This post appeared Tuesday on RealMoney. Click here for a free trial, and enjoy incisive commentary all day, every day.
, we looked at how the corporate debt markets operate, and we began to delve into the nitty-gritty of credit default swaps, or CDS. Once we understand the quirks and limitations of the corporate debt markets, we can gain a better perspective on CDS. These instruments don't cure all the bond market's peculiarities, and they have received plenty of criticism. But does this mean they should be banished?
One of the first comments about the CDS market is the increased volatility that seems to follow these instruments. From my perspective, this is probably not so much "increased" volatility but maybe "true" volatility unleashed. Historically, the corporate bond and bank-loan markets had a long-only mentality. (Remember my comment in "
" about the bond manager who was being heroic by only being 70% of the index weight in WorldCom?) Because of the issues mentioned in Part I surrounding the fragmentation of the debt markets (and the resultant effect on the repo market), there was never really a consistent method of going the other way in credit.
The corporate debt market was one of the few markets that did not have this contribution to price discovery. (The foreign-exchange market, the interest rate market, the equities market and commodities markets all have long had two-way price action, with an ability to be short.) So it is only natural that some volatility would be unleashed once a vehicle was established to do so in the corporate credit market. If I could wave a magic wand and somehow get a fully developed repo market for any corporate bond out there, I believe we'd be grappling with the same issue of volatility being released.
There is talk of allowing CDS usage only for "hedging" purposes, but that argument has flaws. If I can use these instruments only to hedge a position, that takes us right back to a long-only market. The only takers of risk in synthetic form would be long investors, and depth would suffer. Arguably, this could lead to even more price "undiscovery." There would be no feedback loop in the markets process.
Imagine if Jim Chanos could not have established his view on Enron equity in a position and only could have just "not owned" Enron. Without him giving price feedback to the market, the stock could have gone higher, thus creating greater losses for those involved on the long side. If short-sellers are wrong, they create better opportunities for the longs. By having both longs and shorts, we achieve price discovery from all market participants.
There is a lot of talk about "naked shorting," but that does not apply here. CDS contracts have physical settlement. Therefore, there is a limitation on deliverables that should provide a boundary for hedgers and speculators -- they could be adversely affected by squeezes, should a credit event occur. This same thing can happen when there is more open interest in an oil futures contract for a given month than there is available supply. Physical settlement provides a degree of honesty to the process -- we are not manufacturing more debt for a company; the derivative is based on what is already outstanding.
However, the futures analogy highlights a shortcoming of the current system. With a future, we can actually observe the open interest, but since CDS are bilateral contracts, we cannot observe positioning.
This strongly suggests that CDS need to be traded on an exchange. And unlike the fragmented corporate bond market, CDS lends itself well to exchange trading, because the multiple deliverables lead to unified pricing. With an exchange, we could all observe pricing, open interest and net positioning and get valuable feedback, which should lead to greater price discovery for everyone. Right now, because that is all largely secret, traders in the product are skittish and perhaps afraid to go against the tide, and that leads to more momentum continuation. When people cannot see what is going on, they are more reluctant to buck a trend... but if they had the information available to them regarding short interest, open interest, etc., they could make more informed decisions.
An exchange also solves the margin issue. Currently, because these are bilateral contracts, margin is determined individually between counterparties. This is a major problem, because firms compete not just on pricing of the swap but also on the pricing of the margin; if I don't have the franchise information to give me the best bid or offer, I can still compete on the basis of my margin terms. Having margin completely unregulated could lead to problems in terms of speculative activity, stability of brokers and market integrity.
(As an aside, from my experience, the firms with the best handle and discipline on margin issues -- setting, collecting, processing -- are all still standing. That should tell you something.) I believe a lot of the AIG issue surrounding the company's CDS/synthetic CDO exposure was from banks and dealers who did not adequately margin AIG, relying more on their credit rating than on actual collateral. This allowed AIG to take on bigger positions than if it had been adequately margined.
