This report was published by the former Department of Families, Community Services (FaCS).
- 1. Introduction
- 2. The sociological literature
- 3. Economic studies
- 4. New family types
- 5. Conclusion
The purpose of this review is to study the literature on within-household attitudes and behaviour vis-à-vis money, in order to understand the implications for the wellbeing of individual household members, particularly children.
These studies probe couples' attitudes to money and analyse decision-making in terms of who manages the household's finances and who actually controls spending decisions.
- They find that the ideology of money within marriage is one of privacy and of jointness.
Conflict between the ideology of equality and sharing between husband and wife, and the traditional idea of the man as breadwinner, is hidden.
Privacy conceals power relations within marriage.
Although this view of money in marriage is particular to the life-cycle stage, income level and race of the couple, it has dominated and shaped thinking in policy and research.
Associated with this ideology is the idea of lack of fungibility of income. Women's income is regarded as their own by their spouses, who trivialise its importance to the household's finances. Men, in their role as breadwinners, are supposed to provide the necessities while women's income is for discretionary spending. In reality, however, studies show that a woman's income is more likely to be spent on household, rather than personal, goods; men are more likely to increase personal consumption if their wives earn.
- Management of household finances, as opposed to the control of spending decisions, is largely with the wife.
Several different typologies of management and control have been proposed, most notably by Pahl, and by Edwards in Australia. Edwards proposed a schema that distinguished between wife, husband, joint and independent managements.
In Edward's study, low-income family finances were more likely to be managed by the wife; almost all (95 per cent) males, however, had either joint or exclusive control.
Few households had joint control of finances. Such households displayed the lowest gender inequality with such inequality being highest in both low and high- income households with male control of finances.
- Factors like income and contribution to household income were not as important as factors associated with particular management systems.
Research opinion is divided about the link between level of income and management system. Studies report that when it is a burden, rather than a source of power, it is the wife who does the managing.
Is an individual's contribution to household income important? A study in Canada found that while it meant greater control for men, for women other factors were more important in determining their influence on household finances. These factors included 'better options', like the woman's age and the couple's age difference, which affected the woman's probability of remarriage.
In 'status-reversal' couples, that is those where the wife is the main breadwinner, making more money did not translate into more power. In fact, women in such relationships tend to compensate for the role reversals, pointing again to the strong hold of conventional views of gender roles and responsibilities.
Families do tend to regard income from benefits differently. They tend to accept the official definition of the income. Women tend to have sole responsibility for the family allowance, while joint access and decision-making is more common with income support payments in Australia.
Previous experience, either of one's parents' or of a previous marriage, influences choice of financial management system. Young couples give equality a greater priority to avoid bickering like their parents. Remarried couples are more likely to keep finances separate.
- Most studies focus on gender inequity while analysing the impact of different management systems on the wellbeing of individuals, though later work has begun to address the impact on children. Women bear the brunt of low incomes; in households with low incomes they tend to have less personal spending money than their partners. Moreover, their income is more likely to be spent on children.
Men are unaware of this inequity. Children are least likely to suffer a reduction in basic necessities. The evidence on whether this results in 'hidden' female poverty within non-poor households is, however, mixed.
Can anything be gleaned from large-scale survey data about the extent of income pooling within households and its impact on expenditure?
- The common preference model, or the one that implies pooling of income within a household, has been challenged by models that attempt to incorporate the divergent and conflicting preferences of household members.
- Empirical evidence shows that pooling is incomplete.
A study in the United Kingdom found strong evidence of a shift toward greater expenditures on women's clothing and children's clothing after payments of child benefit began to be made to the mother.
In Canada, a study found that couples were more likely to use their own incomes for personal consumption, and that expenditure on household goods reflected gendered spheres of responsibility. Thus, an extra dollar of the wife's income was more likely to be spent on child care; while an extra dollar of the husband's income was more likely to be spent on transportation flows.
There exists the possibility that redistributions in favour of women could lead to a decline in their welfare, if the husbands reduce their own voluntary spending on domestic public goods by more than the amounts their wives receive as additional income.
- While little work has been done on financial management practices and their implications for the wellbeing of individuals, particularly children, in the non- traditional households, there is considerable indirect evidence from other studies.
Responsibilities in remarried or stepfamilies are allocated along biological lines. The stepparent, usually the stepfather, does not, and is not expected to assume responsibility, financial or otherwise, for the children of his new partner.
Couples in such families are more likely to keep finances separate; and money is a major source of conflict among such couples.
Evidence shows that children in remarried families suffer numerous disadvantages that manifest themselves in low educational outcomes, psychological problems, and so on.
Interestingly, mothering behaviour too is different in remarried and cohabiting families with the levels of supervision and care being lower than in the traditional two-parent family. Children in single-mother families report the highest degree of supervision. This points to the need to consider the range of mothering behaviour rather the altruistic model being universal. This might have implications for spending from family allowances.
Daughter: You have never been fair with Mother about money. You've had everything—your poker games, your cigars, your holiday fishing trips! … And that is why I'm asking you to pay her a definite salary instead of making her come to you like a beggar for every penny she gets.
Father: Most wives have to ask their husbands for money and I've noticed they don't stutter much.
Daughter: There isn't a wife in the world who wouldn't sooner have less money to spend if only she knew it was her own and she could do what she liked with it without answering those everlasting questions: 'What do you want it for?' and 'What did you do with the last I gave you?'
(From the play Chicken Feed, which opened in Chicago in 1923).
The battle over money between couples has been more than theatrical entertainment. Since the late nineteenth century, it has played out in the privacy of homes, sometimes spilling over into the public arena in divorce courts. Money has been subject to negotiation among affluent as well as poor couples (Zelizer 1994).
The question of how households manage money and make consumption decisions is of great significance to policy concerned with equity. The methods that households employ to allocate income for consumption between individual members have implications for their wellbeing. The longstanding assumption among economists and policy-makers, that income would be equitably shared among all members, is increasingly being questioned in the light of evidence to the contrary.
While the objective of welfare policy is usually the wellbeing of the individual, most policies have been targeted at households, despite a tacit understanding that individual welfare is based on a complex set of interactions within the household. The primary objective of many income support programs and taxation policies, for instance, is to improve the living standards of children, although income transfers are paid to parents, rather than the children themselves. It is, therefore, important to analyse the decision-making processes within households and how they might be affected by factors like income levels, demographic composition, labour market participation of individual members, life-cycle stage, and so on.
The literature on decision-making within the household follows two distinct strands. One is the growing body of economic literature, both theoretical and empirical, which questions the traditional assumption of the household being the unit of analysis and attempts to model the household as comprising members with different consumption and production functions. The other is that of qualitative sociological studies which explore in depth the concept of money management and control among couples.
The purpose of this review is to study the literature on within-household attitudes and behaviour vis-à-vis money with a view to understanding the implications for the wellbeing of individual members, particularly children. Section 2 draws on interview-based studies to understand the attitudes and processes involved in decisions regarding money in families. In particular, it looks at the ways in which couples manage their incomes and control different spending decisions, as well as individual household members' feelings about their current arrangements.
The assumption of the household as the unit of consumption is firmly embedded in mainstream economics. It has, however, been increasingly challenged by the work on intra-household economics, questioning the idea that households pool all their incomes. The theoretical work in this field has sought to incorporate the divergent and conflicting preferences of household members into economic analysis. Section 3 discusses the empirical studies that apply these models to large-scale survey data1.
There is a growing interest in the kinds of financial management likely to arise in the non-traditional family types, such as the so-called blended families. There are very few studies that directly address this question. One can, however, survey the literature on such family types in general and attempt to glean some implications for our study. This literature is discussed in Section 4.
In order to disentangle attitudes about money from the reality of sharing, Section 2.1 begins by investigating the literature on the ideology of money in a marriage. In Section 2.2, the focus shifts to works that attempt to investigate actual money management practices in a household. Section 2.3 studies the impact of the different patterns of management and control on the wellbeing of the members of the household, particularly children.
The term ideology refers to a certain set of ideas and beliefs held by people and groups that aids in understanding 'reality' and in turn shape social behaviour. At any given time in society, however, there are likely to be numerous different and conflicting ideologies. This conflict is often concealed by the dominance of one particular ideology or by the creation of what Therborn calls a discursive order (Pahl 1989). The nuclear family is one such ideological construct.
As one begins to open the 'black box' of the household economy, one first encounters the question of money in a marriage in an ideological form. Any investigation into the nature of money and relationships immediately comes up against beliefs about how money should be negotiated in a relationship. In order to understand the way in which households handle money, one must then first disentangle what people think about it from what they actually do.
Why is it important to distinguish between ideology and reality? First, the manner in which a couple construct their ideas about money has implications for their actual practice. Dissonance between the ideology and what researchers are able to discern about actual behaviour would reflect the difference between what the couple want to be seen to be doing and what really happens. It would also point to the difficulty of understanding what really happens within a family; couples might be aware of the dissonance and cloak their responses accordingly. It is only when the relationship breaks down, and social norms and ideological talk are abandoned, that the outsider gets a glimpse of what might be going on within the couple.
Second, to the extent that ideology affects behaviour, any changes in ideology over time could point to real changes in differential access to household income. With the greater awareness of gender equality and the rights of children, ideas about money in the household could be expected to change. This could in turn affect behaviour.
Money in marriage
How do married people regard money? Supriya Singh studied attitudes towards money among couples in two suburbs of Melbourne in 1991-92 (Singh 1997).
One suburb, 'Woodville,' was predominantly middle-income. She conducted open- ended interviews with 37 people from 21 households, nearly half of them between 35 and 44 years old. The respondents, who were randomly selected from the electoral rolls, were mostly in first marriages, Anglo-Celtic, Australian-born, with average household income above $50 000 per annum. In order to study the impact of differences in ethnicity, education and income she extended her sample to another suburb where the average annual income was less than $15 000 per annum and the residents came from 30 different countries. English was a second language and the group was younger, with 30 per cent being between 25 and 34 years old.
In Woodville, most respondents felt that joint control of marriage money was of prime importance. Pooling of resources and sharing of expenses was seen as essential to the marital relationship. It was taken for granted that joint accounts follow marriage; couple after couple spoke of it in ritualised terms. Trust in marriage was symbolised by the joint account (Singh 1997). Money was considered private and most respondents did not discuss it even with their children. Moreover, the boundaries of marriage money were set by the marital unit with no outflows to, say, elderly parents or inflows from children.
