"Workfare" schemes that offer poor participants unskilled jobs at low wages have become a popular alternative to cash or in-kind handouts. Yet little is known about a key determinant of the cost effectiveness of such schemes in reducing poverty: the behavioral responses through time allocation of participants and their families. These responses affect the foregone incomes of participants and, hence, the net transfer benefits. This paper estimates how time allocation within sampled households responded to new rural employment opportunities under the "Employment Guarantee Scheme" (EGS) of the State of Maharashtra, India. Data were obtained from households surveyed over 6 years in two villages, Shirapur and Kanzara, that differed in wealth, land ownership patterns, agricultural factors, and occupational structure. Participation in the projects was affected by gender, village, and household wealth and caste. Literacy and education had little effect on EGS employment except in Kanzara, where head of household's education was significant and positive. In both villages, EGS employment was generally exogenous to time allocation, suggesting that the ideal of providing work on demand was not being met. Significant gender cross-effects in time allocation were observed, as well as displacement of different activities for different genders: unemployment for men, leisure and domestic activities for women. Overall, the projects generated sizable net income gains to participants, certainly far greater than implied by using market wages to value foregone income. Transfer benefits alone led to a reduction in poverty, of almost the same magnitude as a uniform and undistorting allocation of the same gross budget. Contains 34 references, 13 data tables, and statistical equations. (SV)