It is now widely recognized by financial institutions, corporate investors, insurance firms and regulators that an over-reliance on low-quality legacy credit ratings was a major contributor to the credit crisis of 2006 to 2011. The Levin Report (United States Senate, April 13, 2011, pages 243 to 317) summarizes these findings in detail. How can investors of all types make the transition from legacy credit ratings to modern default probabilities? This blog summarizes a 10 step transition that is practical, detailed, and efficient in replacing the 100-year-old rating concept with best practice in credit assessment: modern default probabilities.
Objectives of Credit Assessment
It was once common to hear banks, insurance firms and other institutional investors argue that it was too expensive for such investors to do their own credit analysis; reliance on ratings, it was said, was much more efficient. A more modern view, however, comes from legendary fixed income investor Bill Gross:
“No one or no one company has a monopoly on investment or ratings expertise. Second grade intelligence and a high [common sense quotient] are a rare combination for an individual rating agency or an investment management firm as well. Still, the rating agencies in recent years have displayed little of either. In addition, they have brazenly sold their reputations for unbiased judgment to the very companies they were standing in judgment upon. Don’t bury them however; like vampires in the dead of the night they will outlast us all. Those looking to profit at their expense, however, will dismiss them. They no longer serve a valid purpose for investment companies free of regulatory mandates that can think with a teaspoon of IQ and a tablespoon of CQ.” (William H. Gross, “Investment Outlook: Lovin’ Spoonful,” May 2010, PIMCO)
In light of the 2006 to 2011 credit crisis, common sense corporate governance says that, if an investor cannot afford to do its own credit analysis or cannot afford to buy high quality credit assessment tools, the investor should not buy the security at issue. We have seen in the credit crisis that it is much more expensive to rely on conflicted third parties like legacy credit rating agencies when the resulting losses from rating agency errors are taken into account. Institutional investors need the tools to answer the basic credit questions with a high degree of accuracy:
What is the primary source of repayment and its probability?
What is the secondary source of repayment and its probability?
How do the risks to the security’s cash flows vary with macro factors?
Can these risks be hedged?
In light of these observations, what does the term “investment grade” mean in the 21st century? Fortunately, the U.S. government has recently improved on its definition of that term.
What Does Investment Grade Mean Today?
The 2010 Dodd-Frank Act requires the U.S. government to remove mandatory reliance on legacy credit ratings from its regulations. The new definition of investment grade that applies to national banks in the United States was published in June, 2012, and it is clear and explicit:
Section 1.2 Definitions
(d) Investment grade means the issuer of a security has an adequate capacity to meet financial commitments under the security for the projected life of the asset or exposure. An issuer has an adequate capacity to meet financial commitments if the risk of default by the obligor is low and the full and timely repayment of principal and interest is expected.
Source: Federal Register, Vol. 77, No. 114, Wednesday, June 13, 2012, Rules and Regulations, page 35257
How does a bank, insurance company, investment management firm or corporate treasury operation go about using this definition after decades of reliance on the legacy credit ratings concept? We present a 10 step transition plan in the next section.
A 10 Step Transition Plan from Legacy Credit Ratings to Modern Default Probabilities
There is a very direct and efficient 10 step transition from legacy credit ratings to modern default probabilities. We summarize them briefly here and then discuss each step in more detail.
- Recognize the facts about ratings.
- Revise all investment policies.
- Use default probabilities, not ratings, internally.
- Understand macro-factor linkages.
- Eliminate “transition matrices.”
- Use stress testing.
- Use dollar risk limits and default probability risk limits.
- Use these limits for portfolio risk levels.
- Conduct on-going “re-education camps” internally.
- Ask for help.
These 10 steps will help firms make the transition in an efficient manner and result in best practice credit assessments. We now discuss each of these steps in turn.
