The most recent recession from December 2007 to June 2009 (U.S. Bureau of Labor Statistics 2012) has been called the worst economic downturn since the Great Depression (Baek and DeVaney 2010; Mosely 2009). As members of the “sandwich” generation, mid-life couples are caught in a precarious balancing act between caring for children in the home and aging parents, while saving for retirement and their children’s college educations and marriages (Nema and Bansal 2015). During an economic downturn, these same couples may struggle meeting these demands economically while developing doubts about whether they will have adequate job security and health coverage, sufficient financial reserves for retirement, or enough funds to cover rising living costs amid lower returns on investments (McDaniel et al. 2013). These concerns may manifest as financial distress (perceived economic hardship) over a period of several months or years, which are likely to impact marital interactions and quality (Conger et al. 1990; Gudmunson et al. 2007).

One way financial distress may influence marital interactions and marital quality is through marital conflict and the tactics couples use to manage this conflict. Gottman’s (1999) research on couple conflict identified both positive and negative conflict tactics. Positive tactics may include turning towards one’s spouse instead of turning away from the spouse. However, when facing conflict, couples often engage in negative tactics, namely criticism, contempt, defensiveness, and stonewalling, which tend to push partners away from each other. Gottman’s research has ultimately shown these negative conflict tactics to harm marital quality.

Research has shown that conflict about money reduces relationship satisfaction and increases the probability of divorce (Britt and Huston 2012; Dew et al. 2012). Additionally, Papp et al. (2009) noted conflict about money was generally longer lasting, more recurrent, and more difficult to resolve than other sources of marital conflict. Papp et al. further suggested conflict about money was related to more anger from husbands and more depressive behaviors from both husbands and wives, which may promote a self-defeating cycle of unresolved conflict and increasingly negative conflict tactics, including aggressive behaviors (e.g. hostility, psychological aggression, etc.), which may create further distance between spouses, especially if occurring during times of economic hardship and financial distress (Conger et al. 1990; Falconier and Epstein 2010).

While researchers have examined hostility and psychological aggression in the context of economic hardship and financial distress (Conger et al. 1990; Falconier and Epstein 2010), they have not examined relational aggression within this context. Relational aggression consists of behaviors that create distance in couples, including withdrawing one’s love and affection when upset with a spouse or spreading rumors about or trying to embarrass one’s spouse in front of other people (Carroll et al. 2010). Relational aggresion may reflect what Anderson and Sabatelli (2011) referred to as triangulation, which occurs when a spouse seeks out a third person to help relieve tension within the relationship or as a way to vent about the situation. Relational aggression has been shown to be common within long-term and relatively stable marriages (Carroll et al. 2010) and to be harmful to marital outcomes (Carroll et al. 2010; Coyne et al. 2017). However, relational aggression has not been examined within the context of economic hardship and financial distress. As such, the current study adds an important element to the literature by exploring the degree to which relational aggression mediates the relationship between financial distress and marital quality in a sample of mid-life married, heterosexual couples following the recent economic recession.

History of Research on Economic Hardship and Financial Distress

Research examining the relationships between economic hardship, financial distress, and marital outcomes began in the 1980s when Liker and Elder (1983) examined couples’ experiences during the Great Depression. This early research suggested that lost income was associated positively with irritability and moodiness among husbands and financial conflict and marital tension among couples, all of which were associated with weakened marital relationships. Additional research during the Midwestern United States’ Farm Crisis of the 1980s further validated the harmful effects that economic hardship and financial distress can have on marital relations (Conger et al. 1990; Johnson and Booth 1990).

From this latter research from the Farm Crisis, Conger et al. (1990) developed the Model of Economic Hardship (also known as the Family Stress Model; Conger and Dogan 2007). This model included economic hardship, measured as unstable work, change in income, and income-to-needs ratio, which represented objective stressor events with limited variability (i.e. hardship either occurs or does not occur), and economic strain (also referred to as financial distress), which represented a wider range of subjective emotional, cognitive, or behavioral responses to this hardship (Conger et al. 1990; Valentino et al. 2014). Conger et al. found a positive relationship between economic hardship and economic strain, which supported research based on the ABC-X Model (Hill 1949), which posited that distress is not felt unless the stressor event is recognized as being stressful. Conger et al.’s model also expanded Hill’s research by noting this response of economic strain influenced marital processes, having found husbands’ economic strain was related positively to hostility and negatively to warmth towards their wives, which, in turn, were related negatively to marital outcomes.

The Model of Economic Hardship (Conger et al. 1990) has been validated among varying family structures, nationally and internationally (e.g. Aytac and Rankin 2009; Conger and Dogan 2007; Falconier and Epstein 2010; Kwon et al. 2003). Despite the validation provided for this model, uncertainty remains about the relationships involved as they pertain to specific groups and situations. Related to the current study, less is known regarding whether this model applies to dual-income, mid-life married couples following the most recent economic recession. One area where prior research in this area is mixed is whether financial distress is linked directly to marital outcomes, or whether specific mechanisms account for this link indirectly.

