The main objective of this study is to develop a credit risk management instrument of the Bank for Agriculture and Agricultural Cooperatives, an important organization of the Thai rural financial market. The paper highlights the application of the credit scoring system in managing risk in agricultural lending activities. The logit model is first developed to identify the probability of default from the economical and geographical risk factors. The results verify the payment history, collateral types, loan to value ratio, deficit irrigation, saving, landsuitability, natural disasters (flood and drought) and income to expense ratio in the household are important factors in determining of the probability of default in the farmerslending. The model is tested for reliability and validity of the prediction power in discriminating the debtors. Results from the logit model are subsequently employed to formulate the credit scoring system and internal credit risk rating system with reference to the Basel II criteria. The results show how agricultural exposures can be managed on a portfolio basis which will enable the bank to diversify the risk in each of portfolio share, determine the interest rate on the basis of risk, and analyze for the minimum capital requirements and optimal returns in agricultural loan portfolio.