Efficiency improvements via monitoring in medical group practice.

摘要

This dissertation tests the hypothesis that medical partnerships can effectively respond to moral hazard by monitoring their members. While it has long been believed that partnerships are inefficient due to inability to control the opportunism of members, widespread occurrence of this organizational form belies that belief. This project examines the methods used by firms to address potential shirking and the relative effectiveness of these methods.; A partnership model is developed which describes the decision-making process in medical groups. The model was inspired by Lee (1990) and Kandel and Lazear (1992), with the structure adapted from that in Gaynor and Gertler (1995). To evaluate the impact of group structure on choice of compensation scheme and surveillance system as well as the impact of these choices on technical efficiency, empirical analysis is performed using data collected by the Medical Group Management Association.; Physician responses to groups' choices of incentives and monitoring methods are examined by estimating individual physicians' demand equations using Random Effects Instrumental Variables estimation. Group choice variables are instrumented to correct for endogeneity and random effects is used to account for sorting by physicians on the basis of group culture. The methods used are similar to those of Gaynor and Pauly (1990) and Gaynor and Gertler (1995).; Results indicate that multispecialty groups and physician-owned groups are more likely to engage in monitoring and to base compensation on productivity. Physician owners use more support staff and charge higher prices on average while multispecialty groups tend to be larger than single-specialty groups and to hire more non-physician labor, but to charge lower prices.; Results support the hypothesized non-jointness of physician production. Productivity-based compensation encourages quantity of output but not effort. The use of peer monitors has a positive impact on physician output and monitoring productivity improves physician effort but not necessarily the quantity of output. It is found that physicians sort themselves into medical groups based on group choices over compensation scheme, size, support staff, price, and surveillance system.; Results further indicate that efficiency improvements result mainly from increased effort, which shifts the production function outward. Firms who monitor more intensely operate more efficiently. Monitoring of productivity and resource use leads to efficiency improvements and firms that employ peer monitors are likely to achieve higher levels of technical efficiency.