Interestingly, Singh found that her respondents from Woodville 'block' questions about equality and power in relation to money. One of the ways in which they avoid seeing the links between money and power is by presenting money as not being meaningful. This trivialisation of money is done by both men and women, and by those who have control and those who do not. Most persons in Woodville do not admit to any problems over money, even when asked pointedly about tension over money. Nor do they admit to thinking about the financial repercussions of divorce. They reduce the importance of money by minimising personal wants or seeing personal wants satisfied in collective expenditure. Obscuring questions of equality and power helps to negotiate the differences between two different yet coexisting ideologies about marriage—one of equality and the other of the more traditional view of the man as the provider and the woman as the home-maker (Singh 1997).
A comparison of marriage and cohabitation (Singh & Lindsay 1996) shows a similar absence of discussion of power and control in relation to money between cohabiting couples, even though in cohabitation the emphasis on separateness and equality is greater than in marriage.
As Pahl (1989) notes, 'the secrecy with which couples protect their financial affairs is as great as the secrecy surrounding sexual relationships; yet compared with the enormous literature…on sex, little has been written about money in marriage'. Given the importance of money in contemporary society, 'the fact that it is so private is perhaps a testimony to its importance'.
Normative expectations about the allocation of money could be embedded in the socialisation of the couple, be held by the couple's social network or arise out of the occupational structure of the local labour market (Pahl 1983). Edwards found that many wives thought their husbands were 'rightfully' in control of the family's finances (Edwards 1981). And even though paid employment of the wife could engender confidence and independence, 'the wife's inclination towards deference may not be affected by her work experience, particularly if she is in a relationship of authority and deference in the work place' (Cass 1978). Studies of couples where the woman rather than the man is the chief earner show the force of ideologies surrounding the conjugal contract (Pahl 1989). Female breadwinners somehow have to compensate their husbands for the loss of breadwinner status.
Writing 20 years ago, and referring to what Poster called the 'bourgeois family' (Poster 1978), Pahl argued that one particular historical configuration, which was powerful among the middle classes in the first half of the twentieth century, has become the norm. In her view, much of present social and economic policy has been based on the assumption that the normal family is composed of the breadwinner husband and a dependent wife, with the husband giving the wife an allowance for 'housekeeping' being the norm. This allowance system reinforced the inequality between the sexes, and keeping the financial arrangements within marriage essentially private was a way of concealing power relations within the household. Pahl goes on to liken it to the concealment of violence within the family. It is noteworthy, therefore, that the increased interest in resource allocation within the household, and in the contribution of women to the family income, has been accompanied by a demand for disaggregation of financial arrangements within families and a decrease in the predominance of the allowance system (Pahl 1983).
In contemporary societies, this egalitarian demand on the family is challenged by two other ideologies about money in marriage (Volger 1998). Along with the notion that money in marriage ought to be based on equal sharing, regardless of who contributes what to the household, goes the opposing notion of the individual's right to what they earn. This latter idea feeds directly into the second belief, that the breadwinner has a right both to more power over household money and to more money for their own use (Burgoyne 1990). In practice, it is the ideology of power that dominates; for although husbands in sole-earning couples do not overtly restrict their wives' expenditure, the wives feel inhibited about spending money on themselves and limit their expenditure from the joint account (Burgoyne 1990).
These findings echo those of Pahl (1989) who reports that many of her respondents spoke eloquently about the importance of equality in marital finances and presented the sharing of money as an expression of deeply held values. Many couples described their financial arrangements in normative terms, describing the pooling system as 'how things should be done'. Yet at the same time, the interviewers noted there were many ambiguities between the claims made by couples in joint interviews and their admissions in separate interviews. These disparities between ideology and reality make the pooling system another 'black box'.
Singh's study of the two Melbourne suburbs (Singh 1997) did not extend itself to probing the lower income suburb's money management practices, focusing simply on the respondents' banking experiences. The picture of marriage money presented in the studies above is very obviously the Anglo-Celtic, middle-income, middle-age view. As evidence from other cultures shows, these are not universal. In South Asia, for instance, it is the family that is ideologically dominant, not the marriage (Agarwal 1994). This encourages a flow of money from adult children to parents, as a sign of filial duty. Ewen (1985), in her study of Italian and Jewish immigrants in New York analyses a similar flow pattern between children and parents. Thorogood (1987) describes how black women in the United States continue to have financial responsibilities to their mothers. Nor is money private in all cultures. Yet, as Thorogood says, it is the discursive order of privacy and jointness of money in marriage that has dominated and shaped thinking in policy and research.
Associated with the ideals about money in marriage is the idea that the money is not fungible. Pahl (1980) reports that men regard increases in overtime earnings as more personal spending money while any increase in basic wages is regarded as going into the collective consumption of the household. To the married woman, the money given to her by her husband to run the household is different from what she might earn herself (Pahl 1995).
This lack of fungibility is intricately tied up with the idea of the man as the breadwinner who must provide for the basic needs of the family. While the woman's income, if any, is to be used for the 'extras', the money earned by the husband is regarded differently—it is considered the family's income. Zelizer (1994) talks of the social meaning of money and argues that it is embedded in social relations. She contends that so long as the husband is regarded as the breadwinner, the wife's income is 'different, less significant and ultimately dispensable'. This trivialisation of the wife's income is evident in the man's claiming not to know what she does with it (Edwards 1981). While on one level it can be seen as his indulgence and letting her do as she pleases, on another it reflects his perception of its unimportance to the family's true income.
Zelizer's research shows that money is not fungible in the sense that it is highly differentiated according to its source, the uses to which it can be put to and the person who has the right to use it. 'Monies are set aside by linking certain sources of money to selected uses: income earned by the woman may be reserved for her children's education, while her husband's income pays the mortgage'.
Several studies support this conclusion. In Australia, Edwards (1981) found that family allowance was regarded as women's income; most husbands reported that they did not know what their spouses did with it. Most women, however, regarded it as important, especially those in low-income families or those who did not have incomes of their own. It was also important to those women whose husbands managed the household's finances; they were also opposed to the idea of it being paid to their husbands instead of them (Edwards 1981).
Interested in how couples felt about money that entered the household, Pahl (1995) asked them whether they felt it was their own income or their husband's or wife's as well. Most couples amended the question to say that it belonged to the family. There was considerable gender difference in perceptions. Most husbands (93 per cent) saw their own income as belonging to the family, and 76 per cent of the wives agreed—evidence that the idea of the male breadwinner is powerful. A majority of the husbands (52 per cent) felt their wives' incomes remained the property of their wives.
This lack of fungibility associated with money is significant for its implications for spending. Thus who earns the money and in what manner is important for how it gets spent. This is an issue that recurs often in the literature, and is discussed further at several points below.
Challenges to ideology?
Have there been any changes in perceptions about money in the last decade? As noted above, most of these studies have focused on mainstream, married older couples in the 1970s and the 1980s. Are there any more recent challenges to the ideology of the nuclear family as the dominant construct?
Differences in financial power within a couple are usually obscured until the relationship breaks down. Wilson (1987) reports that inner-city women who were not living with their partners or were contemplating leaving them, were much more open about financial disagreements.
Some authors have noted that one reason younger people give for managing their finances differently was that they did not want to be like their parents who constantly bickered about money (Glezer 1996). Does this mean that money has declined in importance as an issue in marriage? Or does it mean that these younger couples have separate spheres of consumption?
A study in Sweden (Nyman 1999), where equality in consumption within the married couple is enshrined in law, reports very similar ideals to those found in the Singh study (Singh 1997). The Swedish study, too, found a discrepancy between the couples' view of what should be and what they perceived was actually happening. Although she did not attempt to measure actual consumption, Nyman reported that, while all 10 couples in her study agreed to the importance of sharing equally, the women perceived they had less access to entitlements and access to money for personal spending than did their husbands, a view supported by the husbands' accounts.
The striking similarity between Nyman's findings on the ideology of sharing and those of Singh is not surprising given that both sets of respondents were of European backgrounds, although in different locales. But while Singh perceived caginess in talking about money, Nyman's respondents displayed no such inhibitions. The women expressed their dissatisfaction with their husbands' spending money on personal consumption, for example, hobbies. They shared explicitly and were not wary of calculating shares and estimating contributions.
The age difference between Singh's and Nyman's groups could be one explanation for the willingness on the part of the younger women to challenge the dominant ideology of the family.
Interestingly, even though eight out of the 10 couples pooled their incomes, there was great variation in the ways in which they organised their finances, pointing to the low association between management and control (see below).
A study by Singh and Lindsay (1996) develops Singh's earlier work and contrasts marriage money with cohabitation money. Of the 15 couples in the cohabiting group, 14 had separate main bank accounts. The only couple with a joint account saw their relationship as permanent; while over the 12-month period of the study, two couples got married and opened joint accounts. This predominance of separate finances among cohabiting couples could be seen to reflect generational change and a rejection of the 'ideals' of marriage, as much as it reflected differences in degree of commitment to relationship—more than half of Singh's group was aged 35-54 married with children, while in the Lindsay study most couples were in their twenties, had never been married and did not have children.
This distinction is significant when one attempts to define money in more nebulous relationships. All the studies that we have dealt with so far have focused almost entirely on married couples with dependent children. Given that such households are an increasingly small subset of the family types in which children live, it is important to understand the differences in ideology about money.
How does that ideal of sharing mesh with the observed inequality within a couple which is in turn engendered by social norms and income inequalities? To answer this, one needs to peel the next layer—that of management of money and its control.
Early work on household financial management practices focused on the distinction between management and control. The studies aimed at refining typologies or classification of management system in order to separate day-to-day management of household incomes from the final decision-making authority. This focus was based on the observation that while women seemed to be running the household, paying bills and making purchases, true power to decide how much to spend and on what or which member of the household did not lie with them. This literature is relevant today for its insights into the manner in which households manage their finances and the gender-based differences in control over resources.
Drawing on previous work and on her own study of a group of women whose marriages had broken down because of their husbands' violence, Pahl (1980) found that while the proportion of women performing the daily functions of paying bills, buying food, and so on, was large, the percentage who made the final decisions about how household income was to be spent was small. Part of the data for her study came from interviews with 50 battered women at a Women's Aid refuge in Britain. All the women had children. Only a small part of the study had focused on financial arrangements within the household, but it soon emerged that money played a significant role in marital difficulties.
