Introduction

Radical changes in the nature of romantic relationships are reflected in the decoupling of what were once considered “marital behaviors” from the institution of marriage. Cohabiters increasingly engage in many of the same practices of shared living as do married couples, benefitting from economies of scale, risk pooling, production of relationship-specific capital, and joint consumption without marrying (Lundberg and Pollak 2013). These changes have increased the attractiveness of cohabitation, whose financial and social barriers are considered lower than marriage (Sassler 2004). Cohabitation rates continue to rise and unmarried coresidential living has become the modal pathway to union formation for women and men in early and young adulthood (Addo 2012; Sassler 2010). And yet, in spite of high rates of cohabitation dissolution and serial cohabitation (Guzzo 2014; Lichter et al. 2010) young adults still marry, and nearly two thirds of recent marriages were preceded by cohabitation (Kennedy and Bumpass 2008; Manning 2013).

Much of the literature on the cohesion and relationship progression of cohabiters has focused on either implicit signs or subjective measures of commitment and relationship intentions (Brown and Booth 1996; Sassler and McNally 2003) rather than explicit or purposeful actions of relationship cohesion. Romantic commitment, or dedication, is important for establishing the emotional connection between partners, whereas constraints provide permanence to a relationship shaping transitions from one relationship phase to the next (Stanley et al. 2010). Constraints that are taken on with purpose (such as establishing a shared bank account) as opposed to ones adopted by accident (e.g., unplanned pregnancy) are equally important to study because they provide insight into the sort of scaffolding couples build around their relationships. When couples engage in joint financial practices they may be doing so as a matter of dedication (e.g., trust, commitment to the future, etc.) or for short-term convenience (e.g., costs savings). In either event, they believe that these practices will make their relationship and life together better (Addo and Sassler 2010; Treas 1993).

Yet, empirical research on financial management and income pooling behavior of couples in the United States has predominantly focused on married couples. The purpose of this study was to explore the individual and couple-level motivations for assuming financial integration practices among young adult cohabiters and within cohabiting unions, and the relationship between integrated financial practices and subsequent relationship status two years later. The analysis used data on 691 heterosexual current cohabiters interviewed over two waves from the Young Adult Survey of the National Longitudinal Study of Youth 1979.Footnote 1 Understanding how cohabiters arrange their households can help explain the subsequent relationship steps, whether it is to marriage, dissolution, or remaining in a cohabiting union. It may also assist with distinguishing cohabiters who are engaging in cohabitation as an alternative to single from the alternative to marriage and precursor to marriage groups. Does engaging in joint financial management practices serve as a precursor to matrimony, or are some young adults these days playing house with little intent of transitioning to marriage?

This study contributes to a growing body of research examining the financial lives and the accumulation of relationship-specific capital within cohabitation. Young adult cohabiters are disproportionately drawn from populations with less than a college degree, earn less, and experienced family instability during adolescence (Sassler and Miller 2011; Kennedy and Bumpass 2008). It is equally important to consider the ability to participate in a practice that has high exit costs when studying economically vulnerable populations, such as low-income young adults and those who come from disadvantaged backgrounds, a large and growing share of a policy-relevant population (Gibson-Davis 2009; Kenney 2004; Waller and McLanahan 2005). Family scholars who generalize relationship behaviors to all cohabiters may misrepresent the processes shaping the behaviors of less advantaged, more marginalized groups.

Collectivized Systems of Money Management in Marriage and Cohabitation

Family scholars studying how couples manage their financial situations have relied, for theoretical grounding, on the notion of transaction costs similar to those engaged in by firms (Williamson 1981). According to this perspective, couples engage in various forms of economic organization so as to minimize the costs of conducting exchanges between partners, while maximizing self-interest (Oropesa et al. 2003; Treas 1993). A collective, or joint, approach to family finances assigns precedence to non-market mechanisms of exchange over economic principles of self-interest (Treas 1993). This approach minimizes the costs of coordinating and monitoring arrangements, and potentially can lead to fewer disagreements. But separate fiscal management systems—when partners maintain privatized versus collectivized money management arrangements—increased over the past few decades (Heimdal and Houseknecht 2003; Oropesa et al. 2003; Treas 1993), in part due to increases in marital instability and the rise of cohabitation. Difficulties with ensuring relative exchange costs are borne equitably by both partners increase the likelihood that separate fiscal arrangements will be maintained. Uncertainty with the future of the relationship, previous experiences with failed relationships, or lack of trust reduce the incentives to pool finances, and increase the attractiveness of keeping separate systems.

Previous studies have found that couples’ decisions about who will manage their financial resources and whether they choose to pool these resources or maintain separate pots reflect the level of investment and integration in a relationship (Heimdal and Houseknecht 2003; Oropesa et al. 2003; Treas 1993). Social scientists who study resource management often distinguish between collectivized strategies, in which couples pool their assets, and private strategies that emphasize the separate well-being of individual partners and individual autonomy (Pahl 1989; Vogler and Pahl 1994). Collectivist household management systems represent a more egalitarian system of financial management, and become a means for creating and maintaining a couple-level identity (Bellah et al. 1985). Joint management projects a gendered sense of fairness regarding the division of labor within the household. Driving this belief is the idea that control and management over the resources coincide with decision-making (Bennett 2013).

