This paper attempts to understand the complex nature of how gender relations within the household are affected by extra-household conditions in South Asia. After describing quantitative determinants of household bargaining power, such as land ownership and income, the paper investigates qualitative aspects, including (1) the role of social perceptions and norms, (2) the question of altruism versus self-interest, and (3) the way in which gender relations in the market, community, and state interact with each other and with household dynamics. These multiple conditions can simultaneously reinforce and conflict with each other, and can critically affect the accuracy of theoretical formulations and policy interventions.

The paper builds on the author's work in A Field of One's Own: Gender and Land Rights in South Asia (1994, Cambridge University Press), but also incorporates an extensive literature review and informal interviews and observations from numerous women's coalitions, including the Self Employment Women's Association (SEWA) of India, the Bangladesh Rural Advancement Committee (BRAC), and the Grameen Bank in Bangladesh. It also uses case stories from Hindu and Muslim residents of India, Nepal, Pakistan, and Tibeto-Burmese communities.

Determinants of Intrahousehold Bargaining Power
The paper rests on the premise that household members bargain for their subsistence needs. Although such bargaining is often unspoken, it accounts for the way in which all resources and tasks are allocated among family members. Multiple conditions determine the strength of bargaining power in that they determine fallback position, or one's ability to survive outside the household.

"Better analysis of the complex factors affecting household bargaining power will allow for more accurate models of what households need and how they utilize resources."

A person's power in bargaining for subsistence needs in rural areas depends most on control over arable land (or in an urban context, on property other than land). Closely linked is access to other income or employment. Bargaining power is also enhanced by access to communal resources such as village commons or forests; traditional social support systems such as family, caste, and kinship networks; support from NGOs; and support from state programs. Since gender is a significant basis of inequality and restricts property rights in South Asia, social support systems and support from the state and community, as well as access to the market, are particularly important for women.

These determinants of bargaining power not only interact with each other, but are colored by the following dynamics.

Social Perceptions and Norms
Social perceptions and norms, strongly influenced by gender dynamics, have a significant impact on the fallback position of women, whose contributions to the family are typically undervalued by both family members and policymakers. This decreases expectations about what women deserve, and thus weakens their voice in the home. Most rural Indian women eat last in the family and receive the smallest portions and poorest quality of food. Simi-larly, social norms (standards of acceptable behavior) limit what can be bargained about, justify gender-restrictive laws and practices, and can limit the strength of an individual's voice within a household. Social norms define the gender division of labor in the home, the segregation of labor in urban employment, rights to ancestral land and even who may discuss such matters as inheritance rights.

Social perceptions and norms, however, are subject to contestation and change, particularly through group organizations, which have been effective in challenging such practices as purdah in Islamic societies and restrictions on employment opportunities in Hindu communities. For example, women in the Bangladesh Rural Advancement Committee (BRAC), which provides group credit and employment opportunities, reported significant increases in bargaining power both within the household and the community, as well as better treatment by their husbands. Through collective action, the women were able to stop community elders from criticizing untraditional behaviors such as working outside the home.

Self-Perceptions, Altruism, and Self-Interest
Just as social norms affect bargaining power, so do self-perceptions. The paper challenges the theory that women behave more altruistically than men do; that they undervalue their own worth, needs, and well-being; and that they act as implicit accomplices in an unequal order. Rather, the paper presents evidence of women's everyday resistance by secretly selling parts of crops or raising livestock, hiding rice in other women's homes to conceal it from their husbands, feigning spirit possession in order to receive food otherwise denied them, and similar behaviors. Overt appearances of compliance can simply reflect creative survival strategies.

Policy Implications
An accurate understanding of household dynamics is important for designing policy interventions. Better analysis of the complex factors affecting household bargaining power will allow for more accurate models of what households need and how they utilize resources.

The paper suggests that strengthening women's fallback positions—through improvements in either quantitative or qualitative conditions—is the most effective way to increase their bargaining power. Greater voice in the home means better food, health care, and general welfare for women and children, and can also lead to increased land rights, employment opportunities, and other improvements in women's lives.

The paper calls for the building of gender- progressive coalitions in all areas of bargaining, at the level of the home, the market, community, and state. Although women have clearly developed creative survival strategies, they derive great benefit from gender-progressive programs and coalitions. Women who have recently attained arable land through state or NGO programs, for example, report greater self- esteem, economic security, and improved treatment from their husbands and families.

In addition, the paper calls for further empirical testing of aspects other than income that bear on bargaining power, including intrahousehold coalitions. While some qualitative dimensions such as the value placed on women's contributions and needs cannot be tested empirically, they should nonetheless be studied and incorporated analyses and models as fully as possible.