Do couples share income? Variation in the organisation of income in dual-earner households.

Gray, Edith ; Evans, Ann

Introduction

This paper considers the organisation of income in dual-earner couples with a focus on whether income organisation differs by current relationship type and past relationship experience. Heimdal and Houseknecht (2003) suggest that although income organisation is central to family relationships, it has not received much empirical attention in family research. This is despite the fact that there have been substantial increases in women's labour force participation and their income over the last thirty years.

In order to investigate the income organisation of dual-earner couples in Australia, we examine whether couples keep income totally separate, whether they totally pool their income, or whether they use a combination of pooled and separate income. Our paper starts by examining past research on the distribution of money within married couples. We then discuss how the institution of marriage may impact income organisation, in comparison to 'incomplete institutions' such as cohabitation and remarriage.

The organisation of household money by married couples

There have been important studies focussing on the organisation of money in the household, mostly originating from the UK. The work of Wilson (1987a), Pahl (1989; 1990; 1995) and Vogler (1998; Vogler and Pahl 1993; 1994) is central to what is known about the organisation of income by married couples. This work highlights two key issues. Firstly, money earned by wives is different to the money earned by husbands both in the way it is recognised and in the way that it is spent. For example, Zelizer (1994) finds that women's wages are seen as 'extra' or 'secondary' household money. It has also been found that extra income brought into the household by women is more likely to be used for the family than income brought to the household by men (Pahl 1989; 1990).

Secondly, household income is not distributed equally between individuals within the household. It is not the household that earns and spends money, it is individuals from the household. The management of money then is influenced by what control the individual has over how the money is spent (Pahl 1990; Burgoyne 1995).

Past research has found a variety of factors that influence the organisation of money in married-couple households. These include bow much work the wife engages in, with wives in full-time employment being more likely to pool income than wives without full-time employment (Morris 1984; Vogler and Pahl 1993). Similarly, where both partners are in comparable labour market positions they are more likely to pool income (Vogler and Pahl 1993). Cheal (1993) finds that an increase in female employment increases 'joint householding' rather than 'independent' organisation of income.

Both the total income pool and the homogeneity of the couple income affect the organisation of income within a household. Wilson (1987b) and Pahl (1990) both find that women from low-income households are found to be responsible for household income, but they lack control because the money does not stretch to cover basic needs. Women from the middle-income households, are more likely to have control over some money--and therefore the ability to spend money on household priorities--if they earned money. For women in high-income households there is more variation in the way husbands and wives organise money, however women are able to more freely spend money. Pahl (1990) found that where wives earn more than 30 per cent of their husbands' earnings they are more likely to control the pool of household income than if they earn under 30 per cent or have no earnings.

Vogler and Pahl (1993) also find that the system of financial allocation is associated with: respondent's parents' money management, husband's education, age, and attitudes to the breadwinner model of the family.

Marriage and cohabitation: Different styles of income organisation

There are distinct differences in the treatment of money in cohabiting and married couples. Therefore, it is surprising that a substantial amount of the literature on income organisation focuses on only married couples, or treats cohabiting couples as married. Given recent changes in the organisation of families--particularly the increase in cohabitation before, or instead of, marriage--some researchers have called for an investigation of cohabiting couples as well as married couples (Burgoyne 1995).

An important contribution to the understanding of household income is the work of Singh (Singh 1997; Singh and Lindsay 1996). This work defines the nature of money in relationships by characterising money in married relationships and money in cohabiting relationships. The argument is that money is qualitatively different in cohabiting as compared to married relationships. 'Marriage money' is characterised by a joint bank account, which is often set up upon marriage, or when buying a home (Singh and Lindsay 1996). This pooling of money makes individual money collective, and 'does not necessarily translate to the concept of sharing of income' (Singh and Lindsay 1996: 60), instead the joint account reflects trust. In comparison, cohabiting couples keep finances separate and make equal contributions to expenses and purchases of assets. Cohabiting couples tend to have separate bank accounts and joint accounts are often used for purposive pooling. This separation of finances reflects the independent nature of cohabiting relationships. Singh and Lindsay further find that transitions to joint money occur when couples marry or buy property together.

