Financial Disagreements and Money Management Among Older Married and Cohabiting Couples in Sweden

This study investigates how partners’ money management strategies are associated with the experience of financial disagreements among older couples (60–80 years old). Money management is a broad concept, and this study operationalizes whether the partners pool all money, the partners pool some money, one partner manages the money (and gives a share to the other partner for personal spending), or the partners keep all money separate. The data comprise a subsample from the Swedish Generations and Gender Survey from 2012 to 2013. The descriptive statistics show that 11% of older couples experience financial disagreements and that there is a large variation in how couples manage their money. Contrary to our expectations, logistic regression analyses further indicate that couples who pool all money are less likely to have financial disagreements than couples who either keep all money separate or adopt a lower degree of pooling. Whether some or all money is kept separate does not seem to be important for the likelihood of financial disagreements, as all these couples are more likely to experience disagreements. Among couples with financial hardship, partial pooling and keeping money separate are associated with a higher likelihood of financial disagreements than pooling all money. In conclusion, the greater probability of financial disagreements among couples who do not pool their earnings calls for greater awareness of the potential consequences of various money management contexts among individuals, couples, decision-makers and practitioners. In the worst cases, couples may have poor financial wellbeing.


Introduction
Money is one of the most common reasons for disagreements among couples (Dew, 2011;Henry et al., 2005;Levenson et al., 1993;Stanley et al., 2002). This study addresses the question of how such disagreements are related to the money management strategy used by couples. The amount of money over which each partner retains personal control and the amount that is contributed to a common pool may relate to arguments about money and potentially lead to disagreements. The circumstances under which couples pool money may be especially challenging when one of the partners feels that he or she has less influence over spending than the other and that the other partner makes most decisions (Kirchler et al., 2001).
How young and middle-aged couples manage money has been well documented (Addo & Sassler, 2010;Clarke, 2002;Evans & Gray, 2021;Evertsson & Nyman, 2021;Halleröd, 2005;Kenney, 2006;Knudsen & Waerness, 2009;Vogler, 2005), including a few studies that were conducted in Sweden in the late 1990s (Ahrne & Roman, 1997;Heimdal & Houseknecht, 2003;Roman & Vogler, 1999). Since money and the strategies used for money management change over the life course, the life situations of younger and older couples are not always comparable (Lott, 2017). Older couples have passed the childrearing and role specialization ages, and they often have constrained financial situations (Casey & Yamada, 2002). The financial decisions and priorities that retirement-aged couples face are in many ways unique to this phase of life.
In this study, using the perspective of conflict theory (Hamon, 2016), we investigate the link between money management strategies and the experience of financial disagreements among older couples in Sweden. Money management is a broad concept and can mean different things. In this study, whether the partners pool all money, the partners pool some money, one partner manages the money (and gives a share to the other partner for personal spending), or the partners keep all money separate is operationalized (Pahl, 2008). We also take into account which money management strategy may be the most beneficial among older couples who are in a financially constrained situation.
Understanding older couples' use of different money management strategies and relationship wellbeing is central in identifying the various aspects that may be associated with poor individual wellbeing and couple stability at later ages. Couples' financial disagreements tend to be more intense and last longer than other disagreements (Papp et al., 2009). Disagreements may also have negative outcomes for the relationship and for the partners, such as an increased risk of depression (Sandberg & Harper, 2000), heart disease and poor overall health (Kiecolt-Glaser & Newton, 2001;Levenson et al., 1993). Financial disagreements may also decrease relationship quality, in severe cases leading to union dissolution among both cohabiting and married couples (Dew, 2011;Dew et al., 2012;Henry et al., 2005). However, older individuals are likely to have a higher threshold for leaving a dysfunctional relationship, as the financial margins are often small (Lin & Brown, 2020;McManus & DiPrete, 2001;Swedish Pension Agency, 2018;Tach & Eads, 2015). In addition, an older individual in poor health may depend on his or her partner (Parsons et al., 2021), while a care-providing partner may have concerns over leaving his or her partner due to his or her vulnerable situation (Mutch, 2010).
In the extant literature, there is very little research on the financial wellbeing of older couples, especially in Sweden. Understanding these couples requires further development; we need to examine the diversity of money management and the meaning of money management for older couples' financial wellbeing.

Swedish Context
When studying the link between money management strategies and financial disagreements among older couples, it is essential to identify each couple's context, which is likely relevant for the effectiveness of the couple's money management strategy. In particular, it is important to understand the opportunities of the individuals involved to earn their own money over their life course and to determine the economic resources that may be allocated to the couple. It is also important to establish whether there is economic interdependence between the partners and whether economic independence is valued. The Swedish government's aims for gender equality in the labour market and in the family present a specific situation for older couples. Swedish women's and men's earnings over the life course make women less economically dependent, which is highly valued among couples (Stocks et al., 2007). The country's well-established dual-earner family model encourages women to participate in the labour force and men to participate in childcare, and this model had already been implemented when today's older generations were young (Gustafsson & Jacobsson, 1985;Stanfors, 2014). In 2012, 83% of women and 89% of men (aged 16-64) participated in the labour market, and among those aged 55-64 years old, 70% of women and 76% of men were in the labour force (SOU, 2014). The average retirement ages were 64 for women and 66 for men. Since a reform in the early 2000s, the proportion of both women and men who work at older ages has increased, as such work became more profitable after the reform (OECD, 2022;Statistics Sweden, 2020).
While most women in Sweden work, they more often work part time in female-dominated, low-income occupations and have more work disruptions during their careers. On average, women have an income that is 90% of men's income, and the gender income gap has been stable over recent decades (Statistics Sweden, 2014). The gender gap is much greater for pensions due to women's lower pension contributions, with women receiving on average 70% of men's pensions (The Swedish Social Insurance Inspectorate, 2017). Twice as many retired women compared to retired men have a pension that is below the poverty threshold in Sweden (Swedish Pension Agency, 2018). These circumstances show that, on the one hand, most women and men in Sweden hold power in the form of economic resources in the relationship; on the other hand, women's earnings still lag behind men's earnings, especially in the case of pensions. Both of these trends are relevant for how money is allocated in the couple and for how money is perceived by the partners, and they create different situations for older women and men living with a partner.