Given the circumstances of the respondents and the fact that only one partner was interviewed, the results from the study are likely to be non-representative. Moreover, income questions are notoriously hard to ask and a large percentage of respondents did not, or could not, give estimates of their husbands' incomes. As well, the survey was confined to a particular family type—married with dependent children—and it is likely that rather different allocation systems would be at work in other families, depending upon the nature of their relationship, life-cycle stage, and so on.
Despite these caveats, however, the simple three-part typology of allocation systems Pahl developed has been found useful for most later research. The types she proposed are—the 'whole wage' system, the 'allowance' system and the 'pooling' system. In the whole wage system, the husband hands over all his wages to his wife who manages the family's expenditure, giving him back a certain amount of personal pocket money. This pattern of financial management was common in the past. It was the custom in the countryside in Oxfordshire in the 1880s, when the farm workers would hand over their 10 shillings to their wives and receive a shilling back (Thompson 1954).
Many husbands boasted that they never asked their wives what they did with their money as long as there was food enough, clothes to cover everybody…and they seemed to make a virtue of this and think what generous…fellows they were. (Thompson 1954)
Pahl is quick to note that rather than being a significant source of power for the spouse whose job it was to manage the family's income, it became an onerous chore when money was short.
Other studies, too, have reported this system. Land found that it was most likely in families on social security (Land 1969). A modified system, in which the husband first took out his pocket money before handing over his pay packet, was described by Rowntree (1954) and Brennan (1959), among others. The amount the husband kept as personal spending money would in most cases be more than in Pahl's unmodified system; some of the husband's pocket money might also be used for collective spending or saving.
The second system described by Pahl was the allowance system under which the husband gave his wife an allowance, sometimes called her wage, based not so much on his income as on some norm of what was an appropriate sum. In this situation the wife gets a fixed amount each week to run the household. But under this system, the allowance tends not to keep pace with inflation. Moreover, the lower the husband's income the less likely the wife would be to see an increase in the housekeeping allowance, while the range of items of household consumption that she would be expected to pay for would also be greater.
The pooling system seems to be more characteristic of families where both husband and wife have earnings. Pooling of incomes, however, is not always complete—the wife's earnings usually go to paying for household expenses, and having an income might not give her the same degree of financial independence that it does her husband (Hunt 1968). In fact, some argue that the power of women vis-à-vis their husbands actually declines when the former earn an income, for their earnings become earmarked for collective expenditure while the husband keeps more for his personal consumption (Jephcott, Seear & Smith 1962)2.
In Australia, the pioneering work on financial arrangements within families has been undertaken by Meredith Edwards (Edwards 1981; 1982). In what she called a 'small and exploratory' study, Edwards surveyed 50 married couples with dependent children under the age of 16, in Queanbeyan, New South Wales. The couples were divided into three groups, one of 20 single-income couples, one of 20 dual-income couples where the wife earned at least 30 per cent of the family income, and one of 10 dual-income couples where the wife earned less than 30 per cent of the family income. Husbands and wives were interviewed simultaneously but separately. Respondents were asked about their housekeeping expenditure, personal spending money and use of and attitude to credit, as well as the use and importance of family allowances and their attitudes to financial arrangements.
Edwards distinguished explicitly between financial control and management. Control referred to the function of decision-making, while management was the implementation of these decisions. The person making the decisions, she argued, was the one who had control over all but the very minor expenditures within the family.
Adapting the typology developed by Pahl (Pahl 1980), Edwards proposed a schema with four categories—a 'husband-managed system' in which the wife gets a small personal allowance; an arrangement where she gets a 'housekeeping allowance' with which to manage the household's day-to-day expenses; 'joint management' where both partners pay the non-personal bills; and the 'independent management system' whereby the partners keep their incomes largely separate. She thus dropped Pahl's whole wage system, and added instead the husband-managed system. She also added the idea of independent management. These changes were perhaps a reflection of the fact that 60 per cent of her sample comprised dual-income households.
Half the families reported the wife as being responsible for managing the family's finances. These families were more likely to be in the lower income bracket, regardless of the labour force status of the wife. Most husbands expressed their satisfaction with the arrangements—the lower income ones were happy to be relieved of the day-to-day worries while the better-off felt it saved them time. In 14 per cent of the families, husbands managed the finances completely; more than 70 per cent of these were in the high-income bracket. Here, the labour force status of the wife appeared to have an influence. In another 14 per cent, the couple managed jointly, while in 22 per cent the couple practised independent management of finances.
In terms of control over money, of the 30 women in paid employment, only nine (less than a third) had no control over family finances; of the 20 who were not in paid employment, the number rose to 11 (more than half ). Nearly all the men (95 per cent), however, had either joint or exclusive control of family finances. Interestingly, of the three instances where the wife had control over the family money, two were lower income families where the wife did not work, and the other was a family where the wife's income was less than 30 per cent of the total income.
One must note that the high proportion of women classified as having joint control are those cases in which there was ambiguity between the husband and wife responses. Edwards does not report on the extent of discrepancy between husband and wife responses—several studies have found a great degree of dissonance between partners' responses (see Blumstein & Schwartz 1983; Pahl 1989)
Other studies have sought to replicate Pahl's findings by adapting her typology. Burgoyne tested Pahl's revised typology (Pahl 1989) by means of semi-structured interviews with mainly middle-class couples where most of the women were in the process of returning to the workforce. She found that the typology could be used to trace major shifts in the balance of economic power as a function of the life-cycle and of the associated changes in the source of income. The findings, however, also showed that a couple's report of the pooling of incomes could give the erroneous impression that resources were equally shared. In particular, the 'rights' of ownership associated with having earned the income may remain hidden, and lead to patterns of control with potential consequences for a non-earning partner (Burgoyne 1990).
The growing body of research on the intra-household economy suggests that in couple households there are significant associations between control over household finances and more general power within the household. However, most of the early research has been based on relatively small samples. Volger and Pahl (1993) revisit the issue with a study based on a major British data set, produced by the Social Change and Economic Life Initiative, to examine the relationship between money, power and inequality within marriage.
This study identifies six different systems of financial allocation, a development of the initial three-part typology proposed by Pahl (1980). The results suggest that, even when couples nominally pool their money, only one of them controls the pool in practice. In only one-fifth of couples was the pool actually jointly controlled, and these were households characterised by the highest levels of equality between husband and wife in terms of decision-making, experience of deprivation, and access to personal spending money. Findings indicate a complex pattern of relationships between household income level, household allocative system and gender. Female control of finances, though it was associated with greater decision-making power for women, did not protect them against financial deprivation; however, male control of finances, especially when it took the form of the housekeeping allowance, did serve to protect the financial interests of men in comparison with women. Gender inequality was least in households with joint control of pooled money, and greatest either in low-income households or in higher income households with male control of finances (Volger & Pahl 1994; Pahl 1995).
Criticism of allocative typologies
By constructing a typology, Pahl's and other works are open to criticism of being rigid and ignoring the reality of the fluid nature of household relations. Wilson (1987) commented that the whole wage system was almost indistinguishable from the allowance system. Snape Molloy & Kumar (1999), too, were of the opinion that methods of financial organisation do not consistently conform to rigid typologies and are instead fluid and variable within these typologies.
As Pahl herself points out, 'there is an infinite variety of different allocative systems…[the] focus [is] on that particular household type which is composed of two parents and their dependent children' (Pahl 1983). She stresses that her proposed typology represents points on a continuum of allocative systems, but points out that previous research suggests that it has considerable validity in Britain (Pahl 1983).
Citing Saflios-Rothschild's (1976) work on financial control and management, Woolley (2000) presents a diagram that conceptualises different modes of financial management in a two-dimensional space rather than as part of a rigid typology (see Figure 1). Control is measured along the y axis and management along the x axis. Positive values reflect male control or management and negative values female control or management. For example, the whole wage system is in the quadrant denoting female management and male control, while the whole wage system is located in the male control-male management quadrant.
Another criticism levelled against Pahl's work in particular has been that it neglects the importance of the source of income (Wilson 1987; Molloy & Snape 1999a). Ritchie (1990) argues that the model is built around earnings and does not have room for smaller sources of income such as Child Benefit, which might be managed differently. While the primary focus of the model is indeed income from earnings, given that it would be the bulk of the family's income, Pahl's model does allow for women's earnings. Thus, the whole wage and the allowance systems are prevalent when the male is the main or sole earner, while the pooling and the independent management systems (especially the latter) are more likely when the female has her own income too.
Edwards' and Pahl's studies were conducted in the 1980s and it can be argued that the numerous changes in the labour market since then, including the increased participation of women, the decline of manufacturing jobs, greater use of electronic banking, the significant proportion of both women and men in the labour force with dependent children (40 per cent), and so on, would have implications for within-household management.
In a review of three studies—two conducted by the Australian Institute of Family Studies called the Australian Family Formation Project (1981-82) and the Australian Family Formation Case Studies (1982), the third by Australian National University called the Australian Family Project (1986)—Glezer (1995) commented on the changes that have occurred 'with a shift in ideology away from traditional family values and an associated decline in the breadwinner model of the family'. She argued that, 'accompanying these changes are a greater tolerance for a variety of family forms and a rise in an egalitarian companionate model'. She also examined the behaviour of couples within households with respect to control and management of finances.
The first wave of the Australian Family Formation Project was conducted in 1982, comprising 2544 interviews with adults between 18-34 years of age. Of these, 1923 were contacted again in 1991-92 and telephone interviews conducted with 1536 respondents.
The nature of questions asked was direct, for example, who has control over money? Given the paramount importance most married couples attach to the ideology of jointness and of sharing money in marriage, and given the manner in which they shy away from any discussion of the power imbalances that income inequalities might engender in a couple, it is likely that most couples would profess to joint management. And indeed, 75 per cent of the couples said that financial control in the household was shared or split 50/50 between the two partners.
With respect to two-income couples, Glezer reported there was no statistically significant relationship between demographic variables like gender, age, religion or education and financial management patterns. Household income, being a social security beneficiary, or husband's occupational status were similarly not significant. In relation to the wife's contribution to the household income, no significant relationship was found when the proportion of her income to her husband's was 30 per cent or less, but when her income was equal to or more than her husband's, there was slightly less pooling of incomes.
In 1981, as part of the Australian Family Formation Project, case studies were conducted of 58 married respondents. Joint control was reported by 75 per cent of these married couples. By 1991 this proportion had risen to 81 per cent. This was supported by the Australian Family Project which found that, in 1986, 85 per cent of the couples reported joint decision-making on major purchases. On the basis of these findings, Glezer concluded that there was a high degree of pooling among two-income couples (as evidenced by the 83 per cent keeping joint accounts), as well as of joint decision-making.
