HOW TO IMPROVE CREDIT REPORTING EFFICACY? VALUE ADDED SERVICES. Lynn Wang - Transunion Kuala Lumpur November, Session n.

Transcription

1 HOW TO IMPROVE CREDIT REPORTING EFFICACY? VALUE ADDED SERVICES Lynn Wang - Transunion Kuala Lumpur November, Session n. 13

2 Agenda About TransUnion Credit Bureau Product Evolution Bureau Scoring Customized Acquisition Score BCC (Bureau Credit Characteristics) & Application Beyong Decisioning Identity Fraud Solution Market Trend Report Q&A Exercise and Discussion 2

3 China Identity decisioning and fraud solution services Developed the generic score for Shanghai Credit Information Services India Consultancy (Credit Information to PBOC Bureau of India Ltd. CIBIL) Shareholder, technology provider, score and solutions provider to CIBIL, India s dominant and only operational consumer and commercial credit bureau Philippines Developed the country s first international private positive credit bureau Hong Kong TransUnion s Asia hub 100% of Hong Kong market share 99.9% CMS score subscription upon bureau inquiry. Korea Developed the generic Score for Korea Information Service Singapore (Credit Bureau of Singapore) Developed the country s first generic score with both POS and review functionalities for Singapore s dominant credit bureau Thailand (National Credit Bureau) Shareholder and technology provider to the only operating bureau in Thailand Developed the country s first generic score 3 3

4

5 STAGE 1 STAGE 2 STAGE 3 STAGE 4 STAGE 5 STAGE 6 CREDIT BUREAU DATABASE CORE VALUE ADDED SERVICES ANALYTICS & DECISIONING CONSUMER MANAGEMENT TOOLS DIRECT-TO- CONSUMER SERVICES MULTIPLE LINES OF BUSINESS Negative Data Positive Data Credit Reports Add-On Services Bureau Characteristics Scores Fraud & Identity Management Marketing Services Consumer Reports Consumer Scores Commercial Credit Report Auto Information Solutions Alerts Decision Tools Collections Management Credit Monitoring Rental Screening Custom Analytics Data Repository Consumer Education Other Lines of Business South Africa Hong Kong Canada India Mexico Chile Philippines TransUnion LLC All Rights Reserved 5

6 Account Acquisition Portfolio Management Collections / Debt Recovery Evaluate the consumer s risk Portfolio Quality Monitoring / BASEL II Find missing debtors Approve the right consumer Line Management / Top-up Prioritize collection accounts Customer Challenges Line Assignment Appropriate interest rate and fees Verify the consumer s identity Automated workflows into credit processing and instant decisioning Monitor consumer behavior and act accordingly Benchmark credit portfolio against the industry Stress test portfolio against expected and unexpected changes Lower fraud related loses Provide debtors an easy way to pay on past-due accounts Revenue Potential Cross-selling & Retention Analytics (Bureau score / customized acquisition, behavior or collection scores) Value-add Solutions and Platforms Data (Consumer, Commercial, Alternative) 2012 TransUnion LLC All Rights Reserved 6 6

7

8 Bureau Generic Risk Score Benefits Regulator: Stimulate economy growth. More consistent, objective and accurate risk assessment of portfolios. Lenders: Decreased loss rates / reduced credit costs Operation efficiency Responsible lending Consumers: More accessible credit at more affordable rate for responsible consumers. Less accessible credit for less responsible consumers so they don t go beyond their means. 8

9 How does credit scoring work? The design objective of a credit score is to to minimize the overlap area. The more effective the credit score is, the better it is at differentiating between these bad and good accounts. BAD ACCOUNTS Maximize the separation between good and bad accounts GOOD ACCOUNTS Low Scores OVERLAP AREA High Scores 9

10 Characteristic Values Points Utilization of open accounts updated in the past 12 months Months since most recent delinquency Average number of months account on file Cannot Calculate 0% 1% - 12% 13% - 28% 29% 39% 40% - 53% 54% - 59% 60% - 71% 72% - 77% 78% - 87% 88% - 92% 93% or more or more or more The score is comprised of characteristics that are predictive of future credit performance. Each scored consumer will have one value for each characteristic, based on his or her credit report at the time of scoring. The final score is the sum of the points awarded to each characteristic

11 New data elements evaluation: Incremental lift on existing score Impact to existing score lift Odds Development Recalibration: score odds change through different economic environment Present Score Redevelopment: score prediction deteriorates through different market dynamics Score fine-tuning plan recommendation Odds Development Present Score 11

