Why there’s no such thing as high returns with low risk
‘When it comes to money, we¹re a fairly unshockable bunch here at MoneyWeek. We’ve grown accustomed to markets’ apparent inability to recognise risk in any way, shape or form, whether in the form of a US property crash or a North Korean nuclear weapon.
But every once in a while, something lands on the desk that surprises even us. And most recently, that thing has been a new piece of derivatives wizardry called the constant proportion debt obligation or CPDO.
Essentially, a CPDO is a piece of financial engineering that carries a top-notch credit rating (AAA or equivalent), but promises to pay its investor a higher return than can be got on a conventional AAA-rated bond.
It does this by using credit default swaps (CDSs), one of the key innovations in credit markets in recent years. A CDS is essentially an insurance policy against a company defaulting on its debt.
In a CDS transaction, one party – the CDS seller – sells this insurance to another – the CDS buyer. The seller receives an up-front premium from the buyer. But in the event that the company covered by the CDS defaults, the seller must pay the buyer the value of the defaulted debt.
Note that it¹s not necessary for the company to owe money to with the CDS buyer – there may be no financial connection between them. While CDSs were originally devised for insurance purposes, today they are widely used to speculate on a company¹s creditworthiness.
So what does a CPDO do? Rather than sell insurance against a single company, it sells it against all the names in an index of companies – either the European iTraxx or US CDX indices of investment-grade debt (ie debt rated BBB or above).
That means that it¹s liable if any of the firms in the indices default, so at first glance a CPDO looks pretty risky. But it doesn¹t have to be. The solution lies in how it sells this insurance. Every six months it closes out the CDSs it has already sold on the indices (effectively it buys them back) and sells new ones.
This works in its favour because the indices are updated every six months.
Because the indices cover investment grade only, any companies that are at any significant risk of default should have been downgraded to junk by the ratings agencies and so will be kicked out of the indices. So the chances of any one of the companies defaulting while the CPDO is exposed to it are theoretically low.
To multiply its returns from CDS selling, the CPDO also uses plentiful amounts of leverage – typically ten to 15 times. And it regularly adjusts its portfolio and its leverage in an effort to hit a predetermined target that will leave it with enough assets that it can be certain of meeting all its future payments to its investors (if it makes losses it can increase its leverage to try to win back the shortfall – rather like a gambler doubling up on each losing bet). Once it hits the target, it can close out all its CDS positions, stick the proceeds in safer assets such as government bonds and make its payments out of those.
Sounds wonderful doesn¹t it? So why does this tool worry us? Well, as the old adage goes, there¹s no such thing as a free lunch. If something is AAA-rated, but pays substantially more than another AAA-rated investment there¹s probably a flaw somewhere.
We see two obvious concerns. Firstly, because CPDOs are so new, we have no real idea now they¹ll behave when things get tough. (The same is true of many of the other recent innovations in derivatives.)
Modelling by the ratings agencies suggests that the ones launched so far are pretty durable – hence the agencies¹ willingness to give them AAA-ratings.
But it still looks as if a wave of defaults or a series of sharp rises in CDS premiums could put them under stress.
More importantly, if liquidity in the not yet crisis-tested CDS market dried up at the same time, things could get very tricky. CPDOs might find themselves unable to close out their positions at the six-month rollover point.
Most people clearly think these risks are unlikely. We think that they¹re looking at the benign conditions we¹ve experienced for the last few years and extrapolating them off to infinity. Unless the cycle has been abolished, the credit market will eventually turn, defaults will increase and CDS premiums will go higher. And history suggests that when that happens, it could do so a lot faster than people expect.
But if a few CPDOs default, that¹s the investors¹ problem. The bigger issue is what these instruments are doing to a credit market already infested with complacency and plagued by low yield.
Already-low CDS premiums have been sliding further in recent weeks, with the likely culprit being the new CPDOs flooding the market with offers as they set up their portfolio. This is troubling; it seems highly probable that sellers are not being compensated for the risks they¹re assuming. Past experience suggests that some of them will take very big losses and that the knock-on effects from those won¹t stop with CDSs and CPDOs. In times of crisis, markets often panic together.
But unless an upset happens very soon, we doubt that this is the end of the story. Lower CDS premiums also mean that the returns that new AAA-rated CPDOs can get may be rather less than the current crop, since their premium income will be lower. Indeed, we¹re already seeing signs of that. The first handful of CPDOs paid LIBOR + 2%, but the latest prospectus we¹ve seen promised just LIBOR + 1%.
With that in mind, we¹ll suspect that very soon someone will try to build a CPDO that plays with junk-grade debt, but technically has a low-investment-grade rating, just to eke out a bit more yield. Of course, that will depress premiums and yields further, and leave markets scrabbling for the next miracle innovation to compensate.
And so the merry-go-round continues. Low yields beget more risk, begetting still lower yields and yet more risk. It’s a wild ride – but someday the brakes will go on.