Indirect Versus Direct Influences of Financial Distress on Marital Outcomes

Existing research provides examples of financial issues having both direct and indirect influences on marital outcomes. Financial stressors (Archuleta et al. 2011), financial distress (Johnson and Booth 1990), financial disagreements/financial unfairness (Dew 2011), and frequency of money arguments (Britt and Huston 2012) are related directly to lower relationship satisfaction and more relationship dissolution. Alternately, research also supports the notion that finances are associated with marital outcomes indirectly through a variety of intervening variables, including hostility (Conger et al. 1990), locus of control and depression (Dew 2007), couple disagreements, conflict, and quality time together (Gudmunson et al. 2007), and emotional distress and marital conflict (Kwon et al. 2003).

Some studies have found only indirect associations between financial distress and marital outcomes (Conger et al. 1990; Falconier and Epstein 2010). For example, Conger et al. examined whether there was a direct relationship between economic strain and marital quality, but they did not find such a relationship and excluded such pathways in subsequent analyses, focusing instead on the indirect associations operating through hostility and warmth. Subsequent research on financial distress and marital outcomes has followed this indirect model without describing whether researchers examined the possibility of direct pathways between financial distress and marital outcomes (e.g. Gudmunson et al. 2007; Kwon et al. 2003). Given the limited research, it may be warranted for studies to test both direct and indirect pathways to determine whether and when these associations exist between financial variables and marital outcomes. Aytac and Rankin (2009), for example, found both a direct relationship between financial hardship and marital problems, as well as an indirect relationship operating through emotional distress. The current study proposes that economic hardship will be related directly and positively to financial distress, as earlier research has supported (e.g. Conger et al. 1990; Hill 1949). Additionally, it is expected that financial distress and marital quality will be related directly and negatively as found in recent research (Britt and Huston 2012; Dew 2011). This direct relationship also will be mediated by an intervening variable, namely perceptions of relational aggression.

Gender Differences in Behaviors Associated with Financial Distress

Boss (2002) suggested that each member of a family system may exhibit different responses when exposed to a shared stressor event, such as economic hardship, thus creating a need to examine multiple members of that family system within the same analytical model. Existing research has noted that when financially distressed, men demonstrate more irritability and moodiness (Johnson and Booth 1990), hostility (Conger et al. 1990), and psychologically aggressive behaviors (Falconier and Epstein 2010) than women do. Some men may engage in conflictual behaviors when distressed, in part, because of their traditional role as the primary household provider (Gudmunson et al. 2007), and may feel added pressure to provide adequately for their families during times of an economic downturn. However, as dual-income households have become more prevalent in American society, in part because of the rising cost of living and housing, does this distinct gender difference still apply to married couples in which the majority of couples are dual-income households?

In addition to pressure to provide adequately for one’s family, it has been suggested that, when financially distressed, men are more likely to exhibit externalizing behaviors, whereas women are more likely to exhibit internalizing behaviors (Gudmunson et al. 2007; Leinonen et al. 2002). Conger et al.’s (1990) research partially supported this claim showing that men engaged in increased hostility when financially distressed. Falconier and Epstein (2010), on the other hand, found both men and women engaged in more psychological aggression when men were financially distressed, which supports Conger et al.’s original hypothesis, but it does not support the claim that externalizing and internalizing behaviors are gender specific. Furthermore, financial distress has been shown to be related to higher levels of emotional distress and depression among both women and men (Conger et al. 1999; Gudmunson et al. 2007). The results of these studies challenge the claim that externalizing and internalizing behaviors are gender specific within the context of financial distress.

A more plausible explanation for when gender differences may or may not emerge is that responses to financial distress are dependent on the specific externalizing behaviors being examined. Falconier and Epstein (2010) found women engage in externalizing behaviors in the form of psychological aggression when their partner was financially distressed. This engagement may be in response to their partner’s increased aggressiveness (e.g. reciprocated aggression). However, it also may be that the measures of psychological aggression used by Falconier and Epstein (2010) overlap with relational aggression, a form of aggression common to marriage, even among well-established and relatively stable marriages (Carroll et al. 2010). This overlapping element consists of hostile withdrawal (refusal to talk to the partner about a problem; Falconier and Epstein 2010), which is similar to the relational aggression behavior of giving a partner the silent treatment or withdrawing affection when upset (Carroll et al. 2010; Crick et al. 1999). Research on relational aggression within a marital context has shown that wives engage in these behaviors more than husbands do, and that these behaviors have a negative association with marital outcomes (Carroll et al. 2010; Coyne et al. 2017). Earlier research has found hostile and psychologically aggressive behaviors operate through husbands’ financial distress, but not wives’ financial distress (Conger et al. 1990; Falconier and Epstein 2010). However, if women engage in more relationally aggressive behaviors than men under non-distressed circumstances (Carroll et al. 2010; Coyne et al. 2017), the question arises, might husbands and wives engage in higher levels of perceived relationally aggressive behaviors when either spouse is financially distressed? This question is explored in the current study, which is the first to explore relational aggression within the context of financial distress.