12 Customized Acquisition Score

13 Unique Advantages No Other Service Providers can Compete: Speed to Market Reject Inferencing at Consumer Level, Score Vintage w/ Unit and $ Bad Rate% Possible Swap Set Analysis at Consumer Level Sufficient Bad from Rejected Turn Around Time Bureau Data Knowledge Early MOB Development Monitoring Possible Early MOB Monitoring Possible 13 * Using only bureau info Bad Rate by Score Simulation* on Other Creditors Apps Feasible

14 Unique Advantages No Other Score Vendors can Compete No Data Security Concern Existing Data Exchange Platform Existing Confidentiality Agreement Multiple O/S, Payment and Limit Snapshots as Predictive Attributes Existing IT Infrastructure Only Demographic Info Required Facility by Member Cluster Available / Line Sensitivity Min IT Effort on System Enhancement 14 Bureau Attributes Already Available (Atts Never Coded by Banks)

15 BCC (Bureau Credit Characteristics) & Application

16 A list of predefined characteristics that summarize various aspects of a customer s credit information. These characteristics can be used for detecting fraud, model development, data analysis, customer profiling, and other account management analyses. DETECT FRAUD Detect suspicious fraud by gaining visibility into suspicious behavior: , Telephone, Addresses used by different customers at different banks EX: # of consumers with the same cell phone number Number of new account being enquired or opened in a short period 16 Utilize predefined calculated variables to perform analysis with no IT development required: Delinquincy, outstanding balances, inquiry, payment ratios, etc. Ex: # of accounts with 90+ DPD w/in 12 MOS ELIMINATE MANUAL CALCULATIONS: Fool proof / Control Operation efficiency

17 Credit: manage credit risk Discrete criteria (# of presently delinquent account >=1 then reject) Scorecard Application workflow (with own bank account, route to team A; without, route to team B) Line assignment (time at bureau >XXX and # of open account>yyy then 2.5 times of monthly salary; time at bureau >XXX and # of open account<=yyyy then 2 times of monthly salary) Fraud: manage fraud risk Which application / transaction needs what level of investigation Which application / transaction needs phone verification Which application / transaction needs employment verification 17

18 BCC Fraud Related Trend Related Inquiry Related Account Related Demographic Related Phone # # of FID with the same cell phone # Bal Related Balance change % within last 3 / 6 months Count # of new app inquiry within Last 3 / 6 months Count # of credit card account open Within last 3 / 6 months Address # of FIDs with the same address Limit Any credit line increase within last 3 / 6 months Age # of month since last new app inquiry Age # of months of oldest trade Age # of months of latest income Employment # of FIDs with the same Company name Utilization Any credit line increase within last 3 / 6 months Amount Max inquiry amount within Last 3 / 6 months Amount Sum of total credit limit / OS Bal # of FIDs with the same address Payment # of Account with partial Payment Within last 3 / 6 months Delinquency Utilization of delinquent accounts 18

19 Consumer Mobile # Address Employment Work phone # Different people with the same mobile number?? Different people with the same / similar home address?? Different companies with the same business phone number?? 19

20 Beyond Decisioning Not only decision engine, also workflow engine, image storage, reporting, data capture

21 80% ready made solution with standard features + 20% customization to meet your particular requirement. Our solution is developed based on point and click applications so the development timeframe is much shorter compared to our competitors Risk Management Solutions Scoring & Analytics Solutions Fraud and Identity Management Solutions Marketing Solutions 21 21

22 Credit Data Customer History Data Interface Workflow Demogra phic Business Rules Engine Bureau Access Control Reports & Analysis Collector s IVR Call Center POS Web TM Direct Mail Decision Channel Interfaces 22

23 Key Customer Benefits Improved management of acquisition strategy to address key business objectives: Achieve lower losses Increase profit Increase productivity Improved controls and accuracy of execution of credit policies Automated access to credit data within process Improved feedback and reporting of performance of acquisition strategies 23

24 Sales Person enters Credit Application Fail Pre Bureau Decision? Pass Pull Credit Report Calculate Variables Credit Database Customer Sales Person requests additional information or co-applicant Fail Post Bureau Decision? Pass Calculate Scores Determine Required Documents Investigation Queue Check Submitted Documents Fail Multiple Approval Queues based on delegation and nature of deviation approval Senior Credit Officer Credit Officer Credit Approver Pass Document Decision? Manual Decision? Document follow up ID / Address / Employment Verification Pass Automatic Decision 24