Other recent research includes Heimdal and Houseknecht (2003) who investigate financial organisational differences between cohabiting and married couples in Sweden and the United States. They also consider income organisation for people who have experienced divorce. They find that cohabiting couples are more likely than married couples to keep some money separate. They also find that ever having been divorced impacts on the likelihood of keeping some money separate. However, they do not find a difference between the United States and Sweden in the likelihood of cohabiters keeping money separate even given the countries' differing legal and normative contexts.

In Australia it is estimated that cohabitation currently precedes 76 per cent of marriages (ABS, 2006). While cohabitation prior to marriage is widespread, cohabitation is also an alternative to marriage. There are many similarities between cohabitation and marriage, but it is also argued that there are fundamental differences. It is argued that the institution of marriage is socially recognised and that people know the normative 'rules' within marriage.

Cherlin (1978), in considering the effects of remarriage, argued that in comparison to the social institution of marriage, remarriage is an 'incomplete' institution. As marriage is such a strong social institution, an individual's behaviour becomes organised in line with the institutional expectations. In comparison to a first marriage, the proscribed behaviours are not as well defined for remarriage, and that linguistically and culturally there is a lack of institutionalised support (Cherlin 1978).

Nock (1995) has applied the argument of 'incomplete institution' to cohabitation. He suggests that if remarriage is an incomplete institution, surely so too is cohabitation. Nock, in agreement with Cherlin's (1978) earlier work proposes that, as cohabitation is not governed and is less socially recognised than marriage, it suffers from an ambiguity about simple issues, even everyday issues, such as what to call your cohabiting partner. However, this also means that as cohabitation is different from marriage--because those involved are not influenced by institutional constraints--cohabiting couples are able to negotiate different rules and roles. Cohabiting couples have been found to be 'more egalitarian and less traditional than married couples' (Bianchi and Casper 2000: 17). This is demonstrated in recent work conducted in Australia (Baxter 2005). Baxter (2005) examines the household division of labour of married and cohabiting couples, and finds that couples who cohabit prior to marriage have a more egalitarian division of labour than those who do not cohabit prior to marriage. She attributes this to the 'incompleteness' of the cohabiting relationship, which allows people to negotiate alternative roles and responsibilities (Baxter 2005: 320).

One recent study on attitudes to family formation finds that Australians are more likely to view a cohabiting couple as a 'family' if children are present than if there are no children present (Evans and Gray 2005). If, as Singh and Lindsay suggest, marriage is a representation of trust (reflected in combining money), then the presence of children in the household may be another expression of trust in a relationship that could precipitate the combining of money. Another important aspect of the relationship between children and income organisation, is the transfer of money when women reduce their involvement in the labour force in order to have and raise children. It is suggested that as women often earn less during the childbearing years, a transfer of money must take place (Pahl 1995: 365).

The literature clearly demonstrates that people in cohabitating and married relationships exhibit different behaviours. One explanation for this is the idea that cohabitation is an incomplete institution without the necessary social rules to govern behaviour. We would also expect people who have experienced repartnering following divorce to be different from those who are in first marriages because repartnered relationships also lack social guidelines influencing behaviour. To investigate these issues we look at the organisation of income within couple households looking at the effect of previous relationship experience and marriage.

Data and Method

We investigate differences in the way dual-earner couples organise their income. Our investigation focuses on the organisation of income by relationship type. This is done by comparing cohabiters and marrieds, and also by examining the way the previously divorced (and previous cohabiters) organise their income once they are remarried (or repartnered). Due to the institutionalised expectations of marriage, we expect that relationship factors influence the sharing of income in the following ways:

* Married people are more likely to combine their income than cohabiters; and

* Previously divorced respondents are likely to keep some income separate.

We distinguish between respondents who have experienced divorce and those who have previously been in a cohabiting relationship.

In order to analyse the effect of current and past relationship on organisation of income, we use data from the 1997 wave of the Negotiating the Life Course (NLC) survey. The NLC is a nationally representative panel study conducted in Australia (McDonald, Evans, Baxter and Gray 2000). In 1997, respondents were aged 18 to 54, with one participant per household randomly selected for interview. Sample weights are available for population estimates (Breusch 2003). The sample used for this analysis is partnered men and women who each have an income, providing a sample of 883.

