The only way to drive up the cost of protection is to buy a ton of protection. Doing that will “raise eyebrows.” Second, you assume that whatever effects the cost of protection have on the cost of debt will last long enough to actually inhibit the ability of a firm to finance with debt. If you’re a firm that relies heavily on financing, you should have stable lines of credit in place for exactly this reason.

A fatal flaw in your argument is one needs to gain control to sink a company. If a company’s life blood is its financing rate, it is easier to sink it by increasing the financing rate until it is uneconomical to continue operating. Take an independent broker dealer – if its spreads start widening, momentum guys (fast money, hedgies…) get on board and buy protection and supply and demand being what it is, spreads go wider. Financial firm goes to raise $$ and have to pay a higher premium. Margins approach zero, confidence waivers > BSC and LEH occur.

As well, while swap dealers net their positions, they often use correlation books to do so. Hasn’t worked out well so far. Then there is always counterparty risk and the sunday netting party.

Is there some particular reason why you publish contracted versions of your postings in the RSS feed? Would you consider publishing your postings in their entirety?

Thanks.

–Samir

1) On definitions.

A Credit Default Swap in my world is a contract where the underlying is either on a single reference entity / obligation, or on one of the Markit Indices such as CDX or iTraxx. A CDS is a straighforward contract and almost fully automated via the DTCC Warehouse and CLS.

A CDO (whether vanilla, hybrid or with caramel swirls) is a highly structured bespoke product, and quite different from a CDS.

I do agree that hedging a CDO can be achieved using CDS contracts, and there is a direct trading link between the two.

2) On pricing

If you and I went down to a fruit market to buy oranges, and there were multiple sellers, given that an orange of a similar size, is a standard product, you would expect that the sellers would arrive at a common price as the price differential would soon evaporate as the cheapest seller would win all the business.

If on the other hand we look to buy a complex multi-room home entertainment system with support for multiple file formats, wireless connectivity, integration with your XBox 360 and PS3, plus it made a nice cup of tea, you would expect quite different prices as each seller would provide different technology solutions from Panasonic, Sony etc. leading to a non-standard product and customised pricing.

I guess what I’m saying is that the raison d’etre of the OTC market is to offer a customised price for a customisable product.

I contradict myself though as the CDS contracts in the market now *are* standardised

If I am an OTC market maker, I might offer a better price or worse price to the counterparty, depending on a) credit lines b) their ‘riskiness’, c) whether we provide prime brokerage d) “other”

Is it right that protection on a single reference entity should be the same for all sellers in the global market? I don’t know the answer, but an empirical answer is contained within the trade warehouse which has the data to show the variance in pricing over the past 5 years or so.

Good points though, Bill.

I have to admit, I find your definitions misleading. It’s my understanding (feel free to correct me if I’m wrong), that the vast majority of 2006 and 2007 vintage CDOs were hybrid and therefore that recently CDOs and their managers sold large quantities of CDS protection. Now I’m not claiming that they ever sold a big fraction of the CDS on the market, but there is a connection between CDOs and CDS that needs to be acknowledged.

If there’s any published data on the fraction of hybrid vs. cash CDOs on the market or on the size of the synthetic exposure in hybrid CDOs, I would love to know about it.

Bill: “If all traders knew a reference price, why would anyone quote differently? therefore prices would converge, and the supply of prices diminish, leading to the end of OTC trading in credit protection.”

I don’t understand the problem. Very few people seem to feel that the stock market is a failure. What’s wrong with having prices converge to a single price? In a mark-to-market world wouldn’t that be a lot better than having to guesstimate the value of your balance sheet by picking and choosing between different dealer quotes?

Nothing prevents a market maker on the stock market who believes that the posted price is fundamentally wrong, from posting a better one.

What is lost by converging to a single price — besides the profits market makers can make by trading on proprietary price information? In economic terms, this proprietary information is just a kind of monopoly power that our regulations have granted to market makers for reasons I don’t understand.

PS They’re so good, I really do have to ponder them for a while.

The price for protection on a single name CDS is freely available in the professional market, where the professional traders execute trades. There seems to be a strand in this discussion that prices are not publicly available, but why would they be?

Lets say a CCP published an end of day protection price for BigCo. The following day, any buyer of protection on BigCo in the professional market, would want to achieve that reference price, or better, given they know when they would be worse off by paying too high a price.

If all traders knew a reference price, why would anyone quote differently? therefore prices would converge, and the supply of prices diminish, leading to the end of OTC trading in credit protection.

Maybe this is what the government in the US and regulators feel should happen. If you take this through to it’s conclusion, this would be the end of the OTC market, and the end of the idea that you can get different prices for the same product (a CDS) from different suppliers who have their own reasons to offer a higher or lower competitive price.

The OTC market needs to remain as-is, or come to an abrupt end. Does this logic make sense?

Bill

1) Amaranth was an active user of OTC derivatives, I would dispute that their *entire* portfolio was on-exchange.

2) AIG hasn’t collapsed yet

3) The reason Amaranth caused little pain in the market was the size of their portfolio, not the product mix. They were a medium sized fund who posted collateral on all their OTC positions to the major dealers. The collateral would have offset some of the losses from their bankruptcy.

4) There are a number of shortcomings of the “move to clearing”:
a) Not all products will move there, especially the structured products such as CDOs, which are too complicated to risk manage and process within a CCP.
b) A single CCP doesn’t process all trades for each firm. A dealer like Lehman will have been trading at many exchanges around the world, posting margin. Hence not one single CCP will have a coherent risk view of Lehman, and therefore would not reduce the possibility of default, nor give any early warning.

5) On ‘understanding the CDS trade’: I haven’t seen any claims in the market that whole swathes of traders mis-understood how a CDS contract operated. The suggestion that ‘people’ executed CDS contracts seems to suggest naivety in the market by professional persons who really should know better. No private individuals can invest in OTC contracts, it’s just not allowed by the regulators and not via any major dealer. Even hedge funds invest as a corporate legal entity, not as Joe Trader personally.

6) I do agree, trading something you don’t understand is foolish. But I disagree this is a major factor in the current crisis.

7) I think there should be some clarity about CDSs and CDOs:

a) A CDS on a single name (such as GM) and a single obligation (a specific GM bond) is a well understood product. Pricing and risk management for these is working properly.
b) A CDS on an index (such as CDX) allow you to take a view on the credit worthiness of a specific market section or region. These are also being priced and risk managed properly.
c) A CDO is a complex structure containing multiple securities, split into layers, sold to many investors, linking them through to the asset back securities such as MBSs. CDOs have gone badly wrong, as the assumptions on default rates were not cautious enough, i.e. the massive rates of default on mortgages undermined the layers of risk a CDO packaged.
d) An SIV is another form of a CDO, suffering the same problems as CDOs.

That’s it for now, I think Don’s posts are thought provoking, and I’m glad he did them. My feeling is that the OTC market wasn’t the primary cause of the current crisis, but has introduced (unwanted) secondary effects which are unfortunate. This doesn’t justify eradicating the OTC market.