M2 and M4 are two important observation points in the risk management of consumer finance. The reason is that the customer may have overdue accounts due to temporary busyness or negligence, but if it is still below M2 after being collected by M 1, it can almost be confirmed as inability to pay or intentional breach of contract. In addition, according to experience, once a customer falls into M4, the probability of transferring bad debts afterwards is very high.
In addition to general risk analysis, before establishing a credit scoring model, it is necessary to determine the definition of bad debt rate. The so-called bad debts can be interpreted as "non-target customer groups", which cover a wider range than overdue accounts. The general definition of bad may include all kinds of debt agreement households and high-risk control households except overdue customers.
The percentage of delinquencies is purely based on the number of days overdue, while the percentage of bad debts escapes this limit, which in many cases reflects the risk nature of customers better than the percentage of delinquencies. Therefore, after the dual-card storm, the application of bad% has become more and more common.
Bad debt transfer rate (WO%) is short for write-off rate. Assuming that the bad debt transfer time of a product is from M6 to M7, the calculation method of WOLAGED% is the amount of bad debts transferred in the current month divided by the receivables seven months ago.
Usually, WO% will further process the annualized accounts, because: First, the monthly data is easily affected by various factors (such as a single large single turn into bad debts), and the annualized accounts have an average effect, which can slow down the fluctuations caused by the interference of special factors in a single month. Second, after annualization, the monthly turnover rate is converted into the annual loss rate, which is convenient for comparison with the annual interest rate of products.
Net loss rate (NCL) is the abbreviation of net credit loss, which is defined as the amount of bad debts transferred in the current period minus the amount of bad debts recovered in the current period. This is the concept of net loss. From the perspective of overall risk management performance, the recovery of bad debts is also an important part, so NCL% is often tied with WO%.
The calculation method of NCL% is the same as that of WO%, except that the numerator part is changed from write-off amount to NCL, and the observation of NCL% is mainly in the form of annualized NCL lag%. Although the calculation is complicated, it has reference value.
The first payment is overdue and the first repayment is overdue. After the user's credit is approved, the proportion of customers who need to repay the first bill within 7 days after the final repayment date and have not applied for extension is FPD 7. The numerator is the number of users who placed orders within the observation period and overdue for more than 7 days, and the denominator is the number of users who placed orders in the first place in the current period and met the repayment date of 7 days. The commonly used FPD index is FPD 30.
For example, suppose the user passed the credit line at 10. 1 and generated the first installment loan at 10.5, and set the 8th of each month as the repayment date. Then 1 1.08 is the repayment date of the first bill. After the bill is issued, the repayment before the repayment date is not overdue. If 1 1. 16 is not returned, it is included in the molecules of FPD7, and the period is10.1-10.30. Usually, users who are overdue for a few days may forget to repay or lack of money for a while, but FPD 7 index can be used by users to evaluate the credit risk of the crediter and predict the health status of future assets.
Similar to FPD 7, FPD 30 is also an indicator to observe the overdue situation of the user's first bill to be repaid. For users who are overdue within 30 days, some losses can be recovered by increasing collection efforts. For users who are overdue for more than 30 days, the probability of collection is greatly reduced, and outsourcing collection may be carried out. If the user's FPD 7 is very high for a period of time, and most of the money that is less called back falls on FPD 30, it proves that this part of users have a high proportion of non-starters and have never thought about repaying loans at all. On the contrary, it shows that the credit risk of the user group is more serious.
In the FPD section, refer to Zhihu's column-satire.