The Big Short by Michael Lewis

[CityFarmer]

I know this book when i read Micheal Dee article on Olam. I saw it in NLB few days ago, and decided to borrow and read it.

Not a book i enjoy reading, too much story telling with too much boring detail...

But interesting to know a hedge fund manager, Eisman who managed to bet against and to win wall street major players on subprime mortgage loan before it collapses. The derivative used was credit default swaps.

Interesting read...

[felixleong]

I liked this booked, quite a smooth read
I give 7/10, interesting enough and worth my $$

[CityFarmer]

The book illustrated a good case study for mismanaged risk, even among smart people in Wall Street. Value at Risk (VaR) should be a key part of risk management in financial institute, but it is either ignored or under-estimated. AIG FP owned close to $100 billion liability of CDS alone with peak market cap of $181 billion for whole AIG.

The credit rating agencies also played an important role in creating the 07 sub-prime crisis IMO

[specuvestor]

The problem is that if banks use VAR at 99% confidence interval and crisis volatility, they will be frozen stiff and might as well close shop. That's the problem with statistical probabilistic pricing models which has permeated the finance industry assuming all info are in the price. OTOH fundamentalists will say "know your client" is more important just as analyzing a company's credit risk is better than assuming the CDS curve is right

Similarly from a statistical basis the CDOs was not wrong to be priced AAA by the rating agencies based on statistical probabilitic modeling, and on overcollaterisation and seniority. It is easy to blame institutions or use the rating agencies as scape goat after the fact but they were basing their assumptions on well established academic "facts" and theories.

I have pretty good idea what should be blamed instead

[CityFarmer]

The problem is that if banks use VAR at 99% confidence interval and crisis volatility, they will be frozen stiff and might as well close shop. That's the problem with statistical probabilistic pricing models which has permeated the finance industry assuming all info are in the price. OTOH fundamentalists will say "know your client" is more important just as analyzing a company's credit risk is better than assuming the CDS curve is right

Similarly from a statistical basis the CDOs was not wrong to be priced AAA by the rating agencies based on statistical probabilitic modeling, and on overcollaterisation and seniority. It is easy to blame institutions or use the rating agencies as scape goat after the fact but they were basing their assumptions on well established academic "facts" and theories.

I have pretty good idea what should be blamed instead

I do agree that banks should not manage their VAR base on 99% confidence. But what your opinion on AIG FP case? Is it a case of mismanaged on the issue of CDS?

As for the credit rating agencies, rating AAA over a repackaged BBB rated sub-prime mortgage loan, should not be done by professionals, who received a hefty service charges. Of course, it is always good reasons from a statistical model, but professional work is not just number crunching base on the sophisticated model.

Above are views from a non-financial professional. All comments are welcome.

[specuvestor]

Point is tail risk cannot be hedged statistically. If banks were to always look at 99% CI and crisis volatility, as some academics suggest together with Black Swan author Taleb, nothing would work

Root of problem is simple: the INCENTIVE that exacerbates greed. Structurally it is the problem of leverage and regulation to curb incentives. Number crunching is not the solution. Incentives to these people cannot be "no downside". That's why some suggest defer compensation of 10 years vest.

Hedging is not the problem. Derivatives is not the problem. But when you have derivatives more than 10X of underlying asset, something is seriously wrong. That's why central clearing of derivatives is a good idea: to reduce counterparty risk and to limit leverage positioning. IMHO the problem is not insufficient bankng equity. It is overleverage. Value investors can see through the theoretical mumbo jumbo that is so complex that Buffet calls it financial nukes as it seeks to mask the ever increasing leverage in these structures

Can you imagine if a stock has short ratio of >100%?? it is so ridiculous that when a bond defaults the bond rises because people are scrambling to get the underlying bond for delivery.

THAT to me IMHO is the main problem

[AlphaQuant]

Root of problem is simple: the INCENTIVE that exacerbates greed. Structurally it is the problem of leverage and regulation to curb incentives.

i agree.
banks have definitely shored up their liquidity ratios and de-leveraged the books, but the idea of compensation tied to "performance" is still not changed, which is why some years down the road, a bunch of people are still going to be betting big time using banks' money to profit handsomely from it - the incentive scheme ensures that the optionality nature of the payoff - limited downside and unlimited upside.

unless this culture changes, there will be another crisis down the road.

[specuvestor]

The volcker rule is trying to change that but not very successful from lobbying. The fact that banks can claim they are victims are alarming but not totally untrue: banks and shareholders were hurt, it was the individuals who took outsized leverage and risks that benefited. Their extreme capitalistic self interest tendencies ie Screw the rest mentality have to be curbed

[CityFarmer]

Point is tail risk cannot be hedged statistically. If banks were to always look at 99% CI and crisis volatility, as some academics suggest together with Black Swan author Taleb, nothing would work

Root of problem is simple: the INCENTIVE that exacerbates greed. Structurally it is the problem of leverage and regulation to curb incentives. Number crunching is not the solution. Incentives to these people cannot be "no downside". That's why some suggest defer compensation of 10 years vest.

Hedging is not the problem. Derivatives is not the problem. But when you have derivatives more than 10X of underlying asset, something is seriously wrong. That's why central clearing of derivatives is a good idea: to reduce counterparty risk and to limit leverage positioning. IMHO the problem is not insufficient bankng equity. It is overleverage. Value investors can see through the theoretical mumbo jumbo that is so complex that Buffet calls it financial nukes as it seeks to mask the ever increasing leverage in these structures

Can you imagine if a stock has short ratio of >100%?? it is so ridiculous that when a bond defaults the bond rises because people are scrambling to get the underlying bond for delivery.

THAT to me IMHO is the main problem

I agree with your view on the root of problem. Too bad solution does not seem appearing soon in near future. We have to live with it

[CityFarmer]

There is a relevant article on most recent The Edge Singapore, titled "Winning big by betting on US mortgages"

It is about a hedge fund, Metacapital Management LP, founded by Deepak Narula

He shorted on subprime mortgages about the same time as Eisman did, but he fail. The lacking are
- He did not lock-in his investors. Investor headed for the door before the subprime crisis arrived
- He did not consider the time factor

But he reap in huge gain with his second attempt on subprime crisis. The fund achieved 23.6% return in 2011, and 37.8% in the first 10 month of 2012. The fund price up 520% since started trading in July 2008

He bet on Fed bailout on both FannieMae and Freddie Mac, with a sophisticated model on homeowners behavior, but i really not sure exactly the method

A good read for those interested