Identifying At-Risk Names in Your
Credit Portfolio
David T. Hamilton, PhD, Managing Director, Capital Markets Research
Gro...
“Yes, risk taking is inherently failure-prone. Otherwise, it
would be called sure-thing-taking.”
− Jim McMahon

2
1

Strategies for identifying at-risk
names in your portfolio

3
Developing a Monitoring and Early Warning Toolkit
Public Expected Default Frequency (EDF) measures can be utilized in
four...
Challenges to setting up an effective monitoring and
early warning process
» System must be set up to detect outcome(s) of...
Effective credit risk monitoring using EDF measures
» Monitoring and early warning are problems of
classification: which f...
Assessing risk classification accuracy
A credit risk measure, like an EDF or a rating, makes a prediction that will turn
o...
Developing an optimal early warning signal
We can calibrate an optimal early warning trigger level by measuring the TPR
an...
Optimal EDF triggers for select industry sectors
Global public companies, 1997-2011

Sector

t*
(Pctile)

Banks

0.281%
(7...
Static EDF early warning triggers are not optimal
Box plots of EDF measures for US Firms, January 2008 vs. January 2009
40...
The optimal trigger will vary significantly over the cycle

11
Firms whose EDF measures are high relative to their
industry sectors experience higher default rates
One-year default rate...
Negative EDF momentum signals higher default risk
One-year default rates conditioned on EDF momentum

13
Default rates are sensitive to EDF momentum vs. sector

Firm EDF Level

One-year default rates conditioned on EDF decile a...
Developing a monitoring and early warning playbook
Our research suggests a general approach to effective early warning
usi...
2

Strategies in practice: Case study

16
JC Penney’s EDF metric has increased sharply over the
past year, but how concerned should we be?
8

» Since 1Q2012, its ED...
Is a 6% EDF high? How risky is JC Penney?
8

Caa1

» Mapping PDs into ratings
can be useful rules of
thumb, to grasp level...
As JC Penney’s EDF has risen, its relative EDF level
shows that it is risky for its industry sector

» Over the past year ...
The sharp jump in JCP’s EDF suggests that risk is more
likely to increase than to decrease in the future
8

» Although JC ...
The deterioration in JCP’s EDF was also in sharp
contrast with general improvement in its sector’s EDFs

» JCP’s EDF has
u...
Case study key take-aways
» The four EDF-based early warning strategies add powerful tools for
monitoring credit risk; eac...
3

Conclusion

23
Summary
» Using EDF measures as the key metric, we showed how to
develop an early warning “playbook” using four key
monito...
Putting It All Together: A Traffic Light Approach

EDF Level

25
Q&A

26
Contact
David T. Hamilton
Managing Director
Quantitative Credit Research
Capital Markets Research Group

Irina Makarova
As...
© 2013 Moody’s Analytics, Inc. and/or its licensors and affiliates (collectively, “MOODY’S”). All rights reserved. ALL INF...
Upcoming SlideShare
Loading in …5
×

Identifying At-Risk Names in Your Credit Portfolio

702 views

Published on

Identifying at-risk names in your public company C&I portfolio with enough anticipation might allow you to reduce your exposure or take the necessary actions to minimize the losses you could suffer in the event of a credit event. Developing a risk monitoring and early warning system is a time and resource intensive process. This presentation, from a recently held webcast by Moody's Analytics Capital Markets Research Group, addresses developing strategies to identify the relevant measures that could indicate increased risk in your C&I portfolio.

Published in: Economy & Finance, Business
0 Comments
2 Likes
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
702
On SlideShare
0
From Embeds
0
Number of Embeds
1
Actions
Shares
0
Downloads
44
Comments
0
Likes
2
Embeds 0
No embeds

No notes for slide
  • Its EDF measure began deteriorating rapidly in 2012, after its then CEO Ron Johnson failed to improve sales after implementing a new strategy that eliminated the long-rime practice of frequent discounts and promotions
  • Identifying At-Risk Names in Your Credit Portfolio