The analytical strategy employed to examine factors associated with income organisation is ordered logit. We model the effect of relationship type, relationship length, previous relationship experience, education of female and male partners, presence of children, home ownership, family income, ratio of incomes (male:female), ratio of hours worked (male:female), age, and three gender-role attitude scales on income organisation (model variables are described in Appendix 1). Further, we present the predicted proportions of income sharing with 95 per cent confidence intervals (1) controlling for independent variables.

Organisation of income

In examining the way money is allocated, Pahl has classified the way in which finances are allocated within households. The classification is based on whether partners have joint or separate bank accounts in order to determine whether couples pool resources, and by whether the wife feels she has control over the finances (Pahl 1989). Pahl notes that 'the existence of joint and separate bank accounts offered a relatively objective way in which to assess the jointness or otherwise of a couple's financial arrangements' (1989" 87).

The NLC question on organisation of money in the household (Q231) asks 'Couples make different arrangements about the income each brings into the household. Which of the following applies in your household?' Response options are: (1) Only one of us has an income; (2) (2) Our incomes are kept totally separate; (3) Some of our income is separate, some is combined; (4) Our incomes are totally combined. The distribution of this variable is shown in Table 1. Most respondents (68 per cent) have their incomes totally combined. Just over one-fifth have some separate income and some combined income, while 10 per cent have their income totally separate.

Description of variables

A statistical summary of the variables used in the model is provided in Table 1. Relationship type, that is, whether a person is cohabiting or married, is significantly associated with income sharing. As would be expected, married people are more likely to report that their incomes are totally combined than are cohabiting people. For both relationship types, those who keep their incomes totally separate are in the minority. However, one-fifth (21 per cent) of cohabiters report a separation of individual incomes, and a further 43 per cent have some separate and some combined income. For those people who are married, almost three-quarters have their income totally combined. Sixteen per cent of the selected sample is cohabiting (n=119), and the remainder are married (n=764).

The extent of income sharing might be affected by past relationships as discussed above. To test this we categorise people as having ever experienced a divorce and as having ever experienced the break-up of a cohabitating relationship. Thirteen per cent of the sample report ever experiencing divorce (n=112). People who have never experienced divorce include people still in their first marriage and people who have never been married. Fourteen per cent of the sample also reports ever experiencing the break-up of a cohabiting relationship (n=122). Table 1 indicates that people who have ever been divorced are less likely to totally combine incomes (57 per cent) than are those who have never been divorced (70 per cent). A similar pattern is found for those who have experienced a cohabitation break-up.

We use two variables to represent the education level of a couple: Education of female partner and education of male partner. Where the female partner has a tertiary qualification, respondents are more likely to have at least a portion of their income kept separate compared to those without a tertiary qualification. This pattern is also evident where the male partner has a tertiary education.

Presence of children in the household is associated with income organisation. Children are most likely to interrupt mother's employment in the pre- and primary school periods. We therefore classify the presence of children in the household to refer to those households where there is a child aged less than 13 years. Couples with children aged less than 13 are more likely to totally combine their incomes (76 per cent) than are those with no children or older children (62 per cent).

In the models we control for the effect of purchasing a home, as home purchasing is associated with having a joint bank account. We expect that home purchase is associated with sharing income as this represents a significant financial commitment within the couple. In addition, mortgages often require a joint account in the names of the home purchasers. This administrative requirement could be a significant driver of the creation of joint accounts in couples who did not previously pool income. Fifty-four per cent of the sample are currently purchasing a home. Home purchasers are more likely to have at least some income combined than those not purchasing a home.

Family income is also associated with the way respondents organise their income. Average family income is larger for those who keep income totally separate and for those who keep some income separate than the income of those who totally combine income. There is no pattern of difference in income organisation by a couple's ratio of incomes or hours of work. Neither length of relationship nor age are statistically related to how a couple organises their income.

