Unitary versus Collective Models of the Household: Time to Shift the Burden of Proof?

Until recently, most economists viewed the household as a collection of individuals who behave as if in agreement on how best to combine time and goods (purchased or produced at home) to produce commodities that maximize some common welfare index. This model has been extended far beyond standard demand analysis to include the determinants of health, fertility, education, child fostering, migration, labor supply, home production, land tenure, and crop adoption. The appeal of the unitary model is the simplicity of comparative statics generated and the diversity of issues it can address. But, argue the authors, its theoretical foundations are weak and restrictive; its underlying assumptions are of questionable validity; it has not stood up well to empirical testing; and it ignores or obscures important policy issues. They argue that economists should regard households as collective rather than unitary entities. They make a case for accepting the collective model (with cooperative and noncooperative versions) as the industry standard - with caveats. The unitary model should be regarded as a special subset of the collective approach, suitable under certain conditions. The burden of proof should shift to those who claim the unitary model as the rule and collective models as the exception. Implicit in the authors'argument is the view that household economics has not taken Becker seriously enough."A household is truly a'small factory,'"wrote Becker (1965)."It combines capital goods, raw materials, and labor to clean, feed, procreate, and otherwise produce useful commodities."The authors, too, perceive the household as a factory, which, like all factories, contains individuals who - motivated at times by altruism, at times by self-interest, and often by both - cajole, cooperate, threaten, help, argue, support, and, indeed, occasionally walk out on each other. Labor economists and industrial organization theorists have long exploited the value of going ins(This abstract was borrowed from another version of this item.)