The Current Study

The purpose of the current study is to apply the model of economic hardship (Conger et al. 1990) to heterosexual, mid-life married couples following the recent economic recession. Specifically, this study explores the relationships between economic hardship, financial distress, and marital quality and whether perceptions of relationally aggressive behaviors mediate the relationship between financial distress and marital quality. To explore these relationships a modified and dyadic longitudinal mediation model (MacKinnon 2008; Reyes et al. 2015; a conceptual model can be found in Fig. 1) is used to test the following hypotheses:

Fig. 1
figure 1

Conceptual modified longitudinal mediation model (MacKinnon 2008; Reyes et al. 2015). Although not shown in the conceptual model, this model examined couple data dyadically, using an actor-partner interdependence model approach, and controlling for couple-level income, race/ethnicity, wife and husband level of education, and length of marriage

Full size image

H1

Couple-level economic hardship will be related positively to both spouses’ financial distress concurrently in 2009 and 2010, and economic hardship in 2009 will be related positively to both partners’ financial distress in 2010.

H2

Financial distress will be related negatively to both spouses’ marital quality concurrently in 2009 and 2010 and demonstrate longitudinal associations 1 year later.

H3

Both wives’ and husbands’ financial distress will be related positively to both spouses’ perceptions of partner relational aggression concurrently in 2009 and 2010, and financial distress in 2009 will be related positively to both spouses’ perceptions of partner relational aggression in 2010.

H4

Perceptions of partner relational aggression will be related negatively to both spouses’ marital quality concurrently in 2009 and 2010, and demonstrate longitudinal associations 1 year later.

H5

Perceptions of partner relational aggression will mediate the relationships between financial distress and both spouses’ marital quality concurrently in 2009 and 2010, and demonstrate the mediated associations 1 year later.

Method

Data for this study came from the Flourishing Families Project (FFP), a longitudinal study begun in 2007 that examined family and parental processes. The original data were collected from 500 families living in the Pacific Northwest, all of whom had a focal child between the ages of 10 and 14. Using the Polk Directories/Info USA telephone survey database and targeted census tracts mirroring local school districts, 692 eligible families were randomly selected and contacted by letter regarding participation in the study (Day et al. 2013). Phone calls and home visits later confirmed participation in the study and established eligibility and consent. Of the 692 families originally contacted, 423 (61%) agreed to participate. Recognizing that lower socioeconomic families were underrepresented, additional recruitment efforts (e.g. referrals, fliers, etc.) added 77 additional families to the study. Of these 500 families, 165 consisted of single-parent households, while 335 consisted of two-parent, heterosexual, married households. These married couples are the focus of the current study. The average age in 2009 was 45.5 years for wives (SD = 5.38, range: 27–59) and 47.3 years for husbands (SD = 5.97, range: 27–62). Couples were married an average of 19.8 years (wives: SD = 5.2, range: 2–40; husbands: SD = 4.94, range: 2–37). Nearly 70% of both wives and husbands had earned a Bachelor’s degree or higher, and fewer than six percent had received less than a high school diploma. The average yearly combined income fell within the $90,000–$100,000 range, with less than one percent of couples reporting an income under $20,000 per year. Although not a primary focus of the study in its beginning, pertinent financial variables were brought into the study starting in 2009 in response to the economic recession that began in 2007. The current study used data collected from waves 3 (2009), 4 (2010), and 5 (2011).

Measures

Marital Quality

Marital quality at all three time points (2009, 2010, & 2011) consisted of 5 of the original 6 items created for the Quality Marriage Index (Norton 1983). Responses ranged from 1 (very strongly disagree) to 6 (very strongly agree). An example item included “We have a good relationship” (All items can be found in the Appendix). Scores were averaged for each partner. Higher scores indicated higher levels of marital quality. Scales demonstrated good reliability (α = .97 for both wives and husbands). Marital quality in 2009 and 2010 were used both as outcome variables and to account for stability in marital quality across the length of the study (Table 1 contains descriptive statistics for variables of interest).

Table 1 Descriptive statistics for all variables exploring the relationship between economic hardship, financial distress, relational aggression, and marital quality (N = 335)
Full size table

Economic Hardship

Conger et al. (1990) originally identified three potential measures of economic hardship: unstable work, changes in income, and income-to-needs ratio. However, Conger et al. (1994) also developed a 13-item questionnaire for use in the Family Transitions Project, which measured economic hardship by asking questions regarding cutbacks in salary or work hours, being laid off, having to sell property due to financial difficulties, losing property to foreclosure or repossession, and receiving government assistance. While responses to all 13 of these questions were collected in the Flourishing Families Project, the current study used only two of these items to measure economic hardship in 2009, namely work cutbacks during the previous 12 months: (a) taking a cut in wage or salary (1 = yes, 0 = no) or (b) having work hours reduced (1 = yes, 0 = no). These items reflect changes in income, reflecting one aspect of Conger et al.’s (1990) original concept of economic hardship. Scores were combined into a single couple-level measure of economic hardship (ranging from 0 to 4), with higher scores indicating greater economic hardship.

