Reprint: Tyson-Douglas, the Mirage Hotel Sportsbook and the AIG Debacle

PREFACE:
Not only is it Super Bowl weekend, but this past Friday was the 32nd anniversary of one of the greatest upsets in sports history; Buster Douglas getting up off the canvas in the eighth round and knocking Mike Tyson out in the tenth round.  Before the fight, Tyson had a well-deserved aura of invincibility, and Douglas entered the fight as a 42:1 underdog.  At those odds, a $100 bet on Tyson to win would pay a little over $2. 

This poor risk versus return ratio on a Mike Tyson victory was very similar to the risk versus return ratio faced by AIG in its credit default swap trade.  Mortgage bonds were considered as invulnerable as Mike Tyson, and the 'premiums' AIG received to insure them against losses were a tiny fraction of AIG's exposure to losses if the bonds ever did go bad.  As everyone now knows, AIG risked tens of billions in losses to earn a few hundred millions of dollars a year in premiums.  AIG based their entire mortgage derivative trade on the 'insights' of a Wharton finance professor, Gary Gorton.  (Dunce #27)  AIG's blunders in 2008 were virtually identical to those made by the MIT PhDs at Long Term Capital Management (LCTM) in 1998.

PhDs in finance and economics should come with a warning label, just like cigarettes. 

ORIGINAL ARTICLE:
"Oh!  A nice uppercut by Buster Douglas!  Look at this!  He has knocked Mike Tyson down for the first time in his career.  Mike Tyson hits the canvas.  He's in big trouble!  He may not be able to recover.  Its up to seven and eight....he's not going to make it!  Unbelievable! Unbelievable! Unbelievable!  Buster Douglas is the new heavyweight champion of the world.  This has got to be - if not the - one of the biggest upsets in the history of boxing!
Bob Sheridan - the Voice of Boxing, February 11, 1990: The Tokyo Dome - Tokyo, Japan

I am sure many people look back on the time when they were growing up and conclude that everything was better back then.  For anyone like me who grew up in the 1980s, the advantage we have is the time we were growing up was the best!  Perhaps the best evidence of this is in sports, boxing in particular.  The 1980s were a golden age of boxing and one of the most memorable fights of not just the 1980s but of all time was Mike Tyson versus Buster Douglas, (which technically took place in the 1990s).  In fact, the fight is considered one of the great upsets in not just boxing, but all sports. 

To describe Mike Tyson as the favorite doesn't even begin to describe the aura of invincibility that had - justifiably so - been built up around Mike Tyson.  For a heavyweight he had unbelievable fast hands and yet still hit with incredible power.  Even now he remains universally acknowledged as one of the hardest hitting heavyweights ever.  Tyson was unquestionably at his peak on June 27, 1988 when he demolished an undefeated Michael Spinks in just 91-seconds in what was then the richest purse in boxing history.  By the time of the Buster Douglas fight in February 1990, few people in the US were willing to spend any money to see Tyson fight Douglas and the fight had to be held in Tokyo Japan.  (1)

Tyson's aura of invincibility didn't just undermine the prospects for holding the fight in the US, it also undermined the prospects of large amounts of money being wagered on the fight.  Jimmy Vaccaro managed the sports book at the Mirage casino in Las Vegas and placed 27:1 odds on Douglas; basically a $1000 bet on Tyson to win would pay less than $40.  To Vaccaro's astonishment, the vast majority of bets being made were Tyson to win.  As a casino doesn't want to become an unwilling participant in a bet, it tries to place the odds - or the point spread - so equal amounts of money are bet on both sides of the wager.  As bets continued to pour in on Tyson to win, Vaccaro kept pushing the odds higher and higher; first 35:1, then 37:1 and finally, the legendary 42:1. (2) 

All the considerable sporting aspects of the fight notwithstanding, the betting around the fight provides tremendous insight into the housing crisis; the market for mortgage derivatives in particular.  Mortgage derivatives were essentially side bets on whether a particular mortgage bond would collapse.  The parties to a mortgage derivative contract had no vested interest in the mortgage bond itself or any of the individual mortgages that made up the bond.  Instead, the parties to a derivative contract placed a bet - via a financial derivative - on whether the mortgage bond would default or not.  However, it was not an even money bet.  Because mortgage bonds were considered so secure and so unlikely to default, a very small premium could be used to bet on a very large mortgage bond collapse.  For example, for annual premium payments of approximately $25-million, a return of $1-billion could be earned if the mortgage bond collapsed.  Effectively, an investor who was willing to bet on a $1-billion mortgage collapsing, was given odds of 40:1.     

The biggest player in the market for mortgage derivative was AIG.  However, AIG was on the side of taking the bet on the mortgage market collapsing; they weren't making the bet on the collapse.  AIG was willing to accept $25-million in payments even though accepting this money exposed them to losses as high as $1-billion!  Essentially, AIG viewed mortgages in exactly the same way the sports world viewed Mike Tyson as he prepared to fight Buster Douglas; there is no chance that a mortgage bond will ever default.   