Factors influencing financial organisation
According to Pahl, the system of allocating resources within the family was linked to life-cycle stage, income level and occupational, regional and economic culture (Pahl 1980).
Drawing on her observations in the late 1970s of the women who came to the Women's Aid refuge, Pahl (1980) found that disputes about money formed a significant part of the account of their relationship with the men who had battered them; 70 per cent mentioned money as one of the problems. Interestingly, the interviews revealed a wide variety of management systems (pointing possibly to little association between management and control). Most women's primary task was looking after their homes and children; most families could be classified as belonging to the lower end of socio- economic scale. One-fifth of the women interviewed were responsible for managing their families' daily consumption expenditure while another fifth did not receive any money from their husbands. Two couples operated a pooling system while in the others the husband delegated responsibility for parts of expenditure to the wife while retaining control of the income.
Wilson (1987) lists level of income as one factor determining how a household organises its finances. Thus at low-income levels, the wife manages the money because it is her job to make do, while at high-income levels, the husband retains control over the money (Wilson 1987). This observation corresponds roughly to what others report (Brannen & Wilson 1987; Pahl 1989; Volger 1994). This link between household allocative systems and income levels has been challenged by other studies which show that different systems may be adopted by households at similar income levels (Molloy & Snape 1999). Despite this, research in this area has indicated that when financial management is a burden rather than a source of power, the responsibility is the wife's. This has been found to be true particularly of couples on benefit (Goode, Callender & Lister 1998). It has been suggested that this enables the men to avoid facing the fact that their income may be unable to support the family; it also might help them blame the family's financial troubles on their wives' lack of financial prowess (Volger 1994; Goode et al. 1998).
Sometimes, it is argued that women might be more skilled at financial management than their partners (Molloy & Snape 1999). Research evidence seems to suggest that they are more able to exercise restraint and it is thus they who manage household finances when money is short. Related to this was the perception among women that their partner's spending priorities were less practical. In some cases, women on income support described how the men would always spend money on themselves and how the men would not pay bills (Molloy & Snape 1999).
Number of children
Analysis of data from the second wave of the Australian Family Formation Project in 1990-91 (Glezer 1996) shows that couples with only one child were more likely than others to share financial decision-making, while in couples with three or more children, men were more likely to have a greater say in decision-making. Foreman and Wilson (1995), however, reported that wife-only arrangements for financial decision-making with respect to family allowance and income support payments increased as the number of children in the household increased.
Wife's share of income
The exact nature of the relationship between share of income and power in a relationship has been the subject of debate. Drawing on resource theory, it has been argued that the power to make decisions about money in the household will be related to the contribution the individual makes to the household income. That this is true for the male partners has been documented: Ritchie (1990) reports that the males in her study saw themselves as entitled to greater authority over decision-making due to their status as the principal earner. Other studies have shown that women who are employed have greater control over decisions (Volger 1994; Pahl 1983). Edwards (1983) found that women who had an income, or who could claim benefit in their own right, had greater control over expenditure and saving than women who did not.
The literature discussed thus far has been primarily qualitative with little quantitative analysis of the data. Woolley (2000), in a preliminary analysis of a study of 300 households in the Ottawa-Hull region in Canada, looked at English-speaking couples with dependent children, median incomes of $65 000-$70 000; 88 per cent were married; the median length of the relationship was 10.5 years; and more than 15 per cent were in second marriages. She finds most couples (63 per cent) report a joint account as their primary account. When they do hold separate accounts, women are just as likely to hold these accounts as men; this she attributes to the former undertaking day-to-day management functions for the household. Women are responsible for cheque writing, while men make more cash withdrawals. Woolley points out that cash, by its nature, does not leave a paper trail and confers more discretionary spending power while not involving time-consuming administration and management.
She then tests the hypothesis whether the management of the family finances by either the husband or the wife is influenced by each partner's economic position and opportunities, both inside and outside marriage. With control over money as the dependent variable, she found that male control over money was greater the greater his income. In the case of women, 'better' outside options, like age and age difference, which affects their probability of remarrying, were found to improve their control over money, since younger women were in a relatively more advantageous bargaining position. If the male partner had been married before, the main account was likely to be in his name only. This replicates the finding by Treas (1991) that remarried couples were more likely to keep their finances separate3. The variable, female income, had the expected sign, that is higher incomes increased the degree of female control, but was not significant.
That money management and control are related more to gender than income or status was explored in a study by Tichenor (1999). This study investigates so-called status-reversal couples in which the wife either earned more than her husband or worked in a higher status job, or both. She found that, not only did most of these wives still tend to do the bulk of the traditional woman's work of housekeeping, and so on, they also felt guilty about not doing enough, and lauded their husbands for taking over some of the housekeeping duties.
Making more money did not translate into more power for these women. It did not give them more decision-making power, nor did they adopt an 'I make more, I'll decide how it is spent' attitude. While most spouses, especially husbands, completely dismissed the idea of a link between money and power, status-reversal wives seemed more uncomfortable with the idea and went out of their way to minimise the social anomaly of their situation. No status-reversal wife in the sample demanded extra authority over money or decision-making, even when some realised that, in the conventional situation, their husbands would assume control automatically; this did not seem to be a problem for them. One wife reported that she backed away from exercising control over decision-making to avoid being labelled a 'bitch' by her husband and his friends. Decision-making patterns within status-reversal households did not seem to be related to income and status.
What emerges from this study is the strong hold of conventional ideas about gender roles. More than half (64 per cent) of the status-reversal couples were dissatisfied with their relationship at some level. The chief complaint of the wives was that their husbands were not living up to their potential. Some wives wanted their husbands to bring more money into the household; over a third wanted a more conventional relationship. One wife said, 'I feel resentful that I make more. Like I wish it was the other way around'. For both husbands and wives, the tension revolves around the fact that husbands are not meeting their traditional obligations.
The study challenges the assumptions of resource and exchange theory as they relate to marital power. Variations in occupational status and income appear to have little impact on marital power, and couples tend to organise their lives in ways that hide or ignore these variations.
Other studies report similar findings. Stamp reports that breadwinning wives held back from exercising as much power as they might, given their financial contribution, and that they tended to involve their husbands in financial responsibility (Stamp 1985). Another study describes how a husband who chose to swap roles and be financially supported by his wife still maintained a strong bargaining position with regard to housework, because of the ideology supporting the belief that housework was her responsibility rather than his (Hunt 1978; 1980). In another study, the tension created by the wife being of a higher social class was resolved by maintaining joint accounts. Here, the ideology of equality was used to conceal the reality of inequality (McRae 1987).
Income from benefits
Although not much work has been done on this issue, there is evidence that income from benefits is perceived differently from income from paid employment, and that notions of individual entitlement are weaker. Such income is more likely to be perceived as being for the collective use, as Wilson reports (1987). Additionally there is some evidence that people may tend to accept the official definition of the benefits as collectively earned (Goode et al. 1998). These views might still be mediated by ideas about the rights of the breadwinner (Volger 1998). Thus unemployment benefit, which makes eligibility dependent on being available and actively seeking work, confers a notion of the benefit having been 'earned' and therefore there being a right to personal spending money (Goode et al. 1998).
A study by Foreman and Wilson (1995) is important because it directly addresses the question of whether Australian women do indeed have greater control over the family allowance, since the 1991 change in policy that directed it 'from the wallet to the purse'. In a paper based on data from surveys of families receiving family allowance and income support payments, they found that women had sole responsibility for decisions about how to spend the family allowance in 38 per cent of the families (58 per cent reported joint decision-making). In the majority of cases, family payments were deposited into bank accounts to which the wife had sole access or the couple had joint access. They compared the family allowance with income support payments typically made to males and found that couples do indeed treat the two government payments differently. With income support payments, joint access (68 per cent) and decision-making (80 per cent) were more likely than with the family allowance.
In a study commissioned by the Department of Social Security in the United Kingdom, Molloy and Snape (1999b) also looked at the ways in which couples in receipt of either income support or Jobseeker's Allowance managed their incomes. It is interesting that their findings are similar to the older studies. Thus women mostly manage the day-to- day finances; there are few instances of complete male control, and no examples of the independent management system. They found it much more difficult to disentangle control, and this led them to argue that the balance of power between partners is often a grey area, with no clearly defined boundaries.
The decision by a couple to use a particular allocation system has been found to be influenced by their previous experiences—either the system followed by their parents or their own prior or present relationships. Volger and Pahl (1993) report that nearly half of those using the female whole wage and the housekeeping systems had been brought up in households that had used these systems. Other respondents spoke of deliberately not choosing to be like their parents. Some respondents spoke of the desire to retain greater control over their finances in a new relationship where they were being pressured to adopt pooling. Ritchie (1990) reports that women in second relationships spoke of their determination to retain some individual income after having been kept on a tight rein in their previous partnerships.
Glezer (1996) reports that between 1981 and 1991 the percentage of interviewees reporting joint management of finances rose from 75 to 81 per cent. She argues that this indicates a shift towards egalitarianism in marriage and shared decision-making in comparison with the parents' generation. She quoted one man saying, 'We have joint everything because I had seen both sets of parents conflict over money. I wasn't going to fall into that trap'.
What are the implications of these patterns of management and control for different members of the household? Most studies have focused on the inequity along gender lines, though later work has begun to address the impact on children's wellbeing as well.
Access to resources
While Pahl did not explicitly analyse her typology for management and control, she did explore the implications for individual living standards of different patterns of financial allocation and responsibility. She found evidence of poverty among the wives and children of even fairly well-to-do men. Where marital discord was rife, husbands often withheld money from their wives. Among the 25 women in the follow-up group from her study of women in the Women's Aid refuge, only one had been given a specific amount of money as her own pocket money. These findings reinforce the observation that resources within the household were inequitably distributed. Some women reported that they were better off living on Supplementary Benefit (government payments) than they had been while living with their husbands. As Supplementary Benefit levels are taken to represent the poverty line in Britain, by implication these women and their children had been living below the poverty line. As one woman commented after she began receiving the benefit:
She [her daughter] has started to sleep all night…we were that short of money that I couldn't give her as much to eat as I can here [in the refuge]…he used to spend 4 pounds a week on cigarettes…more or less half our food money. (Pahl 1980)
If there is unequal sharing of resources within a household, against women, then conventional methods of poverty measurement will likely lead to serious underestimates of the intensity and incidence of female poverty (Findlay & Wright 1996). Simulations on micro-data from Italy and the United States as part of the Luxembourg Income Study show that, if one assumes that women transfer 50 per cent of their household income share to men and children, their poverty share would increase from 50 per cent in Italy and 60 per cent in the United States to around 80 per cent in both countries. Other researchers too report similar findings.