Increased rates of pre-marital cohabitation and delayed marriage have driven interest in examining economic resource allocation behavior and the role of joint investment within cohabiting unions. In the United States, cohabiters do not have the legal protections afforded married couples (Brines and Joyner 1999; Perelli-Harris and Gassen 2012). They have no legal rights to a partner’s assets, and cannot sue for alimony or debts accumulated within the union upon dissolution (Bowman 2010).Footnote 2 , Footnote 3 Family researchers believe the increased risk of financial loss after separation reduces the initial amount of investment cohabiters contribute, compared to married couples. It also deters the desire to invest in relationship-specific capital, like joint financial practices. The lack of financial integration may reflect partners’ beliefs that the relationship will not last, and a subsequent desire to not increase further the costs of eventual dissolution (Burgoyne et al. 2006). Young adult cohabiters who experience high rates of relationship turnover (Lichter et al. 2010) may be hesitant to pool finances and feel the need to maintain individual systems. Nock (1995) argued that such practices indicated couples were less committed because they had lowered the perceived costs of ending the relationship.

Cohabiters are consistently less likely to pool finances than are married couples (Addo and Sassler 2010; Heimdal and Houseknecht 2003). Yet for those who do, pooling has been associated with relationship cohesion and better relationship quality (Addo and Sassler 2010; Brines and Joyner 1999). Integrating finances while living together in an informal union can also allow partners to become familiar with their partner’s financial status and consumption behaviors; sometimes this can alter the trajectory of the relationship (Dew and Price 2011). Whereas financial integration can work in tandem with positive relationship quality measures, negative financial behaviors can also work in the opposite direction, with negative information leading to relationship dissolution. Carrying lots of debt, for example, has been found to increase the likelihood of divorce among married couples (Dew 2007). To summarize, integrated financial practices reduce transaction costs and are associated with better relationship quality. However, for cohabiters, the associated costs of joint financial practices in the event of dissolution may be greater than with separate financial management systems reducing the incentive to engage.

Financial Integration as a Relationship Constraint of Current Cohabiters

According to Stanley and Markman (1992) the pooling of finances to share expenses or combine assets constitute commitments that constrain the relationship. Such constraints are linked to deeper interdependence in a relationship, they argued, and assist with transitions from one relationship phase to the next. Constraints that have low entry costs but high exit costs facilitate these transitions by making it easier to stay together and more difficult to break up once acquired. Stanley et al. (2010) believed that the difference between cohabiters who marry and those who remain cohabiting was the degree of risk and the willingness to become increasingly constrained. Cohabiters can either actively choose to commit (via marriage) or passively acquire constraints; the difference between the two is in how much control they exert over the process (Stanley et al. 2010). The distinction between active and passive accumulation of constraints is associated with subsequent relationship success.

A similar argument can be made for integrating finances, which requires the agreement of both partners. Financial integration practices are constraints that are purposely adopted because they involve active decisions as opposed to sliding or passive acts. Matters of convenience largely drive adoption of joint household financial practices as a means to decrease transaction costs of everyday living (Treas 1993). In addition, the sharing of finances and joint money management practices require some minimum threshold of couple-level communication and consensus (Addo and Sassler 2010). The adoption of financial integration practices within a cohabiting union may reflect preparation for the transition to a more formal union (Ashby and Burgoyne 2008). It may, however, indicate a desire to engage in more cost and time efficient practices formally associated with marriage without having to marry.

Cohabiters may also selectively choose to adopt a particular practice based on its associated degree of autonomy. They may desire to have something shared to help legitimize the union, but a lack of interest in bearing the legal and social costs of marriage (Ashby and Burgoyne 2008). Assuming the level of integration is defined by the perceived costs incurred in the event of relationship dissolution, the risk of dissolution decreases as cohabiters form a stronger interpersonal bond, or couple-level identity. Couples accumulate more constraints over time, driving the progression from individual to couple-level identity as relationship duration lengthens. One qualitative study found that even when cohabiting couples professed to having independent management systems to preserve autonomy, they still implemented practices that bordered more closely to pooling with some degree of shared management (Evertsson and Nyman 2014). Tension may arise between the appeal of structural constraints, such as an integrated financial system that decreases transaction costs and the desire to maintain an individual identity. A particular integration practice may be well suited to address this issue. For example, cohabiters wanting to maintain their autonomy may find credit cards appealing because they offer the option of joining accounts while also maintaining a separate personal account (Pahl 2008). It may also be the case that young adult cohabiters who exhibit high rates of relationship churning and a lower likelihood of marriage may have shorter relationship horizons and prefer short-term low-cost transactional arrangements. As a result, practices associated with fewer and lower exit costs are expected to be adopted more often and earlier in the relationship.