Our three scales relate to gender-role attitudes. The first measures respondent's attitudes to the breadwinner model of the family. The second represents the extent to which the respondent agrees with independence within couple relationships, and the third measures attitudes to gender-equity in the workplace. The scales relating to the breadwinner model and independence within couple relationships are both significantly related to income organisation. People who have higher agreement with the breadwinner model are more likely to combine their incomes. In comparison, people who believe in independence in relationships are more likely to keep some or all income separate.

Multivariate model

The results of our ordered logit regression model are presented in Table 2 (Predicted proportions are presented in Appendix Table 1). We find that relationship type is significantly associated with income organisation. Married respondents are significantly more likely to totally combine their income than are cohabiting people. Seventy-four per cent of married respondents totally combine their income, compared with 44 per cent of cohabiting respondents, controlling for all other factors in the model. Similarly, cohabiters are more likely to keep their income separate (20 per cent) than are married respondents (six per cent) (Figure 1).

Although divorce is significant at the bivariate level, after controlling for other factors in our model, there is no longer an effect on income organisation. We estimate that 68 per cent of divorcees totally combine their income compared with 71 per cent of those never divorced (Figure 1).

Other factors that impact on income sharing include length of relationship, presence of children, purchasing a home, age and the independence in relationship attitude. The effect of length of relationship is still significant when controlling for other factors in the model. The longer the relationship, the more likely that respondents combine their income with their partner. Presence of children also impacts on combining income. Three-quarters of couples who have children under thirteen in the household totally combine their income, and a further nineteen per cent keep some income separate and pool some income (Appendix Table 1). For couples that do not have a child under thirteen only two-thirds totally combine their income.

[FIGURE 1 OMITTED]

Purchasing a home is associated with increased likelihood of combining some or all of a couple's income. For those who are currently purchasing a home, three-quarters of respondents totally combine their income compared with two-thirds of those who are not currently purchasing a home.

Two of our gender-based attitude scales were associated with income organisation at the bivariate level. After controlling for other factors we find the only attitude scale that significantly impacts on income organisation is the one that represents independence in relationships. Those who have stronger agreement that independence should be maintained in relationships are less likely to totally combine their income and more likely to maintain some or all income separately.

Discussion

This paper investigates the extent to which dual-income couples combine their incomes. Using ordered logit regression we model the effect of relationship type and previous relationship breakdown on whether couples maintain separate incomes or pool their incomes. We also control for the effect of length of relationship, education, presence of young children, home purchase, family income, ratio of couple income and hours of work, age, and gender-based attitudes.

We find support for the argument that the institution of marriage affects income sharing. Married people are less likely to keep their incomes totally separate and more likely to have their incomes totally combined. This fits proscribed behaviour that married couples' organisation of income is shared. Even after controlling for other factors, particularly having young children or buying a home, this behaviour holds.

The combination of money can also be viewed as an expression of trust (as argued by Singh and Lindsay 1996), much like marriage itself. Most cohabiters, like most married couples, combine some or all of their income. We suggest that this expression of trust indicates that cohabitation is somewhat institutionalised in Australia. This is also supported by attitudinal research finding that many Australians view cohabitation as similar to marriage (Evans and Gray 2005). Cohabitation is legally recognised for many purposes, which may make cohabitation more like marriage. However cohabiters are more likely than married people to have their incomes totally separate, and a substantial proportion keep some income separate and have some income combined. This supports arguments that cohabiters are more egalitarian, and may have more power to negotiate different rules (Baxter 2005).

Certainly people who cohabit are not a homogeneous group. Although cohabiters are much more likely than married people to organise their income by keeping some separately and having some combined, we do not find, as Singh and Lindsay (1996) do, that most cohabiters organise their money this way. It is certainly an important way of organising income for cohabiting couples, but we find that so too is having income totally combined. It should be acknowledged that cohabiting couples may pool income for specific reasons whereas married couples pool money as a representation of trust (Singh and Lindsay 1996).

