on Primus Guaranty: Wacky and Uninformed

Feb.21.08 | About: Primus Guaranty, (PRSG)

Until earlier this month, I’d never heard of columnist Nicholas Yulico of And for good reason, I believe. If the February 13 article he posted on Primus Guaranty Ltd. (PRS) (a company near and dear to my heart) is any indication, Yulico’s understanding of the companies he writes about is tenuous at best, and the investment conclusions he arrives at are ill-informed to the point of being preposterous. Who gave this guy a laptop?

Primus, as you may already know, writes credit default swaps for a living. It usually holds its positions to maturity, and rarely if ever hedges them. The company’s main operating unit is Primus Financial Products, rated AAA by Moody’s and S&P. I have gone out of my way over the past several years to understand Primus and its business very, very well. The investment partnership I run owns 10% of the company.’s Yulico, by contrast, seems not to have a clue. If you doubt it, take a look at some of the things he had to say about the company, along with my comments:

Yulico: “Over the past two years, derivatives veteran Tom Jasper bet the farm going long credit risk.”

TKB: Tom Jasper is the CEO of Primus; I’ve met many, many CEOs over the past 30 or so years, and would rank Tom Jasper among the very top, especially as regards his integrity, business prudence, and conservatism.

In particular, Jasper is a legend in the credit-risk management industry. He started up and ran Salomon Brothers’ interest-rate swaps desk in 1982, and later co-founded and helped run the International Swaps and Derivatives Association, the industry’s self-regulatory entity, in 1985. Jasper was elected to Risk magazine’s Risk Hall of Fame in 2002.

All of which is to say that the guy is no gunslinger. For Yulico to say the company has “bet the farm going long credit risk” is true only in the sense that, say, Allstate has bet the farm on property risk or that Toyota has bet the farm on auto demand. Yes, I suppose it’s true in some insanely trivial way. What Yulico ignores is that Jasper & Co.’s long experience in the credit-underwriting business make Primus (I believe) a superior underwriter of corporate credit risk. The company uses its expertise--and makes its shareholders a lot of money--by expanding its CDS book when at times when it believes it will be more than compensated for taking on additional risk. As you’ll see in a minute, one of those times is now.

As of year end, Primus’ portfolio of CDS was 79% on single-name corporate and sovereign issuers, and 21% in tranches consisting of multiple investment-grade issuers. Ninety-seven percent of its book is investment grade. Notably, Primus only suffers a true economic loss if there is a corporate default or if it elects to close out the contract early at a loss. Mere downgrades don’t count: a reference entity’s credit risk can deteriorate and Primus will likely still make a fortune on the contracts it has written on it.

Yulico: “Now, his firm’s $23 billion credit portfolio is losing hundreds of millions of dollars per quarter as corporate credit fears multiply.”

TKB: Irrelevant. Primus is required by GAAP to mark its CDS portfolio to market every quarter, even though the company intends to hold its positions to their full, five-year maturity. Over time, the marks should net out to zero—and thus be meaningless over the long term. In the meantime, remember that the company is AAA-rated and so does not have to post additional collateral despite having to take any negative marks. This is a case when GAAP does not reflect the true economics. Yulico’s comment is utterly beside the point.

As evidence of this, Primus’s GAAP book value at year was a negative $2.08 per share but its economic book value (which excludes the effect of quarterly marks) was $9.10 per share. After the company’s earnings release on February 13, S&P affirmed the Primus Financial Products’ AAA rating and noted,

The reported fluctuations in the market value of PFP’s (Primus Financial Products) credit swap portfolio didn’t have a direct effect on our view of PFP’s issuer credit rating due to the “continuation” structure that allows PFP to hold these positions through maturity rather than forcing it to liquidate them over a short time period.

Even Standard & Poor’s, which seems jumpy about nearly everything lately, understands that Primus’s quarterly marks don’t matter. Yet, oddly, Yulico makes a big deal about them. Either he doesn’t know the difference between GAAP and economic accounting (in which case he shouldn’t be writing for or any other financial publication) or he does know it and is being deliberately obfuscatory. I for one would like to know which, and look forward to his response.

Yulico: “Primus’ main problem is that most of its credit default swaps were sold at a time when credit risk was low.”

TKB: Actually, corporate credit risk is still low. It’s the prices of CDSs that have soared, thanks to the crisis of confidence has roiled the credit markets in recent months. Primus has taken aggressive advantage of this broad mis-pricing. That’s a good thing, not a bad thing. Longer term, the company has a demonstrated record of selling fewer contracts when spreads are tight and then selling more when spreads widen. The table below shows the new business written by quarter, compared with the CDX, an index of swap spreads on investment grade companies. Note that as spreads have widened, the company has written a ton of new business. Primus sells CDSs when spreads meet its hurdle rates of return, but has historically maintained dry powder so it can write even more business sell more when spreads widen to extraordinary levels, which is where they are now. Makes sense to me!

click to enlarge

Yulico: “‘If Primus were a hedge fund, it would be gone,’ says one industry observer.”