To manage this enormous amount of risk associated with its bets on mortgages, AIG heavily relied on computer models developed by Gary Gorton, a professor at the Wharton School of the University of Pennsylvania.  (Gorton now teaches at the Yale School of Management)   The models used reams of historical data to model the mortgage market.  Like nearly all statistical models of this type, the models – at their absolute best – can only be descriptive of things that had happened in the past.  However, models of this type have no predictive value for what might happen in the financial future whatsoever.  The failure of Long Term Capital Management in the late 1990s should have proven this simple concept conclusively.  Because it is still widely believed that computer programs can inerrantly manage highly leveraged investments, the enormous risks latent in AIG's CDS trades were never considered as they should have been.

The best evidence of this must be a press conference held by Prof. Gorton and AIG senior management in December 2007,.  At this time, mortgage bombs had been detonating all over Wall Street.  In July, two Bear Stearns funds invested in sub-prime mortgages had collapsed.  Stan O'Neal had been sacked at Merrill Lynch in October and Citigroup had announced enormous mortgage losses in November.  In spite of all this housing related carnage, Gorton and senior AIG management expressed complete confidence in Gorton’s models and the company’s mortgage bond insurance trades.  Of his models, Gorton claimed they “… are guided by a few, very basic principles, which are designed to make them very robust and to introduce as little model risk as possible.  We always build our own models.  Nothing in our business is based on buying a model or using a publicly available model.”  AIG’s CEO, Martin Sullivan claimed Gorton’s models gave the company a “high level of comfort,” while the head of AIG’s financial products division, Joseph Cassano believed Gorton’s models were “simple, they’re specific and they’re highly conservative." (3)

Of course, as it turned out, the models AIG relied on had holes in them large enough to drive a armored division through sideways.  On February 28, 2008 - just two months after expressing complete confidence in its strategy – AIG disclosed in its end of the year regulatory filing that it was carrying a total of $11.5-billion in CDS losses on its books.  In the same filing, AIG stated it had posted $5-billion in collateral against these losses.  On February 29, 2008 Joseph Cassano was forced out of AIG.  The investment community was shocked but these losses were a mere harbinger of all the losses to come.  In a little over six-months, AIG would require an $85-billion bailout, and by the time the bailouts ended, AIG had received well over $100-billion.  

Concluding Remarks:
In reminiscing about the Tyson-Douglas fight, even thirty-years later Jimmy Vaccaro is still amazed at the bets that continued to be made on Tyson even as the odds kept climbing higher.  One gambler placed a $160,000 bet on Tyson right as the odds made it to 42:1.  This bet - had it won - would have paid less than $4,000!  Stripped of its Ivy League veneer, this was the madness that AIG was engaged in and it went on for years and it involved tens of billions of dollars.  Moreover, the derivative products that AIG used to place its bets, didn't contribute to a single house being built or a single mortgage being made.  The 'bets' that AIG placed could only be made on mortgage bonds that had already been issued.  There is no better example of the financialization of the US economy and the Wall Street casino that has supplanted real productive enterprise than the market for mortgage derivatives.  Whatever you do, don't be confused by the complexity of derivative products or the strategies developed to invest in them.  If you can understand the betting that surrounded the Tyson-Douglas fight, then you can understand everything that went wrong with AIG and derivatives.  

 

 

Peter Schmidt
Sugar Land, TX
August 16, 2020

PS - As always if you like what you read, please consider registering with the site.  It just takes an e-mail address, and I don't share this e-mail address with anyone.  The more people who register with the site, the better case I can make to a publisher to press on with publishing my book!  Registering with the site will give you access to the entire Confederacy of Dunces list as well as the Financial Crisis timeline.  Both of these are a treasure trove of information on the crisis and the long-running problems that led to it.

Help spread the word to anyone you know who might be interested in the site or my Twitter account.  I can be found on Twitter @The92ers

ENDNOTES:
1.  Tyson had already signed to fight Evander Holyfield later in the year and the Douglas fight was largely seen as a tune-up for Tyson.  Tyson had won his previous five fights all by knockout and only two made it past the second round.

• 22 JAN 1998:  Larry Holmes; KO Round 4. Atlantic City, NJ
• 21 MAR 1988: Tony Tubbs, KO Round 2. Tokyo, Japan
• 27 JUN 1988: Michael Spinks, KO Round 1. Atlantic City, NJ
• 25 FEB 1989: Frank Bruno, KO Round 5. Las Vegas, Nevada
• 21 JUN 1989: Carl Williams, KO Round 1. Atlantic City, NJ

2.  Tim Dahlberg, "42:1 a Line That Lives in Las Vegas Betting History," Associated Press, February 14, 1990 https://lasvegassun.com/news/2020/feb/11/42-1-a-line-that-lives-in-las-vegas-betting-lore/

3.  Carrick Mollenkamp, Senea Ng, Liam Pleven and Randall Smith, "Behind AIG's Fall, Risk Models Failed to Pass Real World Test," Wall Street Journal, October 31, 2008 https://www.wsj.com/articles/SB122538449722784635