Financial Arrangements and Relationship Quality in Low-Income Couples

Abstract

This study explored the association between household financial arrangements and relationship quality using a representative sample of low-income couples with children. We detailed the banking arrangements couples utilize, assessed which factors relate to holding a joint account versus joint and separate, only separate, or no account, and analyzed the association between fiscal practices and men’s and women’s relationship quality. The majority of couples held joint accounts, though over one-quarter also have separate accounts; nearly one-tenth have no account. Joint bank accounts were associated with higher levels of relationship quality on numerous dimensions, though more consistently for women than men. Individualistic arrangements appeared to undermine women’s relationship satisfaction and reduce feelings of intimacy, sexual compatibility, and satisfaction with conflict resolution.

Tensions over finances are a key predictor of marital distress (Dew, 2008) as well as the dissolution of both marriages and cohabiting unions (Amato & Rogers, 1997; Smock, Manning, & Porter, 2005). Even though the body of literature on how American couples arrange their fiscal resources is expanding (Heimdal & Houseknecht, 2003; Kenney, 2004; Treas, 1993), to date relatively few studies have explored the associations between financial management and relationship quality. Furthermore, the research that exists has focused disproportionately on married (or remarried) couples and those that are middle class (Coleman & Ganong, 1989; Dew; Papp, Cummings, & Goeke-Morey, 2009; see Kenney for an exception). Yet financial strains are pervasive among low-income families, who often have few assets to weather unexpected employment gaps, medical bills, or other emergencies (Bucks, Kennickell, & Moore, 2006; McKernan & Ratcliffe, 2008). Clearly, a better understanding of the ways fiscal arrangements might be associated with or contribute to family strain, especially among less advantaged families, is warranted.

The Deficit Reduction Act of 2005 set aside $150 million per year for programs to promote marriage and responsible parenting (U.S. Department of Health and Human Services, 2006), through the Administration for Children and Families’ Healthy Marriage Initiative. Educating disadvantaged populations about fiscal management is one component of marriage promotion activities. Funding is available to provide education in high schools on the value of budgeting and marriage education, and relationship skills programs are encouraged to include financial management in their curriculum. Federal and state governments have also implemented programs to encourage saving among low-income families. The Individual Development Account program promotes asset building among low-income families by providing incentives to establish bank accounts (McKernan, Ratcliffe, & Nam, 2010). Such efforts are motivated, in part, by research finding that most low-income families have minimal savings to tide them over in the face of a financial exigency (Caner & Wolff, 2004; McKernan & Ratcliffe, 2008). This shortage of assets can translate into difficulty meeting basic needs, less stable relationships, and inadequate investment in children’s development and human capital (McKernan et al; Nam & Huang, 2009).

This study examines the association between men’s and women’s reports of household financial arrangements and various dimensions of relationship quality, using data from a representative sample of low-income couples with coresident children. We detail the banking arrangements utilized by married and cohabiting parents, including a category for couples where neither partner has a bank account. We then explore which factors are related to couples’ banking arrangements. Finally, we analyze the association between fiscal practices and men’s and women’s relationship quality. Our goal is to enhance understanding of how differential approaches to money management can influence relationship quality among low-income couples with parenting responsibilities for minor children.

Money Management in Intimate Relationships

Since the early years of the 20th century there have been dramatic changes in how American couples arrange their finances, driven in part by public policies granting women greater individual rights and emerging demographic behaviors (Geddes & Lueck, 2002; Kenney, 2006; Zelizer, 1989). Up through the mid-19th century, the doctrine of coverture gave husbands control over all aspects of their wives’ lives, including her assets (Geddes & Lueck, Kahn, 1996). By the early 20th century, almost every state had passed laws granting women greater economic rights (Kahn), though husbands largely controlled family resources. As the 20th century advanced, women advocated for a greater role in administering the family money, as well as the ability to claim some portion for their own use (Zelizer). Adherence to the belief that husbands and wives should share access to a joint account gradually replaced views of the naturalness of the male-controlled resource model (Zelizer).

New factors arose to challenge the primacy of the shared marital account in the closing decades of the century, including the expansion of women in the labor force, the growing prevalence of divorce, the rising age at first marriage, and increases in cohabitors (Bumpass & Lu, 2000; Schoen & Canudas-Romo, 2006; Spain & Bianchi, 1996). These changes affected how couples arranged their finances (Kenney, 2006; Pasley, Sandras, & Edmonston, 1994). Separate fiscal management systems have become more evident over the past few decades, with some couples maintaining both joint and separate accounts, and others holding only separate accounts (Ashby & Burgoyne, 2008; Elizabeth, 2001; Pahl, 1995; Pasley et al.).

Despite a growing body of research on couples’ money management strategies, much of the current literature is based on data from European countries, and the bulk of it consists of nonrepresentative studies (Ashby & Burgoyne, 2008; Burgoyne, Reibstein, Edmunds, & Dolman, 2006; Pahl, 1995; Pasley et al., 1994; Vogler, 2005; Vogler, Lyonette, & Wiggins, 2008). Research by Treas and Widmer (2000) has found that systems of money management vary widely by country. Furthermore, even within the United States cultural differences distinguish how families arranged their finances; for example, Black women are more likely than White women to utilize independent money management (Treas, 1993).

