Cohabitation

Although sociologists treat cohabitation as a novel phenomenon, it is generally recognized that it has existed long enough to predate marriage. Until the mid-eighteenth century, the difference between marriage and cohabitation was unclear in many countries. In England, for example, the distinction between these unions remained fluid until Lord Hardwicke codified marriage in 1753. However, common-law marriage remained popular well after the passage of Hardwicke's Marriage Act. The lack of officials to oversee formal marriages and jurisdictional nuances kept marriage and cohabitation indistinct (Holland 1998; Seff 1995). Marriage really developed into the institution as we understand it today in the nineteenth century. During this period, marriage transformed from a more or less religious practice to one commonly formalized under civil law, and thus became the norm. But this should not be taken out of context. As Winifred Holland (1990) remarks, the family has always been a flexible organization and has responded to changing social circumstances in a dynamic manner, and marriage became the norm for specific historical reasons. Hence, it is incorrect to suggest that cohabitation is "deviant" behavior because this implies that marriage has always been the norm.

Although cohabitation has existed for a very long time, modern trends in cohabitation are qualitatively different from those of the past. The significance rests in the fact that cohabitation has increased in a context where conventional marriage is a clearly defined and dominant social institution. What sets contemporary patterns of cohabitation apart from historical patterns is not simply numerical preponderance. Cohabitation after the 1960s has special importance because it indicates a clear shift in normative behavior related to how families are formed and perceived. Although common-law unions were not unusual or contrary to social values before the nineteenth century, the same cannot be said about cohabitation in contemporary times. The lack of distinction between marital and non-marital union in England before the nineteenth century meant that cohabitation was not abnormal behavior. By contrast, today's nonmarital families represent a clear break from social convention.

Prevalence. One of the most salient facts about cohabitation is how much more prevalent it has become since the 1970s. In the United States, there were 523,000 households with two unrelated adults of opposite sex in 1970, compared with 1,589,000 in 1980—a 300 percent increase in just one decade (Spanier 1985). U.S. survey data on marital status and living arrangements indicate that cohabitation grew by 12 percent per annum between 1970 and 1980 (Davis 1985). This trend continued in the following decades. In 1990, there were 2.9 million cohabiting couples in the United States (Seltzer 2000), and the U.S. Bureau of the Census recorded over 4.2 million cohabiting households in March 1998.

Although these figures account for small percentages of the total pool of marital and nonmarital households, the increase in the number of people who have ever cohabited shows that cohabitation is a much more prevalent social trend than the proportional figures reveal. According to U.S. estimates of cohabitation, almost half of the population in their late twenties to early thirties had at some time been in a cohabitational relationship (Bumpass and Sweet 1989). Larry Bumpass and James Sweet's data also show that cohabitation has rapidly become an important antecedent to marriage. Of first marriages in the 1965–74 cohort, 9 percent of married respondents reported that they had cohabited with their future spouse. In the 1975–79 cohort, 26 percent had cohabited with their future spouse, and in the 1980–84 cohort, 34 percent had cohabited. In the 1990–94 cohort, almost 60 percent of all marriages formed had begun as nonmarital unions (Bumpass and Lu 2000).

In Canada, remarkable increases can be charted since 1981, the year in which the Canadian census began to record data on cohabitation. Over the period 1981–96, the number of cohabiting households increased from 356,600 to 920,640. Today, one in seven families is composed of unmarried couples in comparison to one in seventeen only fifteen years ago. As a percentage of all unions, cohabitations accounted for 13.7 percent in the mid-nineties, a significant difference over the 6.3 percent recorded in the early eighties. In the early 1970s, over 16 percent of all first unions were Cohabitation involves a shared household between intimate partners and thus has characteristics in common with marriage. Although division of labor according to gender prevails, women may look for partners who are willing to share domestic work in what they perceive as a "trial marriage." SCOTT ROPER/CORBIS cohabitational relationships. By the late 1980s, more than 51 percent of Canadian first unions were cohabitational relationships (Wu 2000).

Similar patterns have been observed in many European countries. In Sweden, which has the highest prevalence of nonmarital union in the world, cohabitation has been the norm since the 1970s. For Swedes, cohabitation is nearly a universal experience—for example, 96 percent of married Swedish women had previously been in a cohabitational relationship by the late 1970s (Hoem and Hoem 1988). In France, about 65 percent of all first unions were cohabitational by the early 1980s, more than double the level of one decade earlier (Leridon 1990). In 1994, the percentage of unmarried cohabiting couples was about 10 percent of all family units in Denmark, 13 percent in Finland, and 9 percent in Iceland (Yearbook of Nordic Statistics 1996). In England, cohabitation before marriage grew from one in four in the 1960s to about seven in ten in the early 1990s (Kiernan and Estaugh 1993). Data show that these trends prevail throughout Western Europe. In rough correspondence with declining marriage rates, the Eurobarometer Surveys conducted in 1996 show a preference for cohabitation among youth. For women between the ages of twenty-five and twenty-nine, for example, 40 percent of first unions were cohabitations in Austria, 37 percent in Switzerland, and 46 percent in West Germany (Kiernan 2000).