    1. 1. Identifying At-Risk Names in Your Credit Portfolio David T. Hamilton, PhD, Managing Director, Capital Markets Research Group Irina Makarova, Assistant Director, Capital Markets Research Group Originally presented as a part of a Moody’s Analytics webinar October 2, 2013
    2. 2. “Yes, risk taking is inherently failure-prone. Otherwise, it would be called sure-thing-taking.” − Jim McMahon 2
    3. 3. 1 Strategies for identifying at-risk names in your portfolio 3
    4. 4. Developing a Monitoring and Early Warning Toolkit Public Expected Default Frequency (EDF) measures can be utilized in four ways to more effectively monitor risk and signal early warning: 1. EDF level – EDFs have been shown to be powerful at rank ordering risk – How can we choose the right EDF warning trigger level? 2. Relative EDF level – Peer group or industry sector comparisons add incremental predictive power to detect credit risk 3. EDF change – EDF measures exhibit momentum 4. EDF relative change – Like relative level, comparisons to a peer or industry group add incremental predictive power 4
    5. 5. Challenges to setting up an effective monitoring and early warning process » System must be set up to detect outcome(s) of interest/importance – Default – Rating change – Spread movement » Filters are likely to be very sensitive to time period, peer group, time horizon, etc. » Triggers should also be a function of the economic importance of the exposure » Process can only filter results, cannot dictate actions 5
    6. 6. Effective credit risk monitoring using EDF measures » Monitoring and early warning are problems of classification: which firms in a portfolio should be considered relatively more risky, and therefore merit deeper investigation? » Estimating the benefits and costs of classification may be difficult, imprecise, or impossible » We may therefore attempt to maximize predictive accuracy of classification instead 6
    7. 7. Assessing risk classification accuracy A credit risk measure, like an EDF or a rating, makes a prediction that will turn out to be correct or not. There are thus four types of classification error. Actual Outcome Low Risk Predicted by Credit Measure High Risk Low Risk True Negative False Negative High Risk False Positive True Positive We can also define two useful measures of accuracy from this table: TP TPR TP FN FP FPR TN FP The overall predictive power of any risk measure is determined by both the TPR and FPR 7
    8. 8. Developing an optimal early warning signal We can calibrate an optimal early warning trigger level by measuring the TPR and FPR for a given trigger value. Finding the optimal trigger levels entails three steps: 1. Choose a candidate EDF trigger level 2. Measure the TPR and FPR associated with the EDF trigger level 3. The optimal EDF trigger level is the one that maximizes (TPR – FPR) Optimal EDF early trigger level, global public companies, 1997-2012 t* Percentile Rank 2.673% 82.701% Median 25th Percentile 75th Percentile 0.251% 0.084% 1.224% 8
    9. 9. Optimal EDF triggers for select industry sectors Global public companies, 1997-2011 Sector t* (Pctile) Banks 0.281% (75.464%) Consumer Products 0.247% (65.045%) High Tech 13.715% (95.489%) Transportation 4.344% (83.923%) 9 9
    10. 10. Static EDF early warning triggers are not optimal Box plots of EDF measures for US Firms, January 2008 vs. January 2009 40 EDF % 30 20 10 Review Trigger Level 0 Jan 2008 Jan 2009 10
    11. 11. The optimal trigger will vary significantly over the cycle 11
    12. 12. Firms whose EDF measures are high relative to their industry sectors experience higher default rates One-year default rates conditioned on EDF quartile and relative EDF quartile Relative EDF Level Quartile 1 2 3 4 1 0.02% 0.10% 0.35% 0.72% EDF 2 0.05% 0.31% 0.45% 1.71% Level 3 0.00% 0.29% 0.77% 2.78% 4 NA 2.64% 2.68% Default risk rises with EDF level 6.65% Default risk increases when EDFs are relatively high 12
    13. 13. Negative EDF momentum signals higher default risk One-year default rates conditioned on EDF momentum 13
