Replacing the unitary family model, the household bargaining model has attracted more and more attention from economists recently. This model describes a family[1] as a collection of individuals who have distinct self-interests and negotiate to collaborate[2] with others to achieve better outcomes than living alone. Proponents of this theory, Amartya Sen, Shelly Lundberg et al., and Bina Agarwal have identified factors that contribute to one’s bargaining power[3], and, more specifically, how the interactions between social norms and bargaining power explain gender inequality within households. In this essay, after giving a brief introduction to the household bargaining model and current study about bargaining power, I will argue that Agrawal’s theory that “social norms as limits of bargain” is ambiguous and self-conflicting. I will propose to revise it as “social norms as bargaining force” by comparing it with Lundberg’s separate sphere theory. In the end, I will use an interesting case study of housework share in women-breadwinning families in the US and Australia, to show the advantage of my proposal.
I. Household Bargaining Model
The household bargaining model is one of the efforts that economists take to grape the complexity that was not addressed in Gary Becker’s unitary model for the family. (Bennett, 2013) Beaker’s model assumes that all resources or income a household has are pooled, and then allocated by an altruistic head to maximize the sum of family member’s utility. More specifically, bonded by love and commitment, family is considered a “glued-together” unit whose interests are as one. (Sen, 1997) According to this model, family members specialize in different fields according to their comparative advantage to maximize overall efficiency.
However, many detailed field studies pose challenges to this unitary model. Different family members have different interests and tastes in consumption, also. The fact that about half of marriages end with divorce in Western countries[4] also makes doubt if family members can be truly altruistic. Even if money and other resources are allocated equally, some scholars argue that “it is problematic to assume that it does not matter in a well-being assessment whether a person has earned this money herself, or obtained it from her partner,” because money comes from a spouse may come with expectations; in our capitalist society, one’s prestige is linked to one’s income. (Robeyns, 2003 & Bennett, 2013) To explain the unfair dynamic between the breadwinner and the dependent, economists raise several different models, including the exchange model and household bargaining model.
The social exchange model acknowledges each family member has their distinct self-interests. It sees family as a small market, where (traditionally, and in an oversimplified model,) women do housework in exchange for men’s paycheck, and vice versa. Person A is called more dependent on person B, or B is called to have power, if A has limited access to resources except through exchange with B. In this case, B will give A less in the exchange, because, even if the exchange rate is unfavorable for A, A cannot access the necessary resource without trade with B. Though this theory is appealing, it does not get much attention in recent research about intra-household dynamics, probably because power and dependency are hard to measure and analyze[5].
Many scholars, and this article, pay more attention to the household bargaining model. Acknowledging that different family members have their own interests also, the household bargaining model argues that members form and maintain a family because, in general, cooperation and specialization are more beneficial than living alone. It logically follows that divorce happens when at least one of the family members does not see the collaboration as beneficial or potentially beneficial. Yet, the specific detail of cooperation is shaped by the contest between family members. The difference between family members’ power to bargain will determine who, and to what extent, the decision will favor. If we suppose people are reluctant to do housework, for example, then the one who has more bargaining power will be responsible for less housework.
The household bargaining model is successful, in terms of its capacity to explain the detailed dynamic within the household. Government and gender-progressive NGOs could indirectly promote intra-household gender equality and optimize resource allocation, which was considered a black box or private realm freed from external interference by the unitary model, by enhancing women’s bargaining power. However, to achieve this goal, we must study how one’s bargaining power could be changed by various personal and social factors.
II. Factors Contributing to Bargaining Power
In this part, I will explore factors contributing to one’s bargaining power, with special attention to the gendered perspective. I will start with the standard game theory model, then provide a brief overview of Amartya Sen’s perception model, Shelly Lundberg et al.’s separate sphere theory, and Bina Agarwal’s four ways social norms and bargaining power interact.
