Testing for the Existence of Bargaining in Rural Households: a Study of Decisions on Labor Market Participation in the Cordillera Region of the Philippines

In this paper we derive testable implications of a unitary farm household model and a non-unitary, i.e. bargaining, model. In the unitary household model the impact of spouse specific resources and non-labor income on household decisions should not be different from that of the resources and non-labor income common to the household. In a bargaining model we expect to find a specific impact of spouse specific resources and non-labor income. Our empirical tests are based on a small survey of households in the Cordillera region of Northern Luzon (Philippines). In this region each spouse retains specific rights on her/his inherited land, although within marriage this land is treated as part of the household farm. Inherited land is a truly exogenous variable, which we use as the indicator of bargaining power. We perform probit regressions in which the spouses’ inherited land is a determinant of the probability that a husband or wife participates in the labor market. The statistical results provide some evidence of a specific impact of spouse specific land on labor market participation decisions and therefore cast doubt on the unitary farm household model. They are compatible with a bargaining model of household behavior.