Differentials. In a study of U.S. and Dutch couples, Geertje Wiersma (1983) points out that contemporary patterns of cohabitation are unique because they developed in circumstances characterized by individual choice. Previous patterns emerged mainly because of social prerogative or constituted an imperative, and because of this, cohabitation tended to be isolated to particular segments of society (e.g., the poor classes). Having come about in a context of free choice, contemporary cohabitation is not restricted to any subdivision of society. However, although cohabitation occurs across social categories, important differentials exist that affect its prevalence. Not all people have the same propensity to cohabit. Rates of cohabitation are influenced by factors such as age, gender, marital status, education, employment status, religion, and others. Certain people are selected into cohabitational relationships, or, in other words, cohabitors tend to be certain types, even though cohabitation has spread throughout society. Researchers have isolated numerous factors that determine the selection process. Among the many traits that affect who cohabits, the basic differences are characterized by age, gender, race, socioeconomic status, and personal attitude.

First, age is certainly the most important factor. Although cohabitation is found in every adult birth cohort, cohabitors are primarily selected from younger groups. U.S. census data shows that 38 percent of all cohabitors were between the ages of twenty-five and thirty-four and another 20 percent were between the ages of thirty-five and forty-four. At the other end, very few were in cohabitational relationships in the over-sixty-four group. This group accounted for only 4 percent of all households with two unmarried, unrelated adults (U.S. Bureau of the Census 1998). Comparable trends are visible in other countries. In England, the General Household Survey from 1989 recorded that more than 47 percent of cohabiting men and more than 39 percent of cohabiting women were between the ages of twenty-five and thirty-four. In sharp contrast, only about 7 percent of cohabiting men and 8 percent of cohabiting women were between the ages of forty-five and fifty-four (Kiernan and Estaugh 1993). In Canada, cohabitation is also rare in the older age cohorts. About 13 percent of cohabitations in 1996 were between the ages of forty-five and fifty-four, compared to 31 percent between the ages of twenty-five and thirty-four and 20 percent ranging from thirty-five to forty-four. However, like all cohorts, the older ones have become increasingly more likely to enter cohabitational relationships. In 1981, only a little over 4 percent of cohabitations involved those between the ages of forty-five and fifty-four, which means that the 1996 figure represents a threefold increase (Wu 2000).

Second, a greater proportion of men cohabit than do women. One would expect the numbers to be equal in this regard since every cohabitation includes one man and one woman. However, the higher prevalence of men is not calculated in terms of absolute numbers—it is a ratio of cohabiting men to the total male population. For example, the rate of cohabitation for Canadian men in 1996 was slightly more than 17 percent. For Canadian women the rate was about 15.5 percent. Evidence suggests that this trend has deepened over time. Men tend to marry women younger than themselves, and therefore enter marriage at later ages than do women. Men are thus selected into cohabitation at a greater rate than women because being unmarried longer places them at a greater risk of being drawn into a cohabitational relationship (Wu 2000).

Socioeconomic status is a third factor that affects who cohabits. Although cohabitation first came to scholarly attention because of the living arrangements of 1960s college students, these persons were the imitators, not the innovators (Cherlin 1992). It was been widely observed that lower education levels and poorer employment status positively affect the propensity to cohabit (Cherlin 1992; Raley 2000; Seltzer 2000; Smock and Manning 1997). Researchers argue that economic security is a key factor for the formation of marriages (e.g., Oppenheimer 1994). People from poorer backgrounds often delay marriage because of insufficient economic resources. This makes them more likely to form cohabitational relationships than well-educated people. U.S. evidence shows that of nineteen to forty-four-year-old women who have ever cohabited, the probability for high school dropouts was 60 percent versus 37 percent for college graduates (Bumpass and Lu 1999).

Fourth, many cohabitors are self-selected because of their personal attitudes toward nonmarital unions. Because cohabitation occurs against the norm, cohabitors are partially rejecting the society's dominant value system. Those people who enter cohabitational relationships tend to perceive social rules in flexible terms. On the other hand, people with traditional perceptions of the family and religious backgrounds that prohibit premarital unions are unlikely to enter into cohabitation because of their conservative values (Axinn and Thornton 1993).