Economists employ game theory to begin the discussion. The formation and maintenance of a family is attributed to the fact that cooperation between family members is more beneficial than the best alternative, or the threat point, such as divorce. If one has an attractive threat point, her spouse must offer a more attractive contract to beg her[6] to stay in the collaboration. Therefore, given that one’s income, or one’s potential income, largely determines one’s quality of living alone, it is a significant factor that influences bargaining power within households. Other related factors include how the person is welcomed in the remarriage market, support from her parents and friends, and so on.[7] (Sen, 1988)
However, to fully explain the gendered patterns[8], we must take cultural factors into consideration. Sen, pioneering this project, added perceived contributions and interests into the model. (Sen, 1988) He noticed that, in different cultures, some contributions to the family are (unfairly) considered more or less valuable, which, thereby, gives the contributor more or less bargaining power within the household. More specifically, women’s unpaid housework is usually considered less valuable because it is monetary invisibility. He also argues that some subordinate people, such as women in India, may not have enough perception of their own interests, and, consequentially, lack intention to bargain for their well-being. However, this view is challenged by Agarwal, who, by more detailed field research, discovered that Indian women resist oppression and pursue their own interests in more secret ways, and Kandiyoti, who argues that Sen denies their agency. (Agarwal, 1997& Kandiyoti, 2005)
Later, Shelly Lundberg and Robert Pollak treat social norms as a reference and threat point for bargaining. They point out that divorce is not a realistic alternative to trivial daily bargaining. They claim that, if collaboration is more beneficial, failures of negotiation are caused by “transaction cost.” (e.g., cost to enforce one to keep promise.) A Failure of negotiation, then, will lead to a Nash equilibrium with minimal transaction cost, which means a division of labor according to paradigmatic gender roles, or “separate sphere” in authors’ terminology. In this situation, the cost of enforcement is taken by the society. For example, though taking care of children is usually the wife’s responsibility, and fixing plumbing is the husband’s, if the wife has a piano lesson on Saturday, this couple could, after some kind of bargaining, have a beneficial agreement that the husband will take care of children on Saturday, while the wife must be responsible for fixing the plumbing. However, if one fails to perform the extra responsibility, the other will refuse to do the extra also. Then, the agreement will be demolished, and they will come back to the paradigmatic gender division[9]. In summary, in this separate sphere theory, the social norm serves as (1) the reference point of bargaining, or where the bargaining starts from, and (2) the threat point of bargaining, or where the bargaining ends up if negotiation fails. (Lundberg, 1993)
Recognizing the former discussion on this subject, in her famous work “Bargaining” And Gender Relations: Within And Beyond The Household, Bina Agarwal gives a more detailed study of the interactions between social norms and bargaining power, which conceptualizes them into four categories. (Agarwal, 1994)
(1) Social norms may impact one’s bargaining power indirectly by influencing one’s game theory factors. Women’s income, for example, might be damaged by the gendered division of labor in the labor market, biased hiring decisions, and the social expectation of doing unpaid housework. Divorced women might also be unwelcome in some cultures[10], which increases the cost of threatening divorce. Moreover, as Iversen argued, in many societies, women’s assets tend to be more household-specific and cannot be taken to her next family, such as a well-raised child[11] or a tidy house, but men’s assets, such as land or his own labor force, can be transferred to a new family if he divorce. This cultural expectation would also damage women’s threat point, and, thereby, her bargaining power. (Iversen, 2011)
(2) Social norms regulate the method of bargaining. It decides the method of communication. For example, Women are expected to be quiet and less assertive in certain cultures, and women who disobey the expectation will get punished. In these cultures, when a bargain happens verbally, women may get a poorer result. It decides the underlying principle of the bargain[12]. For example, if we should allocate resources by a need-based, merit-based, or egalitarian method[13].
