FT: Insight: Credit markets face high price for sleight of hand

Insight: Credit markets face high price for sleight of hand

By Satyajit Das, risk consultant and author of Traders, Guns & Money: Knowns and Unknowns in the Dazzling World of Derivatives

Financial Times Published: December 11 2007 18:42 | Last updated: December 11 2007 18:42

Banks have been indulging in their version of the shell game ­ and this sleight of hand will have long-lasting implications for the credit markets.

To recap: the game requires three shells, under one of which is placed a small pea. The shells are shuffled around and bets are taken on the location of the pea.

³Risk transfer² is the shell game of the credit markets.

Regulators believe erroneously that if credit risk is distributed widely it reduces the chance of a crash. Risk transfer does not decrease risk but increases it in complex ways.

Banks originate loans that are securitised. Alchemy is used to convert risky debt, with the help of rating agencies, into AAA/AA asset-backed securities. The ABS is then sold to conduits that issue highly rated ABS-backed commercial paper to investors.

Alternatively, the high-quality ABS is sold to structured investment vehicles which then issue AAA-rated debt to fund the purchase. SIVs even purchase highly rated debt from other SIVs in an astonishing chain of risk. High-quality ABSs are sold to hedge funds that leverage the AAA-rated assets (up to 20-30 times) via banks willing to lend against the value of the securities. Debt seems to buy more debt in a spiral of borrowing. At each level, banks charge fees and earn margins from money they lend.

Risk transfer encourages declines in credit standards as the banks do not intend to hold the risk. The bank receives the difference between the interest on the loan and the return demanded by the investor ³up front². As loans are sold off, ever larger volumes are necessary to maintain profitability.

The growth in the subprime market resulted from the necessity for banks to seek out new markets to generate volumes to fuel their high fixed-cost securitisation platforms.

Banks frequently don¹t actually sell off their real risks. They sell off less risky loans. In a collateralised debt obligation, the bank typically takes all or a portion of the riskiest securities ­ the equity tranche. This is ³hurt money² or the ³skin in the game² to reassure other investors. Banks must hold the loans until they can be sold. In a market disruption, if the bank is unable to sell then the risk remains with the bank.

But risk also returns to the bank via the back door. Where it acts as a prime broker ­ executing trades, settling transactions and financing hedge funds ­ the bank lends to investors using CDO securities as collateral.

The bank assumes risk to the value of the collateral. Banks provide liquidity ­ standby lines of credit ­ to the conduit vehicles to cover funding shortfalls. If commercial paper cannot be issued, then the banks end up holding the assets that they have supposedly sold off.

Risk transfer assumes that the risk is transferred to adequately capitalised investors. About 60 per cent of credit risk in recent years was transferred to leveraged hedge funds.

In modern credit markets, about one dollar of ³real² capital might support between $20 and $30 of loans. Banks must hold $1 of capital against $12.50 of loans.

The higher leverage can only be sustained in a very low default rate environment. Credit risk also moves from a place where it was regulated and observable to a place where it is less regulated and more difficult to identify.

Thus, ³transferred² credit risk is now finding its way back on to bank balance sheets as off-balance sheet structures and hedge funds are forced to sell. The total amount that will be re-intermediated by banks is unknown ­ probably, in the range of $1,000bn to $2,000bn.

The financial shell games that banks, with the tacit support of regulators, have played mean that credit markets are likely to remain sclerotic for years. Banks will reduce lending and need to raise new capital urgently to support the returning assets. As the risk transfer business winds down, the outlook for earnings from this activity will be reduced. Regulations for ³risk transfer², that allowed this to happen, also need urgent reform.

There is a need for real rather than ³false² risk transfer.

formatting link

Reply to
Faubillaud
Loading thread data ...

The more I read by this guy, the more impressed I become. He's the one of the "Mark to Make-Believe" quote isn't he?

Hmmm, mebbe Santa will bring it...

I did a back-of-a-beermat calculation not so long ago and didn't find it hard to get to where 200 Dollars of a toxic waste tranche of a CDO could be supported by a Dollar of real cash. That gets pretty scary when you consider that in a credit crunch, this can, at the limit, mean that each failed mortgage can stop 200 new mortgages being made.

[rest of article elided, but it accords 100% with everything else Ive read]

One thing missed in the article is that many CDOs have clauses in them which require the loan originator to take back the mortgage if payments have been defaulted upon within X period of time (typically 6 months or a year) or if fraud has been found in the way the loan has been made. I know that there have been many subprime mortgages defaulted on the second payment (that's the second ever payment, not the second of the 28 part of a 2-28 reset) and they're required to make the first payment when they sign the papers. Thus just as many mortgage-brokers were unable to sell on mortgages, they were getting stuck with bad loans being returned with demands for a full-value refund. No wonder they're going bankrupt in droves.

In fact there's a theory that this is what the Paulson deal is really about. So many US subprime mortgages involved people lying about their incomes (and even being encuraged to do so by brokers) that the injured CDO investors may take to returning them to the originators for a full refund. The theory runs that there may be so many of these that were they returned to banks, and the banks stuck with compensation and then holding all the bad paper, would fall below capital requirements. That really would turn the screw on the credit crunch and so the Paulson Deal amounts to politicians and American banks getting together to agree that the banks will et protection from foreign CDO investors trying to exercise their contractual rights.

Of course if this should succeed, the US mortgage markets may find themselves completely shut down, as opposed to partially shut down. Who wants a politician to be able to rewrite a contract at whim?

I've wondered for a while how house prices would do if there wer no mortgages. I expect it will be fascinating to find out...

FoFP

Reply to
M Holmes

"M Holmes" wrote

Looking at the price of a new house, part of it is for the land, part for the labour/materials required to build it, and part a "developers premium". What proportion currently pays for the labour & materials?

Presumably - if house prices fell below that, then no-one would bother building houses!

Reply to
Tim

Well most of the fall will be in land prices, but yes, a fall in house prices will lead to less new builds, as has happened in the US.

Reply to
Jonathan Bryce

"Jonathan Bryce" wrote

But is there scope for the sometimes-touted "upto 90% drop" in house prices, just from the land price part? I suspect probably not...

"Jonathan Bryce" wrote

Have their house prices fallen (much) below the cost of labour & materials?

Reply to
Tim

Why not? In many ways it makes no sense to consider the value of land separately from what is on it. It only makes sense to consider the value of land in a market where land is traded as bare building plots.

Otherwise, i.e. if considering the market for house-with-land packages, that market will determine the price of the whole package, usually without any regard whatever for what the land-only value might be.

It would perhaps be more interesting to ask the same question in relation to Japan.

Reply to
Ronald Raygun

"Ronald Raygun" wrote

Well, I was expecting that the "labour & materials" part of UK house prices to (still) be more than 10% of the total...

Do you think the L&M part is less than 10%? [Have you seen have any reliable statistics on it?]

"Ronald Raygun" wrote

AIUI, before Japan's crash in property prices, they had increased to mind-boggling amounts -- talk of "200-year mortgages" (children & grandchildren continue paying!) etc.

So I suspect that at the peak there, the "labour & materials" proportion was probably less than 10%. Am I wrong?

Reply to
Tim

BeanSmart website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.