Couples today encounter a wider range of options about whether and in what ways to use shared financial instruments. Over the past thirty years, political changes have made it illegal to discriminate against credit applicants based on gender and marital status (Blakely 1981) and expansions of the credit market in the United States have increased the availability of financial products and options for consumers (Dynan 2009). With two-earner households and an increasing number of couples purchasing homes and accumulating savings pre-nuptials (Schneider 2011), interest in studying joint investments and resource allocation has grown. Exploring financial practice management now involves examining relationships beyond marriage and incorporating the diversity of financial options currently available. Adding someone to a bank or credit account requires a much lower entry cost (or fixed cost) than applying for a mortgage together (Beverly et al. 2003). Research also suggests that assets and debts are treated differently within marriages, and debt or negative household assets increases risk of divorce (Dew 2007). Behaviors that individuals exhibit in one relationship can also spill over into new unions undermining future relationship success (Coleman et al. 2000).

Relatively little empirical attention has been paid to the myriad ways financial integration exists within cohabiting unions or its role in relationship progression. This is surprising given that participation in shared living increases the odds of adopting structural constraints independent of the legal status of the relationship (Stanley and Markman 1992). The need versus the desire to integrate finances may be tied to the living situation that cohabitation facilitates, but it may also reflect the relationship dynamics exhibited within the union.

Classifying Financial Integration Practices of Current Cohabiters

Integrating finances differ according to their associated exit costs and the degree of individual autonomy. Based on these criteria financial integration practices are grouped into three categories: necessitating, progressive, or investment, as outlined in Table 1. Integrating finances can stem directly from the shared living arrangement. Necessitating integration practices usually appear on the list of reasons that make cohabiting more appealing than maintaining separate residences (e.g., Sassler and Miller 2011). They are inherently interdependent and require a reliance on a partner to provide economic support and assist with financial (and housing) stability. Participation is not tied to increases in couple-level identity above and beyond the one that was established when the cohabiters moved in together. Necessitating practices are hypothesized to be associated with the current socioeconomic status of cohabiters (e.g., education, income, employment status). They are also low-cost practices to adopt, in that the costs of the constraints are linked only to (short-term) residential expenses. In the event of relationship dissolution, the associated costs with necessitating practices of moving out of cohabitation are low when compared with exiting a marriage. Potential financial costs include, for example, remuneration for leases terminated early, the need to find a new place to live and associated costs, and the cost of relocating items, as well as the costs associated with setting up new utilities, etc. at the new location. Then there are the non-financial costs such as the stress associated with housing searches, or associated with remaining in the abode while waiting for the contract or lease to expire. It is hypothesized that necessitating practices will be positively associated with staying together as a cohabiting couple, since shared living (and its resulting economies of scale) is incumbent on participation in the practice.

Table 1 Financial integration practice classification, measures, and hypotheses
Full size table

Both the progressive and investment practices distinguish themselves from necessitating in that they are representative of classic pooling practices. Engaging in progressive practices of joint financial management allows couples to learn more about each other’s finances and financial behavior. They are defined by the ability to access and control a partner’s finances, and consequently reflect increased levels of trust. In doing so, couples also relinquish some autonomy by linking their short-term financial fates. Decreased autonomy over one’s finances increases the overall costs associated with participation and disengagement. These traits and their connection to relationship quality attributes such as trust lead to the hypothesis that cohabiters who engage in progressive practices are more likely to remain together as cohabiters and increase their probability of marriage.

Investment integration practices signal long-term attachment and commitment. They consist of the acquisition of assets (like home ownership), and reflect the belief that the practice will reap more than pecuniary benefits to the partnership collectively. Cohabiters that engage in investment practices are signaling publicly a couple-level identity and investment in the future. Both entry and exit costs are high, and exit costs grow with the level of financial commitment. Because of these properties one expects investment practices to be positively associated with marriage relative to remaining as cohabiters or separation.

The Role of Marital Intent

Stanley and Markman (1992) argued that informal coresidence increased constraints prior to the establishment of an emotional bond (which they termed “romantic commitment”) and the discussion of marital intentions. Sliding—the unplanned gradual progression into shared living—has been associated with what has been labeled the premarital cohabitation effect (Manning and Smock 2005; Stanley et al. 2006) or the greater likelihood that marriages preceded by cohabitation dissolve relative to direct marriages. Although some qualitative research suggest that cohabiters rarely discuss marital plans prior to moving in (Sassler 2004), cohabiters who actively decide to move in together, with expressed marital intent, are much more likely to marry and stay married than cohabiters who move in without being engaged and who subsequently marry (Brown 2000). Therefore, marital intent becomes as important to the adoption of a constraint as the decision to be constrained. For some cohabiters, planning to marry and beginning the process of financial integration practices serve as preparation for the formal union (Ashby and Burgoyne 2008). If cohabiting couples that plan to marry are more likely to engage in similar behaviors as married couples do, then marital intent could be driving the association between joint financial practice and transitioning to marriage. This would also mean that certain financial integration practices are not constraints that increase the probability of marrying, but rather works in tandem with preparation for marriage.

What we do know is that cohabiters are optimistic about their current relationships; they tend to believe their relationships will last and transition into marriage (Brown 2000; Waller and McLanahan 2005). In general, those who report being engaged or plan to marry at the time they moved in with a partner are more likely to transition from cohabitation to marriage (Brown and Booth 1996). For those who do transition to a marital union, marital intent may explain the adoption of joint financial practices, especially progressive and investment, which are not necessarily tied to coresidential living but embody relationship attributes such as trust and long-term commitment. The analysis will test the hypothesis that marital intent mediates positive and significant associations between financial integration practices and a marital outcome.