(3) And, most innovatively, Agarwal claims that social norms are result of bargaining. Agarwal does not consider social norms as a given fact, a pre-determined factor in the analysis; rather, she realizes that norms are result of bargaining. Several examples show that women who work outside are capable of challenging traditional norms because of their economic independence. In a broader scope, Iversen argues that during the transformation from hunter society to agricultural society, then to industrial and post-industrial society, the change of women’s and men’s relative bargaining power determined the norm in these eras. Because of economic motivation, each individual family makes decisions to disobey the old norms. When enough families make the same decisions for enough time, after the next generation grows up, new norms are formed. (Iversen, 2011)
(4) Social limits to what can be bargained about. She seems to argue that, social norms determine that certain practices, probably justified in the name of tradition, are “not open to questioning or contestation,” such as the “category of persons cannot intermarry (e.g. most patrilineal north Indian Hindus forbid marriages to close kin or within the village); the gender division of labor within the home (e.g. housework and child care are usually seen as women’s responsibilities)… and so on.” However, her argument needs more clarification and elaboration, because it seems that she is conflicting with herself, or, at least, was ambiguous about an important distinction.
Agarwal first says, “Norms set the limits to bargaining. They can define which issues can legitimately be bargained over.” Then, she cited Pireer Burdieu’s term “doxa” to explain the role of norms here. She says that “some decision would fall in the realm of… the term ‘doxa’ – which… is not open to questioning or contestation.” Clearly, she means certain actions or decisions are doxa, and they are not bargainable here. But in the later part, she says doxa is “undiscussed, unnamed, admitted without argument or scrutiny,” and “doxa could include widely accepted norms.” Here, she seems to lean towards the idea that doxa is norms about action, instead of actions or decisions themselves, and norms about action are not bargainable.
To make the distinction clear, suppose there exists a social norm that “you ought X,” where X is an action or a practice, such as a division of labor. Agarwal is indecisive between saying that X is a non-bargainable doxa and saying the norm about X is a non-bargainable doxa.The difference is that, if the action X is a doxa, then everyone will do X without any hesitation; but if the norm that “you ought X” is a doxa, people can refuse to do X with the feeling of guilt, or do X reluctantly, while the guilty is not questioned. In the examples mentioned above, the practice of forbidden marriage seems to be a doxa that everyone follows without questioning, but the division of labor seems to be more variable: though most people still agree that certain division of labor is normal, they may not follow the paradigmatic division of labor, or, at least, they may have to doubt it to some extent.
I accept the second explanation, that certain norms are accepted without questioning. In the following section, I will elaborate on why the second explanation is better by connecting it to other theories mentioned above.
III. Social Norms are Independent Bargaining Force
First, saying X is not bargainable contradicts the fact that many people fail to comply with a variety of social norms[14]. Romantic relationships between cousins, for example, are strictly forbidden by social norms but are still (secretly) practiced by many people. The norm about forbidden marriage does influence bargains, but it does not influence bargains by setting the method of bargaining, influence one’s bargaining power, or changing one’s perceived contribution/interest. It does set a reference point for the bargain, (e.g., if one wants to marry her cousin, she has the burden to ask her parents to allow her, but her parents are burdened to explain the reason for prohibiting it) as the separate sphere theory captures, but the punishment for cousin marriage is much beyond the transaction cost the separate sphere theory predicts. This implies that we need a new theory to explain the effect of this social norm.
Second, some social norms are indeed formulated as “you must X” or “you should never X”; yet some other social norms are formulated in softer ways, such as “you would better X” or “you would rather not X”. Saying that X is not bargainable or norms set limits cannot conceptualize these soft norms. And these soft norms also cannot be conceptualized in other categories. To be more precious, soft norms are usually linked to a “good” woman/husband/citizen. Following soft norms will give some progress in becoming “good.” Though it is hard to measure how much benefit one will get from following a single norm[15], soft norms obviously come with some attraction.