25 Pre-bureau Policies Bureau Policies Post Bureau Policies Product Terms Rules based on applicant information Saves cost of pulling credit if basic policies are not met Criteria: Minimum Age Maximum Age Minimum Household Income Resident status Employment Type Minimum Time of Employment Regional Exclusions Set separate limits for both applicant and co-applicant Rules based on the credit data for the applicant and/or co-applicant Includes most commonly used credit data variables Criteria: Age of credit file Minimum number of trade references Maximum Delinquency Max delinquency in last 12 months Max delinquency in last 6 months Credit Utilization Total Debt Minimum risk score Set separate rules for each applicant. Customizable Customer internal policy rules Available tools Implement Custom internal risk scorecard(s) Debt to Income, capacity calculations Internal Database rules Criteria Minimum internal risk score value(s) Max Debt to Income Presence of Internal Database hit Rules to set terms of approved applications Provides uniform product terms based on policy rules Each product and subproduct may be set independently or shared Configurable product matrices. Preconfigured to use: Monthly Household Income Percent of monthly Debt to Income Available Product Terms Max Loan Amount Interest Rate Max Loan Term 25 TransUnion, Confidential

26 26

27 Allows customers to store loan documents electronically in a central repository Risk manager can manage required loan documents as part of the credit policy rules. Required documents may be tailored based on product type and channel. Generate loan documents to complete the loan approval process: Application forms for signature Offer Letters Letters Requesting additional documentation Amortization tables Truth in Lending forms Loan documents can be created in any MS Office formats (Word, Excel, etc.) 27

28 ABC 28

29 29

30 30

31 Maker / checker / approver control User module that allows customer to make changes to their risk and business policies. Real-time changes to UAT environment Move to production within 24 hours Five control modules Pre bureau policies Bureau policies Product Terms Document Requirements Approval Authorities EXPRESS SOLUTION FOR ACQUISITIONS TransUnion DecisionCentre Home New Application View Reports Rules Manager POLICY MANAGER Pre-Bureau Policies Credit Bureau Policies Pre-Bureau Policy Rules Minimum Age 18 Applicant Product Terms Submit Changes Credit Authorities Action Co-Applicant Decline Refer Referral Queue 18 TransUnion DecisionCentre Credit Referral Maximum Age Credit Referral Home New Application View Reports Rules Manager Minimum Monthly Credit Referral Household Income Minimum Employment Credit Referral Time in Months Credit Exclusions Residential Status Employment Type Select EXPRESS SOLUTION FOR ACQUISITIONS Select EXPRESS SOLUTION FOR ACQUISITIONS POLICY MANAGER Permanent Resident Regions Excluded Select Temporary Resident TransUnion DecisionCentreNon-Resident Residential Pre-Bureau Status Policies Credit Bureau Policies Select Excluded Residential Statuses Assign Product Matrix Home New Application View Reports Rules Manager POLICY MANAGER Pre-Bureau Policies Credit Bureau Policies Percent of Debt to Income Assign Product Term Constraints Min Loan Amount Max Loan Amount Min Length of Loan (months) Max Length of Loan (months) Product Description Row Range 3 <= 40% Product A Product B Product C Row Range 4 Standard Gold PlatinumA 50% <= Row Range 5 100% 500 <= CHANGE HISTORY Log Out History of Parameter changes with Notes should be stored here. Credit Authorities <= 2000 <= 5000 <= <= <= A 500 A Product Terms Select Product Group for each Cell Row Range 1 Cancel Changes <= 20% A A B Row Range 2 C Product Terms Credit Authorities <= 25% A A Monthly Household Income Col Range 1 Col Range 2 Col Range 3 Col Range 4 Col Range B B B A A B B C A A B C C A B C C C CHANGE HISTORY CH Log History of Paramete changes with Note should be stored he Product APR% 18% 14.5% 9.98% 31

32 Score-Card Set-up: Maker / Checker Intercept or Starting Score Reason Codes Score Details Linear Score-card or Raw Score Logistic Score-card or Final Score Multiplier x1, x2 - Characteristics 32

33 Transactions Reports Business Performance Daily/Monthly Processing Report Monthly Billing Report Daily Bureau Performance Report Daily/Monthly Approval Report Total, By Branch/ User/ Channel Daily/ Monthly Decline Report By Decline Reason Daily Bureau Performance Report Monthly Override By reason code Score Distribution Report 33