An exchange would have an added benefit of a dramatic reduction in counterparty credit risk (which has led to greater hedging activity, thus greater pressure on credit spreads). In September and October, I was getting up in arms every time the media would talk about the "information content" provided by broker CDS spreads. There was a flaw in looking at spread reactions simply as the market pricing increased probability of default, because, again, the fixed-income markets are principal-based businesses.
Dealers (who are the market makers of CDS) have exposure all over the place to one another -- they may have a forward mortgage or currency trade on with one another. These trades take up credit lines. These trades may also have their own inherent benefits, in terms of risk reduction or P&L potential. So a $1 billion forward yen trade may take up $10 million of credit line exposure, but it may have significant risk or profit benefits. Contrast this with, say, selling $10 million of protection on that counterparty, or buying $10 million of that counterparty's bonds, where the credit line is used, but the P&L potential is lower and there is no risk mitigation -- that trade just chews up line.
A bank that is managed holistically will manage all of its credit exposure, usually leaving the credit trader with limited ability to take on exposure to other dealers. Thus, if someone asks for a bid on another dealer's bonds, or asks to be offered CDS protection, the market is set up so that there is limited availability (and this is compounded by fewer market makers still standing). If a dealer gets lifted on protection of another dealer, the first instinct is to get flat -- the new bid was the old offer. I would venture to guess that there were people who leaned on this limited liquidity to cause a bear raid.
A move to an exchange would alleviate a lot of this. We would no longer be reliant on the dealer community to be the sole provider of liquidity in this area. The dealers could manage their credit lines more efficiently, freeing up capacity for other business. And the "Lehman" or "AIG" scenario interdependencies could be diminished.
Another issue with bilateral margining is that it addresses counterparty credit risk, but not necessarily market integrity issues. Consider the risk profile of buying and selling protection. If I buy protection from someone, my risk is potentially for the notional par value of the swap--which is much larger than if I sell protection, where my risk is for the premium payment owed. Plus, if someone stops paying premium, then I stop protecting them. From a counterparty perspective, my risk is lower, so I charge less margin. However, from a market integrity standpoint, this makes the market susceptible to bear raids.
George Soros, in his recent piece in the
, brought up a valid point about asymmetry of payout in bonds; bonds either mature at par or "mature" at recovery in the event of default. This may also lead to higher proclivity to be short rather than be long. This could be addressed in margin requirements. The ability to control margin requirements is a powerful check and balance.
If the margin process is brought into an exchange, then it can also be adjusted to control excessive speculation -- on the long or the short side -- just as we can do with futures or options or common stock. We will have position reports and could determine short or spec interest, and tweak accordingly. There could be position limits put in place, and you could even put in the equivalent of an uptick rule on short credit positions.
Moving CDS to an exchange would not necessarily mean the end of profit potential for the brokers in derivatives, either. There will always be a spot for bilateral trading to allow for customization. After all, the equity markets have both over-the-counter and exchange-driven derivative markets, and both thrive. The interplay between the exchange and the OTC market may very well enhance the profit potential, because the vanilla does not tie up valuable resources.
And the margin regime of the exchange could provide guidance for margining OTC trades that a regulator could observe. If we could move the vast majority of vanilla CDS to an exchange, put under the purview of a regulatory body, my guess is that a lot of the issues we are discussing can be alleviated, and the positives of CDS can still be available to complete the fragmented credit market.
With that, please allow me to address the first question I am sure I will get:
Hold on a minute...aren't you the guy who is on the soapbox about these short-sided ETFs? This seems a little inconsistent...
Actually, I don't think this is inconsistent at all. With equities, we have proven market mechanisms for establishing leverage and shorting stock via margin accounts and hypothecation agreements. We have not really had proven mechanisms for the credit markets, for the reasons outlined in
. I believe short-selling is a valuable contributor to price discovery -- I have never been "anti-short." Furthermore, you will note that I am advocating a means of establishing consistent margining so that leverage terms do not give one an excess "edge"-- this is exactly the same position I take on these leveraged ETFs. Let's establish rules of engagement and then have at it.
At the time of publication, Oberg had no positions in the stocks mentioned.
Eric Oberg worked in fixed income, currencies and commodities for Goldman Sachs for 17 years before retiring as a managing director.