TKB: What hatchet job would be complete without a scary quote from an anonymous source? There’s just one problem with the blind quote above: Primus is not a hedge fund. It is a Credit Derivatives Product Company (CDPC) which, as a AAA-rated entity, does not have to post additional collateral on mark-to-market losses as a hedge fund would. That completely changes the economics of what the company does. It’s yet another irrelevant point from Yulico.

Yulico: “Billionaire super-investor Warren Buffett had been vocal about the dangers inherent to the derivatives market – which includes CDOs.”

TKB: Ah, yes, Warren Buffett of Berkshire Hathaway. This is the same Berkshire Hathaway, by the way, that disclosed in its latest 10-Q that its “maximum exposure of other derivatives contracts” totaled $35 billion up from $12 billion at year-end 2006.

These other derivative contracts “primarily pertain to credit default risk of other entities as well as equity price risk associated with major worldwide equity indices.”

So, the Warren Buffett who has called derivatives “financial weapons of mass destruction” owns a derivative portfolio that’s 50% larger than Primus’s and is awfully similar to it. Clearly, Buffett doesn’t think all derivatives are accidents waiting to happen. Neither, suffice it to say, should Yulico.

Yulico: “. . . widening spreads mean more writedowns are coming at Primus as it reflects the worsening market value of the swaps.”

TKB: Here again, Yulico confuses temporary mark-to-market losses from true economic losses. As I say, the marks are beside the point. If anything, they are the sign of an incredible opportunity for Primus. Let’s look at some numbers, and you’ll see what I mean. At the start of 2006, the price of a CDS on General Electric’s debt was 12 basis points annually. By mid-2007, the price had risen by roughly 50%, to17 bp. Assuming Primus sold CDS on G.E. debt at mid-year, 2007, today it would have to take a considerable mark-to-market loss on those G.E. swaps, since recent prices on CDSs on GE debt have ballooned to around . . . 120 basis points. Those mid-2007 contracts would still have 4 ½ years remaining; Primus would only suffer an actual cash loss if G.E., rated AAA, were to default. Is G.E. ten times more likely to default on its debt now than it was 18 months ago, or is the runup in G.E. CDSs a reflection of broader turmoil in the financial markets? I say the latter. That’s a good thing for Primus, not a bad thing, since it puts the company in position to write some ridiculously profitable new business. Any intervening negative marks Primus might have to take until its contracts expire are basically meaningless.

Or take Berkshire Hathaway itself. CDS spreads on Berkshire debt were 8 basis points at the beginning of 2006 and 8 basis points at mid-year 2007. Hypothetically, if Primus sold CDSs on Berkshire debt in mid-2007, the company would also suffer a large mark-to-market loss, since recent prices on CDSs on Berkshire debt are around 91 basis points now. Again, remember that Primus only suffers a loss on a CDS if the company defaults within the five-year time period of the contract. Do you think the odds of G.E. and Berkshire defaulting on its corporate debt has risen tenfold since the beginning of 2007? I don’t!

In fact, I like the business of receiving $91,000 in premiums annually for every $1 million in principal to insure that Berkshire Hathaway will not default on its debt over the next five years. Which is why we own 10% of Primus.

If I like the business so much, why don’t I go into it myself? Because we are not a AAA-rated Credit Products Derivative Company; we are an unrated hedge fund. That means the additional collateral requirements we’d be required to put up whenever we marked our positions to market would render the whole enterprise uneconomic. Primus, as noted, is not a hedge fund; it can take advantage of the craziness in the market, and is doing so. Too bad market commentators like Nicholas Yulico are so confused!

Don’t Forget the Internet Bubble

This is not the first time a financial market has gone insane, don’t forget. Remember the Internet search software company, Inktomi? It went public in the late 1990s and rose by 1,900% to its peak in March, 2000. How many pundits were wildly bullish on Inktomi by then? Even some who wrote for, I suspect. At its peak, Inktomi traded at 250 times revenues, and hardly anyone batted an eye.

In the end, though, the stock fell 99%, even including the premium Yahoo! paid when it acquired Inktomi in 2002. Inktomi was of course just one of scores of Internet companies that soared and then crashed during the Internet bubble. Such bubbles happen all the time.

My point: great discrepancies between value and price take place in the stock market not just on the upside, but also on the downside. I think many great discrepancies exist in the financial services sector today because of fear, overgeneralizations, misunderstanding, and simple trend following. Columnists such as Nicholas Yulico help create those discrepancies. You, as an investor, can profit when such poor analysis create these opportunities.

Disclosure: We have a position in Primus


Tom Brown is head of