Financial Distress

Financial distress was measured with an 11-item scale from Spilman and Burzette (2006), which assessed two domains of financial distress: financial concerns (5-items; e.g. “I have trouble sleeping because of my financial problems”) and financial constraints (6-items; e.g. “I have enough money to afford the kind of food that I need”). Responses ranged from 1 (strongly disagree) to 5 (strongly agree). After reverse coding responses to the financial constraint items, all eleven items were combined into a single financial distress scale for each spouse (Cronbach reliability scores ranged between .90 and .92 for both waves of data), where higher scores indicated higher levels of perceived financial distress.

Perceptions of Relational Aggression

Perceptions of relational aggression measured the respondent’s perception that his/her spouse engaged in relationally aggressive behaviors. It was measured using 12-items adapted from the Self-Report of Aggression and Victimization in Marriage (SRAV-M; Nelson and Carroll 2006; all items can be found in the Appendix), with responses ranging from 1 (not at all true) to 7 (very true). Following Carroll et al.’s (2010) precedent, relational aggression was separated into two distinct domains: social sabotage and love withdrawal. Social sabotage consisted of 6-items (reliability scores ranged from .86 to .89 across both waves) that measured the degree to which spousal behaviors were used to embarrass the respondent during times of conflict (e.g. “My partner tried to embarrass me or make me look stupid in front of others”). Love withdrawal also consisted of 6-items (reliability scores ranged from .87 to .89 across both waves) that measured the degree to which spouses withdrew affection and support from the respondent during conflict (e.g. “My partner gives me the silent treatment when I hurt his/her feelings in some way”). Scores were averaged for both husbands and wives. Higher scores represent greater perceived relational aggression being used against the respondent by his/her spouse.

Control Variables

The current study controlled for combined household income, whether both spouses shared a common race/ethnicity [measured as both partners being white (1) or non-white (0)], wives’ and husbands’ education levels, and the length of marriage.

Plan of Analysis

All frequencies and descriptive statistics were calculated using SPSS Version 25 (Table 1). A modified and dyadic longitudinal mediation model (MacKinnon, 2008; Reyes et al. 2015) was estimated in MPlus (Muthen and Muthen 2011) to explore the relationships between economic hardship, financial distress, relational aggression, and marital quality. Full information maximum likelihood (FIML) was used to estimate models in the presence of some missing data. Model fit was assessed with indices of chi square, comparative fit index (CFI), Tucker-Lewis Index (TLI), and the root mean square error of approximation (RMSEA) (Hu and Bentler 1999). To test mediation, we examined the relationships between financial distress and marital quality first without the mediating variables (i.e., relational aggression) in the model and then again with the mediating variables in the model. If a significant relationship between financial distress and marital quality existed in the first model, then perceptions of relational aggression could be tested as a mediating variable. However, if no significant relationships existed between financial distress and marital quality, then no mediation could take place (Ul Hadi et al. 2016). In the model with indirect effects a bootstrapping approach was used (Preacher and Hayes 2008), to adjust the standard errors and to estimate bias-corrected confidence intervals for indirect effects. Without the bootstrapping approach, standard errors of indirect effects can otherwise be artificially inflated (Preacher and Hayes 2008). Additionally, Wald chi square parameter tests were used to examine whether certain pathways should be constrained to be equal, such as whether couple-level economic hardship influenced husbands’ and wives’ financial distress equally.

Results

The descriptive statistics (Table 1) showed participants experienced low levels of economic hardship and financial distress in both 2009 and 2010. Perceptions of spousal engagement in relationally aggressive behaviors were also relatively low at both waves, although husbands reported their wives engaged in these behaviors more frequently than wives reported their husbands engaging in these same behaviors. Finally, participants experienced relatively high levels of marital quality across all three waves of data.

An examination of the bivariate correlations (Table 2) offered preliminary support for the proposed hypotheses. Couple-level economic hardship was related positively with both spouses’ financial distress across both waves, but it was not related to either spouse’s perceptions of relational aggression or marital quality. Both wives’ and husbands’ financial distress was related positively to both spouses’ perceptions of relationally aggressive behaviors and negatively to both spouses’ marital quality, again at all time points. Perceptions of social sabotage and love withdrawal from either spouse were related positively to each other and negatively to both spouses’ marital quality across time. Spouses’ reports of marital quality were positively associated with each other. All variables were relatively stable across waves.

Table 2 Bivariate correlation coefficients for variables of interest
Full size table

To address the study hypotheses, a modified and dyadic longitudinal mediation model with indirect effects was estimated. The model, presented in Fig. 2 and supplemental Fig. S1, fit the data well (chi square = 288.16, df = 159, p < .001; CFI = .97; TLI = .94; RMSEA = .05, ns).

