| The peer monitoring problem is an important issue in the study of group lending. While asymmetric information structure induce problems of "adverse selection" and "moral hazard", group lending can overcome these problems. With joint liability, group members spontaneity form a structure of peer monitoring. This can help solving the problem of "adverse selection","moral hazard" and credit rationing.Our paper first set up a long-term dynamic model to discuss the optimum path of borrowing and peer monitoring of each borrower. Then we proved that the optimum peer monitoring intensity may decrease with the growth of the number of group members, which indicated that the number of group members created an adverse incentive for peer monitoring. In some extreme cases when the number of group members is large enough, no peer monitoring might be an optimum choice for each borrower.In the latter part of this paper we discussed how to solve the adverse incentive problem that the number of borrowers creates for the peer monitoring. In some situation when the cost of vertical monitoring for extent of peer monitoring is little, we can solve it by inducing vertical monitoring. While the "rotative 1 on 1" peer monitoring structure can eliminate the "moral hazard" problem in members' monitoring, it may create some side effects. Later we discussed the "rotative 1 on N" peer monitoring structure. It can alleviate the "moral hazard" problem with fewer side effects., which raises the stability of group lending. Comparatively speaking it is a better structure of monitoring.In the last part of this article we gave several political suggestions for alleviating the "moral hazard" problem of peer monitoring in group lending. We suggested that lenders should persist in "Long term dynamic lending mode", avert "moral hazard" in group lending, and keep flexible monitoring structure aiming at solving this problem. |