A second major drawback of the research on couples’ money management is the lack of consistency in how couple’s fiscal arrangements are classified. The preponderance of extant research relies on the typology detailed by Pahl (1995) that examines not only whether monies are pooled or held independently but also the sex of the partner responsible for money allocation (i.e., bill paying; Ashby & Burgoyne, 2008; Kenney, 2006; Vogler, 2005; Vogler et al., 2008). Few studies, however, ask respondents if they and their partner have joint or separate personal checking or savings accounts (Pasley et al., 1994), though qualitative studies suggest that is what couples understand income pooling to be (Ashby & Burgoyne; Burgoyne et al., 2006).

Notwithstanding the growing emphasis on those who maintain independent management systems, the vast majority of American couples pool their monies in a joint account. An early study by Treas (1993) that used data from the 1984 panel of the Survey of Income and Program Participation (SIPP) found nearly two-thirds of couples with bank accounts kept joint accounts only; the other one-third had at least one separate account. Later studies report even higher levels of income pooling. Using data from the 1994 International Social Survey Program (ISSP) sample in the United States, Heimdal and Houseknecht (2003) found that 80.6% of couples pooled their resources, whereas only 19.4% kept their money separate. Data from the 2002 ISSP sample also revealed that four-fifths of U.S. couples pooled their money; of those who kept money separate, 10.2% engaged in partial pooling—that is, maintaining both a joint bank account and at least one separate one—and the remaining 9.8% utilized completely independent accounts (Hamplova & LeBourdais, 2009).

Even studies that focus on low-income samples find that the majority engage in income pooling, though at a somewhat lower rate. Kenney (2006), examining data from the Fragile Families and Child Well-Being Study, found that nearly 60% of the women in her sample reported some form of income pooling (though they do not indicate if joint accounts were held). Studies of whether economically disadvantaged families have bank accounts report similar findings. Researchers utilizing data from the 2003 SIPP found that 56.5% of low-income families had a bank account (McKernan & Ratcliffe, 2008). This figure is somewhat lower than the results obtained with the 2004 Survey of Consumer Finances, where 75% of families in the bottom earnings quintile had a checking or savings account (Bucks et al., 2006). Unspecified in these reports is whether both partners have access to these accounts.

Of note is that none of these prior studies report the proportion of couples where neither member had a bank account, despite their presence, particularly among the economically disadvantaged (Garasky, Nielson, & Fletcher, 2008). An estimated 10% of the U.S. population is considered unbanked (Kennickell, Starr-McCleur, & Surette, 2001), defined as having no checking or savings account with a depository institution or no transaction account with a money market fund or brokerage firm. Our study also examines whether the absence of any kind of bank account is associated with relationship quality.

Despite country-level variation in couples’ approaches to money management, some factors emerge as consistent predictors of how couples arrange their finances. Divorced individuals are more likely to maintain separate rather than joint systems of money management in subsequent unions (Burgoyne & Morison, 1997; Treas, 1993). Cohabitors are also less likely to pool their money (Heimdal & Houseknecht, 2003; Kenney, 2006). Although some cohabitors purposefully work at maintaining fiscal autonomy as a signifier of equality (Elizabeth, 2001), for others separate money management results from less commitment to the relationship or beliefs that the current arrangement is not permanent (Ashby & Burgoyne, 2008). The likelihood that cohabitors’ pool income also increases among those who share children (Winkler, 1997). Furthermore, many cohabitors go on to marry their partners (Kennedy & Bumpass, 2008), highlighting the need for additional attention to whether couples’ money management systems change over time.

Theoretical Perspectives on Collective Versus Individualistic Money Management

How couples manage assets provides insight into the importance assigned to individual desires—for autonomy or control—versus a willingness to pursue the interests of a larger collective (such as the family). Social scientists have long examined this struggle, particularly as it relates to family functioning and gender roles (Bellah, Madsen, Sullivan, Swidler, & Tipton, 1985; Coontz, 2005; Lundberg, Pollak, & Wales, 1997; Pahl, 1995). A collectivist approach is one that assigns value to sharing resources with others, as a means of establishing a sense of solidarity and group purpose (Bellah et al.). According to this perspective, couples focused on shared goals establish joint accounts because collective strategies are more beneficial than individualized approaches to fiscal management; they are more efficient, minimizing exchange costs between members, and maximize self-interest because they stabilize and strengthen the family (Browning & Chiappori, 1998; Oropesa, Landale, & Kenkre, 2003; Treas, 1993). A common pot approach to family finances may also signal greater commitment to a relationship’s future and can strengthen bonds by emphasizing common interests over separate ones (Oropesa & Landale, 2005).

Conversely, couples where one or both members retain separate accounts are engaging in utilitarian individualism (Bellah et al., 1985). Those who pursue their own material autonomy, even when partnered, assign precedence to self-interest over collective goals (Treas, 1993). Uncertainty in the future of the relationship, previous failed relationships, or residing in less formal relationships with fewer protections from the state reduce the incentives to pool finances and increase the attractiveness of maintaining separate accounts. There is some evidence that those who pursue individual autonomy over bank accounts (i.e., the “separate pots” approach) do so as a means of ensuring equality (Elizabeth, 2001); but doing so can exacerbate inequality due to gender disparities in earnings and family responsibilities (Vogler, 2005; Vogler et al., 2008).