Additional relationship attributes that have been associated with relationship commitment and transitioning to marriage include having shared children in a cohabiting household (Tach and Halpern-Meekin 2009), the length of the cohabitation, and relationship quality. Couples that report poorer relationship quality are less likely to combine finances (Burgoyne and Morison 1997; Kenney 2006; Oropesa and Landale 2005). Age at the start of the cohabitation (Kuperberg 2014), previous cohabitation, and a prior marriage (Gibson-Davis 2009) have all been associated with relationship transitions among cohabiters. Race/ethnicity, maternal education, parents’ marital status, and a cohabiter’s birth to a teenage mother consistently predict marriage and relationship dissolution of cohabiters. Indicators of current socioeconomic status, such as educational attainment, post-secondary school enrollment, employment status, and income should be associated with the ability to engage in a financial practice and dictate current relationship status (Sassler 2010).

Methods

Sample data were from two recent waves of the National Longitudinal Survey of Youth 1979 Young Adult cohort (NLSY79-YA). The NLSY79 is an ongoing nationally representative cohort study of young adults who were between the ages of 14 and 21 as of December 31, 1979. In 1986 a separate survey began following the biological children of the NLSY79 women, and in 1994 an additional questionnaire was created and administrated to these children as they aged into adolescence and young adulthood. Participants of NLSY79-YA are interviewed every 2 years, and questioned on a variety of topics including employment, education, family, and relationships.

As of 2015, 11,512 children have been born to the mothers of the NLSY79. Of those, 6305 had aged into the young adult sample and were interviewed in 2008, and 5311 of these 6305 were interviewed again in 2010.Footnote 4 There were 860 current cohabiters in the 2008 sample, but only 711 were re-interviewed in 2010. At the time the survey was fielded same-sex marriage laws varied by state in the US. Therefore, twenty respondents who reported being in same-sex relationships were removed from the analysis. The final sample consisted of 691 heterosexual young adults who were cohabiting in 2008 and interviewed again in 2010. Multiple imputation techniques were employed in order to maintain maximum sample size and reduce the influence of observations missing data. Multiple imputation is the preferred method for dealing with missing data because it is less likely to bias the sample and results (Rubin 1996). Imputation models were generated using the mi impute chained command in STATA 13, and standard errors of the imputed estimates were corrected according to Rubin’s combination rules, which account for both within and between imputation variances (Rubin 2004).

Because 149 cohabiters were lost to attrition between waves, sample selection could potentially bias the estimated coefficients by inflating the transition outcomes of the cohabiters (Sassler and McNally 2003). Logistic regression models of cohabiter characteristics in 2008 indicated non-random attrition. Attritors were more likely to be older (25–35 years old), have lived with their married parents at 14, and previously married. The inverse of the predicted probabilities from the attrition model were used to re-weight the data for all estimations (Robins et al. 1994). The weighted sample was representative of children born to women who were between 14 and 21 in 1979 and currently cohabiting in 2008. Selecting the sample based on relationship status, however, reduced its representativeness; as a result, the analytic sample included a high concentration of young adults with less than a high school degree, an overrepresentation of Black and Hispanics, and youth from relatively low socioeconomic backgrounds. Therefore, the results should be interpreted as applicable to this subgroup of young adult cohabiters.

Current cohabitation status was constructed from the survey question that asked the respondent’s relationship status in 2008. The relationship status of the cohabiter with that same partner was then assessed in 2010 to create the dependent variable for the multivariate analysis. The cohabiter could have been in one of three current relationship status states: still living together as a cohabiting couple, married to that same partner, or no longer together.

