Credit failure mode
Dec. 21st, 2007 01:05 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
drscottIt's often a bad idea to invest based on the drumbeats of daily news. A few weeks ago, the press was filled with gloomy stories about the decline of the dollar and a possible crisis; people were advised to hold more stable currencies and buy further into emerging markets (which, BTW, look a little bubbly at the moment.) Since then the dollar has strengthened and a flood of exports and reduced domestic consumption have started to reduce the current account deficit. So much for following the Common Wisdom.
One might look at the slew of stories about falling house prices and further troubles in banking and credit the same way, but one would be wrong. What very few people understand (outside of a minority of professional bankers and economists) is that the daily news stories reflect the increasing likelihood of a serious collapse of credit; every few days the ECB or Fed tries new ways to inject money into the system to get credit flowing again, showing that they see the hazard.
There are a large number of clever ways to reduce the risks of a portfolio of assets; one is to diversify, adding together many small risky assets whose risks are independent of each other to create a less-risky whole. Another is to insure risks; to find another party who will accept some of the risk of the asset in return for a payment.
Most of our recent financial crises and collapses involved misunderstanding and misuse of these techniques. The Crash of '87 came about because of overreliance on portfolio insurance, which supposedly removed some of the risk of holding stocks -- but it didn't; the LTCM hedge fund collapse in 1998 came about because some supposedly brilliant people (Nobel Prize-winners and the like) didn't see that their models would come apart when the many, many assumptions made in formulating them were violated. The current mortgage crisis stems from similar causes: projecting future results based on past bahaviors, not recognizing a systemic risk when assumptions ("house prices will not fall nationwide") prove to be false, and creating supposedly safer products out of risky ones by spreading the risks so broadly they can't be identified.
The packaging and commoditization of risk management has had many good effects, including reducing the cost of credit and improving the efficiency of many industries as money could more easily go to where it would improve efficiencies and be invested wisely. Bank regulations allow credit to be extended to a ceiling which is based on the presumed risks of assets held by the bank; when a bank has a pile of AA-rated bonds, it is allowed by the regulations to extend a multiplier of that amount in new loans. The creation of many seemingly low-risk bonds out of risky mortgages and consumer loans allowed banks to lend far more, raising the supply of money in the economy and pumping up the values of assets worldwide.
The trouble began years ago with the use of insurance on bond issues. In its simplest form, an issuer (let's say a community college district) would sell bonds to build new buildings backed up by its promise to pay them back out of its local tax revenue. A rating agency would have given this a rating of A or so based on the unlikeliness of default, but the issuer would then get the rating upgraded by seeking insurance -- in the event of a local problem creating default, the insurer would make good on the bonds. This backing decreased probability of default sufficiently to raise the bond rating to AAA, and so they could be sold for more (i.e., at a lower interest rate), with a small payment for the insurance easily made up for by the higher price.
Insurance as a risk management tool relies on spreading risk far enough so that it touches enough deep pockets to be covered. To achieve this, the risk is diced up into many small pieces and combined with other small pieces to create a less-risky whole. But this does not work as intended when the small risks are not independent of one another; pooling house insurance contracts from one coastal city in Florida doesn't much mitigate the risk since a large number of the insured houses can be trashed because of one hurricane.
The bond insurers began to go astray when they started to accept contracts to insure pools of mortgages and commercial loans, where the possibility of recession or declining house prices would affect many of the loans at the same time. While the good times rolled, competitive pressures and hunger for profits led them down a path of taking too little in premiums for accepting correlated risks. Now the bond insurers themselves are in deep trouble, having to acknowledge that the losses they might face are vastly larger than the capital they have to cover them. As a result, their own credit ratings are being downgraded and every issue they cover looks riskier, too. A downgrading of a large fraction of all outstanding debt as the market discovers suddenly that pooling and insuring risks doesn't make them go away is occurring.
Remember the banks? Having spread new credit far and wide based on their supposedly AAA-rated assets, they now discover that the downgrading of much of their portfolios of assets means they must reduce lending abruptly or find new capital to meet the regulations. Major banks are selling pieces of themselves to new investors (many of them foreign) to try to meet these capital needs. The Fed and other central banks are trying to unfreeze the flow of credit by loaning billions to the banks to get them through the crisis.
As in the dot-com bust of 2000, the central banks may succeed in pushing enough money into the economy to pump up some sectors to make up for the sudden collapse in values in another; the Fed at that time reduced short-term rates enough to start the creation of the housing bubble, and enough people found work in building-related trades to make up for the lost jobs in technology.
So despite what you might glean from the babble of news and candidate debates, the real hazards to the system now are far more serious than just a few million stupid or gullible investors and homeowners who got loans they shouldn't have; we are crossing a chasm that if poorly handled could lead to a sharp contraction in the world economy, leading to more pressure to de-globalize, which if politicians bow to it would lead to worldwide depression. The chance of this is not large, but it's there.
One might look at the slew of stories about falling house prices and further troubles in banking and credit the same way, but one would be wrong. What very few people understand (outside of a minority of professional bankers and economists) is that the daily news stories reflect the increasing likelihood of a serious collapse of credit; every few days the ECB or Fed tries new ways to inject money into the system to get credit flowing again, showing that they see the hazard.