Money Management in Couples
To determine how couples manage their money, we used a commonly used typology (Pahl, 1989;Vogler & Pahl, 1994;Vogler et al., 2006). Money management strategies differ from each other in terms of the partners' degree of control over and access to household money, the degree of togetherness and interdependence between the partners, and how personal and household resources and spending are defined. The strategies are embedded within social and cultural institutional contexts, and thus, they may shift in importance over time and space. The typology has been used in many studies over time and across different contexts (Evans & Gray, 2021;Stocks et al., 2007).
Among pooling couples, both partners' earnings are pooled into a joint account in which all money is considered to belong to the couple and, in theory, both partners decide on expenditures together. This strategy indicates dependence in the couple. It has often been applied among couples with traditional gender role attitudes, lower educational levels, and lower incomes as well as couples who have been together for longer durations (Evans & Gray, 2021).
In contrast, partial pooling couples operate as single economic units while also keeping some money separate. Specifically, the partners pool some of their money to pay for collective expenses and keep the rest of the money in separate accounts (Vogler & Pahl, 1994). Partial pooling has been argued to combine two conflicting principles of togetherness and autonomy (Fleming, 1997). A recent crossnational study, including Sweden, indicated that these are predominantly couples of short duration and that they are often younger, childless, and cohabiting (Evans & Gray, 2021). There may be many reasons for partial sharing: These couples may not have the many types of shared expenses that often come with, e.g., children and duration, and they may therefore feel that they must keep some money separate (Vogler, 2009).
Compared with the other strategies, female money management is more dominated by institutionally separated gender spheres, and it tends to reflect more traditional gender roles in which the man is the main breadwinner (Pahl, 1990). The partners put their incomes into a joint pot; subsequently, the female partner is solely responsible for the money and expenditures. The male partner receives a share of the money for personal spending, and he often has little or no control over the money that comes in to the couple and household expenditures (Pahl, 1995;Yodanis & Lauer, 2007). In couples adopting this strategy, the partners tend to have weaker economic resources or labour market constraints, such as unemployment (Roman & Vogler, 1999;Vogler & Pahl, 1994). Today, female money management is one of the least practised money management approaches in Sweden (Evans & Gray, 2021).
Male money management is similar to female money management, but here, it is the male partner who is solely responsible for both partners' money and finances and who gives a share to the female partner (Pahl, 1990). It is also strongly dominated by institutionally separated gender spheres and often reflects traditional gender roles in the couple in which the woman takes the most responsibility for the home and family, while the man is the main breadwinner (Pahl, 1990). The female partner has little or no control over money and finances, and the inequalities between the partners are often large (Pahl, 1995;Yodanis & Lauer, 2007). Empirical findings have shown that in couples adopting male money management, the husband is often the only full-time employed (Roman & Vogler, 1999;Vogler & Pahl, 1994). Similar to female money management, male money management is fairly uncommon in contemporary Sweden (Evans & Gray, 2021).
In independent management, each partner keeps his or her own money in separate accounts and spends how he or she chooses. The partners operate as two autonomous individuals with rational exchanges and calculations of costs, and each partner takes responsibility for different household expenditures. These couples are often described as having low interdependence (Kenney, 2006;Vogler et al., 2006) and a high need for financial autonomy and privacy (Ashby & Burgoyne, 2008;Fleming, 1997). Compared to the other strategies, this strategy has the highest degree of independence and autonomy. These values are highly prized in Sweden over the life course, not only in relation to couplehood and money (Björnberg & Kollind, 2018) but also in relation to other life domains (e.g., Björnberg, 2012;Haak et al., 2007). Empirical findings in Sweden and Norway have predominantly found independent management in young cohabiting couples and in couples with greater income (Evans & Gray, 2021;Heimdal & Houseknecht, 2003;Lyngstad et al., 2011).
To better understand how the typology may relate to financial disagreements, we adopted conflict theory as a theoretical lens (Hamon, 2016).

Conflict Theory and How Disagreements Arise in Couples
Conflict theory emphasizes the self-interested and competitive side of humans. This emphasis is in line with the dimensions of negotiation and bargaining proposed in economic theories on the division of labour in couples (Hamon, 2016;Sprey, 1969Sprey, , 1979. Such characterization of individuals in a union is not in accordance with a common perception but may help understand underlying mechanisms. Partners' real actions are most often not as calculated as conflict theory assumes, and partners act differently in different situations, often considering the interests of both partners. In couples, there are many ways in which the partners interact (e.g., independence theory, Van Lange & Balliet, 2015; social exchange theory, Nakonezny & Denton, 2008;doing gender, West & Zimmerman, 1987), but in regard to money, conflict theory is often relevant. This theory provides a perspective that considers the dimension of the couple's relationship that helps us focus on the importance of money management, which can be quite concrete and negotiable. This may explain why money is one of the most common topics identified by partners as a source of conflict (Dew, 2011;Henry et al., 2005;Levenson et al., 1993). Disagreements (about money) are also likely to occur because coresidential partners spend much time together and have many common activities and responsibilities (Bartos & Wehr, 2002;Hamon, 2016). Such activities and responsibilities often imply financial prioritization, and partners will not always have the same opinions in this regard. Such different opinions on priorities are likely to be more relevant when resources are more limited, as they tend to be among retired couples.
Conflict theory assumes that cooperation between partners is most often possible; however, for cooperation to take place, the situation must be structured in an equal and non-competitive way. When the partners are unable to reach a consensus, they bargain by using their power, which is translated from personal, material or symbolic resources, often at the expense of the other partner (Bartos & Wehr, 2002;Hamon, 2016). Such bargaining may be more or less present in different money management strategies, as the way in which different ideas about priorities are treated is institutionalized differently, potentially relating to different propensities for financial disagreements. At the same time, partners often have different interests, understandings and perceptions that may create more or less challenging circumstances.