Financial integration practices were determined using study questions asked only of current cohabiters or married couples. Shared household expense participation was used to determine that couples engaged in necessitating financial integration practices. Respondents who reported sharing at least some of the household expenses with their partner were coded equal to one, zero otherwise. The analysis contained two progressive financial integration practice measures: holding a joint bank account and a joint credit card account. Both financial practices include the ability to access, withdraw, or deposit funds into a joint account. They also offer additional incentives, such as a means to build savings and acquire credit history (Xiao and Anderson 1997). Although the bulk of finances are still shared, couples may employ alternative practices, such as maintaining separate and joint accounts, which allow each partner to maintain some financial autonomy (Ashby and Burgoyne 2008). Bank accounts determined access to joint assets, i.e., positive income and savings, and joint credit card accounts reflected access to joint debt. It is also possible to accumulate debt in bank accounts (e.g., overdrawing) and build up an asset (a higher credit score) through good stewardship of a credit card account. Recent studies suggest that the value of assets and debts lead to differential outcomes within relationships. For example, consumer debt was associated with marital distress, and increased conflict and risk of divorce (Dew 2007, 2011). Consumer debt and education loan debt was associated with lower likelihood of transitioning to marriage among young adults (Addo 2014; Dew and Price 2011), whereas net wealth and financial assets were positively correlated with movements into marriage (Addo 2014; Schneider 2011). There was no information distinguishing checking and savings accounts, and the analysis did not differentiate between types of credit cards (e.g., charge, store, etc.). Respondents that reported buying a house together were considered to engage in an investment financial integration practice, with interviewers allowed to clarify whether the respondent had a joint mortgage on a house with their current partner. For most US households, homeownership is considered an expensive long-term investment (Rohe and Watson 2007); it has also been found to increase the odds that low-income cohabiters with children wed (Gibson-Davis 2009). According to the US Census in 2007, the homeownership rate for 24–29 year olds was 40.6 %. Among low-income families with children, homeownership rates were around 35 % (Gibson-Davis 2009). Entry costs for homeownership are high, requiring financial transparency and proof of employment, steady income, and good credit histories. Steady employment with stable income is inherent in the ability to pay one’s mortgage. The financial integration prediction and relationship transition models included additional controls for demographic and family background, current socioeconomic status, and relationship characteristics linked to cohabiters’ relationship transitions and household resource allocation decisions. Respondent’s gender and race (non-Hispanic Black, Hispanic, and non-Hispanic White) were assessed from their first young adult interview. Family background characteristics consisted of maternal education and the marital status of the mother both at age 14 of the young adult. Education variables included highest degree completed and whether the cohabiter was currently enrolled in a post-secondary school or program at the time of the 2008 interview. Labor market controls include an employment dummy for paid work since the young adult’s last interview and logged income from the previous year. These same characteristics were provided for the cohabiter’s partner. Also included was whether there were any other adults living in the household with the cohabiters. Relationship history was captured with whether the cohabiter previously cohabited with someone else (Brown 2003) or had a prior marriage (Burgoyne and Morison 1997), as well as if their partner was previously married, and if the partner had a child in the household that’s not the respondent’s child. Current relationship covariates included presence of shared children in the household (Brown, 2003; Waller and McLanahan 2005) and current length of cohabitation in years (Brown, 2003; Sassler et al. 2010). Relationship quality was measured from the survey question: “Would you say your current relationship is very happy, fairly happy, not too happy?” and the marital expectations variable was based on the question: “Do you and your partner have definite plans to get married?” Very happy responses and definite plans to marry were coded as one, and zero otherwise.

The results section is divided into a two-part analysis. The first analysis characterizes financial integration practices using summary statistics including an examination of their relationships with the respondent’s background, current socioeconomic and relationship attributes. The second analysis uses multinomial logistic models to assess the relationship between the cohabiter’s relationship status in 2010- still cohabiting, married, or separated—given their financial integration practices in 2008. All tables report average marginal effects, which evaluate the average contribution to the outcome for a given change in the variable of interest when evaluated at the sample mean.

Results

Descriptive Statistics

Table 2 summarizes the descriptive characteristics of the analytic sample. Non-Hispanic Blacks and Hispanics were over represented in this sample of young adult current cohabiters, composing approximately 53 % of the full sample. About 12 % of the young adults were born to mothers 20 years old or younger, and just over 15 % had mothers who had not earned a high school diploma. Low educational attainment reflects the relative youth of the sample (the average age was 21); the largest share, 43 %, had less than a high school, while an additional 34 % had received their high school diploma. Approximately 14 % were currently enrolled in a post-secondary school or program at the time of the 2008 interview. Nearly the entire sample (94.5 %) reported being currently employed, with average wages of $18,000 in 2007 dollars. Partners of the respondents were also employed, but to a lesser extent, (79.3 %) and made comparable average salaries.

Table 2 Descriptive statistics of current cohabitors (N = 691)
Full size table

The young sample already exhibited high rates of relationship churning. Almost 20 % reported previously cohabiting and 2.4 % had a prior marriage compared with 7.8 % of their partners. An advantage of this dataset is that this sample includes cohabiters both with and without children. Just over half of the sample had young children living with them (not shown), and 33 % reported that at least one child was also the biological child of their current partner, whereas only 5.1 % reported living with partners who were parents to a child that was not their own. Cohabiters reported being fairly happy with their relationships with 72.2 % stating that they were very happy. Almost all (94.3 %) of the cohabiters were optimistic about the relationship and reported plans to stay in their current relationship, and close to 70 % of the sample reported intentions to marry their current partner. As of the 2010 follow-up interview, 49.2 % of the cohabiters were still cohabiting with the same partner, 16.6 % had married, and 34.2 % were no longer together.

Adoption and Determinants of the Financial Integration Practices of Current Cohabiters

Table 3 displays the mean financial integration practice measures for the full sample organized by final relationship status. The prevalence of necessitating, progressive, and investment strategies varied across cohabiters. Sharing household expenses (necessitating) was most common, with close to 90 % participation. And yet, there were still 11.5 % of the cohabiting sample that reported they did not share expenses with their partner. For this small group, cohabitation did not necessarily entail this particular financial integration practice. There was little variation across relationship outcomes; the participation rate for cohabiters who married was 91.6 % and for those eventually separated was 89.0 %.

