Today's news that Countrywide has been named in one more lawsuit makes an analysis of how the firm communicated its own risk to investors an important issue. As of today, legal websites report that former Countrywide CEO Angelo Mozilo is named in 78 lawsuits and the firm's directors are named in 28-30 federal lawsuits each. We evaluate Countrywide's public statements about its own risk management by looking at its answers to the 4 question pass/fail test on risk management posted on the Kamakura blog on April 27, 2009. We suggested that any institution that could not answer these four questions would fail the test:
Question 1: What happens to the market capitalization and net income of the firm if any of these risk factors change: home prices, foreign exchange rates, commercial real estate prices, stock index levels, interest rates, commodity prices?
Question 2: Using an insider's knowledge of the assets and liabilities of the firm, both "on balance sheet" and "off balance sheet," what is the best estimate, monthly for the next ten years, of the probability that the firm will fail in each of these 120 monthly periods?
Question 3: Using only information available to an outsider, what is the best estimate of the probability of the failure of the firm in both the short run and the long run?
Question 4: If the firm is able to answer Questions 1, 2, and 3, what hedging position is necessary to insure that the macro factor sensitivity of the firm and default probability of the firm reach the target levels set by the Board of Directors?
In the rest of this post, we use Countrywide's own financial reports to show that management did not ask themselves at least 3 of these four questions. Even if they had asked questions 1, 2 and 4, Countrywide was faced with an archaic risk systems infrastructure that could not have produced an answer to the questions if they'd been posed by the Board. The firm's failure was just a matter of time, because management "confused a tail-wind with horsepower" in the words of Z-man, a sophisticated and modest risk expert.
Countrywide Risk Expertise
Countrywide, in theory at least, had a lot of people at the firm who knew a lot about risk management. One prominent member of the Board had a Ph.D. in economics, more than 30 years experience revolving around risk, and a long a career at the Federal Reserve. Another had a Ph.D. in economics but also served as CEO of a major savings and loan and Vice Chairman of two major commercial banks where he dealt often with risk issues. The head interest rate risk analyst was another well-regarded MBA with more than 20 years of risk expertise and was the author of many books on interest rate risk management. The firm relied on a commonly used legacy interest rate risk system also used by IndyMac, New Century, Lehman Brothers, Washington Mutual, and FNMA. Why, then, did the company fail? The firm's own 10-k and annual reports tell the story.
Countrywide Risk Management Comments from its 2006 Annual Report
We start with some quotes from the firm's 2006 annual report, which was published just as the credit crisis was beginning to be obvious to all.
Page 1: "...We aim to continue building and refining a business model that can deliver stable earnings growth and shareholder value through a variety of business cycles."
The CEO's letter recognized that the economy was turning down:
Page 5: "The past year was characterized by slowing home sales, falling industry production volumes, and increasing pressures on the credit quality of home loans."
The CEO's letter confirms that senior management and the Board were oblivious to the home price risk the firm was facing in this quote:
Page 5: "During the year, we launched the first stock buyback in Countrywide's history, repurchasing 38.6 million of our common shares." In the 2007 10-k, the company reveals that it spent $2.5 billion repurchasing equity capital less than a year before it had to access $11.5 billion in credit lines from Citibank and Bank of America (drawn down August 15, 2007). $1.5 billion had been spent to repurchase common stock by December 31, 2006 (2006 10-k, page 46).
The CEO's letter goes on to emphasize the importance of risk management to the company:
Page 5: "One of the most important objectives at Countrywide is to produce a stable earnings stream throughout different business cycles. Over the years, we have refined our business model to the extent that it has enabled us to achieve strong earnings growth from one cycle to the next. Coming out of each cycle, Countrywide was substantially stronger than before the cycle began."
The CEO's letter goes on to describe "key strategies," including this phrase:
Page 6: "Optimizing the use of our capital and building a world-class risk management infrastructure."
At the same time the firm was using cash to repurchase its own common stock, the CEO's letter notes
Page 7: "Various factors could adversely impact our industry, including potential legal and regulatory actions, diminished marketability of certain loan types and their associated credit risk, and further housing market declines."
This sentence, as it turns out, was literally the only mention of home price risk in the text written specifically for the 2006 annual report. The 10-k, which is incorporated in the annual report, has a precious few other mentions of it as well.
Home Price Risk Comments in Countrywide's 2006 10-k Filing
The company's assessment of home price risk in its 10-k filing, dated February 28, 2007, were not much more extensive even though page 9 of the 10-k shows that 19.03% of the Company's non-prime loans were already delinquent as of December 31, 2006:
Page 36: "We may experience credit losses due to downward trends in the economy and in the real estate market."
Page 37: "While we estimate and provide for credit losses accruing to our credit-subordinated securities and for the corporate guarantees and representations and warranties, worsening economic and real estate market conditions could negatively impact the value of the credit subordinated securities we retain and increase our liabilities under our corporate guarantees and representations and warranties. Likewise, worsening economic conditions increase the risk that our borrowers will not be able to repay our portfolio loans, and worsening real estate market conditions increase the risk that the value of the properties securing our loans will be insufficient to repay amounts owing t us in the event our borrowers default on the loans."