Conflict Theory, Money Management in Couples and the Study Hypotheses
In line with the previous literature review and theory, we test four hypotheses that we argue for below and list them at the end of this section. Conflict theory predicts that disagreements can occur when there are unequal circumstances between the partners and one partner uses bargaining power at the expense of the other (Kelly, 1983). The distribution of power may be more central and noticeable among couples who pool money. When all money is pooled, the partners may have, in theory, more space for discussion, and all money may be subject to supervision and negotiation by the partners (Vogler, 1998). Other scholars have pointed out the challenges of the pooling system, e.g., pooling is not fixed or has clear boundaries (Sonnenberg, 2008), indicating that the partners continuously need to handle financial decisions.
In addition, it is often assumed that marriage should be based on equal sharing and that all money should be shared equally regardless of who contributes what to the household (Burgoyne, 1990). At the same time and in line with conflict theory, there is the idea that partners in some sense 'own' the money they have earned and that they have a right to spend it how they want. The concept of 'own' money is in conflict with the principle of equality (Bennett, 2013;Bennett et al., 2012). Instead, it feeds directly into the idea that the partner who earns the most, oftentimes the man, is believed to be entitled to a larger share or to dominate economic decisions (Kenney, 2006;Vogler et al., 2008) and consumption (Bonke, 2015).
Women's increased earnings give female partners more negotiating power in a relationship. At the same time, there has been an increasing trend of a greater need for financial autonomy and control over what to spend money on compared to the past, when women were predominantly housewives (Pahl, 2008). This ambivalence between sharing all money and striving for independence may be particularly challenging among pooling couples. What the household's money should be spent on can also be differently perceived by women and men (Lawrence et al., 1993). Hence, female and male partners often have distinct and potentially competing interests and different means of realizing them. Such circumstances may not be advantageous for partners who pool money, as they can lead to financial disagreements that are difficult to avoid. This may be particularly true in Sweden, where economic independence is highly valued. To conclude, all these special circumstances (e.g., more space for negotiation, 'own' money, unequal contributions, the increased need for independent money and financial autonomy) indicate that pooling all money may be more challenging than a lower degree of pooling or keeping money separate.
Furthermore, a key issue for money allocation in couples is how much money each partner retains under personal control, and partners may disagree over the amount of money that is allocated as joint money and for joint spending (Vogler, 1998). When partners partially pool their money, each partner has a share of independent money that he or she can use for whatever he or she wants regardless of the other partner's opinion (Vogler & Pahl, 1994). Partial pooling may therefore lead to less tension and bargaining than when money is pooled, under which partners may struggle more to reconcile their conflicting ideas. At the same time, partial pooling may generate greater tension and bargaining situations compared to when all money is separated, as partners must make decisions about the joint money.
The strategy where the man manages all the money and gives a share to the partner has been criticized for restricting women's bargaining power and access to the household's money, negatively affecting women's financial security and wellbeing (Burgoyne, 2004;Kenney, 2006). Earlier research showed that in these couples, the man tends to be the most resource-strong partner (Roman & Vogler, 1999;Vogler & Pahl, 1994). Earlier research also showed that wives tend to spend more money on household-related items than men (Pahl, 1990(Pahl, , 2008. Relatedly, women do not tend to translate their advantage of being in charge of money into power in other areas (Tichenor, 1999). Instead, wives are often perceived as being capable regulators of emotions in the relationship, which has been shown to be important for marital satisfaction (Bloch et al., 2014;Nolen-Hoeksema & Jackson, 2001). It is not unlikely that these features are also positive for fewer financial disagreements. Husbands tend to have less constructive communication, and the resource imbalance often disadvantages wives and households (Evertsson & Nyman, 2021;Helgeson, 2020;Mirgain & Cordova, 2007). Thus, there may be more potential occasions of financial disagreements in couples in which the man manages all the money because this situation may be less fair for both partners and the household compared with when the woman manages all the money.
In line with women's increased labour force participation and earnings, the use of the independent management strategy has increased, and pooling has decreased (Çineli, 2021;Yodanis & Lauer, 2007). These changes indicate that many couples do not consider pooling to be the most optimal strategy (Hiekel et al., 2014). For example, dual-earner couples in Germany have been found to have a higher likelihood of financial disagreements when they pool all their money compared to when they keep their money separate (Lim & Morgan, 2021). Couples who keep all their money separate have high financial autonomy and, at the same time, limited communication about money (Kenney, 2006;Vogler et al., 2006). Financial privacy, which is important for partners (Bennett, 2013), is also greater when money is kept separate (Vogler et al., 2008). Conflict theory assumes that when two parties hold unequal positions, conflict will surface (Bartos & Wehr, 2002;Hamon, 2016). It is possible to avoid conflict when the situation is structured in a non-competitive way, which is more likely when money is kept separate compared to when it is pooled. Therefore partners should not need to use their bargaining power to negotiate about money because they have individual pots and the responsibilities for different household expenditures are often defined. Such circumstances may create fewer occasions for financial disagreements in relation to any form of pooling of money that requires more interaction by the partners, where the boundaries are more blurred (Vogler, 1998) and where one partner often has a stronger position than the other (Kenney, 2006;Vogler et al., 2008).
Pooling may be especially problematic among couples who experience financial stress, such as insufficient resources, as a result of which spending must be handled more carefully. Financial stress has been linked to lower financial satisfaction and relationship quality (Archuleta et al., 2011;Lee & Dustin, 2021). The situation may be further deteriorated because the partners in such circumstances tend to suffer from bad moods, treat each other poorly and engage in less supportive behaviour (Conger et al. 1994).