Table 3 Financial Integration Practices in 2008, by relationship status in 2010
Full size table

Progressive integration practices were far less prevalent than necessitating practices. Just under one fifth of the sample reported holding joint bank accounts with their current partners, with close to 10 % sharing credit card accounts. The largest share holding both accounts were those who eventually married. Not surprisingly, separate banking accounts were the most common joint management system. For both bank and credit card accounts, the percentage of respondents who reported having none was larger than the percentage who reported holding a separate account and a joint account with their partner, an additional indication of their current relationship status and the disadvantaged nature of the sample. Cohabiters who married held both joint and separate accounts in greater percentages than those who continued to cohabit or separated; they were also underrepresented in the unbanked and no credit card accounts categories. Home ownership or holding a joint mortgage was reported by 13.3 % of the full sample. Approximately 30 % of the cohabiters who eventually married indicated some form of shared homeownership, which was more than twice the amount of those who continued to cohabit or eventually separated.

The second panel of Table 3 lists the mean adoption rates of the practices in combination. The average number of financial practices was highest among those who married (1.765). Financial practices were adopted in order of increasing risk or associated costs of dissolution, consistent with what was hypothesized. And, if cohabiters engaged in only one integration practice it was shared household expenses, the practice associated with the living arrangement, lowest exit costs, and greatest individual autonomy. There was only about 1.1 % of the sample that solely practiced either joint bank accounts or joint credit card accounts, and none of the cohabiters shared only homeownership. Among those cohabiters who reported sharing only two practices, they were most likely to combine a necessitating practice with a progressive, joint bank account, or investment practice. Qualitative evidence that joint accounts ensure that partners contribute their share to household expenses supports this finding (Burgoyne et al. 2007). Not only were joint credit card accounts least likely to be adopted, they were also least likely to be combined with necessitating or investment integration practices.

Logistic regression models were used to analyze the predictors of each financial integration practice (Table 4). Model estimates indicated that the only significant predictors of shared expenses were the partner’s income, having shared biological children in the household, and cohabitation duration. None of the family background, demographic, relationship quality, or marital expectation estimates reached conventional levels of significance. These results were in line with the definition of necessitating practice; adoption and continued engagement relies heavily on the current financial situation. While current income status and educational attainment were not associated with either of the progressive integration practices, a few demographic characteristics were related to participation. Female cohabiters sampled reported being less likely to share bank accounts with their current partner than male respondents and having married parents at age 14 positively predicted engagement. Living in a household with other adults also decreased the probability of a shared bank accounts (p < 0.05) and credit card accounts (p < 0.01). Reporting marital intent had positive associations with joint bank accounts and joint credit card accounts.

Table 4 Logistic regression models of financial integration practices
Full size table

Enrollment in a post-secondary school or program deterred participation in a shared mortgage as did the presence of other adults in the household, and the respondent’s income was positively associated with shared ownership. Cohabiters who expressed definite plans to marry were 14 % more likely to share homeownership. Comparing across the three financial integration levels, definite patterns emerge that were in line with their associated costs and autonomy attributes. For example, as exit costs grew marital intentions became increasingly significant. With the exception of marital intent, relationship attributes including having a marital history and the quality, surprisingly, were not significant predictors across all joint integration practices, whether it was sharing expenses or pooling income or resources.

Financial Integration Practices and Relationship Outcomes

Table 5 presents the results from the multinomial logistic models of relationship status in 2010 on financial integration practices in 2008, net of demographic, socioeconomic, and relationship history controls. The estimates in Table 5 list the predicted probability of transitioning to either marriage or dissolution relative to remaining together as a cohabiting couple with that same partner. Statistically significant (p < 0.05) estimates between marriage and dissolution are also underlined. Because the estimated average marginal effects of the additional covariates are consistent across all five models and the empirical and analytical focus is on the relationship between the financial integration practices and relationship outcomes, the covariate results have been omitted from the tables (available upon request). Models indicated that over the two year period compared to non-Hispanic Whites, Hispanic cohabiters were more likely to stay together as cohabiters relative to separating. Compared to high school graduates, cohabiters without a high school degree were less likely to marry and being enrolled in a post-secondary institution increased separation relative to staying together as cohabiters. A history of cohabitation increased the likelihood of separation as did having a partner that was previously married. Both positive relationship quality and marital intent were strong predictors of marriage. Reporting that one was very happy in their relationship was negatively associated with a subsequent dissolution. Current cohabiters who reported definite plans to marry were 14 % more likely to transition to marriage over the study period.

Table 5 Financial integration practices in 2008 on relationship status in 2010
Full size table

Panel A of Table 5 presents results when each practice was entered separately into the full model. As shown in Table 3, the integration practices, especially progressive and investment, were highly correlated. Including all four practices in the same model decreased the ability to isolate the relationship of a specific financial integration practice on a relationship outcome. Both panels, A and B, consist of results from two models. Model 1 indicates the direct effect of the financial integration practice on subsequent relationship outcome, and Model 2 presents the results when marital intent is added as a potential mediator.