When faced with financial difficulties, who is more likely to cut down their consumption? Adelman, Middleton and Ashworth found, from their analysis of Family Expenditure Survey data in the United Kingdom that children were least likely to suffer a reduction in necessities like food, clothing, and so on, and that most of the brunt of the family's economies were borne by their mothers, and to a lesser extent by the fathers. The survey design was interesting because it asked what household members lacked. Hence, it seems to be a question of their perception of the difference between what they should have and what they actually did have.
Pahl also asked this question of her respondents and found that, not surprisingly, the women reported that they did a greater degree of cutting back of consumption than their husbands did. She also found that, although husbands defined their wives' income as their wives' own money, it was more likely to be spent on the children. Moreover, this was more likely to be the case in households where the wife managed the money, and hence in lower income households (Pahl 1995). Snape et al. also found that, where women manage the finances, they tend to have less money for personal expenditures than their partners do, and to consume fewer resources. They also tend to prioritise the needs of other members of the household above their own. Another of their findings was that men seemed much less aware of inequality of resources. Molloy and Snape found that spending on children varied little with income and that, as a consequence, the proportion poorer families spent was considerably more than that spent by richer families. They also found that mothers were one and half times more likely to claim to go without daily necessities like food, adequate clothing and shoes than their male partners.
The source of income is important in assessing access to resources. Thus, in some cases the male partner used the primary household benefit as money for paying the household bills, while Child Benefit (United Kingdom) was viewed as the responsibility of the female partner to meet children's needs (Snape et al. 1999).
In a move away from small qualitative studies, Cantillon and Nolan (1998) used data from a large sample in Ireland, collected through a specifically designed survey carried out in 1987 by the Economic and Social Research Institute. They measured the difference between spouses using non-monetary indicators of deprivation to look at intra-household differences, and found a quite limited overall imbalance in measured deprivation in favour of husbands. This would seem to suggest that fears about hidden poverty might be overstated, that the incidence of poverty among women married to not-poor men might be low, and that there may not be much greater deprivation among women than men in poor households—at least in the context of this study. Their results, however, might not hold at the lower end of the distribution or in specific cultural contexts.
Spending patterns within households
Spending patterns within households have implications for the wellbeing of family members.
Using data from the United Kingdom Family Expenditure Survey (1993-94), of nearly 3700 married couples, Pahl (2000) reports that spending within the household is highly gendered. Women spent more than men on food, women's and children's clothing, child care and educational courses, while men spent more on alcohol, motor vehicles, repairs to the house, meals out, gambling and holidays. (This does not imply that women and children would be excluded from the activities that the men paid for.) Also, women were responsible for 40 per cent of the amount spent on men's clothing. Women bought more low-cost items, with 75 per cent of items costing less than 10 pounds being purchased by women. Men, on the other hand, bought 60 per cent of the items that cost more than 1000 pounds.
The Family Expenditure Survey found that women's employment status made a significant difference to patterns of spending. Women spent significantly more on food when they were not in paid work, though even when both husband and wife were in paid employment, the wife's spending still accounted for 75 per cent of the spending on food. Women doubled their spending on holidays when they were in paid work, while spending on meals outside the house was greater for both men and women, significantly so for the latter. Interestingly, women in employment tended to spend significantly more on gambling too. Spending on household goods and motor vehicles was not affected by women's employment status.
In her own study, Edwards asked respondents about the manner in which the family allowance was spent (Edwards 1981). The most common use was for buying children's clothes (33 per cent of cases), followed by general household funds (18 per cent), and trust funds for the children (12 per cent). Expenditure on children's clothes was reported across the income strata and was not related to the mother's contribution to household income. In the early 1990s, Foreman and Wilson (1995) reported that the family allowance was mainly spent on items for children (75 per cent). They found that in most families women had sole access to and control of the family allowance. They also found that spending on the children was predominantly a wife-only activity.
Significant in terms of ease of access to resources and greater autonomy with respect to decisions about expenditure, is the finding that women who were not in full employment were much less likely to use credit cards, compared with their employed husbands (Pahl 2000).
Snape et al. (1999) discussed a number of psychological effects of different systems of financial management within the household. One effect is the stress felt by women with responsibility for the money in low-income households, especially if they have children. Their stress was greater when their husbands not only took little responsibility in management, but also made requests for additional resources for personal consumption. This was particularly the case when the wife's management of the finances enabled the husband to ignore the gravity of the financial situation.
On the other hand, women could gain a sense of security through knowing that the bills were being paid. This was particularly true in the case of women whose partners had previously failed to meet their financial obligations.
Managing resources can also be a source of self-esteem, particularly for women who had long experienced working under a male boss; financial responsibility boosted their self-esteem and they felt they were fulfilling an important function within the household. By contrast, men who perceived themselves failing in their the role of breadwinner or provider for their families, especially if they had children, described a loss of self-esteem if their wives handled the family finances.
Like Tichenor in her study of status-reversal couples, Snape et al. (1999) also found that some women were very conscious of the potential threat their taking responsibility for the finances posed to their partner's identity as breadwinner. They tried to avoid this challenge by playing down their own managerial role or highlighting their partner's.
When the household finance system is male-managed, it is common for the husband to retain part of the household income and this leads to perceptions of inequality on the part of the wife. Moreover, having to ask their husbands for even the smallest item of expenditure makes wives feel powerless.
Inequities such as these can create problems for relationships. Women can resent shouldering the responsibility for managing meagre resources. They also sometimes report losing respect for their partners (Snape et al. 1999) but when men are responsible for financial management, women can resent the control they exercise.
The ideology of money in marriage is thus one of privacy and of jointness; it is supposed to be jointly controlled by the husband and wife, with the marital unit defining its strict boundaries. While this dominant view is being challenged by the ideology of the individual's rights to what they earn, it continues to shape thinking in research and policy. Drawing upon the distinction between management of household finances and real control over them, the literature reports a dissonance between ideology and actual practice. Most studies find that management and control roles within the household are largely gendered and have little or no relationship to household income levels, education, age, or even the wife's contribution to household income.
3. Economic studies4
In the qualitative analyses discussed previously, evidence for the pooling of income within married couple households with children was mixed. The studies, however, are necessarily small-scale and localised. Can anything be gleaned from large-scale survey data about the extent of income pooling within households and its impact on expenditure?
Traditionally, most economic work analysing survey data has treated the household, and not the individual members, as the unit of analysis. Decisions about consumption and production are assumed to be arrived at by all members of the household. But there is increasing evidence that this is not the case.
Before going on to discuss the economic studies, however, it must be noted that the term 'pooling' is used in somewhat different senses in the two strands of literature. The qualitative/sociological studies talk of pooling simply as a management practice in which couples put their incomes together. This does not imply, of course, that the partners get equitable shares of the income, since that is the very question at issue in these studies. Households that pool incomes might still have iniquitous distribution of spending patterns. Economic models, however, refer to pooling as the situation where the source of income, whether it be the husband's or the wife's, does not affect the way in which it is spent—an additional dollar of the husband's income will be spent in the same way as an additional dollar of the wife's. These two types of pooling behaviour are not necessarily the same. Couples may manage their money from a common pool, but make expenditure decisions that depend on who brought the money into the household. Similarly, spouses may manage their finances separately, but pay attention to the preferences of the other spouse so that expenditure does not depend upon the source of income.
The common preference model, or the one that implies (economic) pooling of income within a household, has been challenged by models that attempt to incorporate the divergent and conflicting preferences of household members into economic analysis. The allocation mechanism in these individual utility models have included cooperative bargaining (Manser & Brown 1980; Lundberg & Pollak 1993), noncooperative bargaining (Kanbur & Haddad 1994; Lundberg & Pollak 1994) and a generic collective approach that avoids specifying a particular mode but assumes that family allocations obey a Pareto-efficient sharing rule (Chiappori 1988, 1992). A separate spheres version of the cooperative bargaining model is proposed by Lundberg and Pollak (1993). Here the non-cooperative outcomes involve voluntary contributions to household public goods, but social norms have resulted in a gendered assignment of responsibilities.
From the perspective of policy that seeks to influence consumption behavior or the intra-household allocation of resources, the key question is how the intra-household distribution of decision-making power varies in line with observable characteristics, and in particular, characteristics that can be influenced by policy decisions.
There is now a substantial body of research addressing these empirical questions. The general conclusion from this research has been that there are many factors related to the distribution of power within the household that are associated with consumption behavior, though some have questioned the robustness of these conclusions.
The most frequently employed estimation strategy has been to examine how consumption patterns change as the wage income shares of husbands and wives vary. Because consumption patterns may vary with hours of work, these estimates restrict attention to couples with both members working full-time. Reports using this approach include Bourguignon et al. (1993), Browning and Chiappori (1998) and Phipps and Burton (1998). The latter study, using 1992 Canadian Family Expenditure Survey microdata, found that husbands and wives pooled their incomes for certain categories of expenditure but not for others. Using a sample of husband/wife couples in which both the partners were employed full-time/full-year, they estimated Engel curves for 14 categories of household consumption and test the hypothesis that an additional dollar of male income is spent the same way as an additional dollar of female income. The hypothesis was rejected for the following categories: restaurant food; household food; wife's, husband's and children's clothing; child care: and transportation flows. However, they found they could not reject the income-pooling hypothesis for other items like housing. They explore what male versus female income is used for various categories in which they reject income pooling and find that couples were more likely to use their own incomes for items of personal consumption, and that expenditure on household goods reflected gendered spheres of responsibility. Thus, an extra dollar of the wife's income was more likely to be spent on child care; while an extra dollar of the husband's income was more likely to be spent on transportation flows.
Lundberg et al. (1997) have argued that while such tests based on cross-sectional variation in income received by husbands and wives have decisively rejected the pooling hypothesis, these tests are not reliable since they exploit differences in incomes that are not exogenous with respect to prices, preferences and other important differences in consumer behaviour. For example, the within-household income distribution may be influenced by other factors such as relative wage rates, which in turn may directly affect consumption patterns.
