Testing between Competing Models of Sharecropping.

The "Marshallian" approach assumes a prohibitively hight cost of monitor ing the sharecropper's activities while the "monitoring" approach a rgues that landlords stipulate and effectively monitor sharecroppers' activities. The author presents new evidence using detailed data col lected from eight Indian villages. Most tenants own some land of thei r own; this provides a controlled environment in studying the impact of contractual arrangements. The differences in input and output inte nsities on owned minus sharecropped land of the same household are fo und to be sizable and significant, suggesting a rejection of the moni toring approach and supporting the notion of the "Marshallian produc tive inefficiency" of sharecropping. Copyright 1987 by University of Chicago Press.