Invest Skeptically

Overview–
Primus Guaranty Ltd. (PRS) sells CDS (credit default swap) protection on single name corporates, CDO tranches, and ABS. As selling protection is the same as going long credit risk, the business’ success relies on management’s ability to select names that pay a good spread for their level of risk and to trade credit exposures as conditions change.

Despite the recent turbulence in the credit markets they have maintained their AAA rating (in Primus Financial Products LLC) and grown the portfolio. The notional size (amount of credit protection they’ve sold) of their portfolio at September 30, 2007 was $20.4 billion.

Economics–
One way of measuring the value of the business is to estimate the value of their backlog. Essentially the business model is a net interest margin (NIM) or net spread business. To illustrate, the $20.4 billion portfolio pays a CDS premium (like an insurance premium) of 46 bps annually. This portfolio is financed using debt and preferred shares that cost 14.5 bps to service. This makes the NIM 31.5 bps or $64.3 million annually. The value of the portfolio is the present value (PV) of this stream of cashflows: $226 million, or about $5 per share. As a comparison, the stock closed on Thursday at $6.53.

These figures were calculated using a discount rate of 5.35%, the yield on GE Capital bonds (let’s say this is the minimum cost of capital to Primus), and a tenor (or term) equal to 4 years.

The portfolio’s value represents the amount that a similar credit derivatives product company with the same skills, technology, and back office support (to avoid incurring additional operating costs) would pay to buy the portfolio. In other words, this is an absolute maximum value for the business at a given point in time. If operating costs were included in this analysis, the net spread would go from 31.5 bps to 8.2 bps, or 74% lower.

Growth is a risk–
The above analysis indicates that the stock is at least 23% overvalued. But this analysis looks at the portfolio at a point in time. What if it changes? Management can grow the notional size, extend the tenor, and increase the spread (spread ~ riskiness). Over the last four quarters, the portfolio has been growing at an average rate of 7% per quarter, or 31% annually. Can this continue? I think not:

• The portfolio is already quite large at $20.4 billion
• Continued fears in the credit markets may make it difficult to raise new capital at favourable rates
• The ratio of shareholders equity to portfolio notional has been declining over the last four quarters and now stands at 1.5%. This may make it difficult to grow without raising new equity capital which will be costly and dilutive to existing shareholders. Equity capital is undoubtedly required to maintain their AAA rating.
• While CDS spreads have risen dramatically in the last six months, the ability to trade into new, higher spread, contracts will be limited as the rise in spreads has resulted in negative mark-to-markets on existing positions.
• Defaults will change the capital requirements of the business and hinder growth. The historically low default rates enjoyed in recent years can not continue forever… can it?

I’m including a link to a summary of my PRS backlog analysis. The four scenarios at the top try to incorporate a suddenly larger portfolio size and longer tenor on day 1. The average stock price of the four scenarios is $6.42. Again, this backlog figure is the theoretical value of buying the portfolio and collecting the net spread without incurring any operating costs to manage it.

Markit provides authoritative quotes for standardized structured finance and credit indices used in the credit derivatives markets. As these represent the cost of buying/selling credit default swap (CDS) protection against a basket of credits they can be a useful data point in evaluating the level of fear in the marketplace. As 2007 has taught us, fear can lead to low liquidity, high volatility, and high correlation (i.e. fear in the credit markets can affect your stock portfolio as well).

Here’s a brief–and greatly simplified–summary of some popular indices:

• The CDX.NA family of indices represent North American (mainly U.S.) corporate credits. The “.HY” stands for High Yield, “.IG” stands for Investment Grade, “.HVOL” stands for High Volatility, and “.XO” stands for Crossover (credits between high yield and investment grade). They are quoted in spread, which represents the premium in bps you would have to pay to buy protection on the index. For example, a spread of 76 on the CDX.NA.IG means you pay 0.0076% annually to buy default protection on the 125 names in the index. The index is “rolled” every six months which creates a new series.
• The iTraxx Europe indices are similar to the CDX.NA but for European credits.
• The ABX-HE (CDS on Home Equity ABS) indices are for CDS protection on tranches of subprime mortgage securities. As with the corporate credit indices, these are rolled every six months. Because of the nature of the underlying product, the labelling and quoting is a little bit different. The caption, “ABX-HE-BBB 07-2,” means its the ABX-HE index of BBB subprime RMBS tranches created in the second half (July) of 2007. It is important to remember that this is an index of BBB tranches from different subprime securitizations. So the exposure is to, say, 20 individual tranches but possibly thousands of underlying mortgages. Because of the timing of the roll this index would be mainly comprised of RMBS issued in the first half of 2007. If ABX-HE-BBB 07-2 quotes at a price of 22, this means you would pay 78% (=100%-22%) upfront (i.e. on day 1) and then 500 bps (or 5%) annually to buy CDS protection on this index. Sound expensive?
• The CMBX is similar to the ABX but for CMBS securities. It is quoted in spread.

One important caveat about these “indices” is that they are created based on the relatively high liquidity of the underlying names. They do not necessarily serve as a broad credit market proxy. For example, the CDX.NA.IG only has 125 names in it.

On their website, FocusShares states that their mission includes “captur[ing] investors imagination”. I think their product offering has succeeded:

• FocusShares ISE-CCM Homeland Security Index Fund (MYP)
• FocusShares ISE SINdex Fund (PUF)
• FocusShares ISE-Revere Wal-Mart Supplier Index Fund (WSI)
• FocusShares ISE Homebuilders Index Fund (SAW)

24/7 Wall St thinks they’ve gone too far in becoming so specific and offers some new ETF suggestions, including a porn ETF.

The Star ran an interesting series on regulation and enforcement in the Canadian securities markets. The tone of the work is best summed up by the quote, attributed to Barbara Stymiest, calling Canada’s securities enforcement an “international embarrassment.” Here are the links:

• Why the OSC so rarely gets its man
  “More than 450 employees work at the Ontario Securities Commission. About 40% are paid more than $100,000 a year. Their dismal track record begs the question: What on earth are they doing?”
• Who’s in charge?
  Graphic of the “Canadian enforcement mosaic” (David Wilson).
• Why white-collar crime team fizzled
  “Launched four years ago to clean up markets, police squad is now best known for its failures.”
• OSC chief takes it all in stride
  “Under fire for perceived regulatory failings, David Wilson says he ’sleeps pretty well’.”
• Fraud squad lacking credibility
  “The elite teams became bogged down by bureaucracy, staff turnover, lack of leadership and low morale.”
• Richer rogues not on radar
  “New study on corporate crime suggests system is stacked in favour of influential industry insiders.”

For further depressing reading the ROB Magazine did a lengthy article on the Ian Thow (formerly of Berkshire Investment Group, which is now owned by Manulife Financial) case.

And also: Diane Francis’ take on the RCMP’s poor record on white-collar crime.

FT Alphaville presented some scary charts showing that even older vintage (2004/2005) subprime RMBS have experienced noteworthy foreclosures and delinquencies. A commenter on naked capitalism (who covered the FT Alphaville piece) points out that the high loss severity ratio of 35% stated in the article (and attributed to CreditSights) is influenced by the fact that many older vintage deals have paid down significantly.