Hypothesis
Based on the theoretical arguments and previous empirical findings, we formulated four hypotheses on how different money management strategies relate to financial disagreements in the Swedish context.

1.
Couples who pool all their money are more likely to experience financial disagreements than couples who adopt a lower degree of pooling. 2. Couples who adopt some form of money pooling are more likely to experience financial disagreements than couples who keep the money separate. 3. Couples in which the man solely manages all of the money are more likely to experience financial disagreements than couples in which the woman solely manages all of the money. 4. Pooling is more strongly associated with financial disagreements among couples who have financially constrained situations than among couples who are not financially constrained.
Notably, little research has been conducted on how money management strategies operate in practice among older couples and how they relate to relationship dynamics such as financial disagreements, especially in Sweden. We argue for the most logical perspective based on conflict theory, the empirical findings on money management and the Swedish context; however, we are aware that alternative perspectives are also relevant. First, it is likely that there is no association at all between money management and financial disagreements. Second, it is possible that couples who keep all money separate are less likely to experience financial disagreements than couples who adopt some form of pooling.
In support of the latter perspective, LeBaron et al. (2019) found that when a female partner reported that the couple had joint household checking (i.e., bank) accounts, the male partner reported higher relationship quality and relationship stability. In another American study, Addo and Sassler (2010) found that compared to having separate bank accounts, ownership of joint bank accounts was associated with fewer disagreements and higher relationship satisfaction. The study included low-income American couples younger than age 45 with co-resident children surveyed in 2006. Lim and Morgan (2021) found a weak association between joint accounts and less frequent financial disagreements among German couples (aged 20-46), of whom the vast majority were single earners. Their study grouped all couples who did not have joint accounts as 'separate', limiting the conclusions of their results.
These previous studies are not comparable to the present study of older couples in Sweden. First, earlier studies have found that there is a low correlation between having a joint account and being an income-pooling couple (Burgoyne et al., 2007). Second, in the Swedish context, female and male partners are economically more equal, and female labour force participation has historically been high. Consequently, most couples in this context are dual-earner couples, which is also the case among older cohorts. Individualism, independence and gender equality among couples are also more widespread, and they are more generally highly valued in Sweden than in many other countries (Çineli, 2021;Nyman & Reinikainen, 2007). Similarly, Evans and Gray (2021) found that in Sweden and other countries with greater social welfare spending and tax systems that treat couples as individuals, couples had a higher probability of partially pooling their money or keeping their money separate. Therefore, it is likely that in Sweden, with its strong emphasis on economic independence and the widespread acceptance of the ideal of gender equality, it may be more challenging for partners to pool all of their money, as doing so goes against the partners' expectations, behaviour, and social norms.

Data and Sample
We used the nationally representative Swedish Generations and Gender Survey (GGS) conducted in 2012-2013 (Thomson et al., 2015). The Swedish GGS is part of the Generations and Gender Programme (GGP) and includes 9688 individuals aged 18-80 years (response rate of 54%). Because this study focuses on older individuals who are living with their partner, we included respondents who were either married or cohabiting. The age of the respondents was selected as 60 or older, as this study targets retired or soon-to-retired individuals. Hence, the subsample consisted of married or cohabiting respondents aged 60 to 80 born between 1933 and 1953 (n of 1764). Note that the response rate for the Swedish GGS is low compared to the average response rate for countries participating in the first wave of the GGP (54% for the Swedish GGS and 67% for countries participating in the first wave of the GGP). However, the survey was weighted by age, gender, region, country of birth, education, income and family status (Fokkema et al., 2016).

Dependent Variable
The dependent variable was whether the respondent and the partner had financial disagreements, and the survey question was as follows: 'Have you had disagreements within the last 12 months regarding money?' The response options were 'never', 'seldom', 'sometimes', 'frequently' and 'very frequently'. The original variable had a skewed distribution in which a small percentage of the respondents reported financial disagreements. In this study, we distinguished between partners who reported disagreements from partners who did not by categorizing 'never' and 'seldom' responses as 'no' and by categorizing 'sometimes', 'frequently' and 'very frequently' responses as 'yes'. Dichotomizing disagreements on an ordinal scale has been commonly performed in previous research. For example, Van der Lippe et al. (2014) dichotomized disagreements (including financial disagreements) with partners, but they used other categories, that is, 'never', 'less than once a month', 'once a month', 'several times a month', 'once a week', 'several times a week' and 'every day'. Table 1 shows that 11% of respondents reported financial disagreements 'sometimes', 'frequently' or 'very frequently'. A somewhat larger proportion of women reported financial disagreements, but a chi square test did not reveal that the women's and men's reports were significantly different.
Before proceeding to the independent variables, we briefly justify the choice to dichotomize financial disagreements. The variable is on an ordinal scale, and it is very difficult to predict the change from 'never' to 'seldom', from'seldom' to'sometimes', etc. In some cases, dichotomizing variables is justifiable (Farrington & Loeber, 2000). In this study, financial disagreements were severely skewed, and it was quite uncommon for the respondents to report a high degree of disagreements. Hence, it was difficult to predict the values at the end of the tail in the distribution of the variable. Lin and Morgan (2021) kept an equivalent five-scale variable continuous in an ordinary least squares (OLS) model. However, compared to this study, they had a higher mean (1.59 in our study and 2.05 in theirs). In this study, it was difficult to know what level of change would be predicted in an OLS regression. Another reason is that money management does not have a proportional relationship with financial disagreements. We observed two distinct groups, i.e., those who experienced financial disagreements and those who did not, and we argue that the presence of disagreements is the most important issue, not the level of the disagreements. Nonetheless, sensitivity analyses (not presented) using OLS regression models produced results similar to those produced by logistic regression models.

