The reason why the big boys don't want a clearinghouse is because they are able to profit on information asymmetry. A good portion of profits on derivatives is made off of buying CDS, taking a cut, and then selling the CDS to someone else with a lower "premium" payout. So I sell a CDS to Bank A and receive $40,000 per year for every $10M insured. If I can purchase a CDS on the same company for under $40,000, I've just made money risk free. No not really risk free in reality, but it is to these guys. A clearinghouse would stop these transactions because everyone would know that the market price is $40,000.
The reason why there are so many CDS outstanding is because everyone plays the, pass it along and take a cut, game. I can purchase a CDS for $50,000 a year, then I write out a CDS for $60,000 a year, giving me a $10,000 a year profit that is hedged away for risk. My buyer can take my $60,000 CDS and sell his own for $65,000 a year, and get $5,000 in "risk-free" profits per year. And on and on this goes until you get a guy who wants to hold on to the CDS because he actually has the bonds to insure, or you find the biggest sucker who can't sell for a higher markup and is stuck with the CDS.
These type of profits provide no value and are born of inefficiencies that can be eliminated by way of a clearinghouse or exchange that posts the prices of the last trade
Showing posts with label derivatives. Show all posts
Showing posts with label derivatives. Show all posts
Saturday, May 22, 2010
Financial Regulation and Credit Default Swaps
CDS isn't involved in just insurance, it's used in CDOs to make synthetic bonds. If I write a CDS on GM bonds, I get paid the "premium" as long as GM doesn't default. If they do default, I must pay the contract amount and I receive the defaulted bonds in return (or I receive the liquidation value of the defaulted bonds back). I also have to put up collateral that pays the risk-free Treasury yield. Add it all together and you have something very much like a GM bond where I would receive interest payments as long as GM doesn't default and if they do, I lose my initial purchase (contract amount) minus whatever I can receive for the defaulted bonds (liquidation value).
The theory is to separate risk from investment. What if I just want to take on the risk of GM bonds without having to actually lend GM money, thus tying up my precious capital? With a traditional bond, I need to loan GM X money for a period of years and I will get X money back at the end assuming there is no default. What if I don't want to tie up that capital but still want to take on that risk? Well let someone else lend GM the money, I'll write a CDS on it and sell it to them. There, they just transferred the risk to me and I didn't have to lend GM any money (though I still have to put up collateral usually). Now that the other guy has a risk-free asset, he can then purchase more bonds up to the risk limit set by the risk management team. That's how it's supposed to work in theory. Of course, in reality his insured GM bond isn't risk free, if I go under, his insurance is worthless.
I think the problem lies with leverage and leverage ratios. CDS can help firms manage their risk without making big capital intensive moves. Selling GM bonds from your portfolio means that you have to find a buyer with the money, and then you're left with the problem of what to do with the money you just got. Where to put it? Why not just buy a CDS contract instead? That way, you get rid of the risk and don't have the problem of what to do with the excess cash.
When risk is being pawned off in an endless cycle, it can become hard to tell exactly how much systemic risk is left. I think I have sold my risk thanks to my CDS purchase, but so does the guy who sold me the CDS because he purchased another CDS to get rid of his risk. Risk keeps on getting transferred, we know that the total risk to the market as a whole cannot be eliminated or reduced merely by transferring, but each person thinks he's OK because he's sold his risk to some other guy who has sold his risk and on and on. It's hard to tell if the guy I purchased CDS from has too much CDS exposure (too overleveraged) because I don't know how many CDS he's sold or purchased since the time I originally purchased CDS from him.
Some limit has to be set for the amount of CDS exposure a firm has given his capital base. That amounts to a limit on leverage. A clearinghouse would make the market more transparent so I can see that there is way too many CDS out there for the CDS I just purchased to be considered safe. Right now there is no way for the individual actor to see the entire whole and the entire whole has a great impact on the individual actor. That is part of the problem. We need to think about the issues and come up with good solutions, I don't think banning CDS is even remotely close to the best solution.
The theory is to separate risk from investment. What if I just want to take on the risk of GM bonds without having to actually lend GM money, thus tying up my precious capital? With a traditional bond, I need to loan GM X money for a period of years and I will get X money back at the end assuming there is no default. What if I don't want to tie up that capital but still want to take on that risk? Well let someone else lend GM the money, I'll write a CDS on it and sell it to them. There, they just transferred the risk to me and I didn't have to lend GM any money (though I still have to put up collateral usually). Now that the other guy has a risk-free asset, he can then purchase more bonds up to the risk limit set by the risk management team. That's how it's supposed to work in theory. Of course, in reality his insured GM bond isn't risk free, if I go under, his insurance is worthless.
I think the problem lies with leverage and leverage ratios. CDS can help firms manage their risk without making big capital intensive moves. Selling GM bonds from your portfolio means that you have to find a buyer with the money, and then you're left with the problem of what to do with the money you just got. Where to put it? Why not just buy a CDS contract instead? That way, you get rid of the risk and don't have the problem of what to do with the excess cash.
When risk is being pawned off in an endless cycle, it can become hard to tell exactly how much systemic risk is left. I think I have sold my risk thanks to my CDS purchase, but so does the guy who sold me the CDS because he purchased another CDS to get rid of his risk. Risk keeps on getting transferred, we know that the total risk to the market as a whole cannot be eliminated or reduced merely by transferring, but each person thinks he's OK because he's sold his risk to some other guy who has sold his risk and on and on. It's hard to tell if the guy I purchased CDS from has too much CDS exposure (too overleveraged) because I don't know how many CDS he's sold or purchased since the time I originally purchased CDS from him.
Some limit has to be set for the amount of CDS exposure a firm has given his capital base. That amounts to a limit on leverage. A clearinghouse would make the market more transparent so I can see that there is way too many CDS out there for the CDS I just purchased to be considered safe. Right now there is no way for the individual actor to see the entire whole and the entire whole has a great impact on the individual actor. That is part of the problem. We need to think about the issues and come up with good solutions, I don't think banning CDS is even remotely close to the best solution.
Subscribe to:
Posts (Atom)