Fig. 2
figure 2

Simplified fitted modified longitudinal mediation model exploring the relationships between economic hardship, financial distress, and marital quality, as mediated by perceptions of relational aggression (N = 335); Model controls for couple-level income, race/ethnicity, wife and husband level of education, and length of marriage. The complete model with pathway coefficients for all significant pathways can be found among the online supplemental materials (Figure S1). Fit statistics: chi square = 288.16, df = 159, p < .001; CFI = .97; TLI = .94; RMSEA = .05, ns. Solid lines represent significant pathways for both partners; Dashed lines represent significant actor effects; Long dash/dot dot lines represent significant partner effects

Full size image

Our first hypothesis proposed couple-level economic hardship in 2009 would be related positively to both spouse’s financial distress in 2009 as well as 2010. This hypothesis was supported for both spouses at both time points (2009: β = .24, p < .001; 2010: β = .04, p < .05). Additionally, it was proposed economic hardship experienced in 2010 would also be related positively to both spouses’ financial distress in 2010; this part of the hypothesis also was supported (β = .06, p < .05).

Our second hypothesis stated both spouses’ financial distress in 2009 would be related negatively to both spouses’ marital quality in 2009 and 2010. Additionally, we predicted both spouses’ financial distress in 2010 would be related negatively to both spouses’ marital quality in 2010 and again in 2011. This hypothesis was supported partially in that wives’ financial distress in 2009 was related negatively to wives’ marital quality in 2009 (β = − .19, p < .001), demonstrating an actor effect. Additionally, both wives’ and husbands’ financial distress in 2009 was related negatively to husbands’ marital quality in 2009 (wives: β = − .19, p < .05; husbands: β = − .21, p < .001), demonstrating both an actor and a partner effect. However, neither partners’ financial distress in 2009 was related to either partners’ marital quality in 2010, nor was financial distress in 2010 related to either partner’s marital quality in 2010 or 2011. Because direct relationships were found between financial distress and marital quality in 2009, it was possible to test whether perceptions of relationally aggressive behaviors mediated these relationships, which will be discussed in hypothesis 5 below.

Our third hypothesis posited that both spouses’ financial distress in 2009 would be related positively to both wives’ and husbands’ perceptions their spouse engaged in relationally aggressive behaviors in 2009 and 2010. Additionally, financial distress in 2010 would be related to both spouses’ perceptions their partners engaged in relationally aggressive behaviors in 2010. This hypothesis was supported partially. Wives’ financial distress was related positively to wives’ perceptions their husbands engaged in both social sabotage (β = .26, p < .001) and love withdrawal (β = .12, p < .05) in 2009, demonstrating actor effects. Actor effects were also found as husbands’ financial distress in 2009 was related positively to husbands’ perceptions their wives engaged in social sabotage (β = .26, p < .001) and love withdrawal (β = .12, p < .05) in 2009. However, both partners’ financial distress in 2009 was unrelated to both partners’ perceptions of social sabotage or love withdrawal in 2010. Wives’ financial distress in 2010 was related positively to wives’ (actor) reports their husbands engaged in social sabotage in 2010 (β = .14, p < .01), while husbands’ financial distress in 2010 was related to husbands’ (actor) perceptions their wives engaged in social sabotage in 2010 (β = .14, p < .01). No relationship was found between either partner’s financial distress in 2010 or either partner’s perceptions of love withdrawal in 2010.

For our fourth hypothesis, it was expected that perceptions of relational aggression in 2009 would be related negatively to both spouses’ marital quality in 2009 and 2010. Furthermore, perceptions of relational aggression in 2010 would be related negatively to both partners’ marital quality in 2010 and 2011. This hypothesis was supported fully for associations in 2009. Specifically, wives’ perceptions their husbands engaged in social sabotage was related negatively to wives’ (β = − .18, p < .001) and husbands’ marital quality (β = − .10, p < .05). Wives’ perceptions their husbands engaged in love withdrawal was also related negatively to wives’ (β = − .13, p < .01) and husbands’ marital quality (β = − .15, p < .01). Husbands’ perceptions their wives engaged in social sabotage was also related negatively to husbands’ (β = − .27, p < .001) and wives’ marital quality (β = − .13, p < .01). Finally, husbands’ perceptions their wives engaged in love withdrawal was related negatively to husbands’ (β = − .14, p < .01) and wives’ (β = − .13, p < .01) marital quality.

No relationships were found between perceptions of social sabotage or love withdrawal in 2009 and marital quality in 2010. Thus, this part of the hypothesis was not supported. The hypothesis that 2010 reports of relational aggression would be related negatively to either spouses’ marital quality in 2011 was only supported partially. Perceptions of social sabotage demonstrated partner, but not actor, effects in that wives’ perceptions their husbands engaged in social sabotage in 2010 was related negatively to husbands’ marital quality in 2011 (β = − .09, p < .05). Additionally, husbands’ perceptions their wives engaged in social sabotage in 2010 was related negatively to wives’ marital quality in 2011 (β = − .10, p < .05). Perceptions of love withdrawal demonstrated actor, but not partner, effects in that wives’ perceptions their husbands engaged in love withdrawal in 2010 were related negatively to wives’ marital quality in 2011 (β = -.10, p < .05). Husbands’ perceptions their wives engaged in love withdrawal in 2010 were related negatively to husbands’ marital quality in 2010 (β = − .11, p < .05).