One important issue to consider is possible endogeneity between relationship quality and income management. The quality of the relationship, in particular whether a partner thinks it will endure, can influence decisions about whether to engage in a joint versus separate money management system, rather than the relationship operating in the opposite direction. Those less committed to a long-term future with their current partner may not desire a joint account, wary of the entanglement that represents. Kenney (2006) found that women who assessed their partners as fair, supportive, and understanding were far more likely to utilize income pooling than women with poorer reports of their partners’ behaviors. Oropesa and Landale (2005), in contrast, argued that the stronger association is from the system of money management to relationship quality, because of what shared money management systems signify—a collective approach that assigns precedence to the good of the couple over that of both individuals. A willingness to pool assets, they argued, helps reduce the odds of relationship dissolution.

The amount of time couples have lived together is also an important issue to consider with regards to money management. It takes time for two once autonomous individuals to alter their fiscal management systems. That may be one reason why studies find that cohabitors are less likely than marrieds to pool income (Heimdal & Houseknecht, 2003), as cohabitations tend to be of relatively short duration (Bumpass & Lu, 2000). But adherence to separate systems of money management may weaken with increased time together. In one qualitative study of couples interviewed before their first marriage and 1 year later, the majority of couples utilized independent management systems prior to their wedding, but during the first year of marriage the trend was toward more merging of finances among those who had not already been pooling all their income (Burgoyne et al., 2006; see also Oropesa et al., 2003). We therefore would expect the likelihood of having a joint bank account to increase among couples who have been together for longer periods of time.

Money Management and Relationship Quality in Intimate Unions

The body of literature assessing the factors shaping relationship quality is extensive. Even though most are focused on married couples (and marital quality; Knobloch, 2008; Williams, 2003), new union formation patterns have caused scholars to broaden their scope to encompass cohabiting couples as well as those who lived with their spouse prior to marriage (Lichter & Carmalt, 2009; Stanley, Rhoades, & Markman, 2006; Tach & Halpern-Meekin, 2009). Such research tends to focus on the mental health benefits conferred by involvement in relationships, particularly those sanctioned by the state (i.e., marriage).

The few studies assessing how couple’s methods of arranging money is associated with relationship quality do find some limited support for the premise that joint investments result in higher levels of marital quality. In one of the earliest studies on the subject, Fishman (1983) suggested that couples who pooled resources had higher satisfaction and better adjustment than did households that kept all resources separate; nonetheless, her sample consisted of only 16 remarried couples. Using a larger sample of 91 remarried couples, Coleman and Ganong (1989) found that both stepmothers and fathers in remarried families that pooled resources felt closer. Others have also found that income pooling was associated with greater marital satisfaction and happiness, and men and women who maintained all or some separate accounts were least satisfied with their family life (Pasley et al., 1994; Vogler et al., 2008). Yet all these studies are based on small, nonrepresentative samples and do not address the issue of causality. Furthermore, most studies rely on very few measures to assess the overall quality or strength of romantic unions. Relationship quality is comprised of multiple dimensions (Lichter & Carmalt, 2009; Moore et al., 2004), and financial arrangements may exert greater effects on some aspects—such as conflict frequency and intensity or communication—than others (Dew, 2008; Papp et al., 2009). Our study addresses these shortcomings.

The Current Study

We took advantage of newly collected Internet survey data to advance the research on marital quality in several ways. First, we examined the income organization utilized by men and women in our sample of low-income parents and assessed the factors associated with a collectivist versus an individualistic approach. We tested several hypotheses regarding how money is arranged, based on the extant literature. We expected that respondents who have already experienced divorce and those in cohabiting unions (or who cohabited prior to marriage), would be less likely to own a joint account and more likely to manage separate sources of money. To assess if a willingness to merge money evolves over time, we next hypothesized that couples who have been together longer should demonstrate greater odds of having a joint account relative to more independent management approaches. Finally, if the causal direction is from relationship quality to money management systems, then those with poorer quality relationships should be more likely to maintain individual rather than merged money management systems.

We next explored the association between couples’ approaches to fiscal organization and relationship quality. In this analysis, we expected reliance on a joint account would be associated with better quality relationships, measured across various dimensions, than when individuals maintain separate accounts. Accounting for respondents’ characteristics was expected to diminish the impact of fiscal organization, if earlier predictions regarding the differential approaches to money management utilized by cohabitors and the previously married are borne out. We tested to see if cohabitors who share collective money management systems differ from those who married directly in relationship quality, as a means of assessing whether commitment levels vary even among those utilizing joint bank accounts.

Method

Data

Data are from the Marital and Relationship Survey (MARS), a web-based survey of married and cohabiting couples administered by Knowledge Networks (KN; see Lichter and Carmalt, 2009, for a description of the data). The survey includes probability samples of persons who are members of a web-enabled panel designed to be representative of the U.S. population and covers both the online and offline population in the United States. The population is identified from telephone surveys of listed and unlisted telephone numbers. KN provides ongoing household panelists with an Internet appliance, Internet access, Web TV, and a cash payment in return for completing the survey. Panelists receive unique log-in information to access surveys online and monthly follow-up emails inviting them to participate in research. Because Internet accessibility was provided, the use of an Internet survey did not exclude members of disadvantaged backgrounds, who are the least likely to own a computer or have access to the Internet (Fairlie, 2004). The MARS response rate was 80.3%, and item nonresponse was low (less than 4%). Panelists are rotated in and out of the survey to assure up-to-date nationally representative samples.

