
Optimizing Credit Risk Assessment Models for Market Entry Qualifications
ERCOT is evaluating credit risk assessment models to enhance market entry qualifications. The focus is on selecting relevant credit ratios and weightings to improve model accuracy. The assessment involves analyzing sample R-squares and correlation coefficients for rated counterparties. Two scoring models are used for public and non-public power entities, with a tiered weighting system. Through data analysis and methodology evolution, ERCOT aims to enhance the accuracy of its credit risk assessment models.
Download Presentation

Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
5. Counter-Party Credit Risk Assessment Mark Ruane Director, Settlements, Retail and Credit CWG / MCWG ERCOT Public July 15, 2020
Credit Risk Assessment Models Outline MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps 2 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps ERCOT is evaluating credit risk assessment models as part of the market entry qualifications initiative. The focus has been on choosing relevant credit ratios and associated weightings to optimize the r^2 of the model with respect to the scores for rated entities and their agency ratings. At the June CWG/MCWG, we reviewed the ERCOT credit scoring model 3 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps ERCOT model sample R-squares and correlation coefficients, utilizing 73 rated CPs out of 254 total.* R^2 Correl Base case Oliver Wyman 2008 0.076518 0.276619 2014 weights 0.115935 0.340493 First pass optimization 0.348826 0.590615 Evolution methodology 0.348826 0.590615 * Constrained for positive weights only. 4 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps 2008 Weight 2014 Weight Opt. Sol. Ratio Type Working Capital/Sales Cash/Assets Current Ratio EBITDA/Interest expense Debt Service CFO/Sales FCF/Debt Total Debt/Total Capital Equity/Assets EBITDA/Sales Net Income/Assets Total Assets Tangible Net Worth Sales/Assets Alternative Metric 1 Alternative Metric 2 Liquidity Liquidity Liquidity 30% 0% 10% 10% 0% 0% 0% 20% 10% 10% 10% 0% 0% 0% 0% 100% 15% 0% 10% 10% 15% 0% 0% 20% 10% 10% 0% 10% 0% 0% 0% 100% 0% 3% 30% 0% 20% 7% 4% 1% 8% 0% 1% 20% 6% 0% 0% 100% Weights with Excel- optimized solution. Debt Service Debt Service Leverage Leverage Profitability Profitability Size Size Activity Total 5 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps MISO uses two scoring models one for public power entities and another for non-public power. Note that the non-public model uses two-tier weighting. The MISO model was run with data from ERCOT agency-rated counterparties. 6 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps MISO model results for agency-rated Counter-Parties. Count 73 17 56 Case All 0.0251 -0.23362 0.00023 -0.62753 0.510935 0.17553 0.422986 -0.7148 -0.41896 -0.65037 0.565159 0.17562 0.411469 -0.75177 -0.41908 -0.64146 0.642087 0.136 0.411477 -0.8013 -0.36878 -0.64146 0.370243 -0.60848 0.228897 -0.47843 Pub. 5.3E-08 0.208899 Non-PP Base case - MISO weights R^2 Correlation R^2 Correlation R^2 Correlation R^2 Correlation R^2 Correlation R^2 Correlation Optimized non-public power Optimized public power Optimized both segments One model fits all - non-public One model fits all - public 7 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps MISO ratios and weightings. Quantitative Ratios Non-Public Quantitative Ratios Public Effective Weight Ratio Type Weight Category Weights Ratio Weight EBITDA/Interest expense Cash Earnings/Debt Service Free Cash flow/Total Debt Quick Ratio Liquidity Liquidity Liquidity Liquidity 25% 35% 30% 10% 100% 35% 15% 25% 25% 100% 25% 25% 25% 25% 100% 8% 11% 9% 3% Liquidity Leverage Performance 30% 20% 50% Days Cash/SGA+Interest Expense Debt Service Coverage Equity/Total Assets Times Interest Earned Cash/Current Liabilities CFFO/Total Debt Capex/Sales 20% 15% 15% 15% 15% 10% 10% 100% Total Debt/Total Capitalization Short Term Debt/Total Debt Leverage Debt/Net Fixed Assets Debt/TNW Leverage 7% 3% 5% 5% Leverage Leverage Total Total Return on Sales Return on Assets Operating Margin Return on Equity Performance Performance Performance Performance 13% 13% 13% 13% 100% Total 8 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps Optimized non-public power weights. Quantitative Ratios Non-Public Quantitative Ratios Public Power Effective Weight Ratio Type Weight Ratio Weight EBITDA/Interest expense Cash Earnings/Debt Service Free Cash flow/Total Debt Quick Ratio Liquidity Liquidity Liquidity Liquidity 0% 0% 53% 16% 0% 0% 0% 0% Days Cash/SGA+Interest Expense Debt Service Coverage Equity/Total Assets Times Interest Earned Cash/Current Liabilities CFFO/Total Debt Capex/Sales Total 4% 3% 63% 30% 0% 0% 0% 100% Debt/Total Capitalization Short Term Debt/Total Debt Leverage Debt/Net Fixed Assets Debt/TNW Leverage 3% 0% 9% 0% 