Third, Agarwal argued that norms are a result of the bargain. If the action or the practice is not bargainable, then we would never have a chance to change the norm, because everyone accepts the practice without questioning[16]. However, saying norm is not bargainable, is consistent with the idea that norm is a result of bargaining. As Iversen described, when the method of production changes, neglecting the norm becomes economically beneficial. Then, many families, after bargaining about the action, independently and coincidently decide to disobey the norm with guilt in mind. The complex view of female factory workers in Malaysia exemplifies this. Many Malaysian families have a negative perception of women working in factories, but for those who economically depend on female factory workers, their attitude becomes twisted. (Kinnaird, 2002) Later, after enough time, a new norm is formed. This process and the idea that what is non-bargainable is norms instead of actions, successfully reconcile the superficial conflict.
Recall that Agarwal introduced the idea that certain actions are not bargainable to illustrate that certain norms set limits for bargains. If actions prohibited by norms are bargainable, what are the effects of these norms on the bargaining process? Inspired by the separate sphere theory, I will propose that social norms are bargaining forces. Lundberg et al. acknowledge that social norms set the default value of bargaining. It is to say that people will perform according to social norms unless they have specific reasons, such as having a piano lesson on Saturday. The forbidden marriage can be illustrated in this way easily: you won’t have romantic relationships with your cousin unless you desperately love your cousin. In Malaysia, you won’t agree with your wife to work in a factory, unless you have enough economic motivation. However, different from the separate sphere theory, in these cases, the cost of deviating from the norms is not only the transaction cost, but also a strong discouragement from surrounding people, and even internal morality. The more one deviates from the paradigm, the more resilient force will be applied to turn the one back. Therefore, saying that these social norms are independent, unquestionable forces in the bargaining process seems to be suitable. Social norms are like a spring, with one end hanging on the paradigm, and one end hanging on people. I use the term “bargaining force” instead of “bargaining power” because bargaining power belongs to a certain person, and the owner could choose to use it or not; but the resilient force of social norms does not belong to any person, and the advantaged person cannot decide to neglect this force, because (part of) it is enforced collectively. In the following section, I will provide an interesting case study about female breadwinners in Australia and the US to show how this idea can be applied to real-life cases.
IV. Case Study: female breadwinners in Australia and the US.
Share of housework is one of the paradigmatic cases of intra-household bargaining. Evidence shows that, as expected, when wives’ income increases from zero to about a half of families’ total income, the division of housework becomes more equitable; however, surprisingly, when women’s income exceeds their husbands, the more relative share of income they have, the more share of housework they have. (Cooke, 2006) In specific, Bittman et. al. compare data between the US and Australia and find that, though both countries have the same trend, high-income women in Australia have more housework share than high-income women in the US. (Bittman, 2003) If we assume people are reluctant to do housework, more share in housework suggests a lower bargaining power; on the other hand, an increase in personal income should raise one’s bargaining power. How could we understand this abnormal phenomenon?
First, it will be hard to argue that women with higher incomes will have a worse life after divorce or partially non-cooperating[17]. A higher income indicates more materialistic freedom. Statistics in the US show that, in families where the wife has a higher income, they are more likely to discuss the idea of separating, which indicates a higher acceptance of partially non-cooperative alternatives. (Bertrand, 2013) It is also unlikely that any personal or private factor result in a decrease in bargaining power. No evidence shows that her parents and friends won’t give them less support if a woman has an income higher than her husband. We must search the social factors for the cause of this phenomenon.
Bittman describes the women’s housework as compensation for their derivation from the “normative income standard.” Because relatively-high-income women’s femininity is damaged by their relatively high income and economic decisiveness, they must do more housework to recover their femininity. Bittman subsequently argues that the difference between Australia and the US is explainable by the institutional differences that make breadwinning women more “deviant” in Australia, which confirms social norms should be attributed to this abnormal division of labor. But how could we fit this into our household bargaining model?
Existing theories cannot explain this fact. There is no reason to believe that more income means less contribution to the family, yet it is possible that extra money earned by women is not considered a contribution to the family. There is no evidence showing that high-income women will have less perceived interest or that high-income women have to change their method of bargaining, which, thereby, pushes them into an unfavorable situation.