Independent Variables
The key independent variable was how the partners organized their money. The response options to the question 'How is income organized in your household?' were in close accordance with the typology first developed by Pahl (1989). The response options were as follows: 'We pool all money and each take out what we need', 'We pool some of the money and keep the rest separate', 'I manage all the money and give my partner/spouse his or her share', 'My partner/ spouse manages all the money and gives me a share', 'We each keep our money separate', and'Another way'. To further resemble the typology, the 3 rd and 4 th categories identified whether it was the woman or the man who was responsible for the money and who gave the partner a share for personal spending (Vogler & Pahl, 1993;Vogler et al., 2006Vogler et al., , 2008. The final categories were 'pool all money', 'partial pooling', 'the woman manages', 'the man manages', 'keep money separate', and 'another way'. Table 2 indicates that the largest share of the couples pooled all the money (49%), followed by couples using partial pooling (23%). Approximately 20% of the couples kept their money separate. It was relatively uncommon for only the woman or the man to manage the money (4% and 3%, respectively). To put these findings into context, we compared them with findings from Swedish data on younger couples from the 1990s. In the studies from which the findings were drawn, one-person management was the most common, followed by pooling money and keeping the money separate (Ahrne & Roman, 1997;Roman & Vogler, 1999). Another Swedish study from the same period found that pooling was common, but it did not distinguish between different degrees of pooling (Heimdal & Houseknecht, 2003). Additionally, a cross-national study found that one-third of Swedish couples pooled all their money (Treas & Widmer, 2000). Two other cross-national studies that used data from 2012 showed that 50% of Swedish couples aged 18-64 pooled all their money (Çineli, 2021;Evans & Gray, 2021), with Evans and Gray (2021) also finding that 33% partially pooled their money, 13% kept their money separate, and in 4% of couples, one partner managed the money. We conclude that pooling money has been and continues to be common in Sweden, although there is large variation in money management.
Couples who are economically constrained or who experience financial pressure are more likely to report financial disagreements (Dew & Stewart, 2012;Dew & Yorgason, 2010;Henry et al., 2005). Therefore, we adjusted the models for two subjective measures of financial hardship: whether the household normally has some money left for savings and whether the household has difficulties making ends meet. The answers to the question about the couple's possibilities of saving money were 'yes' or 'no'. Overall, 21% reported not having money left for savings. For the question about making ends meet, the answers were 'with great difficulty', 'with difficulty', 'with some difficulty', 'fairly easily', 'easily', and 'very easily'. Overall, 11% had difficulties making ends meet (answers 1-3), and 25% could do so 'fairly easily'. The rest of the couples did not report having difficulties making ends meet (62%, answers 4-5). Ultimately, we would have liked to include both partners' individual earnings, but the survey included only the respondents' joint earnings. Nevertheless, the subjective measures provide a good representation of the couples' financial situation that may produce financial disagreements.
It has often been shown that partners have unequal access to and control over money (Evertsson & Nyman, 2021). We aimed to determine whether the partners in the sample were equally involved in and had control over purchases for the household or whether one of them had a stronger say. Therefore, we included two variables for who makes decisions about routine and expensive purchases for the household. The response options for the corresponding questions regarding routine and expensive purchases were 'always the respondent', 'usually the respondent', 'the respondent and the partner equally', 'usually the partner', and 'always the partner'. We combined the 'always' and 'usually' responses and made the variables gender specific, i.e., the woman decides, the man decides, and both decide. These variables show another dimension of the couples' financial organization and communication in relation to the key indicator. A Cramer's V test for nominal scale factors (Lee, 2003) showed that these variables have moderate correlation, indicating that they are independent measures and predictors of each other (routine purchases 0.11, and expensive purchases 0.09).
We adjusted the models for whether the partners were retired or working (labelled activity status), the partners' educational composition, marital status, and joint children and the respondents' gender and general health (the partners' health was not provided in the data). Other studies have related these factors to couples' financial (and other) disagreements, wellbeing and stability ( Szinovacz, 1996;Van der Lippe et al., 2014). Furthermore, we distinguished between couples in which the partner was at least three years younger, the partner was the same age or ± 1-2 years younger/older, or the partner was at least three years older. Adjusting for age composition addresses relational power dynamics that age composition may produce (Grøntvedt & Kennair, 2013;McKenzie, 2015). The descriptive statistics for the variables in this study are displayed in Table 2. A variance inflation factor diagnostic test for the adjusted variables showed a

Analytical Strategy
We performed logistic regression with the dependent variable of financial disagreements. The key independent variable was money management, and the control variables were money left for savings, difficulties making ends meet, being the primary decision-maker for routine purchases and expensive purchases, the partners' activity status (partners either retired or working), educational composition, marital status, joint children, age composition, and the respondents' gender and general health. Moreover, the results are presented as odds ratios with corresponding significance levels and in the order of the four hypotheses. The first model in Table 3 was gender stratified because we expected women and men to perceive the occurrence of financial disagreements differently, as they tend to have different roles and levels of engagement in the relationship and the labour force. However, the results did not indicate great differences, and the succeeding models were not gender stratified. All models were adjusted for the variables presented above. Additionally, bivariate and stepwise models indicated that the main association between money management and financial disagreements persists in these analyses (not presented due to space limitations). To address hypothesis 4 on financial hardship and money management, we included a separate adjusted model including an interaction between money left for savings and money management. A final and supplementary analysis including an interaction between money management and decision-making in routine purchases further verified the consistency of the findings and the validity of the indicator of money management.