Page 120: "For credit exposures, we maintain a credit risk profile that has been developed to help ensure that instantaneous erosions of housing equity, or a similar shock scenario, does not lead to increases in credit capital beyond established risk tolerances. These equity erosion tests provide valuable information about a key source of risk exposure for the Company."
The page 120 comment notwithstanding, there was no such disclosure in either the firm's 10-k or annual report.
Page 125: "Increasing housing values affect us in several ways. Rising housing values point to healthy demand for purchase-money mortgage financing and increased average loan balances and a reduction in the risk of loss on sale of foreclosed real estate in the event a loan defaults. However, as housing values appreciate, prepayments of existing mortgages tend to increase as mortgagors look to monetize the additional equity in their homes. Over the last several years, the housing price index has significantly outpaced the consumer price index and growth in personal income. Consequently, we expect housing values to increase at a slower rate in the coming years than in the past several years. Although it is likely that certain markets will experience housing price depreciation, we believe that price depreciation will not persist nationwide for an extended period of time. Over the long term, we expect that housing appreciation will be positively correlated with both consumer price inflation and growth in personal income."
Along with a short comment on page 126, this is the complete disclosure of home price risk by the Company in the 151 pages of the text of the 10-k or the 95 pages of exhibits for 2006. There were no tables or stress tests of home prices, although there were a few lines of calculations for small increases in mortgage default rates.
Lack of Diligence in Credit Review
Exhibit F-13 contains this bombshell:
"The Company's loan portfolio is comprised primarily of large groups of homogeneous loans made to consumers that are secured by residential real estate. The company does not evaluate individual homogenous loans for impairment."
There it is in black and white--one of the largest mortgage lenders in the United States, one servicing 7.4 million loans, doesn't take the time to calculate the current loan to value ratio on its holdings or the loans it services. This isn't a big volume of data--one of the largest banks in China, a Kamakura Risk Manager client, processes 92 million transactions PER DAY using the Kamakura system. Countrywide, however, couldn't be bothered to do the most fundamental risk management calculation on a portfolio only 1/13th as large as the Kamakura client's portfolio. The comment above is the only useful piece of information about home price risk in the 95 pages of exhibits to the 2006 10-k.
Interest Rate Risk Myopia
The Company's disclosure in the interest rate risk area is very typical. It has much more analysis than the near-zero analysis of home price risk, including the standard reports on parallel shifts in the yield curve that banks have been doing for 30 years. If Countrywide had simply done a stress test with respect to home prices, it might still be an independent company today.
Conclusion: Management Didn't Ask the Right Questions and Couldn't Have Gotten The Right Answers Anyway
The conclusions from reading the annual report and 10-k for 2006 are inescapable. One of the largest home mortgage lenders in the world didn't think it was important enough to report on stress tests with respect to home prices in either document. These same stress tests would be made mandatory by U.S. bank regulators for the top 19 financial institutions only 24 months after the Countrywide 10-k was signed. In addition to this huge error, the Company states that home price declines are not likely and admits that it does not take the time to compare collateral balances and home prices on a loan by loan basis.
Moreover, the Company's legacy interest rate risk infrastructure relies on a system incapable of making these calculations. For more details on that issue, please contact us at info@kamakuraco.com. Even if management had asked the right question, it did not have a system in place that could have answered the question and avoided failure of the firm.
As we document in Kamakura Risk Manager--In Depth, Countrywide lost 94% of its commercial paper supply between June 30 and September 30, 2007. By December 31, 2007, its 2007 10-k confirms that it was unable to issue commercial paper at any price. The rescue by Bank of America unfolded soon after.
Epilogue
Like any good TV drama, lots of entertaining things happened at Countrywide after its demise was inevitable.
I asked one director, "How did this come about? You guys are smart guys-what happened?" He said, "When the Company started to originate loans for its own account, they had no credit culture. I thought about [asking for outside help] but they had to learn to walk before they could run." Unfortunately, "they" were in the middle of a long train tunnel, the train was coming down the tracks, and walking was not an option. "Did you ever ask about exposure to home prices while you were on the Board?" I asked. "If I had, the director said, "Angelo would have fired me from the Board." Here's what happened...
- The CEO of Countrywide is now named in 78 federal lawsuits.
- The directors of Countrywide are now named in about 30 lawsuits each (it varies by individual)
- The Chief Risk Officer jumped to a new job in 2007. Unfortunately, it was at Washington Mutual.
- The Head of Enterprise Risk Management is now consulting in risk management and presumably putting his graduate degree in divinity studies to good use.
- The Head of Interest Rate Risk Management is now employed at a financial institution 1/50 of Countrywide's size at its peak.
For any risk manager who wants to avoid a similar fate, focusing on answering the 4 key questions in our pass-fail test on risk management would allow the risk expert to save his firm.
Donald R. van Deventer
Kamakura Corporation
Honolulu, May 18, 2009