Our fifth and final hypothesis was that perceptions of relational aggression would mediate the relationships between financial distress and marital quality. Preliminary analysis without relational aggression in the model revealed only three direct and significant associations between financial distress and marital quality in 2009 from which mediation could be tested. The inclusion of perceptions of relational aggression partially mediated these associations as the strength of these associations decreased once perceptions of relational aggression were included in the model. Specifically, the association between wives’ financial distress and wives’ marital quality in 2009 decreased from β = − .19, p < .001 without relational aggression to β = − .12, p < .05 with relational aggression. The association between wives’ financial distress and husbands’ marital quality in 2009 also decreased (without relational aggression: β = − .19, p < .001; with relational aggression: β = − .12, p < .10), as did the association between husbands’ financial distress and husbands’ marital quality in 2009 (without relational aggression: β = − .21, p < .001; with relational aggression: β = − .13, p < .05).

Bootstrapping analysis also allowed for detection of both direct and indirect effects within the current model. As marital quality in 2011 was the final outcome of interest, only indirect associations with this outcome are highlighted here (Tables 3, 4, 5 contain a breakdown of all direct and indirect effects for the current study; Confidence Intervals not reported in Tables 3, 4, 5 are available from the first author). Couple-level economic hardship in 2009 was related indirectly and negatively to both partners’ marital quality in 2011 (wives: β = − .08 (C.I. = − .13 to.04), p < .05; husbands: β = − .11 (C.I. = − .17 to − .07), p < .05). Both partners’ financial distress in 2009 was also related negatively to wives’ marital quality in 2011 (wives: β = − .14 (C.I. = − .25 to.03), p < .01; husbands: β = − .14 (C.I. = − .25 to − .03), p < .05) and negatively to husbands’ marital quality in 2011 (wives: β = − .15 (C.I. = − .27 to .04), p < .05; husbands: β = − .22 (C.I. = − .33 to − .11), p < .001). Both partners’ perceptions of social sabotage in 2009 were related negatively to wives’ marital quality in 2011 (wives: β = − .13 (C.I. = − .24 to .01), p < .05; husbands: β = − .11 (C.I. = − .22 to − .02), p < .05) as well as husbands’ marital quality in 2011 (wives: β = − .13 (C.I. = − .22 to .03), p < .01; husbands: β = − .15 (C.I. = − .26 to − .04), p < .05). Both partners’ perceptions of love withdrawal in 2009 were also related negatively to wives’ marital quality in 2011 (wives: β = − .17 (C.I. = − .27 to .09), p < .001; husbands: β = − .11 (C.I. = − .22 to .00), p < .05). Both partners’ perceptions of love withdrawal in 2009 were also related negatively to husbands’ marital quality in 2011 (wives: β = − .12 (C.I. = − .23 to .02), p < .05; husbands: β = − .17 (C.I. = − .28 to − .08), p < .001).

Table 3 Decomposition of direct and indirect effects on marital quality in 2009
Full size table
Table 4 Decomposition of direct and indirect effects on marital quality in 2010
Full size table
Table 5 Decomposition of direct and indirect effects on marital quality in 2011
Full size table

Economic hardship, financial distress, perceptions of relational aggression, and marital quality all remained relatively stable over time for both partners. Additionally, a positive and significant partner effect was found between respondents’ earlier marital quality and their spouse’s later marital quality, occurring both between 2009 reports and 2010 reports, as well as 2010 reports and 2011 reports.

In summary, the results of the current study indicated economic hardship experienced in 2009 was related directly and positively to financial distress in both 2009 and 2010. Financial distress in 2009 was related positively to perceptions of relational aggression in 2009, but not in 2010. Perceptions of relational aggression were related negatively to marital quality, while these perceptions of relational aggression also partially mediated the negative relationship between financial distress in 2009 and marital quality in 2009. Finally, these 2009 variables of interest continued to influence marital quality indirectly up to 2 years later.

Discussion

The results of the current study indicated economic hardship experienced in 2009 was related directly and positively to financial distress in both 2009 and 2010. Financial distress in 2009 was related positively to perceptions of relational aggression in 2009, but not in 2010. Perceptions of relational aggression were related negatively to marital quality, while these perceptions of relational aggression also partially mediated the negative relationship between financial distress in 2009 and marital quality in 2009. Finally, these 2009 variables of interest continued to influence marital quality indirectly up to 2 years later.

Economic Hardship and Financial Distress

It was expected that couple-level reports of objective measures of economic hardship, namely cutbacks in salary and/or work hours by one or both spouses, would be related positively to both spouses’ subjective measures of financial distress concurrently and 1 year later (Valentino et al. 2014). This expectation was supported and is consistent with Conger et al.’s (1990) original research. While the concurrent association with financial distress was expected, even relatively minor economic hardship continued to influence financial distress in couples 1 year later. This finding suggests that couples experiencing or witnessing economic hardship during or following an economic recession may develop lingering doubts about their future financial position (McDaniel et al. 2013). As such, it is important that married couples consider ways to protect themselves, when possible, from such events. As suggested by Hill’s (1949) ABC-X model, the resources a couple has available, such as an emergency fund, may help to protect couples in the events of an economic downturn. Although the current study did not examine factors that may influence (i.e. mediate or moderate) the relationship between economic hardship and financial distress, future research might consider such directions. Factors, such as having an emergency fund, may influence the level of financial distress people experience when facing economic hardship.

