Re: GOOD READ Credit Default Swap (CDS) question and answer

[David Farber <dave () farber net>]

Begin forwarded message:

> From: Tony Weasler <ippost () tony3 com>
> Date: October 19, 2008 7:19:24 PM EDT
> To: dave () farber net
> Subject: Re: [IP] Credit Default Swap (CDS) question and answer
>
> Dave,
>
>  For IP if you wish:
>
> Referenced in a previous post to IP, the best concise explanation of
> credit default swaps I have seen is in a "This American Life" episode:
> http://www.thisamericanlife.org/Radio_Episode.aspx?sched=1263
>
> Now, let's address the original poster's question of whether we can
> clean up the mess by invalidating the non-bondholder CDS contracts...
>
> First, a quick, simplified example of a CDS triangle:
> SuperMega, an investment bank, buys a $100 million bond from Sound
> Capitol to provide Sound with the liquidity that they need to run  
> their
> daily operations.  SuperMega is confident but not certain about the  
> odds
> of getting paid back so they go to Credit, Inc. and buy a credit  
> default
> swap (CDS) that pays them for any losses if Sound's liquid assets fall
> below a certain amount.
> -- In this case the CDS is similar to home insurance.
>
> Second, a slightly more complicated five-party example:
> MoreMega, another investment bank, decides that Sound Capitol's  
> business
> model is pretty good and decides to invest $500 million with them.   
> They
> are concerned that certain unlikely conditions will lead to Sound's
> collapse.  Because of their uncertainty MM buys 2 CDS contracts from
> Hedge, Inc. and 3 CDS contracts from BIG, a large financial services
> company, on SuperMega's loan to protect themselves against a total
> failure of Sound Capital.
> -- In this case the CDS is similar to life insurance policies taken  
> out
> on people who make money for you (like the CEO, President, and CFO  
> of a
> company.)
>
> Finally, let's add a little arbitrage:
> Crazy Capitol, a large hedge fund, analyzes how Sound Capitol is  
> making
> its money and based on current market conditions determines that Sound
> is headed for a decline.  Crazy Capitol also determines that it can
> manipulate the price of Sound's assets by strategically buying and
> selling the financial instruments that Sound trades.  Crazy buys 7 CDS
> contracts from BIG betting on the demise of Sound's portfolio.  Crazy
> then borrows $300 million from SuperMega to trade against Sound and  
> make
> its assets appear shaky.  Crazy is successful and Sound's investors
> begin to run for cover.  Quickly, Sound's assets shrink and the
> conditions of the CDS contracts (and bond default conditions) are met.
> Sound declares bankruptcy because they can't raise the needed  
> capitol to
> repay the bonds.  Crazy loses the $300 million used to bankrupt Sound
> but makes a paper profit of $700 million from the CDS contracts.  BIG
> doesn't have the cash to satisfy the contracts and turns to the
> government to pay its obligations.  The government obliges and pays
> Crazy and MoreMega.  Credit and Hedge's shareholders shoulder the  
> whole
> CDS loss on their own.
> -- In this case the CDS is similar to a loss-of-use insurance policy
> taken out on sports player who can be placed in a situation where they
> are likely to lose the ability to play the game (they are coerced by a
> third-party into in a riskier environment than when the insurance  
> policy
> was purchased by that same third-party.)
>
> These are all simplistic, purely fictional examples, but I hope that
> they explain a few of the potential scenarios where CDS contracts  
> can be
> used.  Obviously, governments can't invalidate all of them without
> unintended negative consequences.  Imagine what would happen to
> MoreMega's shareholders if they lost their entire investment.  Even
> though MoreMega didn't hold the bonds, MoreMega was still  
> significantly
> affected by Sound's demise and would be "made whole" again through
> payment of the CDS contracts that they own.
>  Obviously, Crazy Capitol's bets were not beneficial to the financial
> system as a whole and potentially criminal, but often the case is not
> nearly as cut-and-dry especially when international transactions are
> involved.  It is often very difficult to determine whether a holder  
> of a
> CDS was diversifying risk or gambling because there isn't enough
> transparency in the markets where these instruments are traded.
>
>
> Here is some food for thought:
> Current estimates of the worldwide CDS market are between $55,000 and