The MARS sample was restricted to couples with coresident minor children, with household incomes less than $ 50,000, and where the female partner was under age 45. This is the population of greatest interest to those forming programs supported by the Healthy Marriage Initiative. The survey was conducted in March and April of 2006 and took approximately 35 – 40 minutes to complete. For this analysis, we utilized data from married and cohabiting couples and also included a small number of respondents whose partners did not participate in the survey. Information was collected independently from both partners, and each was provided with a unique log-in to ensure privacy while completing the survey. Our final sample consisted of 532 male respondents and 563 female respondents.

Measuring Relationship Quality

We rely on a multidimensional construct of relationship quality, taking as our starting point the measurement framework described in Moore et al.’s (2004) report, “What is ‘Healthy Marriage’? Defining the Concept.” We measured eight dimensions of healthy relationships: (a) relationship satisfaction; (b) commitment to the relationship; (c) perceptions of partner’s commitment to the children; (d) intimacy/emotional support; (e) sexual compatibility; (f) communication; (g) conflict resolution processes; and (h) frequency of relationship conflict. Confirmatory factor analysis was utilized to determine that our summated measures are valid representations of latent constructs; a complete list of the questions utilized in constructing each measure and the Cronbach’s alpha are available from the authors upon request. Item scores were reverse coded as necessary, so that higher scores indicate better relationship quality.

Measures of Money Management and Arrangement

Our primary independent variable measured how couples arrange their finances, focusing on ownership of bank accounts. Respondents were asked whether they held joint bank accounts with their current partner and if they held separate personal accounts. The question does not distinguish between checking and savings accounts. For the purposes of this study we generated four mutually exclusive categories: ownership of a joint account only; ownership of both a joint and separate accounts; no joint account but at least one respondent has a separate account; and lastly no joint and no separate account (the unbanked).

We also controlled for a range of other relevant attributes. The means of the variables used in our analyses are presented in Table 1. The control variables are presented in Panel A. The sample was a disadvantaged one; less than two-thirds grew up in intact married-parent families, and over a quarter of respondents indicate their mothers had not finished high school. As for their own attributes, sample respondents were disproportionately White; one-fifth to a quarter were previously married; and over half of the respondents had either cohabited prior to their marriage or were currently cohabiting. Of note is that these relationships were of relatively long duration, with the average time together being over 10 years.

Table 1.

Means (Standard Errors) for Variables Used in Analysis, by Sex

Independent Variables	Men			Women
	M (%)	SE	Cronbach’s alpha coefficient	M (%)	SE	Cronbach’s alpha coefficient
Panel A. Background characteristics
Family structure as child
Married, intact family	0.62	0.02		0.59	0.02
Never married mother/do not know	0.09	0.01		0.11	0.01
Parents divorced	0.29	0.02		0.29	0.02
Maternal education
Less than high school	0.31	0.02		0.27	0.02
High school	0.47	0.02		0.41	0.02
More than high school	0.21	0.02		0.32	0.02
Race
Non-Hispanic White	0.87	0.01		0.90	0.01
Black	0.06	0.01		0.04	0.01
Hispanic	0.07	0.01		0.06	0.01
Educational attainment
Less than high school	0.09	0.01		0.09	0.01
High school	0.35	0.02		0.32	0.02
Some college education	0.36	0.02		0.40	0.02
Bachelor’s degree or more	0.19	0.02		0.19	0.02
Previously married	0.25	0.02		0.20	0.02
Child from previous relationship	0.21	0.02		0.26	0.02
Age at start of current relationship (<20 years)	0.17	0.02		0.34	0.02
Relationship status
Married directly	0.39	0.02		0.39	0.02
Currently cohabiting	0.11	0.01		0.10	0.01
Cohabited prior to marriage	0.49	0.02		0.50	0.02
Number of children in household	2.05	0.05		2.03	0.05
Duration of union (in months)	123.90	3.26		123.29	3.27
Less than 5 years	0.22	0.02		0.22	0.02
5 – 10 years	0.32	0.02		0.34	0.02
10 – 15 years	0.21	0.02		0.21	0.02
15 years or more	0.24	0.02		0.23	0.02
Hardships experienced in past year (0 – 10)	2.09	0.09		2.27	0.10
Current health limitation	0.19	0.02		0.19	0.02
Panel B. Financial arrangements
Bank accounts
Joint, no separate	0.64	0.02		0.61	0.02
Joint and separate	0.17	0.02		0.19	0.02
No joint, separate	0.11	0.01		0.12	0.01
No joint, no separate	0.09	0.01		0.09	0.01
Panel C. Relationship quality
Relationship satisfaction (0 – 10)a	8.38	1.80		8.14	2.01
Commitment to relationship (0 – 12)	10.46	2.13	0.84	10.29	1.91	0.84
Partner’s commitment to child(ren) (0 – 16)	13.79	1.85	0.63	12.98	2.56	0.83
Intimacy/emotional support (0 – 20)	16.56	2.78	0.84	16.28	3.22	0.87
Sexual compatibility (0 – 20)	15.32	3.01	0.73	15.40	3.07	0.73
Communication (0 – 20)	15.95	2.95	0.82	15.66	3.43	0.86
Conflict resolution processes (0 – 12)	8.73	1.94	0.61	8.82	2.09	0.70
Frequency of relationship conflict (0 – 12)	9.83	2.27	0.86	9.84	2.34	0.85
N	532			563

Note: Values in bold denote significant difference by sex (p ≤ .05).

^aQuestion range.