1. Recognize the facts about ratings.
Step 1 is to recognize the reality of ratings’ in-accuracies, vagueness, and conflicts of interest. This table from a recent blog (van Deventer, November 7, 2012) makes a stark contrast between the credit assessment tools of the legacy credit rating agencies and a modern default probability service like Kamakura Risk Information Services:
Rating agencies call their legacy credit assessment tool “ratings.” Kamakura Risk Information Services’ credit assessment tool is a set of default probabilities with explicit maturities at all major time horizons. Unlike KRIS default probabilities, which have maturities that run from one month to ten years, ratings are not associated with any explicit term or maturity. There are 10,000 grades in the KRIS default probability service, which run from a default probability of 0.00% (the best credit grade) to 100.00% (the worst grade) in one basis point increments. Ratings, by contrast, have only 21 grades which run from AAA (the best grade) to D (the worst grade). Unlike KRIS default probabilities, there is no default probability explicitly associated with a rating. Unlike KRIS default probabilities, there is also no term structure of ratings by maturity. KRIS default probabilities are available for 31,000 public firms in 37 countries, but legacy credit ratings are available on only about 2,265 public firms. KRIS default probabilities are updated every business day. As of October 29, 2012, the median time since the last change in rating for 2,265 rated companies was 815 days, or 2.23 years (about 2 years and 3 months). See van Deventer (November 13, 2012) for full documentation.
The KRIS default probabilities are based on the best statistical analysis available. “Ratings,” by contrast, are described as “expert judgment” credit assessment tools. Soneji and King (2012) note that 50 years of research in a wide variety of fields shows that statistical methods “regularly” outperform “expert judgment” methods. Hilscher and Wilson (2012) measure the superiority of modern default probabilities over credit ratings. KRIS default probabilities offer full disclosure of a technical guide that includes the mathematical formula, inputs and coefficients used to determine the KRIS default probabilities. There is no such disclosure for ratings. The Kamakura Risk Information Services default probability service is an “investor pays” business model. Legacy credit ratings, however, are an “issuer pays” business model with a huge conflict of interest fully documented by the U.S. Senate (United States Senate, April 13, 2011). Van Deventer (January 31, 2012) documents the fact that legacy rating agencies have regularly omitted their mistakes from “self-assessments” of their own ratings accuracy. With these facts in mind, we can move on to the next 10 steps.
2. Revise all investment policies.
Step 2 is to revise all investment policies and risk policies in terms of default probabilities, not ratings. Here’s another example from the June 18, 2012 announcement by the Office of the Comptroller of the Currency of the revised regulations affecting U.S. banks:
“Under the revised regulations, to determine whether a security is “investment grade,” banks must determine that the probability of default by the obligor is low and the full and timely repayment of principal and interest is expected.”
A modern investment policy that omits reference to ratings is given here:
3. Use default probabilities not ratings, internally.
Many firms maintain their own “internal ratings.” Historically these internal ratings have been “mapped” to external legacy ratings and finally mapped again to default probabilities. Step 3 eliminates external ratings from the mapping process. A follow on to Step 3 simply replaces internal ratings with explicit default probabilities for each reference name for all maturities at which the firm has exposure.
4. Understand macro-factor linkages.
Step 4 is to understand how the business cycle in general and macro-factors in particular (home prices, oil prices, interest rates, foreign exchange rates, stock indices, and so on) drive default probabilities. For a worked example, see van Deventer (September 24, 2009) at this link:
5. Eliminate “transition matrices.”
Step 5 is to replace “transition matrices” based on legacy ratings with simulations where changes in macro-economic factors drive default risk changes. Similarly, we can incorporate these macro factor movements in all risk simulations: counterparty credit risk, market risk, interest rate risk, credit risk, value at risk, capital allocation. The example in Step 4 is a simplified version of how this works. Transition matrices imply that a movement from one credit rating to another is simply a roll of the dice with no causal factor. Nothing could be further from the truth. For that reason, the macro-factor links need to be made explicit, not ignored.
6. Use stress testing.
Step 6 is make sure that the impact of macro factor movements on default probabilities is analyzed and understood for all credit portfolios and investments, even if these portfolios are managed by different groups. Because default probabilities are now the credit measure, a common process can be applied to non-public firms and retail borrowers as well as public firms and sovereigns. Kamakura Risk Information Services provides default probabilities and their macro factor links for a wide array of counterparties above and beyond public firms.
7. Use dollar risk limits and default probability risk limits.
Step 7 is to convert single name exposure limits from a ratings-based limit to a limit based on the counterparty’s default risk and the sensitivity of that default risk, in both probability terms and dollar loss terms, to macro-economic factors. For example, the risk limit for Lehman Brothers would be expressed as a default probability limit and a limit on dollar loss sensitivity to changes in important macro factors (like home prices and commercial real estate). These exposures are available from Kamakura Risk Information Services.
