Delinquency-moving matrices for visualizing loan collections
The present invention, in one aspect, relates to tools for forecasting cash flow and income from a loan portfolio that are particularly useful in volatile markets. In one specific embodiment, consumer payment behavior is modeled, and account movement is simulated. For each month, actual payment amounts can be compared to delinquency, and frequency of payment can be compared to delinquency. Actual performance is then applied to current contractual payments for forecasting. In addition, the models facilitate determination of where payments are coming from, i.e., who is paying.
1. A method for predicting loan collections for a group of non-stationary asset-based loans using a computer system configured with a collections model and a re-marketing model, the group of non-stationary asset-based loans included within a distressed loan portfolio, an account including at least one of the loans, said method comprising the steps of:
categorizing each non-stationary asset-based loan included within the portfolio based on a prior month's payment of the corresponding loan, non-stationary asset-based loans include at least one of automobile loans, vehicle loans, and credit card loans;
categorizing each loan included within the portfolio based on a contractual delinquency of the corresponding loan;
utilizing the computer and the collections model to predict a payment behavior for a borrower of a non-stationary asset-based loan included within a distressed loan portfolio, the collections model is based on historical payment information of the borrower, loan delinquency assumptions, a plurality of collection strategies that may be utilized for collecting payment from the borrower, and the delinquency category assigned to the loan;
initiating at least one of the plurality of collection strategies with respect to the borrower;
analyzing the borrower's payment behavior after initiating the at least one collection strategy including whether the borrower made a payment and, if so, an amount of the payment;
comparing the borrower's payment behavior after initiating the at least one collection strategy to the predicted payment behavior of the borrower and the delinquency category assigned to the corresponding loan;
comparing the borrower's payment behavior after initiating the at least one collection strategy to the prior month's payment category of the corresponding loan;
incorporating management feedback into expectations of future performance wherein management feedback includes recommending a change in collection strategies used for prompting payment from the borrower associated with the loan included within the portfolio and predicting future payment performance of the borrower based on the recommended change in collection strategies:
updating the collections model based on the payment comparisons and the management feedback, the updated collections model predicts future cash inflows for each loan included within the portfolio, the updated collections model is configured to apply a greater weight to the payment performance of each loan for the current month as compared to the payment performance of each loan for prior months;
utilizing the computer and the re-marketing model to calculate an amount generated and expenses incurred from repossessing the non-stationary asset used as collateral for the borrower's loan, the re-marketing model further calculates a probability that an event will occur impacting payment of the borrower's loan;
generating delinquency moving matrices for each loan included within the group of loans including the borrower's loan based on an output from the updated collections model and the re-marketing model, the matrices displaying for each account a percentage indicating a probability that the account will roll forward into a next classification of delinquency, and a number of months that the account is delinquent; and
predicting which accounts will roll forward into a next classification of delinquency based on information displayed in the matrices.
2. A method according to claim 1 wherein said step of generating delinquency moving matrices further comprises the step of:
assigning probability distributions to loan delinquency assumptions; and
inputting the loan delinquency assumptions and the assigned probability distributions into the collections model and the re-marketing model to predict a payment behavior for a borrower of a non-stationary asset-based loan included within the distressed loan portfolio.
3. A method according to claim 2 wherein said step of assigning probability distributions to loan delinquency assumptions further comprises the step of determining a percentage of loans within the probability distributions that will roll forward into a next period of delinquency.
4. A method according to claim 3 further comprising the step of indicating a number of months an account is delinquent.
5. A method according to claim 1 wherein said step of generating delinquency moving matrices further comprises the step of adjusting loan delinquency assumptions to account for variations in a plurality of forces impacting a payment behavior of a borrower including at least one of time of season, changes in political climate, interest rate changes, and a likelihood that a natural disaster may occur.
6. A method according to claim 5 further comprising the step of adjusting probability distributions assigned to the loan delinquency assumptions to account for adjustments made to the loan delinquency assumptions.