Study Limitations
This study has at least six limitations. First, a multi-item indicator of financial disagreements would nuance the answers, particularly considering that a large share of the respondents did not report disagreements. Similarly, some of the variables were dichotomized, and it would be more informative to use scales or categories (e.g., health and education). However, we tested different categorizations, and ultimately, we chose the most suitable variables for the model fit. Second, it would be informative to have more detailed categories in terms of how the partners managed the household money instead of the somewhat broad categories. However, we grouped couples who were very different, particularly among the partial pooling couples. For example, we cannot address whether the amount that remained under each partner's control was the same or whether one of the partners had a larger share of the income under his or her control.
Third, similarly, it may be restrictive to not consider how and whether the partners' money management may change over time, particularly considering whether the amount that was allocated to the couple was consistent over time or was renegotiated from time to time. Fourth, we do not know whether disagreements or the choice of money management came first. Future research should be sensitive to the fact that money management and financial disagreements may both be causes and effects of each other and the fact that, even on a timeline, it may be difficult to determine which came first. Therefore, we want to be careful to point out that only associations were investigated, not causal effects. Fifth, although the analysis included important factors that may relate to financial disagreements, it omitted many other potential factors that may be central for the association, such as homeownership or other costs of living. A qualitative perspective would also shed light on the co-shaping nature and interconnectedness of the different factors related to financial disagreements. Sixth, financial disagreements were reported by one of the partners, and the partners' ideas of disagreements may be different. Future studies should consider using survey data and qualitative data on both partners in different-sex and same-sex relationships, as doing so may shed light on the gender roles and structures embedded in social and cultural contexts. Table 3 shows the results obtained from a test of the first hypothesis; that is, couples who pooled all their money were more likely to experience financial disagreements than couples who adopted a lower degree of pooling. The results did not support the hypothesis. In Model 1, where all respondents were included, the odds of financial disagreements were 1.96 times higher for those who partially pooled their money compared to couples who pooled all money (reference category). Models 2 and 3, where the sample was stratified by the respondent's gender, show the same patterns for both women and men. Women who partially pooled all money had 1.87 times higher odds of financial disagreements than those who pooled all money, whereas male respondents had 1.20 times higher odds. Model 1 shows that couples were almost twice as likely to experience financial disagreements when the woman managed the money. The association had weak significance (p < 0.10), but in the bivariate model (not presented), it had a significance level of p < 0.001. Furthermore, the odds of financial disagreements were 6.04 times higher for couples in which the man managed the money than for couples who pooled all money. In the gender-stratified models, women had greater odds of financial disagreements when their partner managed all the money (Model 2) than men when their partner managed all the money (Model 3). The cell sizes were relatively small (n = 35 and n = 40), and the results should therefore be taken with caution. Taken together, the results are the opposite of the expectation presented in the first hypothesis, and pooling money is associated with a reduced propensity for financial disagreements compared to the lower degrees of money pooling.

Results
To address the second hypothesis that couples who adopt some form of money pooling are more likely to experience financial disagreements than couples who keep their money separate, the odds ratios in Model 1, Table 3, were recalculated. More precisely, we changed the reference category to 'keep all money separate' in Model 1 and reran the model. The results show that the odds of financial disagreements were 1.49 higher for couples who kept their money separate than for couples who pooled all money. There was no statistically significant difference in the experience of financial disagreements between couples who kept their money separate and couples who partially pooled their money. Additionally, couples in which the woman managed the money were not significantly different from couples who kept their money separate. The results further show that the odds of financial disagreements were 4.1 times higher for couples in which the man managed the money than for couples who kept their money separate (6.11/1.49 = 4.1). To conclude, the results do not support the expectations that couples who adopt some degree of pooling are more likely to experience financial disagreements than couples who keep money their separate. Rather, the results show that the lowest odds of financial disagreements are among couples who pool all money, while the highest odds are among couples in which the man is responsible for the money. There is no statistically significant difference between couples who keep their money separate, who partially pool their money, in which the woman manages the money.
The results in Model 1, Table 3, further show that when the man managed the money, the odds of financial disagreements were 3.2 times higher than those when the woman managed the money (6.11/1.89 = 3.23, p < 0.01). This finding is in line with the expectation in the third hypothesis that couples in which the man solely manages all money are more likely to experience financial disagreements than couples in which the woman solely manages all money.
Next, Table 4 addresses the fourth hypothesis that pooling is more strongly associated with financial disagreements among couples who have financially constrained situations than among couples who are not financially constrained. Whether couples have financial constraints was measured †p < .10, *p < .05, **p < .01, ***p < .001  Table 4 Odds ratios of financial disagreements based on money management and money left for savings from logistic regression Notes: The models are adjusted for difficulties making ends meet, the primary decision-maker for routine and expensive purchases, the partners' activity status, educational composition, marital status, children together, the health status of the responding partner, age composition, and the gender of the responding partner. n = 1764 a The partner who manages gives a share to the other partner for personal spending **p < .01, ***p < .001 by the couples' possibility of saving money, which also indicated the couples' current financial situation and savings for future needs. The adjusted model included an interaction to examine the relationship between money management and the indicator of financial constraints. The reference category was pooling and no money left for savings. The results do not show any statistically significant difference between couples who pooled their money and had no money left for savings and couples who pooled their money and had money left for savings. We wanted to further confirm the results by conducting an additional analysis (not presented) in which the reference category included couples with separate money and no money left for savings. In this analysis, we found that couples who kept their money separate and had money left for savings had higher odds of experiencing financial disagreements compared to pooling couples (or 0.39). The results do not show any statistically significant difference between couples who partially pooled their money and couples who kept their money separate, indicating that partially pooling money or keeping money completely separate has a similarly negative association with financial disagreements for couples with no money left for savings. We conclude that the results do not support the fourth hypothesis that financial disagreements are expected to be more strongly associated among couples who pool all money and have financially constrained situations than among couples who pool and are not financially constrained. Rather, for couples with no money left for savings, partial pooling and keeping money completely separate are associated with higher odds of financial disagreements. Table 5 shows the results from an adjusted model including an interaction between money management and decision-making on daily routine purchases. The interaction supported the validity of the money management indicator, providing a more detailed representation of how money management operates within the couple. The results show higher odds of financial disagreements among partners who made decisions together and partners who either partially pooled their money (or 2.34) or in which the man managed the money (or 7.53) compared to partners who pooled all money and made decisions together. Partners who kept their money separate and mad decisions together were not more likely to experience financial disagreements than those who pooled all money, as these differences were not statistically significant. Additionally, there was no statistically significant difference between couples in which the woman managed the money and couples who pooled all money. The couples in which either the woman or the man was in charge had small cell sizes, making it difficult to draw any firm conclusions. In summary, partial pooling and common decisions about money are associated with a greater likelihood of financial disagreements, but both complete pooling and keeping all money separate in combination with joint decisions are associated with a lower likelihood of financial disagreements.
We end with a note on the adjusting variables in Table 4. Similar to not having money left for saving, couples with difficulties making ends meet had higher odds of financial disagreements than couples without difficulties, and the more severe the financial constraints that the couples had, the higher the odds of having financial disagreements. Partners who shared responsibility for expensive purchases had lower odds of disagreements than those who did not, but there was no statistically significant difference between couples in which the woman or the man mainly made these decisions. In the female sample, retired couples had higher odds of financial disagreements than employed couples, and in the male sample, couples in which both partners were retired had higher odds of financial disagreements than couples in which the man was employed. The variation in financial disagreements was not related to the educational and age composition, marital status, whether the couple had children together, or the respondents' health. Some of these measures are indicators of union stability. As an alternative test for union stability, marital status was replaced with union duration. The results remained the same, presumably Table 5 Odds ratios of financial disagreements based on money management and the partner who decides on routine purchases from logistic regression Notes: Cell sizes in brackets. The models are adjusted for difficulties making ends meet, the primary decision-maker for expensive purchases, the partners' activity status, educational composition, marital status, children together, the health status of the responding partner, age composition, and the gender of the responding partner. a The partner who manages gives a share to the other partner for personal spending. n = 1764 †p < .10, *p < .05, **p < .01, ***p < .001 because the majority of the couples were long-term couples (mean duration of 38 years).

