Credit Reporting and Scoring for MSMEs. Vincenzo Resta Orlando Antonio Rossi CRIF Kuala Lumpur November, Session n. 12

Transcription

1 Credit Reporting and Scoring for MSMEs Vincenzo Resta Orlando Antonio Rossi CRIF Kuala Lumpur November, Session n. 12

2 Credit Reporting and Scoring for MSMEs Serious Game followed by presentation Select 4 groups Make decisions as a bank operator or risk manager, in various situations - first information poor, then information rich 2

3 GAME 3

4 Lessons Learnt The game is meant to be illustrative, not realistic, of some of the challenges that need to be addressed in analyzing risk Goal One: show that access to information and analysis beyond what is presented by the client is absolutely necessary due to asymmetry of credit risk information Goal Two: need access to the right kind of information 4

5 Lessons Learnt There is a case of each of the following aspects of credit risk, and the need for specific types of information Identification info need to ensure the right client Negative data eliminate bad payers Positive data existing exposure Historical data prediction of behavior Request info potential exposure Current payment due liquidity risk Co-borrowers, guarantors, companies linked risk Extensions specific or third party data 5

6 Lessons Learnt Different decisions have different points To achieve a high score you need to reject high risk clients, but also, and perhaps especially, accept the low risk ones Lesson Three: credit risk also includes the risk of rejecting good applicants 6

7 Lessons Learnt Fourth Goal: show difficulty to evaluate MSMEs Link between owners and MSMEs and also between mother company and MSMEs In these cases, your decision likely rested on analytical tools (the score) more than in other cases Fifth Goal: credit risk needs to be managed with the appropriate tools through all stages of the credit cycle 7

8 Lessons Learnt Inquire a potential client Review related individuals Review related companies Inquire existing clients to monitor performance Keep tracks of alerts to manage relationship and avoid erosion Client Acquisition Portfolio Management Market Analysis Collections Management Inquire existing clients to Upsell / Cross-sell Inquire existing clients to assess policy results Identify early collections candidates Identify likely collections targets Determine appropriate collection strategy 8

9 The Science Now we will talk in depth about ways to correctly assess the credit risk of MSMEs, with case studies, examples, and tools 9

10 Index Firm Lending The origination process The role of information The Credit Management Platform Risk evaluation approaches Rating models and PD Credit risk mitigation and LGD Expected loss Different strategies for different segments: some examples

11 Firm Lending A relationship based on different processes Origination Monitoring Collection - Managerial resources with different skills - Different risk assessment models - Tools for modifying the price and collateral conditions - Tools for managing collection processes 11

12 The origination process The origination process aims to identify the firm's risk The origination process can be summarized in 5 general stages (investigation): First: fundamental data verification and data collection Second: analysis on business, strategies and policies Third: index analysis (balance sheet) Fourth: forecasting (mainly cash flows) Fifth: analyst s final report 12

13 SME credit evaluation: the role of info Estimating the credit risk of a company requires monitoring of a variety of sources of information - Public - Financial Statement - Fiscal - Credit Bureau - Commercial Payment - Qualitative And, moreover: - Macroeconomic - Industry Generally, the more information you have the better is the evaluation but information has a cost 13

14 Focus on Public information What do we mean with public information? - Public register of companies - Legal proceedings - Liens, repossessions, etc. - Judicial registry of payment defaults - Corporate structure - Public Repository Banks Real estate companies Financial istitution Domestic companies International companies 14

15 Ranking the information Public information: score 5/5 Historical Up to date Financial Statement : score 4/5 Fiscal information: score 2/5 Historical Credit Bureau: score 5/5 Commercial Payment : score 2/5 Qualitative information: score 3/5 Up to date Historical Up to date Forward looking 15

16 Availability of information Basic economic System Qualitative information Fiscal information Financial statement information Public information Credit Bureau Advanced economic System Commercial payment 16

17 Environment information Macroeconomic Macro Industry - Economic cycle - Benchmarking - Competitive positioning - Industry Trend - Average Risk - Working capital dynamics -. Micro Industry 17

