A perpetual question in trading is how to tell an arbitrage opportunity from a trap. The former is free money, the best thing since inflatable dolls. But if it turns out to be a trap, it'd be like a doll embedded with an IED rigged with delayed detonation.
If you're bothered by the mental image, consider this real-life example.
The difference between bond yields and treasury yields of the same maturity, taking into consideration minor factors such as coupon frequency etc, is the risk premium of the issuer's credit. And this is what CDS premium is. So the two should be equal. If the CDS premium becomes bigger, you can sell the CDS and short the bond; vice versa if the bond extra yield is bigger. There are a few bones here and there but as a simplified version this is one of the most brain-dead arbs. Just lever it up as much as you can get away with when you see a divergence, take the rest of the day/quarter/year off.
And this was what happened in late '08. First it was 100 bps (1%). You thought "well I'd better be careful, too crazy out there." Then it went to 5%. You thought "wow, this can't last." And it lasted. You were tempted, but decided to err on the side of caution and wait. And it lasted. You said to yourself "well all I need is for it to come back 100 bps, so I'll dip in a toe" and went in with a very conservative leverage of 10. And it went to 10%. You said to your neighbor "now c'mon this is ridiculous, how can it be this stupid?" and levered it up to 100 -- after all, all you needed was for it to be just a tiny bit less stupid, just for a day - JUST ONE LOUSY DAY MAN. And it went to 12%...
This was how some Merrill CDS flow traders lost a few billion dollars and almost blew up the BoA-ML (NYSE:BAC) merger in the process. Turns out there was a very simple and rational reason why CDS were cheap -- nobody could trust an insurance peddler so the insurance premium had to be very low but if you hold the bond, there's a chance you'll get a few coupons and some portion of the principal back.
Right now there's a vaguely similar divergence developing. For the longest time bond yields and equity yields (based on either dividend or earnings), of the same company or group of companies, have been broadly tracking each other, with the former slightly lower because it's less risky and volatile. But since the Eurozone crisis was supposedly defused in June, bond yield has become increasingly lower than equities. Conventional wisdom says this means either (or both) of two things:
- Bonds are too expensive.
- Equities are cheap.
And the best way to play this would be to long equities and short bonds, which makes money as long as the divergence shrinks, regardless of direction of bonds or equities.
But what if it doesn't shrink and instead keeps getting wider and stays wide for, say, 20 years? Can you stay solvent longer than the market stays irrational? Or, as a more rational question, is it really irrational?
I think not.
- With the high uncertainty in the global economic outlook, and whether in the form of sovereign debt, fiscal/monetary policy, financial regulatory regime, consumer spending, productivity growth -- any which way you cut it, risk in corporate earnings and dividends will remain high at least for the foreseeable future.
- The shift away from equities and into bonds is caused by global, structural, long-term demographic changes, i.e., babyboomers all over the western world going into or getting ready for retirement. We're just seeing the very beginning of it.
- As the entire developed world transforms in sync from decades of a spending spree to a savings glut, there will be structural global demand destruction and, as a result, corporate earnings will transition into a new domain with lower expectations for years, i.e., lost decades (see here and here).
More specifically, we're entering the babyboomers' Retirement Saving Phase II. Starting in the mid-90's, babyboomers passed their peak spending years while entering peak productivity/creativity years and Retirement Saving Phase I. They had a lot of time. The choice for investing for retirement was equities. Now their time is running out and the pain of the '08 crisis has been especially hard. Their natural (and rational) choice is to shift into the perceived relative safety and income stream of bonds. If you missed the opportunity to frontrun the babyboomers or at least catch the ride in early to mid 90's, now they have given you a second chance.
Love 'em or hate 'em, nobody wins against babyboomers.
If you think the current bond bull run is a bubble, this bubble may have a long way to go.
Disclosure: Long TIPS and treasuries