> $75,000 billion[1][2]. They are backing an estimated $6,000 billion in
> bonds and commercial paper[3].  This means that if an average bond
> defaults, the payments to CDS holders could be over 10x the default
> amount. How much was the bailout amount again?
>
> For those people who still think that this is a mortgage-issuer  
> crisis:
> when 10% of mortgages go into default, the value of the mortgages held
> decreases by 5-10% (because of the residual value of the house.)  When
> the bonds covered by CDS contracts default at a rate of 10%, the debts
> of CDS issuers could increase by over 100% of the total bond market
> value[4].   For a present-situation example, Lehman Brothers bonds
> valued at about $130 billion resulted in CDS payment obligations of
> about $400 billion [5] or about 3:1 over-selling of CDS contracts.   
> This
> is surprisingly low given the figures above, but a clear indicator of
> the lack of good information emerging from these elusive markets.
>  Incidentally, all of these Lehman CDS obligations are payable this
> Tuesday, October 21st[6]. That is when we will see whether the sellers
> of Lehman CDS contracts are capable of paying their obligations.  This
> has the potential to create another cascading failure of the capitol
> markets if enough counter-parties can't pay.  It could be another wild
> ride this week in the markets.
>
>  The larger problem as I see it is that people investing in companies
> like SuperMega, and MoreMega don't have enough information to make  
> good
> investment decisions because they don't know whether BIG, Credit,  
> Crazy,
> and Hedge can pay their obligations.  Additionally, there is enough
> time-delay, opacity and repositories of huge cash stockpiles in the
> world's financial markets that organizations like Crazy Capitol can
> orchestrate failures for their own financial gain.
>  I understand the need to maintain stable international financial
> markets through government liquidity injection, but it is beginning to
> look like we would be better off killing off the current
> translucent-speculation system.  A much better alternative is a
> transparent global financial system that discourages highly-leveraged
> speculation in favor of tangible goods- and services-producing
> investments.  This is illustrated by the fact that investors seeking  
> to
> purchase a private company will pay orders of magnitude more for it if
> they intend to bring it public in the near future.
>
> Best Regards,
> Tony
>
> [1]  I don't use the word trillion because it's not commonly used and
> difficult to understand in comparison to numbers in daily life.  
> Billion
> isn't much better, but at least we have some recent references that we
> can compare it to.
>
> [2] http://en.wikipedia.org/wiki/Credit_default_swap
> http://www.reuters.com/article/rbssFinancialServicesAndRealEstateNews/idUSN1472586720080915
> http://www.bis.org/publ/qtrpdf/r_qa0809.pdf#page=103
> I arrived at the $75 trillion by extrapolating the figures for 2008  
> from
> the September, 2008 Bank for International Settlements report linked  
> above.
> (It is insane that the margin of error of the information readily
> available to the public is larger than the size of the US GDP for a  
> year
> and not nearly current enough to act upon.)
>
> [3] I couldn't find a good reference for this... just blog entries,  
> but
> I'm reasonably sure I saw it in a reputable publication in the last  
> two
> weeks.
>
> [4] I am assuming a 10:1 ratio of debt to credit default swaps so if  
> the
> total value of all bonds is $10 trillion and there is a $1 trillion
> default (10% of the total market,) then the resulting obligations of  
> CDS
> issuers is 10x $1 trillion or $10 trillion* (which equals 100% of the
> current bond market.)  It's hard to understand because is sounds so
> completely foolish and counter-intuitive.
> * It's actually lower because there is a residual bond value which in
> Lehman's case was about $0.0975 per $1 issued.  Since the 10x was a  
> low
> estimate anyway I just left the residual value of the bond out for  
> clarity.
>
> [5] http://www.ft.com/cms/s/ 
> 0/25137702-972d-11dd-8cc4-000077b07658.html
>
> [6] http://www.isda.org/2008lehmancdsprot/docs/Lehman-CDS-Protocol.pdf




-------------------------------------------
Archives: https://www.listbox.com/member/archive/247/=now
RSS Feed: https://www.listbox.com/member/archive/rss/247/
Powered by Listbox: http://www.listbox.com