Negative Basis Trades
January 28th, 2009 by Nick SaintI’ve tried to be good, dear reader. This economic meltdown is serious business, and I feel we all have an obligation to understand it before it kills us all. I have lots of canned goods and zero guns, so I will no doubt die in the first wave. Accordingly, I’ve been working overtime to learn about things like synthetic collateralized debt obligations, super-senior tranches, and contango (the forbidden oil market condition!). But sometimes it makes my head hurt. And what’s the point of having an in-house expert on all things financial if my head is going to hurt? So, when Felix Salmon’s explanation of negative basis trades sent me in search of Advil, I called on our own T. Ballgame. First a little background:
Merrill Lynch lost a disgusting amount of money in the final days of 2008. This raised some eyebrows, because of the sheer size of the losses and the fact that they didn’t have to disclose them until after they were eaten up by Bank of America. The thinking was that they had been hiding huge liabilities off their books until they had nothing to lose. But it turns out there is a perfectly reasonable (ok, maybe that’s going too far) explanation for their loss of $15 billion in one month - negative basis trades. What are those? Take it away, Ballgame (with my notes in bold)*:
Negative Basis Trade has four legs to it so an investor’s position would look like the following:
1. Long a Corporate Bond - asset sitting on your books
2. Receiving coupons from said Corporate Bond - cash inflows quarterly from the bond you own
3. Long a CDS - asset sitting on your books, this is insurance against the default of said corporate bond (it should be equal in amount to the bond you are long as well as have the same maturity)
4. Quarterly payments for CDS - cash outflows, this is the cost of purchasing the CDS (i.e., protection of default against the bond you bought)So what you have on your books is two assets and two cash flows. What happens is the two assets (the corporate bond and CDS) cancel each other out so to speak, leaving you with just the two cash flows. If the coupons you receive on the bond are greater than the quarterly payments for the CDS your purchased, then you make money just sitting on the position…this is apparently called Negative Basis, hence the name of the trade. If the coupons > quarterly payments, then you can earn money risk free. Why risk free?
Let’s say it’s a $100 million corporate bond that you bought and that you also purchased $100 million CDS contract on that same bond. You are now perfectly and fully hedged in theory and hence own a risk free position. Assuming you can find bonds out there for sale with CDS of the same face value and maturity, all you then have to do is see if the coupon payments are greater than the quarterly CDS payments and it’s gravy time. Riskless returns like this rarely exist (and if you are an Efficient Markets Doctrine extremist you believe they cannot exist) so when they do people often pile into them big time. Which happened here.
Two quick points: The thing to note about efficient markets extremists is that they’re very obviously wrong. Also, risk-free investments are just the tip of the iceberg of things that these people don’t believe are possible. If they’re right, it’s impossible to find an investment that has a better risk-return ratio than any other, ever. The dartboard will always be just as good as picking stocks. Being a hard-liner on efficient markets is like marketing bacon as a cure for cancer: you have something truly terrific on your hands, but there are some things it can‘t do.
Secondly, even without the complications our expert is about to explain. This wouldn’t really be risk free. If the borrower defaults and the CDS issuer goes under, you’re out $100 million. This is very unlikely (though much less so if the CDS-issuer sold lots and lots of insurance on the debt in question). But it sure sounds like a risk to me (and to our expert, whom I ran this by). So that free lunch remains elusive.
All was well and good for awhile it seems until the financial system started melting down due to all the horrendous loans people made for the past 4 years and many banks scrambled to reduce risk and raise cash. Savvy readers will note that the CDS you purchased needs to be purchased from someone who has a high probability of actually existing in real life. So who were these sellers? They tended to be large banks or bank like things such as that small AIG group melted down the world’s greatest insurance franchise all by itself. Well, these banks did weird things to people in these Negative Basis trades like asking them post a bunch of collateral (very non-standard behavior in the case of CDS contracts, usually it is the selling party who should be posting collateral, not the buying party). Posting collateral has a very real cost of requiring you to actually have cash on hand as well as an opportunity cost (the return you can no longer earn by investing that cash) that changed the rules of the trade right in the middle of it. Here you were sitting fat and happy on your riskless trade, taking in coupon payments greater than your quarterly CDS payments, when all of sudden you were hit with an additional cost you didn’t see coming.
Now, instead of taking in a steady stream of profits, you are losing them, so you have to unwind your position. And it’s not just you, but tons of people. So now you’re all selling the same bond and there is very few buyers so the price of the bond plummets in the market. Now you are staring at a huge mark to market or realized loss as well. This was the cause of all the ugly numbers.
The point is, it’s fundamentally not a horrible trade. It’s just that an unexpected cost hit when banks made folks in the trade put up additional collateral and such. Felix is saying, why not use the gov’t to step in and make these trades. Buy the bonds from the distressed sellers and put the trade back on. After all, no one is going to call the Treasury and ask them to post collateral. The Negative Basis trade would still be profitable for the U.S. of A.
Salmon would have the government buy the CDSs in question at the new CDS exchange it wants to establish. Ballgame is skeptical, since the exchange in question doesn’t yet exist, which strikes me as very sensible indeed:
Also, I have no idea how likely it is for the gov’t to get a CDS exchange up and running soon but this is hugely crucial to reducing the counterparty risk in the CDS. Having never worked on a trading desk or a hedge fund that deals with these things I can only assume it’s fairly complex to research and execute trades like these. Last time I check gov’t work wasn’t all that lucrative and I am not sure the kind of people with the skills to pull this off can be found or would ever end up at the Treasury.
But again, asking the gov’t to come in and do stuff like this is, as Warren Buffett might say, outside it’s circle of competence. Like much of the ad hoc dealings with the financial crisis thus far, this only further asks Washington to step in and do work that it does not have experience doing. Who’s to say they wouldn’t mess it up as well? Assuage my fears on that count and I’m with Felix.
* excerpted selectively by me, no doubt to the detriment of us all, so the blame for any explanatory gaps lies here