However, the idea that social norms are an individual bargaining force can explain this phenomenon easily. As claimed, when one’s behavior deviates from the paradigm, the one will feel a force, from both surrounding people and internal morality to push her back, which is stronger when the degree of deviation increases. With this in mind, we could explain the decrease in bargaining power of high-income women by the force applied by the social norms. Bertrand discovered that a wife is less likely to work if she has a higher potential income than her husband, and even if she works, her real income is significantly less than her potential income. (Bertrand, 2013) This can be seen as a direct result of the resilient force of social norms. In the case of housework, we must take a broader consideration. Per our discussion of soft norms, the evaluation of a “good wife” happens in multi-dimension. To prevent serious punishment, one must keep her accumulative degree of deviation low. If she has too much income, which deviates from the paradigm to the direction that is beneficial for the wife, she must have a negative deviation from the paradigm to the direction that benefits the wife, i.e., benefiting the husband.
V. Significance and further study
In this essay, I start with a brief overview of the household bargaining model and the factors that impact bargaining power. I found Agarwal’s explanation of interactions between social norms and bargaining power not satisfactory and proposed an alternative that could better explain the abnormal division of housework in women-breadwinning families. It could help us better understand the dynamic within family, and, thereby, allow government to make better policies. However, it should be noticed that this paper has very limited quantitative analysis, and, though realizes this phenomenon, does not fully explain how one’s deviation is accumulated among different dimensions.
[1] In this essay, because of the goal of studying gender inequality, a family is assumed to be a heterosexual monogamic family, which contains a wife, a husband, and maybe a child or children, whose agency and interests are ignored. When I use the term bargaining within a family, it specifically means the bargaining between the wife and the husband. I do feel uncomfortable and apologize for excluding homosexual and non-binary families.
The term “family” and the term “household” are used interchangeably.
[2] The term “collaboration” and the term “cooperation” are used interchangeably.
[3] When I use the term “bargaining power,” I mean bargaining power within the household.
[4] In the US, the marriage rate was 16.3 marriages per 1000 women in 2011, while the divorce rate is 9.7 per 1000 women. (Kennedy, 2014)
[5] I am aware that this is not a well-supported claim. Limit literature is available for comparing these two models. It is worth noting that, these models are very similar to each other: they both acknowledge family members’ self-interest and consider the alternative of the current marriage as a significant determinant of dynamics within the household.
[6] In sentences like this one, the person I addressed as “she/her” is not necessarily a woman. In most parts of my essay (except the case study at the end), I intend to employ gender-neutral pronouns like “one” or “a person.” When it is not convenient to do so, she/her will be the default third-person pronoun.
[7] These factors will be called “game theory factors” in the later part of this paper.
[8] I did not specify what the gendered patterns are, because we have different patterns in different cultures and regions. In the US, for example, there exists a pattern that the husband is the breadwinner and the wife takes more housework.
[9] Noticeably, Lundberg et al. also mentioned that, in a family where negotiation has failed when people provide public good according to the social paradigm, it is likely that they won’t put in adequate effort. For example, the wife may still cook for the family but be less careful about nutrition balance. They do not explain this phenomenon in detail, but my proposal will help explain this. I argue that social norms will impose a resilient force on those who deviate from the paradigm; and the more one deviant, the stronger the force is. The wife still cooks after the negotiation fails, because the social norms will punish her if she doesn’t cook at all; but her selfishness may find it beneficial to put a little bit less effort into cooking because she saves time while the social norms’ punishment is neglectable.
[10] When I mention that something happens “in some cultures” or “in certain cultures” in this section, you may wonder about the source of my claim and which culture I am talking about. I hid those details in the main body because this paper puts more attention on the theoretical side of the discussion, but you can get testimony from Agarwal’s study of Indian culture or Iversen’s description of late agricultural culture and early industrial culture.