Contrary to what was hypothesized, sharing household expenses did not initially appear to be significantly associated with relationship status. Differences emerged when comparing the results for the progressive practices. Progressive practices increased the likelihood of transitioning from cohabitation into marriage. A joint bank account increased the likelihood of marriage by 8.2 % (p < 0.05) while joint credit card accounts were associated with a 9.8 % (p < 0.05) increase relative to remaining together as a cohabiting couple. The underlined average marginal effects for joint credit card holders indicate that these cohabiters were also more likely to break up with their partners than they were to marry (p < 0.05). This provides additional evidence that for some cohabiters a credit card’s associations with debt may be negatively associated with relationship success (Dew 2007). Joint homeownership as an investment practice was associated with an 11.1 % increase in the probability of a marital transition from cohabitation.

All of these relationships became more pronounced within the mediation results reported in Column 2. There were noticeable decreases in the average marginal effects predicting transitions into marriage, and an increase in the magnitude and significance of joint credit card ownership and its relationship to dissolution of the relationship. The bootstrap test on the mean mediated relationship (Preacher and Hayes 2008) indicated that marital intent completely mediated the relationship between engaging in progressive practices and transitioning to marriage. It did not, however, fully explain the strong association between having a shared mortgage and marriage. Results suggest that financial integration practices that have high exit costs and very low individual autonomy such as investment practices embodied the constraint commitment definition best. Joint homeownership remained positively and significantly associated with a marital transition, independent of marital intentions.

Marital intent also appeared to suppress the relationship between holding a joint credit card and relationship dissolution. Similar to the results from Model 1, the underlined marginal effects indicate that after controlling for intent, cohabiters who held joint credit card accounts with their partners had an increased probability of dissolution relative to staying together and marrying. Couples who adopted progressive practices tied to negative financial behaviors or financial instability and who had no intent to formalize their unions were more likely to separate.

Finally, the results in Panel B reveal a bit more about the types of cohabiters that practiced particular financial integration types. Cohabiters that engaged only in shared experiences were less likely to marry; a relationship that remained strong, independent of expressed marital intent. This suggests that cohabiters who only engage in necessitating practices may be committed, or constrained, to their coresidential living arrangement. The probability of marriage increased as more constraints were put in place, with the number of financial practices positively associated with transitioning into marriage as compared to non-marital cohabiting and separation. Similar to joint homeownership, marital intent was only able to partially mediate the relationship.

Conclusion and Discussion

For young adults who believe they need to be financially stable and economically secure in order to marry, cohabitation is a more attractive, less expensive coresidential option. The predominant view is that marital dissolution is more costly than ending a cohabiting union because the former has more constraints to disentangle. Stanley et al. (2010; p. 245) wrote, “Commitment can be considered an act of choosing to be increasingly constrained because of the desire to persist, exclusively on the chosen path.” While not all cohabiters marry, there are mechanisms through which they may decide to reinforce their commitment to each other and the couple. Recent evidence suggests that the economic costs of cohabitation dissolution are increasing for women in particular. Avellar and Smock (2005) found that post-separation former cohabiting women fared worse, experiencing lower income and higher poverty rates. Tach and Eads (2015) showed that income losses to formerly cohabiting mothers were similar to those of divorced women. This study examined whether the costs of cohabitation has increased as a result of the increase in financial constraints young adult cohabiters have chosen to adopt within their unions, and whether these risk-pooling activities were associated with subsequent relationship transitions and marital intentions.

The central finding of this paper was that young adult cohabiters were not only sharing expenses, but that they were also intertwining their credit histories and bank accounts and acquiring assets, like buying a house together. Shared homeownership, however, was adopted only in combination with the other household practices of financial integration. Descriptive statistics suggested that integration practices were adopted in order of increasing exit costs. That such a high percentage of cohabiters shared household expenses is not surprising given its relation to the residential situation; it is also consistent with findings of previous studies (Treas and De Ruijter 2008; Kenney 2004). Joint bank account ownership appears to be necessary in order for the other integration practices to occur, and is reflective of its use as a means to accomplish the other medium to high cost practices of shared credit card accounts and shared mortgages (Ashby and Burgoyne 2008). This is not completely unexpected given joint bank accounts tend to be the primary money management tool for couples that have them (Addo and Sassler 2010; Treas 1993). They may also be a means for cohabiters to handle risk in the relationship. Given that cohabitation does not confer the same legal protections in the event of dissolution as marriage does, joint bank accounts expose couples in financial vulnerable positions. They can serve primarily as an insurance mechanism to protect individual assets during the relationship, as each individual has access to his or her partner’s money to pay joint bills on shared expenses. Joining finances becomes a means of reducing uncertainty surrounding ones financial and relationship status (Romo 2014).

The second main finding concerns subsequent relationship transitions and marital intent as they are related to the financial integration practices. The analyses found that practices with medium to high exit costs and those that displayed couple level identity were more likely to be associated with a subsequent union transition. Specifically, sharing mortgages were associated with an increased likelihood of marriage, whereas joint credit card accounts increased the odds of dissolution. Relatedly, marital intent was positively associated with a marital transition and cohabiters with an intent to marry were much more likely to start integrating their finances prior to marriage. Some studies suggest that marital intent is less likely to translate into marital unions for low-income couples and young adults born from socio-economic disadvantage (Gibson-Davis et al. 2005; Gibson-Davis, 2009). While shared mortgage was associated with an increased likelihood of marriage, marital intent did not fully explain why those who shared homeownership were also more likely to marry. This suggests that for cohabiting populations who are more likely to be disadvantaged, intendedness is not be enough to predict a future marriage, but rather evidence of significant financial investment could be. These findings are in line with previous research on low-income samples of cohabiting couples with children, indicating increased likelihood of marriage among cohabiting homeowners (Gibson-Davis 2009). If homeownership requires savings for a down payment and steady income to pay the mortgage, marriages resulting from premarital cohabitation should increasingly consist of financially secure and stable young adults. In addition, the study findings suggest that cohabiting couples who subsequently marry may have a head start on asset acquisition, and explain some of the wealth advantage married couples who cohabited first have over those who directly married (see Painter and Vespa 2012 for more details).