The initial examination of the fiscal management and organization practiced by the couples in our sample is shown in Panel B. Consistent with prior studies of income management, almost two-thirds of men reported keeping joint accounts only; women were somewhat less likely to mention they kept only a joint account (60.6%). Similar shares of men and women maintained a separate as well as a joint account, 17% and 18.8%, respectively.

About 10% of men and women have only separate accounts, whereas nearly 9% of men and women report having neither joint nor separate accounts—the unbanked. The means for our measures of relationship quality are presented in Panel C of Table 1. The bivariate results revealed that men report higher mean scores, on average, than women, especially on relationship satisfaction and their perceptions that their partner is committed to their children.

Analytic Approach

Our analysis proceeds in two steps. We first utilized multinomial logistic regression to assess the factors predicting how couples arrange their money. Next, we estimated multivariate ordinary least squares (OLS) regression models to determine the relationship between bank account ownership and several dimensions of relationship quality. For the final OLS regressions, we fitted individual-level models with our main indicator of financial organization and then incorporated other measures capturing potential financial strain, individual, and relationship characteristics. We also ran sex-specific models for all analyses. All regression models were estimated using multiple imputed data created from the imputation using chained equations program for STATA (Royston, 2006) in order to maintain maximum sample sizes for all variables utilized in estimation.

Results

Factors Associated With Couples’ Income Management

We turn now to our analysis of the factors associated with styles of money management. The results of multinomial logistic regression, which allowed us to simultaneously estimate the odds of more than two outcomes as a function of a set of regressors (Maddala, 1983), are shown in Table 2. We present odds ratios for ease of interpretation. An odds ratio greater than 1.0 indicates an increased likelihood of utilizing that type of banking system relative to the reference group, those who report only a joint account. Focusing first on results for women, we see that relatively few factors predicted having both joint and separate bank accounts, relative to just a joint account. Only relationship duration attained statistical significance; women in relationships of less than 5 years were 2.25 times more likely than women in relationships longer than 15 years to report that at least one partner has a separate bank account, in addition to their joint account; women in relationships of the shortest duration (less than 5 years) were also significantly more likely (p < .05) to maintain separate accounts only or no accounts than women who have been in relationships lasting 5 – 10 years (results not shown). Women who reported that they only have separate bank accounts were far more likely to be racial minorities, cohabitors (either currently or prior to marriage), and in relationships of less than 5 years. Turning to predictors of being unbanked, we found that women who grew up in disadvantaged families—those born to single mothers and mothers who lacked a high school diploma—were 4.5 and 2.8 times more likely, respectively, to currently lack a bank account. Such women were also substantially more likely to be cohabiting and in unions of shorter duration.

Table 2.

Multinominal Logistic Regression Models on Ownership of Bank Accounts (Relative Risk Ratios)

	Women			Men
	Joint and Separate Accounts	No Joint, Separate Accounts	No Joint, No Separate Accounts	Joint and Separate Accounts	No Joint, Separate Accounts	No Joint, No Separate Accounts
	Versus Joint, No Separate Account			Versus Joint, No Separate Account
Maternal education (high school degree)
Less than high school	1.31	1.98	2.84*	1.12	1.81	1.82
More than high school degree	1.25	1.49	1.17	0.93	0.42	0.26
Family structure as child (married, intact family)
Never married mother/do not know	0.77	0.69	4.50**	1.33	0.25	0.33
Parents’ divorced	1.12	1.10	1.18	1.29	1.53	1.00
Individual attributes
Race (non-Hispanic White)
Black	2.28	12.85**	2.76	5.09**	22.24**	11.69**
Hispanic	1.14	2.99	2.07	0.62	1.09	1.09
Age at start of current relationship (<20)	0.82	0.75	1.31	0.95	2.76	1.15
Child from previous relationship	1.11	1.28	1.27	1.60	2.60	1.86
Previously married	1.13	0.64	1.31	0.74	1.26	0.80
Relationship status (married directly)
Currently cohabiting	3.15	254.00***	47.44***	1.24	59.75***	43.50***
Cohabited prior to marriage	1.56	8.68***	2.43	0.93	2.82	2.19
Duration of current relationship (more than 15 years)
Less than 5 years	2.25*	4.53*	5.13*	1.87	3.61*	3.66*
5 – 10 years	1.36	1.32	1.82	1.85	1.12	1.42
10 – 15 years	1.44	3.11	2.49	1.55	3.06	1.29
Relationship quality measures
Relationship satisfaction	0.98	1.03	1.21	0.84	0.75	0.81
Commitment to relationship	0.88	0.91	0.67*	1.08	1.07	0.98
Partner’s commitment to child(ren)	0.92	0.76*	0.93	0.93	0.82	0.84
Intimacy/emotional support	1.00	0.93	1.04	1.02	1.05	1.12
Sexual compatibility	0.94	1.03	1.06	0.95	1.13	1.07
Communication	1.04	1.10	1.08	1.07	1.03	1.01
Conflict resolution processes	1.13	0.86	0.97	0.92	0.78	0.84
Frequency of relationship conflict	0.91	1.04	0.94	1.11	1.09	1.20
−2 log likelihood		−610.98			−558.02
Pseudo-R2		0.22			0.20
Number of cases		563			532

Note: Robust standard errors are available from author by request. Joint account and no separate account is the base outcome in all models.

^***p < .001.

^**p < .01.

^*p < .05.