18 Bank Credit Management Platform All the above information must flow into the Credit Management Platform (CMP) which must be structured taking into account to ensure: - Availability - Navigability - Adaptability - Shareability - Storability CREDIT DATAWHEREHOUSE CMP Rating / Scoring Tools Decisioning Tool Monitoring Tool Workflows Tool Collateral Tool Collection Tool 18

19 Risk Valuation approaches Three main macro categories of valuation models Simple Subjective Bottom UP Judgmental valuation = risk assessment is carried out in a subjective manner on the basis of the analysis of all the available information for the analyst Scoring Models = the risk is estimated by applying simple models of common sense and experience. The output is a class of risk Advanced Objective TOP Down Internal Rating System: The risk is estimated by advanced models that analyze all available information and determine a class of risk and a probability of default 19

20 Rating models and PD A rating model is an advanced set of automatic tools that evaluate the firm form different points of view (economical, financial, behaviour of the firm on the banking system, qualitative information etc.) to determine the probability of the occurence of a negative event that is the default COMPANY PROFILE INFO CREDIT BUREAU BEHAVIORAL DATA FINANCIAL STATEMENTS INDUSTRY ƒlog (,,, ) = PD Probability of Default in the next 12 months Company dynamics Average debt Type of debt Industry financial ratios Macroeconomic variables Credit Bureau Score and risk indicators by industry 20

21 The structure of a rating model: modules Financial Statement Module Fiscal Module External Behavioural Module Internal Behavioural Module Public Information Module A SME Known SME Unknown B Large Corporate X X X X X X X X X X X Preliminary rating class Qualitative Module Final Rating class 21

22 Different rating approaches for different clients A correct estimate of the credit risk requires to develop different rating models depending on counterparty type Models to estimate credit risk of SMEs CORPORATE MEDIUM ENTERPRISES External SMALL BUSINESS mostly manual process semi-automatic process Automatic Process 22

23 Rating models Introducing a rating system into the organization involves many IT system External Information (financial statement, Pubblic information, Credit bureau ) DWH Back OFFICE Rating Engine Internal Information (behavioural) Decisioning Engine Workflow Engine FRONT OFFICE SOFTWARE 23

24 Credit Risk Mitigation The the ability to reduce the risk of credit exposure of a guarantee depends on: - Market value of the guarantee (at the moment of default) - Time required for recovery procedures Personal Guarantees Collateral LGD = Loss Given Default (secured) Recovery Flows Recovery Cost Flows Time Interest rate 24

25 The expected loss Not only PD and LGD but one last element EAD = Exprosure At Default EAD * PD * LGD = Expected Loss Considering the expected loss cost ensures a balance in economic and financial management of the bank 25

26 Origination process: diversity Small AL Medium AL Corporate AL First X High X High X High Second Opt Medium X Low Third Fourth X High X High X X Medium Low Fifth X High X Low First: fundamental data verification and data collection Second: analysis on business, strategies and policies Third: index analysis (balance sheet, or fiscal and behavioral) Fourth: forecasting (mainly cash flows) Fifth: analyst s final report 26

27 Small segment: ex of simple strategy SME Small Dimension Branch Manager Task 1 Task 2 PD OK Low Task 3 Task 4 Check customer presence in the Black List Non scorable firm Application Score PD Medium Check guarantess values Input of preliminary information PD High KO More 50% Less 50% OK KO 27

28 Medium segment: ex of simple strategy SME Small Dimension Task 4 LOW EL LOW PD OK LP Branch Manager Task 1 2 Instrumental Mortgages LGD CALCULATION LOW EL MDM PD LOW EL HIGH PD OK HP M A N Check customer presence in the Black List Rating calculation OTHER CASES KO Input of preliminary information Task 5 Other technical LGD forms CALCULATION Task 3 LOW EL MEDIUM EL OK LP OK HP OTHER CASES KO 28

29 Take away - Financing a firm means creating a relationship - This relationship must be evalueted in three processes: origination, monitoring and (eventually) collection - Evaluating the risk means evaluating the information - The best method to measure risk of the firm is creating a rating system - The PD is not the only element that give you a measure of the risk: you need to measure the quality of guarantees - Combining the risk of default and the estimation of what you will lose in the event of default you ll obtain a synthetic statistical measure of the risk that is the expected loss - SME segment can be highly automatized by well defined automatic strategies trough decisional and workflow software 29