A more straightforward criticism of the approach of Phipps and Burton (and similar studies) is that many categories of expenditure may contain elements of work-related expenses, which may increase with the wage rate (even among the population of full- time workers). This may explain phenomena such as increases in female clothing expenditure as the female income share increases (Bradbury 2003).
An alternative approach is to examine the outcomes when there is an exogenous change in income received by husbands or wives. Lundberg et al. (1997) took advantage of a 'natural experiment'—a change in Child Benefit policy in the United Kingdom that transferred income to wives. Using Family Expenditure Survey data, they found 'strong evidence of a shift toward greater expenditures on women's clothing and children's clothing relative to men's [that] coincided with this [change in] income distribution'.
Until the late 1970s, the United Kingdom Child Benefit scheme had consisted primarily of a reduction in the amount withheld for taxes from the father's pay cheque. At that time it was replaced by a cash payment to the mother. This represented a substantial redistribution of income 'from the wallet to the purse'—in 1980, Child Benefit amounted to 500 pounds for a family with two children, or about 8 per cent of the average male earnings in the United Kingdom. For many families, the new scheme was a substantial increase in income, because the tax allowance in the old scheme would have benefited only those families with incomes high enough to be taxed.
Focusing on clothing, since such expenditures are assignable to individual members within the household, Lundberg et al. used the Family Allowance Survey data for the period 1973-90, with the period 1973-76 taken to represent the period before the policy change and the period 1980-90, the period after. They estimated linear regression equations with the dependent variable either the ratio of children's clothing expenditure to men's clothing expenditure, or the ratio of women's clothing expenditure to men's clothing expenditure. The independent variables were the number of children, income and dummy variables representing the period after the policy change for different family sizes. They found a significant increase in expenditure on children associated with the policy change. They pointed out, however, that this does not imply causation. One explanation that they suggested could be changes in tastes, for example, brand-consciousness among school children.
The policy change that occurred in the United Kingdom involved two changes. One, in the mode of delivery of the benefit; from being a tax deduction, the child benefit was given in the form of a pay cheque. Two, in the person receiving it; from being given to the main earner, the father, it was now directed at the primary carer, the mother. The resultant change in expenditure patterns thus is the result of both changes. Given the discussion in Section 2 about the lack of fungibility of money, it would be interesting to analyse how much of this effect is due to the shift from a tax deduction (which is largely hidden from the main spenders, that is the wives) to a direct transfer in the form of a payment that is explicitly for the child and how much to the change in recipient. As Snape et al. (1999) find, some male partners tend to use the primary household benefit as money for paying household bills, while Child Benefit (United Kingdom) was viewed as the responsibility of the female partner to meet children's needs.
The literature on the differential uses within the family of male and female incomes was influential in the design of a large new program in Mexico called PROGRESA.
It began operation in August 1997 to fight extreme poverty in rural areas, and one of its innovative features was attempting to transfer monetary assistance to women (Haddad 1999). There are several other examples in policy of income transfer being successfully targeted at women. The Grameen Bank's targeting micro-credit at women in Bangladesh is one such instance. Pitt and Khandekar (1998) show that loans to women have a positive impact on outcomes like children's education, contraceptive use, the value of women's non-land assets, and so on.
What happens, however, when men perceive the decline in their incomes as a result of redistribution of incomes from them? Dasgupta (2001) raises the possibility that small redistributions could lead to a decline in women's welfare. If the husbands reduce their own voluntary spending on domestic public goods by more than the amounts their wives receive as additional income, then, there will be a reduction in welfare. Women may respond by increasing their own spending on children; if this increase is more than the reduction by men, then the children's welfare could increase. Moreover, even if there is a relative increase in women's wage rates, their welfare could still decline if they increase their income by increasing their labour market participation, and hence independent income, but sacrifice leisure. Dasgupta models this as a Cournot model of a two-adult household. The difference between men and women arises due to their attitudes towards spending on children and leisure. For men, the two are complements, whereas for women they are likely to be net substitutes.
There can be unanticipated and costly consequences from programs that are targeted at women. The empirical work needs to take into account the likelihood of a male crowding-out effect. This possibility has been flagged by Wilson (1987) who wrote: 'A few husbands appeared to have quite deliberately calculated the amount their wives got for child benefits and to have reduced what they thought necessary to give by that amount'.
Another consequence could be to put women at the risk of abuse. (Haddad 1999). In the context of Bangladesh's Grameen Bank, Goetz and Sengupta (1996) found evidence of women being used as 'fronts' for men who want to gain access to credit. This points to the need to evaluate the benefits from targeting women.
In the literature discussed previously, households and families are generally considered as synonymous. But increasingly, that is no longer the case. Drawing conclusions, then, about families from household data, or alternatively about households from data about families, is becoming more and more problematic. Edwards (1981) admits at the outset of her study that it has limited applicability for households in general since it examines the behaviour of one particular type of household: that of husband and wife and their dependent children under the age of 16, with no other adults living in the household. Even in 1976, such households were only 30 per cent of all household types (ABS 1976). By 1997, only 72 per cent of families with children between 0 and 17 years, were so-called intact families; the rest were either step, blended or one-parent families (ABS 2002)5.
Data is collected on households rather than families primarily because households tend to be easier to define. Most statistical agencies define a household as a collection of individuals in a dwelling unit (Callister & Hill 2002). The definition of a family, however, contains no reference to dwelling units. Families, unlike households, are not spatially constrained; they are therefore much more difficult to study. The ABS defines a family as two or more persons related in a specific way, for example, by blood or legal ties. Under this definition, a married couple or an adult with dependent or non-dependent child(ren) would count as families. But spatially separated families are the new reality. In the United States, for instance, 20 per cent of fathers and 10 per cent of mothers aged 20-39 were living apart from a child aged less than 18 years. Over 25 per cent of children in the United States under 10 years old live without at least one parent. They are either in single-parent households or, typically, living with their mothers and their new partners (Callister & Hill 2002).
In the United States, there has been an increasing amount of research on remarriages and stepchildren. Prior to 1980, only a handful of such studies appeared in print; in the 1990s, the body of published work was three times larger than before. With the increase in the quantity of work, there has been a concomitant advance in the quality and complexity of issues addressed. Of particular interest are families in which the biological parent of the resident children has remarried. Such families would be difficult to identify in large-scale survey data unlike, say, single-parent families. Yet financial arrangements within such households are likely to be different in crucial ways from those in the first marriage, biological-parent family.6
Approximately half of the marriages in the United States represent a remarriage for one or both partners. Similar trends have been reported in Canada and Europe, although remarriage rates are lower than in the United States (Coleman, Ganong & Fine 2000). In Britain, more than 33 per cent of marriages in 1993 involved a remarriage for one or both partners (Burgoyne & Morison 1997). Another interesting trend is that though remarriage rates are falling in the United States, this does not mean that people are coupling less often; increasing numbers of couples are cohabiting, not as a precursor, but as an alternative, to remarriage (Bumpass, Raley & Sweet 1991; 1995).
In Australia too, remarriage rates have been falling dramatically for divorced women over the past 20 years. Between 1981 and 2000, remarriage rates for divorced women under 24 years fell by 50 per cent while for all other ages, it fell by 31 to 48 per cent (ABS 2002).
While there has been little work done on the financial effects of remarriage on children, analysing other effects can point to possible implications for the ways in which family finances are managed. Relatively greater attention has been paid in the literature to demographic trends, remarriage relationships, stepfamily processes, societal views of stepfamilies, legal issues and the effects on children of living with a stepparent (Coleman et al. 2000). It is with this last issue that this review is primarily concerned. Research on young and adolescent stepchildren has focused primarily on academic achievement, psychological adjustment and wellbeing, and behaviour problems.
How are remarriage families different? Citing a study in 1996-98 in New Zealand, Fleming (1999) argued that remarriage families failed in many ways to conform to traditional notions about the family based on the first marriage family group consisting of a man, a woman and their children. In New Zealand, approximately, one in three marriages end in divorce and in one of three marriages one or both of the partners have been previously married (Statistics New Zealand 1998). The objective of the study was to identify patterns of economic support, living arrangements and kinship-based obligations in the families of couples with children from a previous marriage or marriage-like relationship. It found confirmation for the belief that remarriage families are structurally different from first-marriage families. These differences make it difficult for such families to operate in ways that policy assumes they would act.
In the first place, the family as a unit that lives in a home, earns income and spends it on itself, is fractured in a remarriage family. Children move between two homes, non- custodial parents regard them as part of their families, and children report themselves as being part of two families and having two homes. The income a couple earns is not the only income available to the household, given child support payments.
Second, not all household members need to be related, either by marriage, by adoption or by birth. In a remarriage, there is no legal relationship between the children and the parent's new partner. Most children reserve 'Mum' or 'Dad' for their natural parents, whether they see them or not; most partners resist being called by these terms by their new partner's children. They refer to them as their partner's children and emphasise the lack of parental relationship to them.
Third, the tasks of parenting are not necessarily shared between the two adults in the household, but allotted along bloodlines instead. In terms of economic support, most of the women did not expect their new partners to provide for their children from their previous relationship. While the men of the household did contribute more than the women to general living costs, specific costs for the children tended to be met out of child support. Moreover, almost all the women were in employment to contribute to their children's support.
The typical situation that emerges is that of the children living with their mother and her new partner, while maintaining varying degrees of contact with their biological father.
Adelman et al. (2000) found that cohabiting couples were more likely to keep separate finances (6 per cent compared with 2 per cent of married couples). Possible explanations could be that cohabiting couples are likely to be younger and both at work; older couples might have been in previous relationships where money had been an issue and hence would prefer to share responsibility in the new relationship (Volger & Pahl 1993).
Burgoyne and Morison (1997) explored the patterns of control and management of money in a United Kingdom study of 20 couples in which one or both had been married before. Thirty-eight interviews were carried out with both the men and women (two men refused to be interviewed). Most of the interviews were held with both partners at once, although some were interviewed separately. The ages of the respondents ranged from 28 to 83 years, with the majority in the 30-55 age range. The majority of the families had originally conformed to the common pattern of mother and children with stepfather, although only half the couples still had dependent children, not all of whom were co-resident. There was a wide range of incomes across the sample, with roughly a third below £20 000 per annum and a third above £35 000.