7. A method for determining loan collection data for a group of non-stationary asset-based loans using a computer system configured with a collections model and a re-marketing model, the group of non-stationary asset-based loans included within a distressed loan portfolio, an account including at least one of the loans, said method comprising the steps of:
categorizing each non-stationary asset-based loan included within the portfolio based on a prior month's payment of the corresponding loan, non-stationary asset-based loans include at least one of automobile loans, vehicle loans, and credit card loans;
categorizing each loan included within the portfolio based on a contractual delinquency of the corresponding loan;
utilizing the computer and the collections model to predict a payment behavior for a borrower of a non-stationary asset-based loan included within a distressed loan portfolio, the collections model is based on historical payment information of the borrower, loan delinquency assumptions, and a plurality of collection strategies that may be utilized for collecting payment from the borrower, and the delinquency category assigned to the loan;
initiating at least one of the plurality of collection strategies with respect to the borrower;
analyzing the borrower's payment behavior after initiating the at least one collection strategy including whether the borrower made a payment and, if so, an amount of the payment;
comparing the borrower's payment behavior after initiating the at least one collection strategy to the predicted payment behavior of the borrower and the delinquency category assigned to the corresponding loan;
comparing the borrower's payment behavior after initiating the at least one collection strategy to the prior month's payment category of the corresponding loan;
incorporating management feedback into expectations of future performance wherein management feedback includes recommending a change in collection strategies used for prompting payment from the borrower associated with the loan included within the portfolio and predicting future payment performance of the borrower based on the recommended change in collection strategies;
updating the collections model based on the payment comparisons and the management feedback, the updated collections model predicts future cash inflows for each loan included within the portfolio, the updated collections model is configured to apply a greater weight to the payment performance of each loan for the current month as compared to the payment performance of each loan for prior months;
utilizing the computer and the re-marketing model to calculate an amount generated and expenses incurred from repossessing the non-stationary asset used as collateral for the borrower's loan, the re-marketing model further calculates a probability that an event will occur impacting payment of the borrower's loan;
generating matrices for delinquency, gross value, stock value, roll forward, roll back, amounts due and payment for each loan included within the group of loans including the borrower's loan, the matrices including data generated from the updated collections model and the re-marketing model; and
predicting a portfolio value for the distressed loan portfolio using the matrices.
8. A method according to claim 7 wherein said step of predicting a portfolio value further comprises the step of predicting a cash flow value for a portfolio.
9. A system for predicting loan collections for a group of non-stationary asset-based loans, the group of non-stationary asset-based loans included within a distressed loan portfolio, an account including at least one of the loans, said system comprising:
at least one computer;
a server configured with a collections model and a re-marketing model, said server configured to:
categorize each non-stationary asset-based loan included within the portfolio based on a prior month's payment of the corresponding loan, non-stationary asset-based loans include at least one of automobile loans, vehicle loans, and credit card loans;
categorize each loan included within the portfolio based on a contractual delinquency of the corresponding loan;
access the collections model to predict a payment behavior for a borrower of a non-stationary asset-based loan included within a distressed loan portfolio, the collections model is based on historical payment information of the borrower, loan delinquency assumptions, a plurality of collection strategies that may be utilized for collecting payment from the borrower, and the delinquency category assigned to the loan;
analyze the borrower's payment behavior after initiating at least one of the plurality of collection strategies including whether the borrower made a payment and, if so, an amount of the payment;
compare the borrower's payment behavior after initiating the at least one collection strategy to the predicted payment behavior of the borrower and the delinquency category assigned to the corresponding loan;
compare the borrower's payment behavior after initiating the at least one collection strategy to the prior month's payment category of the corresponding loan;
incorporate management feedback into expectations of future performance wherein management feedback includes recommending a change in collection strategies used for prompting payment from the borrower associated with the loan included within the portfolio and predicting future payment performance of the borrower based on the recommended change in collection strategies;
update the collections model based on the payment comparisons and the management feedback, the updated collections model predicts future cash inflows for each loan included within the portfolio, the updated collections model is configured to apply a greater weight to the payment performance of each loan for the current month as compared to the payment performance of each loan for prior months;
access the re-marketing model to calculate an amount generated and expenses incurred from repossessing the non-stationary asset used as collateral for the borrower's loan, the re-marketing model further calculates a probability that an event will occur impacting payment of the borrower's loan;
generate delinquency moving matrices for each loan included within the group of loans including the borrower's loan based on an output from the updated collections model and the re-marketing model, the matrices displaying for each account a percentage indicating a probability that the account will roll forward into a next classification of delinquency, and a number of months that the account is delinquent; and
predict which accounts will roll forward into a next classification of delinquency based on information displayed in the matrices; and
a network connecting said computer to said server.