    14. 14. Default rates are sensitive to EDF momentum vs. sector Firm EDF Level One-year default rates conditioned on EDF decile and EDF change vs. sector change 1 2 3 4 5 6 7 8 9 10 All 1 0.05% 0.10% 0.10% 0.28% 0.32% 0.62% 0.71% 1.01% 3.14% 6.43% 0.66% 2 0.03% 0.05% 0.06% 0.12% 0.23% 0.44% 0.56% 1.01% 2.22% 4.68% 0.63% EDF Change Relative to Industry Peer Group Change 3 4 5 6 7 8 0.02% 0.00% 0.00% 0.01% 0.03% 0.00% 0.06% 0.06% 0.00% 0.00% 0.02% 0.07% 0.01% 0.03% 0.01% 0.03% 0.07% 0.06% 0.17% 0.15% 0.09% 0.10% 0.08% 0.09% 0.24% 0.32% 0.22% 0.24% 0.21% 0.27% 0.45% 0.34% 0.44% 0.56% 0.44% 0.72% 0.66% 0.80% 0.64% 0.72% 0.73% 1.06% 1.19% 1.25% 1.27% 1.44% 1.58% 1.65% 4.83% 5.16% 5.25% 4.34% 4.87% 5.75% 5.76% 7.70% 7.70% 6.96% 7.67% 9.31% 1.08% 1.73% 1.73% 1.83% 2.24% 2.92% 9 0.00% 0.11% 0.03% 0.17% 0.22% 0.51% 1.18% 2.05% 6.37% 9.99% 3.13% 10 0.00% 0.27% 0.18% 0.30% 0.46% 0.97% 1.63% 3.10% 8.39% 13.70% 5.96% ALL 0.02% 0.05% 0.05% 0.15% 0.27% 0.55% 0.89% 1.68% 5.60% 8.94% 2.16% Default risk rises with EDF level Default risk increases with poor performance vs. industry 14
    15. 15. Developing a monitoring and early warning playbook Our research suggests a general approach to effective early warning using EDF measures: 1. Calibrate an appropriate EDF trigger level for your portfolio. Names in excess of the trigger are statistically more likely to default and should be analyzed further 2. For names that exceed the trigger EDF, check their EDF levels relative to their industry sectors. Names in the 75th or worse percentile are of particular concern. 3. Also check the EDF momentum for these names. Firms whose EDFs have worsened in the past are likely to continue to worsen, and are more likely to default than other firms with the same EDF level 4. Last, when an firm’s EDF momentum is underperforming its industry sector, it is more likely to default than other firms with the same EDF level 15
    16. 16. 2 Strategies in practice: Case study 16
    17. 17. JC Penney’s EDF metric has increased sharply over the past year, but how concerned should we be? 8 » Since 1Q2012, its EDF measure has increased sharply, reaching an all-time high of 7.9% in September 2013 » The optimal EDF trigger for the US department stores group is 0.9% on average Sears 6 JC Penney 5 4 3 2 1 0 Sep-13 Jun-13 Jun-13 Jun-13 Mar-13 Mar-13 Mar-13 Dec-12 Dec-12 Dec-12 Sep-12 Jun-12 Jun-12 Jun-12 Mar-12 Mar-12 Mar-12 Dec-11 Dec-11 Dec-11 Sep-11 Jun-11 Jun-11 Jun-11 Mar-11 Mar-11 Mar-11 Dec-10 Dec-10 Dec-10 Sep-10 Jun-10 Mar-10 Mar-10 Mar-10 Dec-09 Dec-09 Dec-09 Sep-09 Jun-09 Mar-09 Mar-09 Mar-09 Dec-08 Dec-08 Dec-08 Sep-08 Sep-08 Sep-08 » Compared to Sears, which one is riskier? 7 1-Year EDF (%) » Coming out of the financial crisis and recession, JC Penney’s EDF measure fell from 3.5% to 0.5% in 1Q2012 Source: CreditEdge STRATEGY 1: EDF LEVEL 17
    18. 18. Is a 6% EDF high? How risky is JC Penney? 8 Caa1 » Mapping PDs into ratings can be useful rules of thumb, to grasp levels and make relative comparisons 7 6 B3 Sears JC Penney » Here, we map EDFs into ratings using historical default rates » JC Penney’s current 6% EDF maps to a B3 rating, suggesting a relatively high risk of default » Sears’ EDF has improved to 1.5% as of Sept., equivalent to a Ba3 rating 1-Year EDF (%) 5 4 B2 3 B1 2 1 Ba3 Ba2 0 Ba1 Baa3 Source: CreditEdge STRATEGY 1: EDF LEVEL 18
    19. 19. As JC Penney’s EDF has risen, its relative EDF level shows that it is risky for its industry sector » Over the past year its EDF has deteriorated in both absolute and relative terms, and is now trending in the 75th percentile of its industry sector » Firms whose EDF levels are in the 75th percentile of their sectors are 4.5X more likely to default than those below the 75th percentile 75% 1-Year EDF (%, log scale) » Between 2008 and 2012, JCP’s EDF has either outperformed or tracked the median EDF of its industry sector 10.0 Median 1.0 25% 0.1 0.0 US Department Stores Group EDF Source: CreditEdge STRATEGY 2: RELATIVE LEVEL 19
    20. 20. The sharp jump in JCP’s EDF suggests that risk is more likely to increase than to decrease in the future 8 » Although JC Penney’s EDF has fallen recently, its EDF is more likely to continue to increase than to decrease over the next year Sears 6 5 4 3 2 1 0 Aug-13 Jun-13 Apr-13 Feb-13 Dec-12 Oct-12 Aug-12 Jun-12 Apr-12 Feb-12 Dec-11 Oct-11 Jul-11 May-11 Mar-11 Jan-11 » Sears is an interesting contrast: its EDF has been higher longer, but has basically moved sideways since its 2011 spike JC Penney 7 1-Year EDF (%) » JCP’s EDF has increased at an accelerating rate, doubling roughly every six months since the start of 2012 Source: CreditEdge STRATEGY 3: EDF CHANGE 20