Conclusion and Discussion
In this study, we investigate the link between how older partners in Sweden manage their money and their experience of financial disagreements. The basic expectation is that couples who adopt different money management strategies have different levels of interaction and interdependence. Couples also often have different ideas concerning how money should be spent (Stocks et al., 2007). Conflict theory (Bartos & Wehr, 2002;Hamon, 2016) helps in understanding the underlying mechanism between different money management strategies and financial disagreements.
The theory leads to the expectation that pooling all money is associated with financial disagreements because it may be challenging to pool all money. For example, in these couples, all money is subject to supervision, and the partners continuously need to handle decisions about money. We therefore expected that both partial pooling and keeping all money separate would be beneficial, as partners have less to decide on together, consequently leading to fewer occasions for disagreements. These strategies provide more financial autonomy and less interdependence, which is highly valued in Sweden. However, the results are the opposite of our expectations: Couples who pool all their money are less likely to experience financial disagreements compared to both couples who keep their money separate and couples who partially pool their money. Hence, conflict theory does not fully explain financial disagreements among older couples in this study.
Instead, being a couple may include sharing behaviour, and pooling may be a way to indicate solidarity within the couple. Solidarity may particularly be true among older couples who have been together for long durations, where the boundaries between what is 'mine' and 'yours' may be more blurred. It may also be a matter of trust in which pooling couples have reached the stage in life at which each partner full trusts the other. Hence, money may not be a source of tension and disagreements, but how it is handled may better reflect the partners' satisfaction with the relationship. As others have argued (Fleming, 1997), partial pooling couples combine two conflicting views-togetherness and autonomy-and simultaneously operate as single economic units while keeping their money separate. As opposed to being beneficial, as we argue here, it may be difficult and challenging for partners to juggle two different systems. Our finding that partial pooling is associated with a lower propensity for financial disagreements is consistent with that of Lim and Morgan (2021). They found that younger dual-earner German couples who adopted partial pooling more frequently had financial disagreements than dual-earner partners who pooled all money. However, these authors had multiple and more detailed measures of partial pooling, including the partners' share of the contribution to the joint pot. Hence, their measure may not be entirely comparable to this study's measure, which is broader. Furthermore, Lim and Morgan's (2021) study did not include couples aged 60-80.
Moreover, couples who partially pool their money and couples who keep their money separate may face similar challenges in financial decision-making. Older partners often have unequal resources and may not have the same spending possibilities (Swedish Social Insurance Inspectorate, 2017). Hence, striving for financial independence may not be beneficial for couple wellbeing in later life. Instead, a joint financial effort may be a more functional strategy for maintaining a higher quality of life, producing fewer occasions for financial disagreements. This study's results may be discussed in terms of the 'greedy marriage thesis' according to which marriage is perceived as a 'greedy' institution that makes exclusive claims on partners' resources while pressing them to cut demands outside of the union (Coser, 1974). By taking into account couples' characteristics, such as how they manage household money, this study sheds light on why some couples may be more inward-looking than others (see, e.g., Kim & Dew, 2016). In a 'greedier' couple, the partners are more likely to pool their money and invest less in their own needs, such as financial freedom. In doing so, the partners fulfil the couple's financial needs, and they will be happy making sacrifices for their union.
Equal and non-competitive behaviours of partners and a lack of overriding of one of partner by the other seem to be good strategies to avoid relationship disagreements, according to conflict theory (Bartos & Wehr, 2002;Hamon, 2016). The analyses show that scenarios under which the woman manages the money are less likely to be associated with financial disagreements than scenarios under which the man manages the money. Assuming that disagreements arise due to an (gendered) imbalance or inequality within the couple, this finding is also in line with research that has found women to be more egalitarian in money allocation and spending (Tichenor, 1999) and to communicate more constructively with their partners than men (Bloch et al., 2014), which tends to reduce disagreements (Wilmarth et al., 2014).
In line with conflict theory, we expect that financially constrained couples who pool all money are in a particularly challenging situation, as financial issues must be carefully negotiated and discussed. Partners may have different opinions regarding what must be prioritized, and they often do not have the same ideas (Pahl, 1990(Pahl, , 2008. Contrary to our expectations, we find that in couples who did not have money left for savings, financial disagreements are most likely to arise among couples who partially pool their money and keep their money separate. It is plausible that among couples who keep any money separate, conflicts can arise when partners disagree on joint spending, which may be especially challenging when resources are somewhat limited. This finding is in line with Addo and Sassler's (2010) finding that low-income couples with small children in the US have a better relationship quality when they pool money and Evans and Gray's (2021) explanation for why cohabiting couples in Latin America tend to pool money to a greater extent. These couples' economically constrained situation may make pooling a necessity.
This study indicates that couples who decide together on purchases have the same likelihood of financial disagreements in cases where they keep their money completely together or separate. This result is in line with the idea in conflict theory that cooperation and agreement between partners can prevent disagreements. It is also in line with LeBaron et al.'s (2019) finding that couples who approach household financial challenges as a team have greater relationship quality and stability. Team spirit may also explain why partial pooling couples have an increased likelihood of disagreements, even in situations in which both partners decide. Conflict theory assumes that when couples cannot reach a consensus, they tend to have disagreements, which may be especially challenging for these couples.
In summary, we find that in three cases, the theory and hypotheses are rejected, and in two cases, we find support for the theory. Contrary to conflict theory, we find that couples who pool all money are less likely to experience financial disagreements (also in financially constrained situations) than both couples who partially pool their money and couples keep their money separate. In support of the theory, we find that couples in which the man manages the money are more likely to experience financial disagreements than couples in which the woman manages. Additionally, in situations where there is communication between the partners about joint purchases, both pooling couples and couples who keep their money separate are less likely to experience financial disagreements.