Among men, growing up in a disadvantaged family exerted little effect on the type of bank accounts maintained. Consistent with prior research (Treas, 1993), Black men were less likely to have only a joint account than they were to have a joint and separate accounts, only separate accounts, or no accounts. The impact of being Black was greatest on the odds of having only separate accounts; Black men were 22.4 times more likely to have only separate accounts than only a joint one. Those who currently cohabit were also more likely to maintain separate over joint accounts. We also found a duration effect, as men in relationships of less than 5 years were also over three times more likely to maintain only separate accounts over a joint one than were men in longer lasting relationships (5 – 10 years, results not shown) or longer. Finally, men in unbanked couples were more likely than those who maintain a joint account to be Black, to be cohabitors, and to have lived with partners for the least amount of time.

The duration measure highlights that for both men and women, those who have lived together for shorter periods were more likely to either maintain only separate accounts or have no bank account whatsoever. Longitudinal data on couples’ money management, and how it changes over time, are scarce. Our results provide further justification for qualitative research that found many couples who initially engaged in independent money management systems shifted to a pooling system; purchasing homes, childbearing, and time together accelerate this process (Burgoyne et al., 2006).

Some research (Kenney, 2006) has suggested that relationship quality may influence the type of management system utilized. Those in high conflict unions may elect to establish separate money accounts, as may those thinking of ending the relationship. To assess this possibility, we also included our eight measures of relationship quality in the model predicting type of fiscal arrangements. None of the measures of relationship quality reached conventional levels of statistical significance in the men’s model, providing no support for the argument that the causal direction runs from relationship quality to banking arrangements. For women, two of the measures did attain significance (p < .05 level). Women who believed their partner was less committed to their child(ren) were more likely to maintain only separate accounts. Additional tests (not shown) indicated that this was not a function of women with children from a prior relationship being more likely to maintain separate accounts. We did find, however, that cohabitors differed substantially from those who married directly in the association between their belief of a partner’s commitment to children and the likelihood that separate accounts were maintained. These results suggest that women did not initially establish separate accounts because of their attitudes toward their partner’s commitment to their child, as many did not yet share children (and we control for births from prior relationships); rather, there was stronger evidence that the maintenance of separate accounts after the childbirth results in the perception that a partner is less committed to that child. The only other attitude that attained statistical significance, commitment to the relationship, lowers women’s odds of being unbanked (relative to having a joint account). We interpret this as providing additional support for Oropesa and Landale (2005); establishing a shared account demonstrates faith in a future together. Longitudinal data are needed to definitively assert that the causal direction is from relationship quality to income management style, though our results suggest this is the more likely scenario.

We turn now to our final analyses, run separately for women and men, to assess the association between fiscal arrangements and relationship quality. Two models are presented for each measure. The first included controls for our main independent variable of interest (the form of bank account ownership), with Model 2 including the other controls. Results for the women are presented in Table 3. In the reduced model, having only a joint account was associated with better relationship quality, and this advantage held across all the eight measures. Even when couples had a joint account, but (at least) one partner has a separate account, women’s relationship quality was significantly lower. The negative effect of having only separate accounts, relative to only a joint account, was generally even larger. Among women reporting that neither has an account, however, only one measure attained significance: unbanked women reporting significantly lower levels of commitment to the relationship. These differences in relationship quality remain even after accounting for the range of controls (Model 2). And women reporting that they maintained only separate accounts were significantly worse off on two measures—perceptions of partner’s commitment to a child and conflict resolution processes—than those women who indicated that the couple had both a joint account and separate ones. As only one of these measures was initially related to women reporting only separate accounts (perceptions of partner’s commitment to the child), our evidence provides stronger support for the notion that couples’ banking systems affect how women assess their relationship quality, rather than the reverse. Finally, our results indicated that women who reported the couple was unbanked did not differ significantly in relationship quality from women reporting they had only a shared account, net of family background, individual and relationship attributes, and union status controls.

Table 3.