    21. 21. The deterioration in JCP’s EDF was also in sharp contrast with general improvement in its sector’s EDFs » JCP’s EDF has underperformed its sector since 2Q2012 » The change in its EDF relative to its sector places it in the 90th percentile, making it twice as likely to default than companies at the median 10.0 75% 1-Year EDF (%, log scale) » Since 2012, the whole distribution of EDF levels for the US department stores sector has shifted lower 1.0 Median 0.1 25% 0.0 US Department Stores Group JC Penny Sears Source: CreditEdge STRATEGY 4: RELATIVE CHANGE 21
    22. 22. Case study key take-aways » The four EDF-based early warning strategies add powerful tools for monitoring credit risk; each adds incremental information over EDF level alone » The strategies can be implemented as part of a formal early warning process or used in name-by-name judgmental analysis » So, who is riskier? JC Penney or Sears? JC Penney Sears 6% 1.5% Relative Level 75th pctile 70th pctile Momentum +1,227% -17% Relative Momentum 90th pctile 55th pctile EDF Level 22
    23. 23. 3 Conclusion 23
    24. 24. Summary » Using EDF measures as the key metric, we showed how to develop an early warning “playbook” using four key monitoring strategies » These strategies have the advantage of being intuitive and relatively easy to calibrate and implement with available data » In the JC Penney case study we applied these strategies to show how to differentiate between two companies that, by appearance, seem equally risky » The strategies can be used a basis for a synthetic early warning signal, or as part of an automated early warning process 24
    25. 25. Putting It All Together: A Traffic Light Approach EDF Level 25
    26. 26. Q&A 26
    27. 27. Contact David T. Hamilton Managing Director Quantitative Credit Research Capital Markets Research Group Irina Makarova Assistant Director Quantitative Credit Research Capital Markets Research Group Moody’s Analytics 7 World Trade Center New York, NY 10007 Moody’s Analytics 7 World Trade Center New York, NY 10007 +1 212 553-1695 tel +1 212 298-6934 fax david.hamilton@moodys.com +1 212 553-4307 tel +1 212 298-6934 fax irina.makarova@moodys.com Join the conversation on LinkedIn: Credit Edge – Best Practices in Credit Risk 27
    28. 28. © 2013 Moody’s Analytics, Inc. and/or its licensors and affiliates (collectively, “MOODY’S”). All rights reserved. ALL INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR OTHERWISE REPRODUCED, REPACKAGED, FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY PERSON WITHOUT MOODY’S PRIOR WRITTEN CONSENT. All information contained herein is obtained by MOODY’S from sources believed by it to be accurate and reliable. Because of the possibility of human or mechanical error as well as other factors, however, all information contained herein is provided “AS IS” without warranty of any kind. Under no circumstances shall MOODY’S have any liability to any person or entity for (a) any loss or damage in whole or in part caused by, resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of MOODY’S or any of its directors, officers, employees or agents in connection with the procurement, collection, compilation, analysis, interpretation, communication, publication or delivery of any such information, or (b) any direct, indirect, special, consequential, compensatory or incidental damages whatsoever (including without limitation, lost profits), even if MOODY’S is advised in advance of the possibility of such damages, resulting from the use of or inability to use, any such information. The credit ratings, financial reporting analysis, projections, and other observations, if any, constituting part of the information contained herein are, and must be construed solely as, statements of opinion and not statements of fact or recommendations to purchase, sell or hold any securities. NO WARRANTY, EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR OTHER OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY’S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be weighed solely as one factor in any investment decision made by or on behalf of any user of the information contained herein, and each such user must accordingly make its own study and evaluation of each security and of each issuer and guarantor of, and each provider of credit support for, each security that it may consider purchasing, holding, or selling. 28

    ×