Financial Distress and Relational Aggression

Although most research has found a relationship between husbands’ financial distress and their irritability, hostility, and/or psychologically aggressive behaviors (Conger et al. 1990; Falconier and Epstein 2010; Johnson and Booth 1990), the current study found a relationship between both husbands’ and wives’ financial distress in 2009 and husbands’ and wives’ perceptions of relational aggression (love withdrawal and social sabotage) in 2009. Additionally, both spouses’ financial distress in 2010 was related positively to their own, but not their partners’, perceptions of spousal social sabotage in 2010. Furthermore, perceptions of relational aggression were found in the current study to partially mediate the relationships between wives’ financial distress and wives’ marital quality, as well as wives and husbands’ financial distress and husbands’ marital quality in concurrent years. Although the hypothesized longitudinal findings were not supported in the current studies, economic hardship, financial distress, and perceptions of relational aggression in 2009 were related indirectly to both spouses’ marital quality in 2011. The findings from the current study are unique from the existing literature and warrant further discussion and exploration.

Some research has suggested men demonstrate increased irritability, hostility, and/or psychologically aggressive behaviors when financially distressed, in part, because of their traditional roles as the primary household provider (Gudmunson et al. 2007). However, as more households have become dual-income households, this pressure to provide may be reduced for husbands as it is shared more with their wives. Yet both men and women may develop doubts about the future when facing economic hardship (McDaniel et al. 2013). Other research has also suggested men engage in more externalizing behaviors while women engage in more internalizing behaviors when distressed financially (Gudmunson et al. 2007; Leinonen et al. 2002). The results of Falconier and Epstein’s (2010) research challenged this argument when they found women also engaged in psychologically aggressive behaviors when men were financially distressed, although women’s aggressive behaviors may have been in response to men’s aggressive behavior (i.e. reciprocated aggression).

As such, a possible explanation for the current study’s findings, especially to those related to perceptions of social sabotage, are that couples may engage in relationally aggressive behaviors, believing these behaviors are not harmful to marital relationships. Carroll et al.’s (2010) study found relationally aggressive behaviors to be common among married couples. However, as couples become victims of spousal relational aggression or experience reciprocated relational aggression, their own and their spouse’s marital quality begin to decline concurrently (Carroll et al. 2010) and longitudinally (Coyne et al. 2017). The current study found financial distress in 2009 to be related positively to perceptions of social sabotage in 2010, which were related negatively to and partially mediated the relationship with marital quality concurrently. Although financial distress in 2009 was not related directly to later marital quality, it was related indirectly to later marital quality, operating through financial distress in 2010 and perceptions of social sabotage in 2010. These perceptions of social sabotage were subsequently related negatively to the partner’s marital quality in 2011. These indirect effects suggest that perceptions of social sabotage, in the context of economic hardship and financial distress, have potential to harm marital quality concurrently, but continue to do so up to 2 years later.

Social sabotage in the context of economic hardship and financial distress may emerge and be harmful to marital quality, concurrently and longitudinally, when couples do not agree on spousal contributions to the household economy or when couples differ in their views on spending and saving behaviors. Shapiro (2007) noted money reflects a person’s dreams and goals, as well as their shames, fears, and vulnerabilities. If one spouse makes a disparaging remark to others about the amount of income or a spouse’s non-monetary contribution to the household economy, that spouse may become embarrassed and have his or her feelings of self-worth damaged. Furthermore, Britt et al. (2017) found two of the leading predictors of marital conflict are experiencing financial worries (i.e. distress) and the perception by either spouse that wives are too extravagant. However, regardless of which spouse is perceived as being too extravagant with (i.e. spender) or too controlling (i.e. saver) of the finances, differences in spending and saving behaviors may prompt marital conflict and harm marital quality (Rick et al. 2011). With increased conflict comes the potential for engaging in relationally aggressive behaviors. When private information is shared among friends or family, the other spouse may become embarrassed, which, when made in the context of economic hardship or financial distress, may result in further conflict or loss of trust and closeness within the relationship.

Financial conflict has also been reported as being more negative, longer lasting, and harder to resolve than other conflict topics (Papp et al. 2009). When couples are experiencing more conflict or are upset with one another because of financial difficulties, they may turn towards other people for support while simultaneously withdrawing affection from their spouse (Carroll et al. 2010; Gottman 1999). However, turning towards others or withdrawing one’s affection may result in couples drifting further apart when facing financial challenges, rather than drawing them closer to one another. Gottman’s (1999) research has suggested during times of conflict, couples need to find ways to turn towards one another instead of creating distance between them. Ultimately, further research is needed to better understand the context behind the use of and perceptions of spousal usage of relationally aggressive behaviors, especially in times of economic hardship, and how relational aggression influences marital outcomes in the short- and long-term.