Implications for Policy, Practice, and Future Research
This study's results suggest a set of policy implications and future research directions. First, diversity in money management is challenging when older couples are being evaluated for financial benefits. Some individual needs may be hidden, as household income in Sweden (and many other countries) is considered in these cases, and the size of household income can be an indication of a more complex situation. Hence, policy-makers should take into account the fact that a significant share of older couples do not fully pool their money. There may be a need to improve the efficiency of the means-tested benefits for older individuals and households to ensure that the benefits are rightfully distributed.
Second, the relatively high proportion of older couples who experience financial disagreements in relation to how they manage their money shows that there may be a need to evaluate the financial support systems available for older couples, such as support in organizing their finances provided by pension agencies, banks (including debt settlement companies) or municipalities (such as social workers or other social support). Financial educators and planners may also benefit from the study's knowledge by taking into account that it may be important to design programmes that strive to help older retired couples plan and manage their money (see, e.g., Schreiner et al., 2002, on the benefits of having tailored programmes for financially vulnerable groups). In Sweden, such programmes could be included in the consumer advising provided by municipalities (and pensioners' organizations). Third, in the most vulnerable cases, an alternative support may be to offer affordable couple counselling for older couples with the aim of facilitating discussion about their financial situation and wellbeing. Therapists, psychologists and counsellors who work with the wellbeing of older individuals and couples may recognize during the therapy process that money management strategies are strongly related to disagreements and financial stability. In these circumstances, it may also be informative to interview older individuals and/or couples about their money management choices over the course of their relationship. Doing so may be informative because money management may be one area of relationship problems that may be a potential explanation for other problems.
Fourth, researchers and decision-makers may strive to collect more comprehensive data on the association between money management strategies and relationship dynamics, for example, by identifying sub-categories of these broad management strategies. These data may take into account the development of information technology (IT) and webbased and mobile options that may either hinder or facilitate money sharing among older couples (e.g., online banking services and mobile payment systems). In addition, longitudinal data would further advance our understanding of the causality between money management choices and relationship outcomes, and qualitative studies would go beyond the statistical representations.
Fifth, a deeper understanding of financial behaviour and gender dynamics will also become increasingly important as more individualistic and egalitarian age cohorts replace older, more traditional cohorts. As others have observed elsewhere (Hamplová et al., 2014), Pahl's (2005) typology is often applied; however, most current empirical studies primarily distinguish among pooling, partial pooing and the separation of money (Hamplova & Le Bourdais, 2009;Heimdal & Houseknecht, 2003;Lyngstad et al., 2011). In this study, the one-person strategies tended to be relatively uncommon. We recommend that future studies should consider updating the current typologies so that they are more applicable to today's older and younger couples. Future investigations may study the channels through which money management strategies operate and are sourced for relational wellbeing and stability over the life course.

Concluding Remarks
This study indicates that the ways in which older couples manage money relate differently to their propensity for financial disagreements. Pooling couples have the lowest probability of disagreements. Even in a country such as Sweden where, e.g., economic independence is highly valued, these patterns are visible. Clearly, this study does not reveal whether couples who do not pool all money would have the same or even a higher propensity for disagreements if they pooled all their money. Given that individuals act rationally, they should choose the strategies that are the least problematic for them individually and for the couple. However, it may be challenging for decision-makers and practitioners that a large proportion of older couples do not pool their earnings and that these couples also have a greater probability of financial disagreements. In the worst cases, partners may have poor financial wellbeing. Disagreements in one domain are also often interrelated with other disagreements that may co-shape negative outcomes for the individual and the couple.
Funding Open access funding provided by Stockholm University.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visithttp:// creat iveco mmons. org/ licen ses/ by/4. 0/