The effect of relationships on the organisation of income is not, however, limited to a person's current relationship. At the bivariate level, we find that people who have experienced divorce are more likely than people who have not experienced a divorce to keep their incomes separate, but this finding does not hold after controlling for other factors. While people who have experienced a divorce may be more likely to separate their income, this is more likely to be due to their shorter length of relationship, and the decreased likelihood of having a child in the household or purchasing a home together than an effect of a divorce per se. Further study, particularly qualitative interviews, would be valuable in understanding how the process of divorce impacts on the organisation of money, and on the concept of trust and money when divorcees are in new relationships.

While our study demonstrates the relationship between income organisation and marriage at one point in time, we propose further research that examines when and why people change the way that they organise their income. What is the process by which people determine when to combine incomes? What life course events are associated with moving to totally combined money, or in fact, what events are linked to putting income in a separate account? We speculate that changing relationships and changing family forms is a large part of the story. Future longitudinal research, of both a quantitative and qualitative nature will assist in illuminating the pathways people choose.

Appendix 1

Construction of variables

Income organisation

Definition: Ordinal variable indicating how income is organised between partners in a household.

Construction: Income organisation is based on Q231, where 0=Our incomes are kept totally separate; 1=Some of our income is separate, some is combined; 2=Our incomes are totally combined

Marital status

Definition: Binary (0,1) dummy variable for cohabiting or married.

Construction: Marital status is based on Q20. Respondents who answered 'living in a relationship but not married' are classified as 0=Cohabiting. Respondents who answered 'married and living with partner' are classified as 1=Married.

Divorce status

Definition: Binary (1,0) dummy variable for whether respondent has ever divorced or not.

Construction: Using variables Q197 (Legal marital status) and Q202a1, q202a2 and q202a3 (How did your first/second/most recent marriage end). Respondents who answered that they were currently divorced on Q197 or that they had ever been divorced on Q202a1, Q202a2 or Q202a3 are classified as 1=Ever divorced. All other respondents are classified as 0=Never divorced.

Ever ended a cohabitation

Definition: Binary (1,0) dummy variable for whether respondent has ever cohabited and the relationship has ended, or not.

Construction: Using variables Q214 (What year did live-in relationship end). Respondents are coded as 0='Never ended a cohabitation' if they never had a live-in relationship end, and 1='Ever ended a cohabitation' if they have had a live-in relationship end.

Education level of female partner

Definition: Binary (1,0) dummy variable for whether the female partner has a university qualification or not.

Construction: Using Q22 (Sex of respondent) and imputed variables HIGHED (Highest education level of respondent) and HIGHEDP (Highest education level of partner), using education of respondent or education of partner depending on the sex of respondent, the education variable is coded as 0=Female partner does not have university qualification and 1=Female partner has university qualification.

Education level of male partner

Definition: Binary (1,0) dummy variable for whether the male partner has a university qualification or not.

Construction: Using Q22 (Sex of respondent) and imputed variables HIGHED (Highest education level of respondent) and HIGHEDP (Highest education level of partner), using education of respondent or education of partner depending on the sex of respondent, the education variable is coded as 0=Male partner does not have university qualification and 1=Male partner has university qualification.

Presence of young child

Definition: Binary (1,0) dummy variable for whether respondent has a child less than 13 years living in the household or not.

Construction: Coded as 1 when positive on Q170 (Number of preschool children present) or positive on Q178 (Number of children of school age but 12 years or less!.

Purchasing a home

Definition: Binary (1,0) dummy variable for whether respondent is currently being purchased or not.

Construction: Using variables Q252 (Own or rent home) and Q253 (Is home fully owned or being purchased), respondents are coded as 1 if currently purchasing own home.

Length of relationship

Definition: Years in current relationship.

Construction: To derive relationship start date: if married (on Q20), relationship start date selected from Q201al, Q201a2 or Q201a3 (In what year did you marry; using Q198 to determine which marriage); if cohabiting (on Q20) Q201al, Q201a2 or Q201a3 (In what year did your defacto relationship begin; using Q209 to determine which defacto relationship). Year of relationship start is subtracted from survey year.

Family income

Definition: Net family income of partners.

Construction: Use imputed variable FINC (Family income).

Income ratio Definition: Ratio of male to female earnings.

Construction: Using Q22 (Sex of respondent) and imputed variables RINC (respondent's income) and PINC (partner's income), respondent's sex is used to determine income of male and female partners. Male partner's income is divided by female partner's income.

