Fooling Some of the People All of the Time, a Long Short (And Now Complete) Story, Updated With New by David Einhorn (tohfa e dulha read online TXT) 📕

bond prices from Chase Securities for Velocita and Startec and compared them to Allied’s historical carrying values. As Table 7.1 shows, Allied did not take timely markdowns.

Table 7.1 Velocita and Startec Bonds

Allied structured its loan to Startec to be pari passu with Startec’s bonds. That means that even though they owned a separate piece of paper, it had the same status as the bonds. As a result, the bond price was a good indication of value. If anything, under SEC rules, Allied should have carried its investment at a discount to reflect the relative illiquidity of its private investment.

One fund manager sent us a copy of a recent BLX securitization prospectus. It showed that 14.5 percent of BLX’s loans were delinquent as of December 31, 2001. We called the credit rating agencies to learn more about BLX’s securitizations. The agencies told us that they got the data used to rate the securitizations from BancLab, which tracked Small Business Administration loans. We licensed BancLab’s analysis of BLX’s portfolio. It would take them about a month to prepare a report.

CHAPTER 8

The You-Have-Got-to-Be-Kidding-Me Method of Accounting

Just two weeks after the speech, Allied held a second conference call on May 29, 2002. Walton came out firing. “We understand [our business] better than any external individual, and we are going to continue to communicate our understanding to you,” he said. “But let’s be clear: We are not having an academic, intellectual debate. The shorts are people with an agenda and a lot of money and reputation riding on trying to get investors to deconstruct our company and to view the Allied Capital glass as half empty.”

Allied first tackled the Audit Guide problem. The company conceded that its argument about the Audit Guide changes was false and acknowledged that the Audit Guide, in fact, changed in 1997. Now, Sweeney asserted on her honor that Arthur Andersen had not identified any problems with the audit. Andersen had shut down due to Enron-related liability, and Allied had just hired a new auditor. Though we will never get to the bottom of why Andersen changed the language, we know Allied’s first explanation was a lie. In any case, whoever opined that Velocita debt was worth par at the end of 2001 didn’t do much of an audit.

Walton continued the call with a valuation discussion. “The other issue is that shorts seem to think that we should write loans down to zero at the slightest sign of trouble, regardless of whether there has been money lost,” he said. Obviously, we don’t believe that loans should be written to zero at the slightest trouble. Walton is smarter than that. Finally, he repeated what Roll told us, “We write loans down to the amount that we believe we will collect.”

Larry Robbins of Glenview Capital pressed the company about the carrying values of troubled credits. Roll explained, “Sure, Grade 3 assets, really we don’t see any real risk of principal or interest loss in what we call [loans] in ‘monitoring.’ In other words, we’re working with the company, and it can be a Grade 3 for a variety of reasons. They can be a Grade 3 because they’re working on something with their senior lenders, they can be a Grade 3 because we’ve had companies go into Grade 3 because we’re working with them and they’re up for sale and, as a result, [they’re not paying us] for whatever reason until they sell. But they can be a Grade 3 for a variety of different reasons.

“Grade 4 is where we get a lot more concerned, because Grade 4 we think we’ve lost contractual interest due to us. In other words, the deal we went into with respect to what they were supposed to pay us in interest, we don’t think we’re going to get that contractual interest. Now, we don’t think principal’s impaired, but we do think we have contractual interest loss. Grade 5 is where we think we’ve not only lost contractual interest, but we’ve actually lost principal. And this is principal, so it’s cost basis. In other words, it’s money that went to them that we don’t think we’re going to collect back. So if an asset is in that situation, [it] is a Grade 5.”

Robbins persisted. “But your carrying values of Grade 4, for example, there is a likelihood that you would not receive guaranteed interest, but you would still receive contractual principal. Where are you carrying those Class 4 loans?”

“It depends on the asset, but most of those would be carried at original cost and . . .” Roll said.

“So, it’s only a Class 5 loan that would get written down below cost?” Robbins asked.

“Right,” she said.

Here we are back to day one of freshman investing class, where beginning students learn that when an investment reaches a stage where the investor would not repeat the investment, it is no longer worth the original cost. As a result, loans to companies performing below plan, violating covenants, or even not making interest payments—including to the point that Allied realized it would never collect the interest—are not worth cost. Nonetheless, if Allied believed it would eventually recover its principal (Grade 4), it valued the investment at cost. This might be called the “you-have-got-to-be-kidding-me” method of accounting. Walton wanted people to believe that we thought loans needed to be written down to zero at the first sign of trouble. What we believed was that these loans might be worth more than zero, but they certainly were not worth 100 cents on the dollar, as Allied’s financial statement indicated.

Sweeney then introduced a white paper (essentially a research paper) that she and Roll authored describing Allied’s valuation strategy. “We have a consistent process we’ve used to determine fair-value, and that process is clearly outlined in our disclosure document,” Sweeney said. “In addition, if you visit our Web site, you will see that we have written a white paper on fair-value accounting and our interpretation of its application. We wrote this