Ordinary Least Squares Regression Results for Relationship Quality: Women

	Relationship Satisfaction		Commitment to Relationship		Partner’s Commitment to Child(ren)		Intimacy/Emotional Support		Sexual Compatibility		Communication		Conflict Resolution Processes		Frequency of Relationship Conflict
	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2
Independent variables
Bank account ownership (joint, no separate)
Joint and separate	−0.69**	−0.58**	−0.71***	−0.63**	−0.90	−0.86**	−1.03	−0.97**	−1.11**	−0.95**	−1.06**	−0.90*	0.45	−0.32	−0.80**	−0.66**
No joint, separate	−1.15***	−0.88**	−1.30***	−1.04***	−1.46***	−1.59***	−1.74***	−1.54**	−1.45***	−0.91	−1.87***	−1.23*	−1.40***	−0.98**	−0.91**	−0.62
No joint, no separate	−0.25	0.07	−0.85**	−0.55	−0.45	−0.5	−0.49	−0.12	−0.37	0.20	−0.71	0.06	−0.61*	−0.19	−0.51	−0.09
Hardships experienced in past year (#)		−0.17***		−0.11**		−0.22***		−0.21***		−0.25***		−0.27***		−0.18***		−0.27***
Current health limitation		−0.44		−0.31		−0.40		−0.81*		−0.89**		−0.66		−0.26		−0.51*
Family background characteristics
Family structure as child (married, intact family)
Never married mother/do not know		−0.38		−0.12		−0.11		−0.46		−0.34		−0.66		−0.16		−0.51
Parents divorced		0.21		0.18		0.16		0.27		−0.08		0.34		0.25		0.16
Maternal education (high school degree)
Less than high school		0.06		−0.02		0.30		0.26		−0.04		0.15		0.04		0.22
More than high school degree		0.05		0.17		0.58*		0.70*		0.07		0.55		0.33		0.49*
Individual/relationship attributes
Race (non-Hispanic White)
Black		−1.04*		−0.93*		−0.27		−0.62		−1.32*		−0.87		−0.31		0.38
Hispanic		−0.04		−0.39		−0.52		−0.72		−0.62		−0.85		−0.35		−0.15
Previously married		−0.01		−0.18		−0.15		0.1		0.43		0.12		0.25		0.21
Relationship status (married directly)
Currently cohabiting		−0.25		−0.32		0.40		−0.27		−0.42		−1.09		−0.65		−0.54
Cohabited prior to marriage		−0.30		−0.23		−0.04		−0.18		−0.13		−0.6		−0.39		−0.41
Number of children in household		−0.03		0.12		0.00		0.04		0.22		0.13		−0.03		0.02
Duration of union (months)		0.00**		0.00**		0.00**		−0.01**		0.00		−0.01**		0.00		0.00*
Constant	8.42***	9.48***	10.64**	11.24***	13.38***	14.29***	16.72***	17.68***	15.81***	16.50***	16.12***	17.43***	9.11***	9.72***	10.14***	11.06***
Observations	563	563	563	563	563	563	563	563	563	563	563	563	563	563	563	563
R2	4.08%	12.65%	6.31%	12.00%	4.39%	12.40%	3.71%	10.35%	3.34%	10.70%	3.69%	12.15%	4.81%	11.70%	2.61%	14.25%

Note: Robust standard errors are available from author by request.

^***p < .001.

^**p < .01.

^*p < .05.

The results for men were quite similar to those found for women (Table 4), at least when only banking arrangements are considered (Model 1). The type of bank ownership system utilized exerted a significant effect for seven of the eight measures of relationship quality. Men who reported holding only separate accounts reported significantly lower levels of relationship quality than men who held only joint accounts. Many of these differences were no longer significant, however, upon controlling for other background characteristics (Model 2). Men holding only separate accounts reported lower levels of relationship satisfaction than those with joint accounts, perceived partners as less committed to their children, and had poorer conflict resolution skills, net of controls. But we found no significant differences between those reporting joint accounts only and men with both joint and separate accounts or between men with joint accounts and those with no bank accounts. The real distinctions, then, were between those men who utilized a joint (collective) approach versus those maintaining a completely individual one.

Table 4.

Ordinary Least Squares Regression Results for Relationship Quality: Men

	Relationship Satisfaction		Commitment to Relationship		Partner’s Commitment to Child(ren)		Intimacy/Emotional Support		Sexual Compatibility		Communication		Conflict Resolution Processes		Frequency of Relationship Conflict
	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2	Model 1	Model 2
Independent variables
Bank account ownership (joint, no separate)
Joint and separate	−0.38	−0.31	−0.11	−0.10	−0.26	−0.23	−0.23	−0.17	−0.49	−0.37	−0.23	−0.05	−0.21	−0.12	0.09	0.21
No joint, separate	−1.10***	−0.70*	−0.78**	−0.40	−0.84**	−0.68*	−0.93*	−0.53	−0.80	−0.19	−1.24**	−0.58	−0.97	0.68*	−0.84*	−0.34
No joint, no separate	−0.55	−0.12	−0.45	−0.02	−0.47	−0.15	−0.49	0.28	−0.33	0.41	−0.43	0.28	−0.33	−0.00	−0.02	0.57
Hardships experienced in past year (#)		−0.07		−0.03		−0.07		−0.08		−0.16*		−0.12		−0.10*		−0.14**
Current health limitation		−0.50*		−0.48*		−0.71**		−0.86**		−0.97**		−0.48		−0.41		−0.54*
Family background characteristics
Family structure as child (married, intact family)
Never married mother/do not know		−0.72*		−0.48		−0.38		−0.18*		−0.83		−1.39**		−0.63		−0.69
Parents divorced		−0.28		0.15		−0.18		−0.36		−0.62*		−0.47		0.11		−0.30
Maternal education (high school degree)
Less than high school		0.10		−0.10		0.14		−0.04		−0.31		0.22		−0.16		−0.04
More than high school degree		−0.22		−0.27		−0.16		−0.20		−0.02		−0.36		−0.32		0.17
Individual/relationship attributes
Race (non−Hispanic White)
Black		0.47		−0.31		0.50		0.15		0.15		−0.43		−0.15		0.14
Hispanic		0.29		−0.31		0.24		0.05		−0.07		−0.09		0.17		−0.23
Previously married		0.00		−0.06		−0.20		−0.09		0.41		0.30		0.00		−0.02
Relationship status (married directly)
Currently cohabiting		−0.95**		−1.02**		−0.36		−0.97		−1.64**		−1.06		−0.53		−0.65
Cohabited prior to marriage		−0.24		−0.22		−0.18		−0.07		−0.37		−0.29		−0.32		−0.24
Number of children in household		−0.02		0.10		0.03		0.08		0.21		0.24*		−0.04		−0.05
Duration of union (months)		0.00		0.00*		0.00**		−0.01***		0.00*		−0.00		0.00		0.00*
Constant	8.61***	9.15***	10.59***	11.12***	13.80***	14.48***	16.70***	17.93***	15.52***	16.39***	16.14***	16.87***	8.89***	9.72***	9.90***	10.46***
Observations	532	532	532	532	532	532	532	532	532	532	532	532	532	532	532	532
R2	3.74%	10.65%	2.30%	9.03%	2.05%	7.60%	1.01%	7.91%	0.93%	9.14%	1.80%	8.46%	2.43%	7.92%	1.33%	7.41%

Note: Robust standard errors are available from author by request.