Hours of work ratio

Definition: Ratio of male to female hours worked.

Construction: Using Q22 (Sex of respondent) and Q110 (respondent's hours worked) and Q143 (partner's hours worked), respondent's sex is used to determine hours worked of male and female partners. Male partner's hours worked is divided by female partner's hours worked.

Breadwinner scale

Definition: Scale ranging from 1 strongly disagree to 5 strongly agree with breadwinner model of family.

Construction: Combines responses to variables Q267a3, Q267a2 and Q234a3: 'A wife should give up her job whenever it is inconvenient to her husband and children', 'People should consider the needs of their spouse and children as more important than their own', and 'It is better for the family if the husband is the principal breadwinner and the wife has primary responsibility for the home and the children'.

Independence in relationship scale

Definition: Scale ranging from 1 strongly disagree to 5 strongly agree with whether a person should maintain independence when in personal relationships.

Construction: Combines responses to variables Q267al and Q267a34: 'Both the husband and wife should contribute to the household income', and 'Having a job is the best way for a woman to be an independent person'.

Equity in the workplace scale

Definition: Scale ranging from 1 strongly disagree to 5 strongly agree with equitable access to the workplace.

Construction: Combines responses to variables Q234a1, Q234a2 and Q234a4: 'If both the husband and wife work they should share equally in the housework and care of the children', 'There should be satisfactory childcare facilities so that women can take jobs outside the home', and 'Ideally, there should be as many women as men in important positions in government and business'.

Appendix

Table 1: Proportion of couples combining incomes predicted by ordered
logit regression model

 Predicted
 proportion SE 95% CI of PP

 Totally separate 0.07 0.01 0.06 0.09
 Some separate some combined 0.22 0.01 0.19 0.25
 Totally combined 0.71 0.02 0.68 0.74
Married
 Totally separate 0.06 0.01 0.05 0.08
 Some separate some combined 0.19 0.01 0.17 0.22
 Totally combined 0.74 0.02 0.71 0.77
Cohabiting
 Totally separate 0.20 0.03 0.14 0.28
 Some separate some combined 0.36 0.03 0.30 0.41
 Totally combined 0.44 0.05 0.35 0.55
Ever divorced
 Totally separate 0.08 0.02 0.05 0.13
 Some separate some combined 0.24 0.04 0.17 0.31
 Totally combined 0.68 0.05 0.57 0.78
Never divorced
 Totally separate 0.07 0.01 0.06 0.09
 Some separate some combined 0.21 0.02 0.19 0.24
 Totally combined 0.71 0.02 0.68 0.75
Children <13 in household
 Totally separate 0.06 0.01 0.04 0.08
 Some separate some combined 0.19 0.02 0.16 0.23
 Totally combined 0.75 0.02 0.70 0.79
No children <13 in household
 Totally separate 0.09 0.01 0.07 0.11
 Some separate some combined 0.25 0.02 0.21 0.29
 Totally combined 0.67 0.03 0.61 0.71
Purchasing a home
 Totally separate 0.06 0.01 0.04 0.08
 Some separate some combined 0.19 0.02 0.16 0.22
 Totally combined 0.76 0.02 0.71 0.79
Not purchasing a home
 Totally separate 0.10 0.01 0.07 0.12
 Some separate some combined 0.26 0.02 0.22 0.30
 Totally combined 0.65 0.03 0.60 0.70

* P < 0.05, ** P<0.01. Source: NLC Data 1997. Predicted proportions
estimated from the model at Table 2 using 'Clarify' (Tomz, Wittenberg
and King 2003; King, Tomz and Wittenberg 2000).

Acknowledgement

Funding for this research was provided by the Australian Research Council (DP0772544).

References

Australian Bureau of Statistics (ABS) (2006) Marriages, Australia, 2005, Catalogue No. 3306.0.55.001 (electronic product), Canberra, ABS.

Baxter, J. (2005) 'To marry or not to marry: Marital status and the household division of labour', Journal of Family Issues, 26(3), 300-321.

Bianchi, S., & Casper, L. (2000) 'American families', Population Bulletin, 55, Washington DC, Population Reference Bureau.

