Is Marriage More Than Cohabitation? Well-Being Differences in 30 European Countries

Article excerpt

The study aims to assess, first, whether there is a gap in well-being between unmarried cohabitants and the married, second, if selection factors can explain this so-called cohabitation gap, and third, if the size of the cohabitation gap differs across countries and how this can be explained. We use pooled data from young adults (18-44) in 3 rounds of the European Social Survey (N [approximate] 31,500). Multilevel regression analyses show that there is a moderate cohabitation gap that can be partly explained with the selection factors material resources and religiosity. Country differences were clear and could partly be explained with the level of institutionalization: In countries where cohabitation is more accepted and more prevalent, the cohabitation gap is smaller.

Key Words: cohabitation, cross-national, life events and transitions, marriage, multilevel models, well-being.

That marriage enhances well-being has consistently been found in research (Coombs, 1991; Waite, 2000; Wilson & Oswald, 2005). In this research, the married are typically compared to the unmarried. Nowadays, however,the unmarried group includes many cohabitants with a marriage-like living arrangement, and the married group includes many ex-cohabitants.

Previous studies do not clearly show whether being in unmarried cohabitation (shortened to cohabitation) has the same well-being benefits as being married: Some studies found a large difference between the union types, with spouses being happier than cohabitants (Brown, 2000; Horwitz & White, 1 998), but in other studies no difference was found (Musick & Bumpass, 2006; Ross, 1995). We call the difference in wellbeing between cohabiting and married groups the cohabitation gap. Our first aim was to examine whether there is a cohabitation gap in Europe. Please note that we refer to a subjective general evaluation of well-being.

Our second aim was to examine which and to what extent selection factors explain the cohabitation gap. Little is known about the factors that explain differences between cohabitation and marriage, if they exist. One possibility is that marriage has certain characteristics that living together unmarried does not have and that enhance well-being. The explanatory factors would then stem from the union type itself. Another possibility is that people who choose cohabitation differ from those who choose marriage. If a certain individual characteristic is negatively related to well-being and among cohabitants there is a larger proportion with this characteristic than among the married, the average well-being level of cohabitants will be lower. This is our selection hypothesis. If selection factors explain the cohabitation gap, the gap would be spurious.

Third, it is likely that marriage and cohabitation have different consequences for well-being in different countries because cohabitation is not institutionalized to the same degree in all countries (Liefbroer & Dourleijn, 2006; Wagner & Weiss, 2006). Research on cross-national differences is scarce, however. Our third aim was therefore to examine and explain country differences in the size of the cohabitation gap. We expected that the gap is smaller, or even nonexistent, in countries where cohabitation is more common and more accepted. This is our institutionalization hypothesis.

To test the hypotheses, data were used from young adults in three rounds of the European Social Survey. As is discussed later, this is a recent large-scale survey in 30 European countries where cohabitation is prevalent and accepted in varying degrees.


On the one hand it can be expected that the well-being benefits of marriage and cohabitation are similar because popular explanations for the higher well-being of the married may be true for cohabitants as well. These explanations are that married people have more material resources and an improved health and that they receive more social provisions than the unmarried (Stack & Eshleman, 1998; Waite & Gallagher, 2000; Wilson & Oswald, 2005). …