In the first part of this post, I framed how CDS could be abused in the market by unethical investors. It is questionable how often these devious plans have worked out or how widespread their fraudulent use has become, but it is pretty clear that it is occurring behind the scenes. Insider trading is difficult to stop in the public markets on the major exchanges, but it is impossible to stop it in opaque derivatives markets between dealers and private investors/institutions.
A CDS contract by itself is not an evil instrument. I would also argue that leveraged credit default swaps on single names, by themselves, do not cause a systemic risk for the financial system. A credit default swap is nothing more than the credit risk of a corporate name. The genesis of this market came about with the benign need from banks to hedge over-exposures to certain credit on their books. The investors willing to take the other side of this transaction were investors who wanted corporate credit exposure from these same particular issuers. A CDS seemed to work for both sides of the transaction: to hedge the exposure of those who did not want it and to efficiently provide credit exposure to those who did. One might argue that these investors who wanted exposure to the credit should have simply invested in the bonds, but the truth is that some corporate names are difficult to gain exposure to because they do not issue debt often and/or the investors who own the debt are hesitant to part ways with it. The CDS market provided a facility for investors to gain exposure to names that they otherwise could not invest in. It also provided an efficient facility for investors and speculators to short the credit of names they did not like, which is one key element of an efficient market with true price discovery.
A credit default swap market on individual names increases the transparency and liquidity of the fixed income markets in general. You no longer have to wait for a cash security of a name to trade on the market, you could now infer the price of the bond based upon the price of the traded CDS. The level of the CDS was chosen by market participants as there are two sides to every transaction. If investor A is happy to sell GE credit risk at 200bps per year while investor A would like to short GE credit risk at 300 bps per year they can easily agree on 250bps as a transaction price. Now we have better and more frequent information regarding the credit worthiness of companies trading in the CDS space.
So what would cause the whole CDS market to break down and stop working? I think the key issues to address are:
- Collateral Management/Exposure Concentrations
When addressing transparency it is easy to see issues such as the one I presented in part I. If market participants do not know the economic exposure of a certain player in the market, outright fraud and insider trading is permitted to occur. There needs to be a watchdog that monitors the activity in the CDS market to ensure that certain players are not butchering companies and spreading rumors for their own benefit or making use of insider knowledge in a market that goes unmonitored. Knowing positions also stops “runaway” markets in that one player can not seemingly drive a market in one direction. If the player was known, then the other market participants would know that any move was driven by the one player and not underlying fundamentals.
Collateral management is a hallmark of derivatives exchanges. The CME opened the day after October 19, 1987 without incident. The key in establishing a stable market is to effectively manage the margins and collateral that should be posted by all participants. The futures exchanges have been effective in managing positions by requiring daily mark to market, initial margin, and variation margin. Right now, many hedge funds and institutions can have outstanding negative balances that are $50M and more between the institution and the investment bank before being required to post anything on a trade. In addition, they can have potential exposures that far outstrip any capital that they could possibly raise in a distressed situation (LTCM).
Leverage is not a bad thing by itself, but as is always the case, it must be utilized prudently. The lack of transparency never allowed the investment banks to figure out how leveraged AIG really was. If you are sitting at the desk of merrill lynch and you bought $200M of protection on GM bonds, how do you know whether AIG sold another $1B to other assorted investment banks? If AIG instead had to go to a central clearing house, then no one in their right mind would allow one firm to sell unlimited protection on corporate names because they would know that the agreements would be worthless; that AIG could never make all of the claims that it was promising. The reality is that AIG was allowed to sell large quantities of protection and leverage its financial products company to a ridiculous level only because no one knew exactly how much AIG was selling except for AIG itself.
Now to a philosophical point – The purchase of debt protection through a CDS contract on GM is very similar to purchasing a very far out of the money put option on GM stock. Most will not say that buying naked put options on a company is inherently wrong, so there is no fundamental reason that purchasing CDS on a company is wrong. It is akin to saying that no one should be allowed to sell a stock that he/she does not own. These rules could be put in place, but I think we would have even larger asset bubbles than we saw in 2000 and 2007. If you are only allowed to buy debt, buy stocks, buy calls, sell cds protection…then prices tend to go up very easily…until things come crashing down again. Be wary of throwing the baby out with the bathwater.