‘ARRANGED’ MARRIAGE, EDUCATION, AND DOWRY: A CONTRACT-THEORETIC PERSPECTIVE

Abstract

(ProQuest: ... denotes formulae omitted.)1.INTRODUCTION: BACKGROUNDDowry is characterised by asymmetry, uncertainty and unpredictability (Srinivas, 1989).This paper provides a framework to explore interconnections among marriage, education, and dowry essentially in an incomplete information setting. The hypothesis is that when the quality of the groom is not directly observable, then education level of the groom as well as dowry transacted at the time of marriage, help to screen grooms of different qualities. In equilibrium, different types of grooms select different education-dowry contracts, with higher/better types being associated with both higher education and higher dowry. Hence by imparting to education an Arrow-Spence-Stiglitz (Arrow, 1973; Spence, 1973; Stiglitz, 1975) kind of screening role, we are able to explain the perverse (positive) relation between education and dowry (which the human capital approach to education, for example, could not, see discussion below). This also has serious policy implications since greater public expenditure on education may then actually exacerbate dowry inflation instead of mitigating the problem, as expected according to the human capital approach. Before we explain the mechanism of the model, we first briefly describe the concepts of 'arranged' marriage, education and dowry, especially in the context of our model.There are broadly two main types of marriage arrangements - marriages that dominate Western nations such as the United States and those in Europe and marriages which are dominant in many parts of Asia and Africa (see Penn, 2011 for example). In the former, partners are chosen by each other and relationships are sustained with notions of romantic love culminating in marriage. An arranged marriage on the other hand, is one where parents (rather than prospective spouses themselves) choose marital partners. Arranged marriages so defined remain typical for large parts of the world like India, Pakistan, and Bangladesh (see Myers et al., 2005).By the very nature of the procedure, information transmitted between the marital parties about each other is necessarily limited in an arranged marriage setting, relative to love matches where the potential spouses get to mix freely with each other.1 Narayan (1993) aptly speaks of arranged marriages as follows:2More and more, parents entrust the lives and safety of their daughters to families they know very little about; they can gain no information or assurance from common acquaintances, and they will not be close by to look out for their daughters ' welfare. Parents experience the marriage of their daughter as losing her to another family....In the Western world, usually men and women can mix freely and date each other before selecting their partners (for instance, see Penn, 2011 for an account in Western Europe). In fact, they can even live together before marrying their partners. Hence information regarding each other is likely to be much more accurate than in many societies where such free mixing is eschewed.3 In contrast to such liberal practices as in the West, stringent social norms in many other parts of the world, allow only 'arranged' marriages where families of the potential bride and the groom meet each other and the marriage.4 Naturally the amount of information that can be gathered about the potential bride/groom is substantially small compared to when couples can meet each other.5Moreover, social pressure is such that marriages are hard to break (see Pothen, 1989 for example). Hence once married the couple is expected to stay together irrespective of how the match turns out to be.6 Hence a spouse, once chosen, remains a life-long partner.7 With virilocality, this means the bride/bridal family makes a decision with quite limited information that is going to affect her for the rest of her life (it affects the groom as well but since the bride goes to live with her in-laws she is likely to be more concerned with the decision). …