^***p < .001.

^**p < .01.

^*p < .05.

Discussion

Supporting and strengthening marriage and couple relationships has been on the forefront of the public policy agenda, with numerous states launching public and community-based initiatives designed to stabilize families. Although many of these programs focus on improving communication and various relationships skills, our work highlights the need for additional attention to family financial arrangements. Our paper is among the first to directly explore how money management and control are related to various dimensions of relationship quality. The results indicate that the way households manage their finances (collective vs. individual arrangements) are highly salient for women, though less so for men. Couples practicing independent money management report lower levels of relationship satisfaction and less agreeable methods of resolving conflicts than when couples share a joint account. Independence in financial matters for the couples in our sample, then, is not beneficial for relationship quality.

A major theme in the literature is that cohabitors are less committed to their relationship than are marrieds (Stanley et al., 2006; Tach & Halpern-Meekin, 2009). But what if cohabitors demonstrate a more collective approach to their relationship, such as having a joint account? To test this, we included an interaction effect to ascertain whether cohabitors with a joint account only have significantly different levels of relationship quality than married couples with only a joint account. Our results (not shown) reveal that when couples with similar financial arrangements (joint accounts only) are examined, cohabiting men actually report higher levels of relationship satisfaction, commitment to the couple, emotional support and intimacy, and are better at resolving conflicts. Given that the analysis of factors related to the type of bank account utilized do not reveal any significant association between men’s reported commitment to the relationship and the maintenance of only joint banking systems, relative to all others, our findings provide additional evidence that the banking system couples utilize influences their relationship quality, rather than the other way around, at least for men. No parallel results are observed for women; that is, cohabiting women who report having only a joint account do not differ significantly in any of the relationship quality measures from their married counterparts, which in itself is notable. These findings provide further verification for Oropesa and Landale’s (2005) suggestion that the association between cohabitation and lower relationship quality found in other studies may be a function of cohabitors’ lower likelihood of maintaining joint bank accounts, among other factors.

Our results highlight the importance of gender as it relates to money management and relationship satisfaction. Women appear to be more concerned with, and more watchful over, men’s purse than men are over women’s. The presence of a separate account—even when there is also a shared one—may trigger women’s concerns that men are holding back from their families, elevate concerns over infidelity (Edin, Kefalas, & Reed, 2004), or signify one has a foot out the door. Our findings suggest some women adhere to the belief couples should share a bank account and cast a distrustful eye on separately held money. Why men are less affected by the bank account situation is something of a puzzle. Men may be better able to obtain work that pays cash, and they may be the ones more likely to have a separate account even when they also share a joint one. Men may also feel the need to hold separate accounts to pay for child support for children from prior relationships or may adhere to hegemonic ideas regarding men’s entitlement to control over resources (Tichenor, 1999).

Conclusion

Previous research has found that money is not the most frequent source of conflict at home, though money-related conflicts often last longer than discussions on other tense topics and are often recurrent (Papp et al., 2009). Our results suggest that although couples may disagree over the assignment of responsibility for bills or why particular spending occurred, dissatisfaction with how the family fiscal situation is arranged may also be reflected in various ways—lower levels of relationship satisfaction or commitment, dissatisfaction with sexual compatibility, or a belief that one’s partner is not there for the children—that may on their face not appear to be about how much money a couple has. For example, couples where neither partner has any form of bank account do not report lower levels of relationship quality, once background factors are accounted for. Our findings call for the need for additional study on how couples arrive at their financial arrangements. Additional support with budgeting, programs designed to build couple-level wealth, and more open dialog about what accounts symbolize could be included in relationship skills courses geared toward less advantaged couples. Growing levels of economic insecurity and unemployment among low-income couples with children make such programmatic efforts increasingly important, given the many ways that fiscal arrangements are associated with different aspects of relationship quality.

Our study is not without limitations. Our data are cross-sectional, making it harder to discern pathways through which relationships are adversely affected or strengthened. Our sample also contained too few minority respondents to determine whether there is much variation among a sample selected to be economically disadvantaged. Furthermore, even though couples may have a joint account, there may not be equal access to or control over how that money gets spent, which we cannot determine with our data. Nonetheless, our results do highlight the need to better understand the ways couples organize and manage their finances, as well as if other asset forms (homes, cars, access to credit) mediate or improve observed outcomes. We find that joint accounts are more common among relationships of longer duration. Furthermore, collectivist, shared money systems appear to be most beneficial to relationship quality, assessed in various ways, especially for women.

The findings from this study must also be interpreted in the context of the broader economic climate in place when the data were collected. The economy was still quite strong in 2005 and 2006, with plentiful service jobs available for respondents with only moderate levels of schooling. The recent economic recession, which economists peg to 2007, has dramatically altered the situation, and disadvantaged couples have increasingly faced home foreclosure, job loss, and a credit crunch. The fiscal arrangement of less advantaged families may now be an even more salient topic with regards to family functioning, particularly for those families experiencing economic challenges, such as home or job loss or underemployment.