You might want to look at Perry Mehrling's biography "Fischer Black and the Revolutionary Idea of Finance." Fischer Black thought derivatives were akin to gambling. With that in mind, Mehrling notes that Black actually envisioned a world of bonds in which you separated them into three parts: the cashflows, the interest rate risk, and the credit risk. Black wrote this in about 1970s, before Bretton Woods collapsed, before interest rate futures began trading on the International Monetary Market at the CME in the mid-1970s, and long before the first CDS ever traded. But he envisioned a world with all three markets working together. CDS should definitely exist. I can't remember if Mehrling things they should be exchange traded or OTC (you should check out his website), but the thing about CDS is that people often want to make unique bets about entities, and that's something the exchanges can't do. Exchanges are best for standardized products, not unique bets. In fact, rights now, it appears that some of the products that the Commodity Futures Trading Commission in the US tried to make tradeable in exchanges, have gone off-exchange, because people don't seem to want standardized products (I heard about a 30 year exchange traded product that people are now trading OTC for 29 years + 360 days). The problem with regulation of financial markets is, the markets emerge to solve problems, but regulators often regulate without understanding why the markets exist. This has happened since the 17th century, when we saw the world's first fully-fledged impersonal financial market spring up in Amsterdam, where the local authorities tried to ban many practices, not understanding that there was a reason (beyond simple greed) why people were doing what they were doing.
In my opinion CDS should be strictly regulated. I don't think that such things must be banned, but the liberalization process had gone too far. And nothing have changed after the crisis 2007-2009.
Banning CDS is a very very bad idea. However, Prof Theissen correctly identifies two serious problems. The systemic risk posed by CDS is hard to quantify. AIG was only a systemic risk because of Lehman - in fact, in the days and weeks immediately after Lehman, almost every disturbance was potentially a systemic threat. The data suggests that banks are dominate both sides of the market as buyers and sellers, so the net exposure is probably relatively small.
The issue of naked CDS buyers does seem to require regulation. I'm not sure whether central clearing could be made to work. If it could, it would help, especially if it made the contracts more liquid
Personally i believe CDS issuance should be strictly limited to the value of the debt it is linked to. As it is now, it's like insuring a car at many companies and obtaining the premium from all if you crash the car. I'm not sure i understand the issue completely, but it does seem like encouraging defaults.
The analogy is misleading. Car insurers would not like you to take multiple insurance policies because they fear you would put yourself in a position where it paid to crash your car. By contrast, the CDS buyer has no control over whether or not the underlying "crashes" - if he did, then yes, the situation would be intolerable.
I think the other aspect of the story is that CDS were not the cause of the crisis, but a symptom of what was happening with all of the mispricing of CDOs because they were misunderstood. CDOs increasingly drove the housing market post 2000, as the investment banks, which previously never looked into housing, suddenly rushed into mortgage lending. As more and more people believed the CDO tranches to be overpriced, the CDS activity increased greatly. The AIG CDS story is also incomplete. My colleague points out http://mercatus.org/publication/securities-lending-and-untold-story-collapse-aig , AIG's Life Insurance group, which was highly regulated, lost lots through their securities lending activities. So it wasn't just the Financial Products group and CDS trades that brought trouble to AIG... Maybe the answer is to treat naked CDS like gambling, and tax accordingly, unless you post collateral, in which case you treat them as standard derivative products?
CDS should never be allowed to cover more than a fraction of the risk. Otherwise it can multiply a risk. Let's take Greece as a case, CDS value was much higher than default value itself. This has multiplied the risk of Greece default to systemic level. Also, CDS seems to be responsible for increasing credit costs. Instead of spreading risks it seems to be a risk multiplication vehicle.
Basically financial markets should never lend money to a lender in high default danger. Emergency lending for governments should be done by specialized institutions like WB and IMF or other governments that may have interest to do so. For companies that pose systemic risk i guess the best way is for the governments to take over and do two things, first stabilize than split those companies.
CDS are "shadow insurance". As such they need to be regulated as much as the insurance industry is.
They are extremely useful instruments for risk sharing and mitigation. For a recent suggestion on using CDS spreads for risk sharing between a sovereign debtor and its creditors see the link below. It discusses the potential for improved stability in the sovereign debt market.
CDS are essential instruments, since we need some way to separate credit risk from market risk. Even naked CDS are probably necessary for the same reason we need speculators in, say, forward markets to balance hedging demand and supply. There is a need for CDS regulation, however, to prevent matters getting out of hand. For example, there have been situations where the volume of long CDS positions far exceeds the volume of underlying bonds issued. The problem, I guess, is HOW to regulate this sort of market, especially across different bankruptcy regimes internationally
CDS have been criticized for facilitating market manipulations in the eurozone crisis and naked trading was banned by the German financial regulator in May 2010 and by the EU since November 2012. However, a European Commission report issued at the time found no apparent mis-pricing in the sovereign bond and CDS market. The empirical investigation of this report finds ``no conclusive evidence that CDS markets increase funding costs for Member States". IMF also presented evidence that also refutes the criticism against their use and argues that CDS have contributed to the deepening and efficiency of the sovereign markets. A nuanced view on the role of CDS in the credit crisis of 2008 is offered by Rene Stultz who states that ``financial derivatives have clearly lost any presumption of innocence'' but argues it would be misguided ``to turn 180 degrees from a presumption of innocence to a presumption of guilt".
I discuss these issues in more depth and give references in a paper where we looked at portfolios of CDS . The paper itself deals with portfolio diversification so I do not think it addresses your question but the introduction and references will be useful. The paper is posted here:
Deleted research itemThe research item mentioned here has been deleted