34 Allow credit operation / approval managers to understand the size of Work-in-Progress (WIP) and make resource allocation / capacity planning / productivity Contents by: Queue type (such as credit review) Username (such as Joe Smith) or overall Team Reporting interval: From and to day with max of XXX days MIS: # of apps processed Result (such as approved / rejected) # of pending processing Time at the queue 34

35 Normal applications will follow first in, first out queue assignment. Urgent / VIP applications will jump queue to the top of queue for priority processing Alert: Automatic pending approval for senior team members 35

36 Other components of acquisition processing in within Express Acquisitions Solution: Negative Hit File Look up against clients database of bad accounts, watch list customers Champion / Challenger testing Production testing of changes in policy rule, scorecards, and product offering provides real-life test results without risking the entire portfolio 36

37 Call to action : MobileCredit 37 37

38 TransUnion Branch in Customer Credit Bureau Bureau Pull Generate predefined variables Workflow Rejection Message Property Appraisal and Draw Down App Form Only Data Validation & Blacklist Dual Matrix of Internal and CMS Scores Generate Internal score Rules based on bureau info No Auto Decision Yes Yes Reject? No Send /SMS to applicant on approval amount / pricing Incremental field capture & Docs Verification Collect Security Documentation Collate Data Reject? No Submit Docs Company Systems Yes Rejection Message Customer Acceptance Admin Control 38

39 Fraud & Identity Management Confirm who is on the other side of your transaction by using an automated environment to access multiple data sources and verify current consumer identities. Predict the likelihood of fraud. 39

40 Data Source 1) Input Consumer Input Pull Credit or Other Data Calculate Characteristics Generate Questions & Exam 2) Exam Exam Output 3) Answers Exam Answers Evaluate Answers Decide Final Score and Decision 4) Score & Decision: (Pass/Fail) Exam Output 40

41 Market Trend Reports

42 Stats Trend Chart as Part of Delivery Chart Window: 15 Months Change from same time last year and latest 2 time periods MoM QoQ

43 Market Share Market Rank Statistics: 1 months Snapshot Total O/S Limit Utilization Cash O/S Interest O/S % of A/C with Interest O/S >0 Unbilled O/S Past Due O/S Payment Amount Payment Ratio Partial Payment Count Partial Payment A/C % Stats 43

44 From individual consumer level, what s the wallet share of his / her usage of your credit facility Average SOW Statistics: 1 months Snapshot on Amount related Stat Total O/S Limit Cash O/S Interest O/S Unbilled O/S Past Due O/S Payment Amount SOW Distribution SOW Distribution Migration

45 Thank you! 45

46 Exercise and Discussion

47 Q1 Scenario: you are now working at a bank and the bank would like to develop an acquisition risk score. Question: 1. Do you think an acquisition score can replace credit officers judgment? Why or why not? 2. What s the pros and cons of fully automated credit decision? 47

48 Scenario: U.S. researchers found correlation between consumers driving and debt repayment behaviors are correlated (responsible debtors are highly likely to be responsible drivers), therefore, consumers credit report and credit score are considered for insurance premium decision. Question: 1. What s the pros and cons from such practice for consumers / insurance companies? 2. What s your view as a regulator? 48

49 Q3 Default Consumer Segment 1: Do nothing: 70% paid 80% paid SMS: 85% paid Dunning letter: 88% Auto dialer: 90% paid Collector: 91% paid Default Consumer Segment 2: Do nothing: 10% paid 40% paid SMS: 60% paid Dunning letter: 70% Auto dialer: 75% paid Collector: 78% paid Question: If you are working as a collection shop manager. Which segment will you take what action? 49

50 Scenario: a country has severe higher education student loan default issue. In order to get repayment from long overdue student loans after graduation, the student loan authority is considering no-payment cases to be included in the credit bureau. Question: 1. Will you as a regulator consider the same? Why or why not? 2. If this has to happen, what will be the prerequisites in your opinion? 50

51 Scenario: a country has severe mobile phone default issue. In order to get repayment, a mobile phone carrier is considering nopayment cases to be reported to the credit bureau. Question: 1. Will you consider the same? Why or why not? 2. If this has to happen, what will be the prerequisites in your opinion? 51

52 Scenario: a country has severe water / electricity default issue. In order to get repayment, utility company is considering no-payment cases to be reported to the credit bureau. Question: 1. Will you consider the same? Why or why not? 2. If this has to happen, what will be the prerequisites in your opinion? 52