One of the more annoying aspects of stock-market commentary is the idea that there is one ideal actual-value price for any given stock, and that the market price is either above that price, in which case the stock is overvalued, or below it, in which case the stock is undervalued. This is the kind of thing that investors do when they look at sum-of-the-parts analyses for companies such as Berkshire Hathaway, and start saying that the stock is trading at the value of its cash and investments, with Warren Buffett and 75 operating busineses thrown in for free.There’s another way of looking at Berkshire Hathaway, however, and that’s through the lens of its credit default swaps, which are now trading at a whopping 535bp — pricing in a probability of default which is much greater than one would ever expect from a triple-A company with barely more debt than cash.

If you take those numbers seriously, then maybe the market isn’t valuing Buffett at zero. Remember that historically Berkshire Hathaway has traded on a price-to-book ratio of about 2: if you gave Buffett $100, and he spent that $100 on shares of American Express or Coca-Cola, then the market would value those shares, as owned by Warren Buffett, at $200.

On the other hand, if you plug in a 60% recovery value, the market is saying that there’s a 13% chance that Berkshire Hathaway is going to default at some point in the next five years. (A handy formula for you: the default probability is the CDS spread divided by 1-R, where R is the recovery value.)* If Berkshire defaults, the equity will be worth zero.

So it seems to me that the market is still valuing Buffett’s investing skills at a premium. If there’s a 13% chance of his company going to zero over the next five years, and presumably a substantial chance on top that the stock will decline markedly but not go all the way to zero, then in order to justify the current share price there has to be a pretty big chance that the stock will double or more over the next five years.

Now it’s entirely possible that the CDS market is overdoing things, and that the chances of BRK going to zero are much less than 13%. But even so, the share price is essentially the result of a probabilistic calculation: look at the range of possible future values of the company, and the probabilities associated with those values, and then discount them at a rate which goes up with the volatility and uncertainty associated with the share price. Given that Berkshire is — like all insurance companies — a leveraged financial institution, and given also that many of its investments (GE, Goldman Sachs, American Express, Wells Fargo, etc) are also leveraged financial institutions, it stands to reason that a sensible investor will apply a pretty high discount rate these days, even if he doesn’t believe the CDS market. What’s more, the lesson of the past 18 months or so is that stock-market investors ignore the CDS market at their peril.

I have no idea how much of Berkshire’s past profitability can be attributed to the fact that it had a triple-A rating, but I know for sure that that rating ain’t worth much any more. And when a company which has grown used to its triple-A then loses it, in name or just in market perception, the consequences are not pretty. So my feeling is that Berkshire has moved from being a safe-and-sound stalwart to being a much riskier stock. It might go up a lot from here — there’s surely as much upside potential as there is downside potential — but it doesn’t really feel like the widows-and-orphans investment that it has been in the past.

*Update: Actually, it’s worse than 13%. I used the formula for a one-year default probability, not a five-year; the five-year default probability is technically up in the 60% range, but at these levels I’m not sure that really means much if anything.