Breusch, T. (2003) 'Negotiating the Life Course, Waves 1 and 2: Sampling Weights for Persons and Income Units', Negotiating the Life Course Discussion Paper Series DP-016, Canberra, The Australian National University.

Burgoyne, C. (1995) 'Financial organisation and decision-making within Western 'households", Journal of Economic Psychology, 16, 421-430.

Cheal, D. (1993) 'Changing household financial strategies: Canadian couples today', Human Ecology, 21(2), 197-213.

Cherlin, A. (1978) 'Remarriage as an incomplete institution', American Journal of Sociology, 84 (3), 634-650.

Evans, A., & Gray, E. (2005) 'What makes an Australian family?'. In S. Wilson, G. Meagher, R. Gibson, D. Denemark, & M Western (eds.) Australian Social Attitudes. The first report, pp. 12-29, Sydney, UNSW Press.

King, G., Tomz, M., & Wittenberg, J. (2000) 'Making the most of statistical analyses: Improving interpretation and presentation', American Journal of Political Science, 44 (2), 347-61.

Heimdal, K., & Houseknecht, S. (2003) 'Cohabiting and married couples' income organisation: Approaches in Sweden and the United States', Journal of Marriage and Family, 65 (August), 525-538.

McDonald, P., Evans, A., Baxter, J., & Gray, E. (2000) 'The Negotiating the Life Course Survey experience', Negotiating the Life Course Discussion Paper Series DP-001, Canberra, The Australian National University.

Morris, L. (1987) 'Constraints on gender: the family wage, social security and the labour market: reflections on research in Hartlepool', Work, Employment and Society, 1(1), 85-106.

Nock, S. (1995) 'A comparison of marriages and cohabiting relationships', Journal of Family Issues, 16(1), 53-76.

Pahl, J. (1989) Money and marriage. London, Macmillan Education Ltd.

Pahl, J. (1990) 'Household spending, personal spending and the control of money in marriage', Sociology, 24(1), 119-138.

Pahl, J. (1995) 'His money, her money: Recent research on financial organisation in marriage', Journal of Economic Psychology, 16, 361-376.

Singh, S. (1997) Marriage Money: The Social Shaping of Money in Marriage and Banking. St Leonards, NSW, Allen and Unwin.

Singh, S., & Lindsay, J. (1996) 'Money in heterosexual relationships'. Australian and New Zealand Journal of Sociology, 32(3), 57-69.

Tomz, M., Wittenberg, J., & King, G. (2003) CLARIFY: Software for Interpreting and Presenting Statistical Results. Version 2.1, Stanford University, University of Wisconsin, and Harvard University, January 5, Available at <http://gking.harvard.edu/>.

Vogler, C., & Pahl, J. (1993) 'Social and economic change and the organisation of money within marriage', Work, Employment and Society, 7(1), 71-95.

Vogler, C., & Pahl, J. (1994) 'Money, power and inequality within marriage', The Sociological Review, 42(2), 263-288.

Vogler, C. (1998) 'Money in the household: some underlying issues of power', The Sociological Review, 46(4), 687-713.

Wilson, G. (1987a) Money in the family: Financial organisation and women's responsibility, Aldershot, Avebury.

Wilson, G. (1987b) 'Money: Patterns of responsibility and irresponsibility in marriage'. In J. Brannen, & G. Wilson (eds.), Give and take in families: Studies in resource distribution, pp. 136-152, London, Allen and Unwin.

Zelizer, V. (1989) 'The social meaning of money: "Special monies"', American Journal of Sociology, 95(2), 342-377.

(1) We present predicted proportions with confidence intervals estimated using 'Clarify'--a Stata add-on program created by Gary King et al. (Tomz, Wittenberg and King 2003; King, Tomz and Wittenberg 2000).

(2) Note that this option is only applicable to single-income households, and hence is not valid for dual-income households. Therefore it is not applicable to this sample.

