of the association between marriage and earnings I use the phrase "marriage effect" to refer to the association between a change in marital status and a change in earnings.
Specifically, this study addresses three broad research questions: 1. What are the short-term effects of marriage on earnings for men and women's earnings?2. To what extent is the impact of marriage on earnings caused by underlying differences in individual and labor market characteristics between those who enter and do not enter marriage?3. Is there evidence that the impact of marriage on earnings varies across contexts for men and women?
In the next sections, I provide a brief description of the literature on marriage earnings differences, with a specific emphasis on issues of selectivity in employment and into marriage.I also discuss the importance of using a comparative perspective for situating the association between marriage and earnings in a broader context.I then describe the data and analytic strategy, before turning to results and conclusions.

Marriage entry and earnings
There are numerous economic and social advantages associated with marriage that have been discussed elsewhere (Ribar 2004; Linda J. Waite/Gallagher 2000), but not enough is known about the short-term economic impact of entering marriage for men and women in comparative perspective.Marriage, for heterosexual couples, is a highly gendered institution that comes with different sets of expectation for husbands and wives.The role of husband is closely associated with that of a breadwinner, whereas being a wife, despite relatively high levels of labor force participation among married women cross-nationally, is not conditioned to labor market success in the same sense.
Research on men has identified a marriage earnings advantage and a specific earnings benefit of marriage entry (Bellas 1992; Cohen/Haberfeld 1991; Kaufman/Uhlenberg 2000; Nakosteen/Zimmer 1997), but recent work by Killewald and Lundberg (2017) has cast doubt on the causality of this association.For women, Light (2004) finds that while household income increases when US women start to cohabit or enter marriage, women's earnings suffer.Korenman and Neumark (1992), however, find no direct effect of entering marriage on women's wages.Some studies examine marriage earnings premiums in longitudinal samples outside the United States (e.g, Coppin 2000), but to my knowledge, no studies include an examination of marriage earnings gaps for both men and women in multiple countries.

Economic Selection
Some studies show that men's employment and favorable economic circumstances increase the chance of marriage for men (D. T. Lichter/Kephart/McLaughlin/Landry, 1992; Daniel T. Lichter/Landry 1991; Oppenheimer 1994; Smock/Manning 1997; Speare/Goldscheider 1987).The observation that men with higher earnings potential are also more likely to enter marriage than those with lower earnings potential has been offered as a central explanation of the marriage bonus for men (Blackburm/Korenman 1994).This implies that observed wage differences between married and unmarried men are explained by the fact that highly productive men with increased earnings potential have better chances of both entering marriage and remaining married, thus raising the average earnings level among those who are married (Nakosteen/Zimmer 1997).Killewald and Lundberg (2017) argue that marriage and rising earnings are merely co-occurring during the transition to the life course for men in the United States.
Recent empirical evidence implies that increased earnings power and economic independence may also increase women's chances of entering marriage under certain circumstances (Ono 2003; Sassler/Schoen 1999).However, the increased chance of entering marriage may be partially offset by the fact that women's financial autonomy reduces the pressure to get married.
Studies have addressed the issue that the marriage earnings gap may be due to selection by using longitudinal data.This allows the researcher to examine the effect of marriage entry within an individual's earnings trajectory, rather than simply comparing earnings across two possibly very different groups of individuals.However, examining the ef-