A key finding of the study was the degree of separateness in financial arrangements. This is in sharp contrast to the findings about married couples. As many as half the couples reported using an independent management system, compared with the 2 per cent of couples in general. So uncommon in the general population was the independent system of management that Pahl (1995) excluded couples using this system from her discussion. In the Burgoyne and Morison study, this degree of separateness in ownership of assets or income was especially marked in the way the interviewees wished their assets to be divided after their death. In line with earlier studies, though, was the finding that the balance of economic power was still in favour of the men, who generally had larger incomes and owned more assets in their own names.
According to Coleman and Ganong (1989) there is ample clinical evidence that money is major source of conflict for remarried couples. Money matters are likely to be much more complicated in remarried families. For instance, the partners may have to cope with ongoing financial links with ex-partners through maintenance or debt payments, fluctuating family composition as children move from one household to another, the problems of merging the financial habits of two different families, and so on (Burgoyne & Morison 1997).
There is some work in the United States suggesting that remarried couples are little different from other couples (Fishman 1983). This researcher found that most remarried couples adopted some form of pooling system. Fishman designated about half of them as 'two-pot' families that tended to share the cost of the joint household expenses, but otherwise divided their spending for children along biological lines. Lown and Dolan (1988) refined the one-pot, two-pot dichotomy by adding a mixed system to the classification. Half the households reported a one-pot system, followed by the mixed management system; only 9 per cent reported totally separate money management.
As Burgoyne and Morison argue, this typology, fails to distinguish between management and control, unlike the simple classification schema proposed by Pahl (1989). Moreover, the methods employed in the studies did not permit a sufficiently detailed investigation of how the money was actually used.
Burgoyne and Morison wondered if the evidence they found of separateness in money matters was just a matter of life stage, rather than a reflection of the desire of second- time marrieds to retain control of their own incomes. In first marriages, joint accounts facilitate management in childbearing years when the wife is less likely to contribute substantially to household income. They speculate that this system persists even after the wife returns to the workforce. In remarried couples, however, the disruption in their money management enables them to adopt a different system. These researchers speculate that remarried couples might be doing what couples in first marriages might like to do.
The finding that financial arrangements might be different among remarried and other non-traditional couple types is reinforced by Vivienne Elizabeth's (2001) study of some cohabitant couples in New Zealand. Based on a study of a small group of cohabiting couples who deliberately 'disrupt…the conventions, both past and present, that govern the management of money amongst heterosexual couples especially those who are married', Elizabeth reports that couples usually followed joint money management or independent management. She focuses on independent money management arguing that this is likely to be the method favoured by the newer types of relationships. Those who engage in the practice of independent money management avoid the dilemmas of financial dependency. However, the goals of autonomy and equality exist in tension with each other. The practice of independent money management defines equality in terms of equal control over individually earned income, but the question of equal access to money is not addressed. She found that problems arose when there were disparities in income.
The Australian Family Formation Project study was one of the few to include cohabiting couples. On the basis of this study, Glezer (1996) reports that they are significantly different from both those in first marriages and those who have remarried. While 72 per cent of first-time marrieds and 66 per cent of the remarried reported combining their income, only 20 per cent of previously married couples who were now cohabiting did so. There is thus less evidence of pooling of resources among cohabiting couples. Glezer and Mills (1991) also showed that while 71 per cent shared housing costs and 66 per cent would buy things jointly for the house, less than half shared bank accounts.
There is considerable evidence in the literature that children living in families with adults other than both their biological parents suffer numerous disadvantages which manifest themselves in low educational outcomes, psychological problems, and so on. Thomson, Hanson and McLanahan (1994), using data from the National Survey of Families and Households in the United States, relate this to the lower level of parental involvement in such families. Paternal support is higher in families with original father, lowest for children living with stepfathers. Income does not explain the differences. Interestingly, problems experienced by children living with their stepfathers or mother's cohabiting partner are associated with lower levels of maternal as well as paternal support. They speculate that this could be due to competition of the relatively newer marital/cohabiting relationship for time and attention that would otherwise be devoted to the parent-child relationship.
Exploring further the changes in mothering behaviour associated with remarriage or cohabitation, Thomson et al. (2001) consider three dimensions of mothering: time and supervision; harsh discipline, and relationship quality. Using data on single mothers from two waves of the National Survey of Families and Households, they find that mothers and children agree that mothers who remained in new partnerships used harsh discipline less frequently than mothers who remained single or whose relationships had ended by the second interview. Children reported less supervision and a better relationship with their mothers if they remained in their new relationship. This research is interesting for it is evidence of the vital link between mothers' partnership experience and their relationship with their children. This brings into question the altruistic model of mothering behaviour that is taken as a given—just as there are a range of couple relationships that need to be taken into account, there might just be as many different degrees of commitment by the mother to her children. This might be especially true in cases where the mother is in nebulous relationships. Although the literature dose not address the question and it would be unwise to dwell on the implications of such findings, it would be interesting to explore the patterns of spending on children by mothers in such situations.
It is in this area of non-conventional household arrangements, specifically, remarried and cohabiting couples and blended families, that one perceives the need and potential for further research into financial management within families. Most of current knowledge is focused on the traditional family where both biological parents live in the same household with their dependent children. The growing likelihood of children living with only one biological parent increases the urgency for such research, particularly, as the above review indicates, such children and households are more likely to be vulnerable—psychologically, socially, as well as economically. Along with small-scale, interview-based studies that explore the nature of financial relationships in such families, there is need, too, for large-scale surveys to incorporate such questions and elicit information on the composition as well as expenditure patterns of families. This would provide an opportunity for the questions raised in the sociological literature to be addressed by economic analysis.
In addition, the focus of research should move from the impact on current consumption by the family to more long-term impact on its wellbeing, particularly that of children. Research in developing countries has demonstrated that greater bargaining power of mothers results in better education and health outcomes for the children. From a policy perspective, such research in developed countries would be invaluable.
The review of the small-scale, interview-based studies of money in marriage highlights the ideology of privacy and of jointness; money is supposed to be jointly controlled by the husband and wife, with the marital unit defining its strict boundaries. Drawing upon the distinction between management of household finances and real control over them, the literature reports a dissonance between ideology and actual practice. Most studies reviewed here find that management and control roles within the household are largely gendered and have little or no relationship to household income levels, education, age, or even the wife's contribution to household income7.
These findings find support in the economic analysis of large-scale survey data on intra-household income sharing. While the evidence on women's income and labour force participation as a determinant of expenditure on children is mixed, these studies also report household spending as being gendered, with the women more likely to spend on children than men.
Finally, policy will need to bear in mind the increasingly complex nature of families and households as a result of repartnering, remarriage or cohabiting relationships. The major work on financial management in Australia relates to marital units a decade or two ago; things have changed since. Family boundaries are increasingly fluid and nebulous. Research on non-traditional family types indicates that responsibility for the children in the family are allocated along biological lines. Money matters are more complicated in remarried families and such couples tend to keep their finances separate. The extent to which this happens is not clearly understood and more research is necessary to ascertain the impact of these differences in financial arrangements in non-traditional families on the children living in them. Thus, if income is pooled only for public goods in such a family, then total household income would be an even more inadequate indicator of the resources available to these children compared with those living with both biological parents.
Overall, the review would indicate that the policy of spreading payments between the two partners should continue. It must, however, be borne in mind that the amount of income involved in the transfer seems to be significant in predicting the outcome. If the husbands reduce their own voluntary spending on domestic public goods by more than the amounts their wives receive as additional income there could be some crowding out of expenditure, which might affect the welfare of the recipient of the benefit. This is more likely to happen if the amount of transfer is large.
Another caveat that should be borne in mind is that the altruistic mother model might not hold in all situations. Almost all the studies that show gendered expenditure patterns involve traditional couples. In these studies the difference between men and women arises due to their attitudes towards spending on children and leisure. For men, the two are complements, whereas for women they are likely to be net substitutes. What happens in situations where the mother has greater control and management of household finances, as in, say, a sole parent or a remarried household? The review of the literature of mothers repartnering indicates that mothering behaviour could follow a continuum of commitment.
2. This is contrary to the ideology that a woman's income is her own. See 'Money and its lack of fungibility' in Section 2.1.
6. The terminology with respect to the different family structures is complicated and unsatisfactory. Several studies emphasise the importance of appropriate terms to describe the increasingly complex family arrangements (Callister & Hill 2002).
7. Although a number of the studies in this review relate to societies other than Australia, given their Anglo-Saxon and European contexts, it would not be amiss to broadly apply their findings about within-family structures to the Australian situation.
Adelman, LS Middleton, S & Ashworth, K 2000, 'Management of household finances and intra-household poverty: Evidence from the Poverty and Social Exclusion Survey of Britain', Working Paper no. 23, Centre for Research in Social Policy, Loughborough University, United Kingdom.
Agarwal, B 1994, A field of one's own: Gender and land rights in South Asia, Cambridge University Press, Cambridge.
Australian Bureau of Statistics (ABS) 1976, Characteristics of the population and dwellings—Australia, Census of Population and Housing, ABS, Canberra.
Australian Bureau of Statistics 2002, Marriage and divorces—Australia, cat. no. 3310.0, ABS, Canberra.
Blumstein, P & Schwartz, P 1983, American couples: Money, Work, Sex, William Morrow and Co., New York.
Bourguignon, F, Browning, M, Chiappori, PA & Lechene, V 1993, 'Intra-household allocation of consumption: A model and some evidence from French data', Annales d'Economique et de Statistique, pp. 137-56.
Bradbury, B 1996, 'Income support for parents and other carers', Reports and Proceedings, Social Policy Research Centre, University of New South Wales, Sydney.
Bradbury, B 2003, 'The Within-Household Income Distribution and Child Consumption Patterns, Module 2', Draft Final Report for the Department of Family and Community Services, Social Policy Research Centre, University of New South Wales, Sydney.
Brannen, J &Wilson, G (eds), 1987, Give and take in families: Studies in resource distribution, Allen and Unwin, London.
Brennan, T 1959, Reshaping a city, House of Grant, Glasgow.
Browning, M & Chiappori, PA 1998, 'Efficient intra-household allocations: A general characterization and empirical test', Econometrica, 66(6) November, pp.1241-78.
Bumpass, LL, Raley, RK & Sweet, J 1995, 'Changing character of stepfamilies: Implications of cohabitation and nonmarital childbearing', Demography, 32, pp. 425-36.