[11] It is understandable to feel trembling to describe children as assets. To describe in a more detailed way, Iversen argues that women spend more time educating children, feeding children, and building emotional ties with children. This means that children would likely support their mother when they grow up. In patrilineal society, divorced women can no longer get support from children. This is considered a potential loss when women divorce.
[12] Agrawal considers deciding the underlying principle as another role of social norm, setting limits. However, I feel that explaining this as either the method or limitation is acceptable. I put it here mainly because the discussion below about setting limits is intensive and putting this part there might blur the main point of my latter discussion.
[13] In these three types of society, the bargain will be about to what extent one needs the resource, to what extent one contributes to the family, or if the quantity one gets is less than others, respectively.
[14] I even cannot come up with a social norm that is never violated.
[15] Notice that following/not following a single norm will not immediately make one good or bad. The reward of being good or punishment of being bad is given for one’s overall performance.
[16] If the norm changes but the practices never changes, then people live later will suffer from endless guilt because they would never change their action even if norm changes.
[17] It is still arguable, though; for example, Bertrand et. al discovered that women with relatively high income are less welcomed in marriage market, which means high-income women have limited alternative. However, it is too extraordinary to claim that higher income will have negative impact on their life after divorce or failure of cooperation, which means extraordinarily strong evidence is required to justify it. And unfortunately, I cannot find enough evidence to support this claim.
Reference:
Agarwal B. (1997) ”Bargaining” and Gender Relations: Within and Beyond the Household, Feminist Economics, 3:1, 1-51, DOI: 10.1080/135457097338799
Bennett, F. (2013). Researching within‐household distribution: Overview, developments, debates, and methodological challenges. Journal of Marriage and Family, 75(3), 582–597. https://doi.org/10.1111/jomf.12020
Bertrand, M., Kamenica, E., & Pan, J. (2013). Gender Identity and relative income within households. The Quarterly Journal of Economics, 130(2), 571–614. https://doi.org/10.1093/qje/qjv001
Bittman, M., England, P., Sayer, L., Folbre, N., & Matheson, G. (2003). When does gender trump money? bargaining and time in household work. American Journal of Sociology, 109(1), 186–214. https://doi.org/10.1086/378341
Cooke, L. P. (2006). “doing” gender in context: Household bargaining and risk of divorce in Germany and the United States. American Journal of Sociology, 112(2), 442–472. https://doi.org/10.1086/506417
Iversen, T., & Rosenbluth, F. M. (2011). Gender Socialization: How Bargaining Power Shapes Social Norms and Political Attitudes. In Women, work, and politics: The political economy of gender inequality (pp. 17–54), Yale University Press.
Kandiyoti, D. (2005). Rethinking bargaining with patriarchy. Feminist vision of development: Gender, analysis and policy, 135-154.
Kennedy, S., & Ruggles, S. (2014). Breaking up is hard to count: The rise of divorce in the United States, 1980–2010. Demography, 51(2), 587–598. https://doi.org/10.1007/s13524-013-0270-9
Kinnaird, V., Momsen, J., & Buang, A. (2002). DEVELOPMENT AND FACTORY WOMEN: NEGATIVE PERCEPTIONS FROM A MALAYSIAN SOURCE AREA . In Different places, different voices: Gender and development in Africa, Asia and Latin America. essay, Taylor and Francis.
Lundberg, S., & Pollak, R. A. (1993). Separate spheres bargaining and the marriage market. Journal of Political Economy, 101(6), 988–1010. https://doi.org/10.1086/261912
Robeyns, I. (2003). Sen’s capability approach and gender inequality: Selecting relevant capabilities. Feminist Economics, 9(2–3), 61–92. https://doi.org/10.1080/1354570022000078024
Sen, A. (1988). Gender and Cooperative Conflicts. World Institute for Development Economics Research of the United Nations University.
Sen, A. K. (1997). Resources, values and development. Harvard University Press.