Limitations, Future Research Directions, and Implications for Practice

Strengths of the current study include being guided by Conger et al.’s (1990) model of economic hardship and applying a dyadic and longitudinal perspective to dual-income couples facing economic hardship. However, this study is not without limitations. One limitation is that the sample consisted of stably married, mid-life couples that were predominately European American, doing well financially, and not experiencing much economic hardship or financial distress. An additional limitation is the inability to identify the context behind a couple’s economic hardship. For example, the current study measured cutbacks in work hours and/or salary, representing changes in income (Conger et al. 1990). However, cutbacks may be forced or voluntary in nature. Forced cutbacks by an employer may be perceived as a greater economic burden than if someone voluntarily cuts back on work hours to pursue other activities. As a result, forced cutbacks may influence financial distress differently (i.e., more severely) than voluntary cutbacks. Further exploration of this topic is needed, as well as a replication of the current findings, among a more diverse sample. Participants in the Flourishing Families Project were predominantly European American and doing well financially, so examining this topic among a more racially/ethnically diverse sample and among a sample that spans the socioeconomic spectrum will be helpful to better understand how married couples experience economic hardship. Furthermore, a greater understanding of the context behind the economic hardship will also be helpful.

Another noteworthy limitation of the current study is the reliance on perceptions that spouses were engaging in aggressive behaviors against the respondent. This reliance may be influenced by attribution bias, which refers to identifying responsibility for why events occur (Fincham and Bradbury 1992). Fincham and Bradbury noted that distressed individuals often place more responsibility on their spouses/partners for negative events taking place within the relationship. While self- and partner-reports on relationally aggressive behaviors would be ideal, caution is also needed in that social desirability may prompt participants to rate their own use of these behaviors more conservatively than their spouse may rate them. Future research may consider using both self- and partner-reported aggressive behaviors, in addition to observational data when possible, to more accurately identify how relationally aggressive behaviors are used within couple relationships.

Further examination of the relationship between economic hardship and financial distress is warranted also, specifically whether mediating or moderating factors help to explain this relationship. The current study demonstrated that couple-level economic hardship influenced both partners’ reports of financial distress. Relying on Hill’s (1949) ABC-X Model or Boss’s (2002) Contextual ABC-X Family Stress Model may allow researchers to explore further these associations, specifically in how coping resources, perceptions of events, or context influence the relationship between economic hardship and financial distress.

The results of the current study not only inform future research directions, but also provide several implications for practice. One implication is a need for educators and therapists to explore how couples are experiencing economic hardship, as they may be experiencing shared stressor events differently (Boss 2002). Research demonstrates that experiencing economic hardship is related to increased doubts, fears, and anxiety about the future (McDaniel et al. 2013), which may carry over into couples’ feelings of financial distress and marital interactions. Practitioners may need to discuss with couples how they are experiencing economic hardship (e.g. doubts, anxiety, etc.) and their levels of financial distress. It may be helpful to have couples identify ways to reduce their financial distress during times of economic hardship, and other times when money concerns or problems arise.

Another implication to consider is the comorbidity of financial and relational issues. Research suggests financial issues are strongly related to marital processes and outcomes. For example, Papp et al. (2009) noted that disagreements about money are more negative, longer lasting, and more recurrent than disagreements about other issues. The emergence of financial therapy recognizes this comorbidity and utilizes a therapeutic approach addressing both issues simultaneously (Archuleta et al. 2012). Therapists may consider helping couples to recognize how financial issues are related to other relational issues, including their marital interactions. Additionally, Anderson and Sabatelli (2011) suggested couples may “rewrite” or revise their history together when things are going poorly and focus on the negative aspects of the relationship. As such, helping couples focus on the quality and positivity of their relationship may be beneficial.

A final implication is that practitioners may need to focus on the content-process-outcome therapeutic approach (Nichols and Schwartz 1998). This approach suggests the content of the problem (financial distress) is less important than the processes (relational aggression) couples use when financially distressed. However, perceptions of relational aggression was found to partially mediate the relationships between financial distress and marital quality, which may harm marital outcomes over time (Carroll et al. 2010; Coyne et al. 2017). As such, educators and therapists may consider encouraging couples to identify ways to cope with financial distress and other stressor events by turning towards their spouse (Gottman 1999), rather than turning towards others that may undermine the spouse (social sabotage) or withdrawing affection (love withdrawal).

In conclusion, the results of the current study demonstrate Conger et al.’s (1990) model of economic hardship remains applicable for exploring how economic hardship and financial distress influence marital interactions and outcomes among married couples experiencing the recent economic recession. Specifically, taking a dyadic and longitudinal approach to analyze data helps to identify potential partner differences both in the short- and longer-term. Relying on stress research guided by the ABC-X models (Boss 2002; Hill 1949) may identify potential mediating and moderating factors associated with the relationship between economic hardship and financial distress. Finally, exploring both direct and indirect relationships between financial distress and marital outcomes will aid in better understanding how content and relational processes together influence marital outcomes, such as identifying how perceptions of relationally aggressive behaviors mediate the relationships between financial distress and marital quality.