10. A system according to claim 9 wherein said server further configured to:
assign probability distributions to loan delinquency assumptions; and
input the loan delinquency assumptions and the assigned probability distributions into the collections model and the re-marketing model to predict a payment behavior for a borrower of a non-stationary asset-based loan included within the distressed loan portfolio.
11. A system according to claim 10 wherein said server further configured to determine a percentage of loans within the probability distributions that will roll forward into a next period of delinquency.
12. A system according to claim 11 wherein said server further configured to indicate a number of months an account is delinquent.
13. A system according to claim 9 wherein said server further configured to adjust loan assumptions to account for variations in a plurality of forces impacting a payment behavior of a borrower including at least one of time of season, changes in political climate, interest rate changes, and a likelihood that a natural disaster may occur.
14. A system according to claim 13 wherein said server further configured to adjust probability distributions based on loan assumption adjustments.
15. A system according to claim 9 wherein said network is at least one of a WAN or a LAN.
16. A system for determining loan collection data for a group of non-stationary asset-based loans included within a distressed loan portfolio, an account including at least one of the loans, said system comprising:
a server configured with a collections model and a re-marketing model;
at least one computer; and
a network connecting said sewer to said at least one computer, said sewer configured to:
categorize each non-stationary asset-based loan included within the portfolio based on a prior month's payment of the corresponding loan, non-stationary asset-based loans include at least one of automobile loans, vehicle loans, and credit card loans;
categorize each loan included within the portfolio based on a contractual delinquency of the corresponding loan;
access the collections model to predict a payment behavior for a borrower of a non-stationary asset-based loan included within a distressed loan portfolio, the collections model is based on historical payment information of the borrower, loan delinquency assumptions, a plurality of collection strategies that may be utilized for collecting payment from the borrower, and the delinquency category assigned to the loan;
analyze the borrower's payment behavior after initiating at least one of the plurality of collection strategies including whether the borrower made a payment and, if so, an amount of the payment;
compare the borrower's payment behavior after initiating the at least one collection strategy to the predicted payment behavior of the borrower and the delinquency category assigned to the corresponding loan;
compare the borrower's payment behavior after initiating the at least one collection strategy to the prior month's payment category of the corresponding loan;
incorporate management feedback into expectations of future performance wherein management feedback includes recommending change in collection strategies used for prompting payment from the borrower associated with the loan included within the portfolio and predicting future payment performance of the borrower based on the recommended change in collection strategies;
update the collections model based on the payment comparisons and the management feedback, the updated collections model predicts future cash inflows for each loan included within the portfolio, the updated collections model is configured to apply a greater weight to the payment performance of each loan for the current month as compared to the payment performance of each loan for prior months;
access the re-marketing model to calculate an amount generated and expenses incurred from repossessing the non-stationary asset used as collateral for the borrower's loan, the re-marketing model further calculates a probability that an event will occur impacting payment of the borrower's loan;
generate matrices for delinquency, gross value, stock value, roll forward, roll back, amounts due and payment for each loan included within the group of loans including the borrower's loan, the matrices including data generated from the updated collections model and the re-marketing model; and
predict a portfolio value for the distressed loan portfolio using the matrices.
17. A system according to claim 16 wherein said server configured to predict a cash flow value for a portfolio.