Table 1: Percentage distribution of income organisation types by
selected factors

 Our incomes Some of our
 are kept income is
 totally separate, some
 separate is combined
 % %

Total sample 9.9 21.9
Marital status **
 Cohabiting 20.5 42.9
 Married 8.3 18.6
Divorce status **
 Never divorced 8.8 21.3
 Ever divorced 17.4 26.0
Ever ended a cohabitation
 Never ended a cohabitation 9.8 21.0
 Ever ended a cohabitation 10.7 27.2
Education level of female partner **
 No tertiary qualification 9.3 18.2
 Undergraduate or higher 10.9 28.1
Education level of male partner *
 No tertiary qualification 9.6 19.9
 Undergraduate or higher 10.6 26.0
Presence of young child **
 No 13.1 25.1
 Yes child <13 years 6.1 17.9
Purchasing a home?
 No 11.6 24.3
 Yes, purchasing a home 8.4 19.8
 Mean Mean
Length of relationship (years) 11.0 13.0
Family income ($) * 82006 77846
Income ratio (M:F) 1.1 1.2
Hours of work ratio (M:F) 0.9 0.8
Age of respondent 39.4 39.1
Breadwinner scale * 2.5 2.7
Independence in relationship scale ** 3.7 3.6
Equity in the workplace scale 4.2 4.1

 Our incomes
 are totally
 combined Total
 % N

Total sample 68.2 883
Marital status **
 Cohabiting 36.6 119
 Married 73.2 764
Divorce status **
 Never divorced 69.9 771
 Ever divorced 56.6 112
Ever ended a cohabitation
 Never ended a cohabitation 69.2 761
 Ever ended a cohabitation 62.1 122
Education level of female partner **
 No tertiary qualification 72.6 555
 Undergraduate or higher 61.0 328
Education level of male partner *
 No tertiary qualification 70.5 600
 Undergraduate or higher 63.4 283
Presence of young child **
 No 61.8 481
 Yes child <13 years 76.0 402
Purchasing a home?
 No 64.1 409
 Yes, purchasing a home 71.8 474
 Mean N
Length of relationship (years) 13.5 876
Family income ($) * 72669 854
Income ratio (M:F) 0.5 883
Hours of work ratio (M:F) 0.7 866
Age of respondent 38.7 889
Breadwinner scale * 2.7 880
Independence in relationship scale ** 3.3 880
Equity in the workplace scale 4.1 877

* P <0.05, ** P <0.01. Sample is weighted to reflect the composition
of the Australian population (see Breusch, 2003). Source: NLC
Data 1997.

Table 2: Odds ratios and coefficients predicting income organisation
(ordered logit regression)

 Odds
 Ratio B Sig. SE

Currently married 3.64 1.29 ** 0.84
Ever divorced 0.86 -0.16 0.23
Ever ended cohabiting relationship 0.79 -0.24 0.18
Length of relationship (years) 1.04 0.04 * 0.02
University educated female partner 0.71 -0.34 0.12
University educated male partner 0.79 -0.24 0.15
Child aged <13 in household 1.50 0.41 * 0.25
Purchasing a home 1.69 0.52 ** 0.27
Family income 1.00 0.00 0.00
Income ratio (M:F) 0.98 -0.02 0.01
Hours of work ratio (M:F) 0.96 -0.04 0.07
Age of respondent 0.95 -0.05 ** 0.02
Breadwinner scale 0.98 -0.02 0.10
Independence in relationship scale 0.69 -0.38 ** 0.06
Equity in the workplace scale 1.23 0.20 0.17

 95% CI of OR

Currently married 2.31 5.73
Ever divorced 0.50 1.46
Ever ended cohabiting relationship 0.51 1.22
Length of relationship (years) 1.01 1.07
University educated female partner 0.51 1.00
University educated male partner 0.55 1.14
Child aged <13 in household 1.08 2.09
Purchasing a home 1.23 2.32
Family income 1.00 1.00
Income ratio (M:F) 0.96 1.01
Hours of work ratio (M:F) 0.83 1.10
Age of respondent 0.92 0.98
Breadwinner scale 0.80 1.20
Independence in relationship scale 0.57 0.82
Equity in the workplace scale 0.94 1.61

* P <0.05, ** P <0.01. Source: NLC Data 1997. Note: 826 cases are
included in the analysis.