Research On Default Behavior Of Loan Farmers From The Perspective Of Heterogeneity

With the rapid development of new agricultural entity,farmers' demand for financing increases,and the problem of “difficult loans” for farmers has become increasingly serious.Currently,many of the financial institutions regard farmers as a homogeneous whole and believe that they have no collateral,which leads to extremely high default risks.Therefore,credit crunch is particularly obvious.In order to solve the difficulties of obtaining loans for farmers,this paper starts with the study of farmers' default behavior by looking into relevant factors that may significantly affect it.These factors are then transformed into quantifiable indicators to categorize farmers into different types and further explore the relationship between farmers' characteristics and corresponding default risk.The study can provide a decision model that can quickly determine which category the farmers belong to for financial institutions,so as to solve the problem of ‘difficult loans' for farmers.The main research contents of the thesis are as follows:Firstly,combining previous studies and the mechanism of loan default,this paper proposes five characteristics indicators that affect farmers' default behavior: general in individual characteristics,family characteristics,assets and liabilities characteristics,loan characteristics and personality characteristics,under which attached to 21 secondary indicators.Factor analysis is later used to merge secondary indicators with strong correlation and reduce the indicator dimension,obtaining 9 heterogeneous factors.In order to ensure that the selected heterogeneous factors are significantly correlated with farmer's default behavior,Logistic model is applied for empirical analysis.Seven heterogeneous factors that are significantly correlated with farmers' default behavior are obtained out of the analysis,excluding solvency factor,which is the most important factor for financial institutions to consider while does not influence significantly on farmers' default behavior.According to the result,financial institutions should not refuse to lend loans due to farmers' unsecured guarantee.Finally,farmers are grouped into four categories by generating cluster analysis through using the 7 heterogeneous factors as the classification indicators.They are high-quality farmers,good-type farmers,general-type farmers and inferior farmers.Among them,high-quality farmers have low risks,and strong stability These farmers are the best loan targets for various financial institutions;As for good-type farmers,they perform well in basic personal qualities,except their high financial risk.As a result,financial institutions can increase loan interest rate measuring with the burden of family and raise the threshold of loans;in contrast,the financial risks of general farmers are higher and the stability and marital status are not good.To provide loans for these types of farmers account managers and the risk control department need to carry out in-depth mining of abnormal indicators so as to better identify customers;the indicators of inferior farmers are significantly abnormal and such farmers have extremely high financial risks and thus high default rates.Financial institutions need to quickly identify such farmers and avoid them.The study also verifies the reliability of the classification model through sample testing.From the above results,it can be seen that different types of farmers have their own unique combination of significant indicators,hence the key to reducing loan defaults depends on quickly identifying the types of farmers who apply for loans and designing corresponding financial policies.