Integration of finances in cohabitation may also weed out incompatible matches. This appears to be true with pooling of economic resources such as consumer debt. Cohabiters who held joint credit card accounts with their partners had an increased risk of union dissolution. Therefore, a potential unintended consequence of adopting certain marital behaviors pre-nuptials, such as integration of finances, is to decrease the negative selection of couples that slide into marriage because of the constraint commitments of cohabitation. Given the low levels of participation as compared to the other practices, sharing credit card accounts may be considered the structural constraint with the greatest perceived exit costs. This is not surprising, given the structure of the credit card market. Problems with paying bills and overdue accounts can negatively affect one’s credit report, which influences one’s future borrowing chances for years. And while shared household expenses was not predictive of union transitions, those current cohabiters who only shared expenses were more likely to stay together as cohabiters rather than marry. This could be an indication that certain financial integration practices, the ones involving short term and limited transactions, have been decoupled from marriage and contribute to the stability of those unions.

One of the unique features of the progressive financial integration practices is the ability to hold both a joint account and a separate account. Supplementary analyses disaggregating the bank and credit card account ownerships explored whether cohabiters who practiced both collective and individualist progressive practices were more likely to remain together, eschewing marriage or subsequently separating. There was evidence that compared to only holding joint bank accounts, maintaining both separate and joint bank accounts was associated with remaining together as a cohabiting couple relative to separating. With respect to only holding joint credit card accounts, maintaining only a separate account was associated with staying together as cohabiters rather than transitioning to marriage; an indication that cohabiters who want to stay cohabiters should not attempt to integrate all forms of finances, especially those connected with debt or that carry stigmatizing effects. Cohabiters with no credit card accounts, joint or separate, were less likely to marry relative to cohabiting and more likely to separate than marry. For some, cohabitations are increasingly becoming similar to some marriages in which the central role of joint financial practices and the integration of resources have decreased (Lauer and Yodanis 2011).

This study is not without limitations. Given the study design, causal claims are outside the realm of this analysis and omitted variable bias remains a concern. There may be unobservable attributes related to shared homeownership that also influence transitioning to marriage. The panel nature of the data, however, does increase confidence in the results based on the temporal ordering of the financial practices and the relationship statuses. It is also difficult to ascertain whether decisions on the pooling finances were made contemporaneously with future relationship decisions. Therefore, it is possible that observed relationship behavior mirrored plans on the part of a select sample of couples. If this were true, then we would expect the findings to be upwardly biased. Another limitation is the representativeness of the sample. Of note is that the analytic sample is necessarily selected on cohabiters that were already in lasting relationships. Many cohabiting relationships of recent cohorts of young adults are short-term ending in dissolution (Guzzo 2014). Study results are more likely to be biased towards the relationship behaviors of young adults in either lasting relationships or recently formed relationships closest to the 2008 interview date (Hayford and Morgan 2008). In addition, the sample data were relatively disadvantaged limiting the generalizability of the findings to all young adult cohabiters. Ten percent of respondents were born to mothers who had not completed high school and a significant proportion had less than a high school degree at baseline. The number of homeowners (19.8 %) is also low when compared to national averages (34.3 %). It is, however, important to note that the study period, 2008–2010, coincided with the Great Recession (2008–2009), during which low-income homeowners experienced some of the highest mortgage default and foreclosure rates in US history. Although this analytic sample was not nationally representative, the NLSY79-YA study is currently the only dataset that contains both questions related to the within household financial behaviors of young adults and longitudinal relationship information.

Despite these limitations, examining financial integration practices can have important implications for how researchers distinguish among the diverse groups comprising current cohabiters. When it comes to integrating finances, some constraints are necessary for coresidential living, like sharing household expenses. Progressive practices can assist with short-term transactions at the same time weed out the precursors to marriage from the alternatives to single and married cohabiters, and investing into the long-term financial integration matters for transitioning into a marital union. Delineating low versus high exit costs, associated autonomy, and short-term versus long-term structural constraint commitments may be more important for relationship transitioning than active versus passive adoption, or sliding versus deciding. In addition to expanding the analysis to multiple periods and including a larger and more national representative sample of current cohabiters, future studies should use couple-level data to address issues of concordance on financial behaviors and control for socioeconomic and relationship homogamy. Future research should also investigate the roles of power and dependency, as played out in cohabiting union by the explicit practices they choose to adopt as qualitative studies find that joint accounts are still controlled by one partner, and in married couples it is more likely to be the female partner (Burgoyne et al. 2006).