0% 0% 0% 0% Leverage Leverage Return on Sales Return on Assets Operating Margin Return on Equity Performance Performance Performance Performance 2% 0% 0% 18% 0% 0% 0% 0% Total 100% 0% 9 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps Optimized public power weights. Quantitative Ratios Non-Public Quantitative Ratios Public Power Effective Weight Ratio Type Weight Ratio Weight EBITDA/Interest expense Cash Earnings/Debt Service Free Cash flow/Total Debt Quick Ratio Liquidity Liquidity Liquidity Liquidity 0% 0% 53% 16% 0% 0% 0% 0% Days Cash/SGA+Interest Expense Debt Service Coverage Equity/Total Assets Times Interest Earned Cash/Current Liabilities CFFO/Total Debt Capex/Sales Total 16% 0% 49% 19% 17% 0% 0% 100% Debt/Total Capitalization Short Term Debt/Total Debt Leverage Debt/Net Fixed Assets Debt/TNW Leverage 3% 0% 9% 0% 0% 0% 0% 0% Leverage Leverage Return on Sales Return on Assets Operating Margin Return on Equity Performance Performance Performance Performance 2% 0% 0% 18% 0% 0% 0% 0% Total 100% 0% 1 0 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps Optimized both segment weights. Quantitative Ratios Non-Public Quantitative Ratios Public Power Effective Weight Ratio Type Weight Ratio Weight EBITDA/Interest expense Cash Earnings/Debt Service Free Cash flow/Total Debt Quick Ratio Liquidity Liquidity Liquidity Liquidity 0% 0% 56% 8% 0% 0% 0% 0% Days Cash/SGA+Interest Expense Debt Service Coverage Equity/Total Assets Times Interest Earned Cash/Current Liabilities CFFO/Total Debt Capex/Sales Total 5% 3% 63% 29% 0% 0% 0% 100% Debt/Total Capitalization Short Term Debt/Total Debt Leverage Debt/Net Fixed Assets Debt/TNW Leverage 4% 0% 8% 0% 0% 0% 0% 0% Leverage Leverage Return on Sales Return on Assets Operating Margin Return on Equity Performance Performance Performance Performance 0% 0% 3% 22% 0% 0% 0% 0% Total 100% 0% 1 1 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps Quantitative Ratios Non-Public Effective Weight Ratio Type Weight Optimized one model fits all non- public weights EBITDA/Interest expense Cash Earnings/Debt Service Free Cash flow/Total Debt Quick Ratio Liquidity Liquidity Liquidity Liquidity 0% 0% 52% 22% 0% 0% 0% 0% Debt/Total Capitalization Short Term Debt/Total Debt Leverage Debt/Net Fixed Assets Debt/TNW Leverage 0% 2% 10% 0% 0% 0% 0% 0% Leverage Leverage Return on Sales Return on Assets Operating Margin Return on Equity Performance Performance Performance Performance 3% 0% 0% 11% 0% 0% 0% 0% Total 100% 0% 1 2 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps Quantitative Ratios Public Power Optimized one model fits all public weights Ratio Weight Days Cash/SGA+Interest Expense Debt Service Coverage Equity/Total Assets Times Interest Earned Cash/Current Liabilities CFFO/Total Debt Capex/Sales Total 0% 0% 2% 5% 69% 0% 24% 100% 1 3 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps MISO model notes Two-tiered weighting eliminated to allow optimization. Note that weightings of some ratios go to zero under optimization scenarios. Base case weighting is effectively random. Negative correlations are a function of the score assignment algorithm. 1 4 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps Moody s reviewed their credit rating tools with ERCOT staff. The primary output of the Moody s credit assessment models is an entity Expected Default Frequency (EDF) that corresponds to an implied credit rating. Supporting results include EDF percentile within segments and risk ratio contribution analysis. 1 5 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps The models (~30) for non-public entities are ratio based. Segments include large firms, not-for-profit, small businesses, etc. The public entity model inputs include market value, debt structure, asset volatility, etc. Outputs are updated daily. Additional features include what-if analysis, stress scenarios, portfolio analytics, trigger warnings, and AI- based credit sentiment reporting. 1 6 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps Moody s allows construction of a unified portfolio (public and private entities) for analytics. The model suite provides for incorporation of customizable qualitative factors. The models are calibrated to be forward-looking, and historically have shown peak EDF somewhat prior to actual defaults. 1 7 ERCOT Public
Credit Risk Assessment Models MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps Internal v vendor model considerations Internal Models Vendor Models Transparent inputs and construction Proprietary Low cost May be costly Limited analytics Greater scope of analytics Little, if any, segment disaggregation Multiple segments Small calibration dataset Large calibration dataset 1 8 ERCOT Public
Counter-Party Credit Risk Assessment MISO Scoring Model Recap ERCOT model Moody s RiskCalc Next Steps Discussion / next steps 1 9 ERCOT Public