8. Use these limits for portfolio risk levels.
Step 8 is to take the same approach to the total default risk and macro-economic factor risk of the entire investment portfolio. Since the portfolio is the sum of its individual exposures, there is no difference between “top down” and “bottoms up.” In most cases, “top down” is a code word for fudging the numbers. There is no reason to fudge portfolio numbers because there is good data at the transaction level.
9. Conduct on-going “re-education camps” internally.
Some people at every institution will be so used to the 100-year-old ratings concept that they cannot conceive that human progress in statistics has resulted in a better mouse-trap. An on-going series of internal and third party educational seminars will help the open-minded people in this group make the transition.
10. Ask for help.
Archaeologist James M. Adovasio was quoted in the New York Times (March 24, 2011) as saying “The last spear carriers [for that idea] will die without changing their minds.” The same is true for many who cannot understand that the era of ratings is coming to an end. The references below will help the open-minded make the transition smoothly. There are many highly qualified third parties who can act as consultants to an institution which is eager to make this 21st century transition sooner than internal resources alone might permit.
We close by quoting the U.S. Federal Register (June 13, 2012) one more time to give the reader a sense that this transition is something whose time has come:
“Under the proposed amendments to parts 1 and 16, a security would be ‘investment grade’ if the issuer of the security has an adequate capacity to meet financial commitments under the security for the projected life of the asset or exposure. To meet this new standard, national banks must determine that the risk of default by the obligor is low and the full and timely repayment of principal and interest is expected. In the case of a structured security (that is, a security that relies primarily on the cash flows and performance of underlying collateral for repayment, rather than the credit of the issuer), the determination that full and timely repayment of principal and interest is expected may be influenced more by the quality of the underlying collateral, the cash flow rules, and the structure of the security itself than by the condition of the entity that is technically the issuer.”
Donald R. van Deventer
Honolulu, November 27, 2012
© Donald R. van Deventer, 2012. All rights reserved
Ayres, Iain. 2008. Super crunchers: why thinking-by-numbers is the new way to be smart. Bantam.
Dawes, Robyn M., David Faust and Paul E. Meehl. 1989. “Clinical Versus Actuarial Judgment.” Science 243(4899, March):1668–1674.
Grove, William M. 2005. “Clinical Versus Statistical Prediction: The Contribution of Paul E. Meehl.” Journal of Clinical Psychology 61(10):1233–1243.
Hilscher, Jens and Mungo Wilson, “Credit Risk and Credit Ratings,” Brandeis University Working paper, January 2012.
Jarrow, Robert A. “Problems with Using CDS to Infer Default Probabilities,” Journal of Fixed Income, Spring 2012.
Meehl, Paul E. 1954. Clinical Versus Statistical Prediction: A Theoretical Analysis and a
Review of the Evidence. Minneapolis: University of Minnesota Press.
Morera, Osvaldo and Robyn Dawes. 2006. “Clinical and Statistical Prediction After 50 Years: A Dedication to Paul Meehl.” Journal of Behavioral Decision Making 19:409–412.
Soneji, Samir and Gary King, “Statistical Security for Social Security,” Harvard University working paper, January 30, 2011
United States Senate Permanent Subcommittee on Investigations (Carl Levin, Chairman), Wall Street and the Financial Crisis: Anatomy of a Financial Collapse, Majority and Minority Staff Report, April 13, 2011.
van Deventer, Donald R. “Modeling Correlated Default in a Reduced Form Model: A Worked Example,” Kamakura blog, www.kamakuraco.com, September 24, 2009. Redistributed on www.riskcenter.com on September 28, 2009.
van Deventer, Donald R. “The Dangers of Using Rating Agency Default Rates in Credit Risk Management,” Kamakura blog, www.kamakuraco.com, January 31, 2012.
van Deventer, Donald R. “Credit Default Swaps and Deposit Insurance,” Kamakura blog, www.kamakuraco.com, October 5, 2012. Redistributed by Riskcenter.com on October 5, 2012.
van Deventer, Donald R. “Banking Risk Assessment for Corporate Treasurers when FDIC Blanket Deposit Insurance Ends,” Kamakura blog, www.kamakuraco.com, November 7, 2012. Redistributed by Riskcenter.com on November 7, 2012.
van Deventer, Donald R., Kenji Imai and Mark Mesler, Advanced Financial Risk Management, 2nd Edition, John Wiley & sons, forthcoming in 2013.