There are a large number of clever ways to reduce the risks of a portfolio of assets; one is to diversify, adding together many small risky assets whose risks are independent of each other to create a less-risky whole. Another is to insure risks; to find another party who will accept some of the risk of the asset in return for a payment.
Most of our recent financial crises and collapses involved misunderstanding and misuse of these techniques. The Crash of '87 came about because of overreliance on portfolio insurance, which supposedly removed some of the risk of holding stocks -- but it didn't; the LTCM hedge fund collapse in 1998 came about because some supposedly brilliant people (Nobel Prize-winners and the like) didn't see that their models would come apart when the many, many assumptions made in formulating them were violated. The current mortgage crisis stems from similar causes: projecting future results based on past bahaviors, not recognizing a systemic risk when assumptions ("house prices will not fall nationwide") prove to be false, and creating supposedly safer products out of risky ones by spreading the risks so broadly they can't be identified.
The packaging and commoditization of risk management has had many good effects, including reducing the cost of credit and improving the efficiency of many industries as money could more easily go to where it would improve efficiencies and be invested wisely. Bank regulations allow credit to be extended to a ceiling which is based on the presumed risks of assets held by the bank; when a bank has a pile of AA-rated bonds, it is allowed by the regulations to extend a multiplier of that amount in new loans. The creation of many seemingly low-risk bonds out of risky mortgages and consumer loans allowed banks to lend far more, raising the supply of money in the economy and pumping up the values of assets worldwide.
The trouble began years ago with the use of insurance on bond issues. In its simplest form, an issuer (let's say a community college district) would sell bonds to build new buildings backed up by its promise to pay them back out of its local tax revenue. A rating agency would have given this a rating of A or so based on the unlikeliness of default, but the issuer would then get the rating upgraded by seeking insurance -- in the event of a local problem creating default, the insurer would make good on the bonds. This backing decreased probability of default sufficiently to raise the bond rating to AAA, and so they could be sold for more (i.e., at a lower interest rate), with a small payment for the insurance easily made up for by the higher price.
Insurance as a risk management tool relies on spreading risk far enough so that it touches enough deep pockets to be covered. To achieve this, the risk is diced up into many small pieces and combined with other small pieces to create a less-risky whole. But this does not work as intended when the small risks are not independent of one another; pooling house insurance contracts from one coastal city in Florida doesn't much mitigate the risk since a large number of the insured houses can be trashed because of one hurricane.
The bond insurers began to go astray when they started to accept contracts to insure pools of mortgages and commercial loans, where the possibility of recession or declining house prices would affect many of the loans at the same time. While the good times rolled, competitive pressures and hunger for profits led them down a path of taking too little in premiums for accepting correlated risks. Now the bond insurers themselves are in deep trouble, having to acknowledge that the losses they might face are vastly larger than the capital they have to cover them. As a result, their own credit ratings are being downgraded and every issue they cover looks riskier, too. A downgrading of a large fraction of all outstanding debt as the market discovers suddenly that pooling and insuring risks doesn't make them go away is occurring.
Remember the banks? Having spread new credit far and wide based on their supposedly AAA-rated assets, they now discover that the downgrading of much of their portfolios of assets means they must reduce lending abruptly or find new capital to meet the regulations. Major banks are selling pieces of themselves to new investors (many of them foreign) to try to meet these capital needs. The Fed and other central banks are trying to unfreeze the flow of credit by loaning billions to the banks to get them through the crisis.
As in the dot-com bust of 2000, the central banks may succeed in pushing enough money into the economy to pump up some sectors to make up for the sudden collapse in values in another; the Fed at that time reduced short-term rates enough to start the creation of the housing bubble, and enough people found work in building-related trades to make up for the lost jobs in technology.
So despite what you might glean from the babble of news and candidate debates, the real hazards to the system now are far more serious than just a few million stupid or gullible investors and homeowners who got loans they shouldn't have; we are crossing a chasm that if poorly handled could lead to a sharp contraction in the world economy, leading to more pressure to de-globalize, which if politicians bow to it would lead to worldwide depression. The chance of this is not large, but it's there.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2007-12-21 09:25 pm (UTC);)
no subject
Date: 2007-12-22 05:27 am (UTC)no subject
Date: 2007-12-21 10:55 pm (UTC)no subject
Date: 2007-12-22 05:31 am (UTC)no subject
Date: 2007-12-22 08:44 am (UTC)no subject
Date: 2007-12-22 08:56 am (UTC)This is similar to and affected by the globalization juggernaut which can be seen as increasing economies of scale and
making labor ever cheaper but dislocating highly paid blue and white collar industrial country workers. The trade-off is cheaper goods for everyone, but those who lose their jobs get a double-whammy of reduced ability to buy even the cheaper goods in an ever increasing spiral of displaced poor workers.
So housing and labor are trying to seek their level of worth on a global scale. The economic forces are too new and too complex to model well at the current time. Maybe this will change. I hope that the dislocations will not be too severe but that the ultimate result is more people prospering the this economy.