(Felix Salmon) I feel I've been having this conversation for over a year now: are bonds too cheap, or are stocks too expensive? Yves Smith points out today that the FT is on the case, with John Authers taking the view that the two are bound to converge soon, and John Dizard taking the opposite case that the disparity is likely to stick around for a while.
Part of the problem is that low prices in the fixed-income market are increasingly reflecting illiquidity rather than low fundamental values. Dizard notes that this has implications for regulatory regimes which are "based on the notion that markets are liquid, continuous, and efficient":
Right now, there is a sotto voce argument in the policy world over the social utility of marking assets to market when, really, there is no market. The effect of mark to market, embodied in America in the form of FAS 157, the accounting standard, is to decapitalise the system more rapidly than new capital can be raised.
(Case in point: AIG.)
Another reason for the disconnect is that the markets have inherent differences when it comes to valuing the future. Andew Clavell, as is his wont, puts it in terms of calls and puts:
Credit instruments (corporate bonds) are essentially risk free bonds with embedded short put options on this pretax corporate cashflow. Equities are call options on after tax corporate cashflow, struck at the present value of the credit instruments.
What this means in English is that when the markets start to anticipate a future recession, bonds fall long before stocks do. Only when a recession starts to actually hit corporate profits are equity prices likely to follow suit...
If bond and stock prices are out of whack, then maybe it shouldn't come as any surprise that bond and CDS prices are also trading miles apart from each other. In fact, you have to be careful when you try to take a measure of the difference between bond and stock prices, since many of the most-used indicators of the fixed-income market are actually indices of credit default swaps, not bonds themselves. And so really the disparity between bonds on the one side and stocks on the other is more of a disparity between the CDS market on the one side, stocks on the other, and bonds somewhere in the middle.
There are good technical reasons why the CDS market is trading wide, which are glossed today by Across the Curve. Structured products like the notorious CPDO are being unwound, which means lots of selling in the CDS market, even as long-only fixed-income investors are perfectly happy with their bond portfolios and have no inclination or desire to sell into a falling market.
At this point in this sorry episode there is no arbitrageur ready to take the other side of the trade by buying the cheap CDS while simulatneously selling the underlying bonds. There is no taker of the trade because no one has an inkling of when the forced selling will be over.
There's another reason, too, why it's not quite as simple as that. If you write protection on a corporate credit while shorting the underlying bond, what happens if that bond does end up defaulting? On many of these credits, there are many more CDSs outstanding than there are bonds. And as a result the credit is likely to end up in a Delphi situation, where the bonds rally in default. You can lose on both sides of your bet, having to pay out on your credit default swap while simultaneously being unable to cover your short on the bond. The arbitrage might be obvious, but it's also very dangerous.