Burgoyne, C 1990, 'Money in marriage: how patterns of allocation both reflect and conceal power', The Sociological Review, vol.38, no.4.
Burgoyne, C & Morison, V 1997, 'Money in remarriage: keeping things simple—and separate', The Sociological Review, pp. 363-95.
Callister, P & Hill, M 2002, Are 'two parent' and 'single parent' families misnomers misdirecting policies and family/work research? Working paper no. 02-18, Bronfenbrenner Life Course Center, Cornell University, Ithaca.
Cantillon, S & Nolan, B 1998, 'Are married women more deprived than their husbands?', Journal of Social Policy, 27, 2, Cambridge University Press, London.
Cass, B 1978, 'Women's place in the class structure,' in EL Wheelwright and K Buckley, (eds), Essays in the political economy of capitalism, vol. 3, Book Company, Sydney.
Chiappori, PA 1988, 'Rational household labour supply,' Econometrica, 56,1, pp. 63-89.
Chiappori, PA 1992. 'Collective labour supply and welfare,' Journal of Political Economy, 100, pp. 437-67.
Coleman, M & Ganong, L 1989, 'Financial management in stepfamilies,' Lifestyles: Family and Economic Issues, 10, pp. 217-32.
Coleman, M, Ganong, L & Fine, M 2000, 'Reinvestigating marriage: Another decade of progress,' Journal of Marriage and the Family, 62, pp. 1288-307.
Dasgupta, I 2001, 'Gender-biased redistribution and intra-household distribution,' European Economic Review, 45 (9), pp. 1711-22.
Edwards, M 1981, 'Financial arrangements within families,' Social Security Journal, December, Department of Social Security, Canberra.
Edwards, M 1982, 'Financial arrangements made by husbands and wives: Findings of a survey,' Australia and New Zealand Journal of Sociology, 18, 3.
Elizabeth, V 2000, 'Cohabitation, marriage and the unruly consequences of difference,' Gender and Society, 14, 1.
Elizabeth, V 2001, 'Managing money, managing coupledom: A critical examination of cohabitant's money management practices,' The Sociological Review. vol. 49, no. 3.
Ewen, E 1985, 'Immigrant women in the land of dollars: Life and culture in the lower east side, 1890-1925,' Monthly Review Press, New York.
Findlay, J & Wright, R 1996, 'Gender poverty and the intra-household distribution of resources,' Review of Income and Wealth, 42, 3.
Fishman, B 1983, 'The economic behaviour of stepfamilies,' Family Relations, 32, pp. 359-66.
Fleming, R 1999, 'Families of a different kind: Patterns of kinship, support and obligation in remarriage families and their implications for public policy,' Social Policy Journal of New Zealand, Issue 12.
Foreman, H & Wilson, K 1995, 'Financial arrangements within families: Recent findings,' Social Security Journal, pp. 24-53.
Funer, K, Harrison, M & Weston, R 1993, Settling Down: Pathways of parents after divorce, Monograph no. 13, Australian Institute of Family Studies, Melbourne.
Glezer, H 1995, 'Financial control and financial management within couples in Australia,' Paper prepared for Department of Social Security, Australian Institute of Family Studies, Melbourne.
Glezer, H & Mills, E 1991, 'Controlling the purse strings,' Family Matters, Issue no. 29, Australian Institute of Family Studies, Melbourne.
Goetz, A & Sengupta, R 1996, 'Who takes the credit? Gender, power, and control over loan use in rural credit programs in Bangladesh,' World Development 24 (1), pp.45-63.
Goode, J, Callander, C & Lister, R 1998, Purse or Wallet? Gender inequalities and income distribution within families on benefit, Policy Studies Institute, London.
Haddad, L 1999, 'The income earned by women: Impacts on welfare outcomes,' Agricultural Economics, 20.
Hunt, A 1968, A survey of women's employment, vol. II, HMSO, London.
Hunt, P 1978, 'Cash transactions and household tasks,' The Sociological Review, 26, 1, pp. 555-71.
Hunt, P 1980, Gender and Class Consciousness, Macmillan, London.
Jephcott, P, Seear, N & Smith, J 1962, Married women working, Allen and Unwin, London.
Joshi, H & Davies, H 1996, 'Financial dependency: Have women born in 1958 broken free?', Policy Studies, vol. 17, no. 1.
Kanbur, R & Haddad, L 1994, 'Are better-off households more unequal or less unequal: A bargaining theoretic approach to "Kuznets effects" at the micro level,' Oxford Economic Papers, 46, 3, pp. 445-58.
Land, H 1969, Large families in London, G Bell and Sons, London.
Lown, JM & Dolan, EM 1988, 'Financial challenges in remarriage,' Lifestyles: Family and Economic Issues, 9,1, pp. 73-88.
Lundberg, S & Pollak, R 1993, 'Separate spheres bargaining and the marriage market,' Journal of Political Economy, 101(6), pp. 988-1010.
Lundberg, S & Pollak, R 1994, 'Noncooperative bargaining models of marriage,' American Economic Review, Papers and Proceedings, 84(2), pp. 132-37.
Lundberg, S, Pollak, R & Wales, T 1997, 'Do husbands and wives pool their incomes? Evidence from the United Kingdom Child benefit,' Journal of Human Resources, vol. 32, no. 3.
Manser, M & Brown, M 1980, 'Marriage and household decision making: A bargaining analysis' International Economic Review 21, 31-44.
McRae, S 1987, 'The allocation of money in cross-class families,' The Sociological Review, 35, pp. 97-122.
Middleton, S, Ashworth, K & Braithwaite, I 1997, Small fortunes: Spending on children, childhood poverty and parental sacrifice, Joseph Rowntree Foundation, York.
Molloy, D & Snape, D 1999, Financial Arrangements of Couples on Benefit: A Review of the Literature, Inhouse Report 58, Department of Work and Pensions, London, United Kingdom.
Nyman, C 1999, 'Gender equality in "the most equal country in the world"? Money and marriage in Sweden', The Sociological Review, pp. 766-93.
Pahl, J 1980,'Patterns of money management within marriage,' Journal of Social Policy, 9, 3, pp. 313-35.
Pahl, J 1983, 'The allocation of money and the structuring of inequality within marriage,' The Sociological Review, 31, 2, pp. 237-62.
Pahl, J 1989, Money and Marriage, Macmillan, London.
Pahl, J 1995, 'His money, her money: Recent research in financial organisation in marriage,' Journal of Economic Psychology, 16.
Pahl, J 2000, 'The gendering of spending within households,' Radical Statistics, Autumn, pp. 38-48.
Pahl, J 2001, 'Couples and their money: Theory and practice in personal finances,' in R Sykes, C Bochel, & N Ellison (eds), Social Policy Review: Development and debates, 13, Social Policy Association.
Phipps, S & Burton, P 1998, 'What's mine is yours? The influence of male and female incomes on patterns of household expenditure,' Economica, 65, pp. 599-613.
Pitt, M & Khandker, SR 1998, 'The impact of group-based credit programs on poor households in Bangladesh: Does the gender of the participant matter?', Journal of Political Economy, 106pp. 958-996.
Pollak, R 1994, 'For better or for worse: The role of power in models of distribution within marriage,' American Economic Review. vol. 84, issue 2., pp. 148-52.
Poster, M 1978, Critical Theory of the family, Pluto Press, London.
Ritchie, J, 1990, 'Household allocative systems', unpublished Working Paper.
Rowntree, G 1954, 'The finances of founding a family', Scottish Journal of Political Economy, 1.
Saflios-Rothschild, C 1976, 'A macro and micro-examination of family power and love,' Journal of Marriage and the Family, 37, pp. 355-62.
Seltzer, J 1994, 'Consequences of marital breakdown for children,' Annual Review of Sociology, Annual, vol.20; pp. 235-67.
Singh, S 1997, Marriage money: The social shaping of money in marriage and banking, Allen & Unwin, Sydney.
Singh, S & Lindsay, J 1996, 'Money in heterosexual relationships,' Australian and New Zealand Journal of Sociology, 32, pp.57-69.
Snape, D, Molloy, D & Kumar, M 1999, 'Relying on the State, relying on each other,' Research Report no. 103, Department of Social Security, Leeds, United Kingdom.
Stamp, P 1985, 'Balance of sociological power in marriage,' The Sociological Review, 33, 3, pp. 546-66.
Statistics New Zealand 1998, Demographic Trends, Statistics New Zealand, Wellington.
Thompson, F 1954, Lark rise to Candleford, Oxford University Press, London.
Thomson, E, Hanson, T & McLanahan, S 1994, 'Family structure and child well-being: Economic resources vs. parental behaviours,' Social Forces, vol. 73, no.1.
Thomson, E, Hanson, T & McLanahan, S 2001, 'Remarriage, cohabitation and changes in mothering behaviour,' Journal of Marriage and the Family, 63.
Thorogood, N 1987, 'Race, class and gender: The politics of housework,' in J Brannen & G Wilson (eds), Give and take in families, Allen and Unwin, London.
Tichenor, V 1999, 'Status and income as gendered resources: the case of marital power,' Journal of Marriage and the Family, 61, pp.638-50.
Treas, J 1991, 'The common pot or separate purses? A transaction costs interpretation,' in The Triple Intersection: Household, Gender and the Economy, Sage, Newbury Park, CA.
Treas, J 1993, 'Money in the bank: Transaction costs and the economic interpretation of marriage', American Sociological Review, 58, 5, pp.723-34.
Volger, C 1994, 'Money in the household,' in M Anderson, F Bechhofer, & J Gershuny, The Social and Political Economy of the Household, Oxford University Press, Oxford.
Volger, C 1998, 'Money in the household: Some underlying issues of power,' The Sociological Review pp. 687-712.
Volger, C & Pahl, J 1993, 'Social and economic change and the organization of money in marriage,' Work, Employment and Society, 7, pp. 71-95.
Volger, C & Pahl, J 1994, 'Money, power and inequality within marriage,' The Sociological Review, vol. 42, no. 2, pp. 263-89.
Wilson, G 1987, Money in the family: Financial organization and women's responsibilities, Gower, Brookfield, Vermont.
Woolley, F 1996, 'Tax-benefit and the intra-household allocation of resources,' Paper prepared for The Family: Legal, Economic and Policy Perspective, Vancouver, June 7-8.
Woolley, F 2000, Control over money in marriage. Department of Economics, Carleton University, Ottawa.
Zelizer, V A 1994, The Social Meaning of Money. Harper Collins, New York.