Credit crisis needs a Churchill
A bulldog spirit is required, says Hong Kong hedge fund manager Paul Sheehan, who puts forward a proposal for addressing the credit crisis.
Paul Sheehan is CEO of Hong Kong hedge fund Thaddeus Capital. He arrived in the alternatives world with a unique background, having in the past worked as a bank examiner at the Federal Reserve in the US, followed by stints as a bank equity analyst at the ill-fated firms Bear Stearns and Lehman Brothers.
Here he presents his manifesto for resolution of the credit crisis.
Letter to America: Action this day
Winston Churchill, upon assuming office as British Prime Minister at the beginning of World War II, would prod a slow-moving government corpus with demands for ôAction This Dayö, the order that was attached to his most urgent missives.
Observing the collective market reaction to the policies of the US government in its attempts to resolve the current credit crisis, it would seem that a box of ChurchillÆs stickers, to be dispensed liberally, is in order.
The G7 and the IMF are anxiously petitioning America for further aggressive action to stem the progress of the credit crisis which began in the US, and is now affecting the entire world.
As we in Asia are not only impacted by the crisis, but know a thing or two about banking crises ourselves, perhaps it is an appropriate time to return the favours of 1997-98 and advise the US on appropriate steps to take.
First, let us review the underlying problem, and the root causes of the crisis:
Credit crisis: Root causes
American consumers are broadly over-leveraged relative to income. This borrowing has been used to finance consumption, and is one of the reasons for the preternaturally-resilient American economy over the past ten years. The US economy is now dependent on consumer spending well above the normal trend as a percentage of GDP.
The major increase in borrowing has been via secured lending with residential property as collateral. This has been encouraged by government promotion of home ownership, regulatory capital calculations which assume only nominal loss on all first-lien mortgage lending, and ratings agency somnolence. It has also been embraced by borrowers themselves, accustomed by steadily-rising property prices to being able to refinance their way out of any problems servicing debt, and thus looking to maximise their gains by taking on as much property as they could possibly finance.
With little recent history of loss and the acquiescence of bank regulators, underwriting standards slipped greatly. At the peak of the property bubble it was possible for borrowers not only to make no down-payment when buying a new home, but actually to take money out immediately from the first mortgage, over and above the cost of the house. This new loan might not only have a low ôteaserö interest rate, but also negative-amortisation features which would cause the principal balance to actually rise each month for the first several years. By the time interest rates reset and the loans began amortising, both borrowers and lenders assumed that the collateral would be worth much more.
The chain of mortgage losses
As these badly originated mortgages season, they are defaulting at a high rate. In addition, negative-amortisation and/or low-teaser-rate loans will be resetting in large numbers through 2010, and are much more likely at that point to also begin defaulting. The initial collateral values backing these loans either were never accurate, or have since fallen substantially. Thus, mortgage-holders will incur substantial loss, even before accounting for the large costs of foreclosure and resale.
Sales or overhang of foreclosed property will further depress property prices. This effect by itself will lead to additional mortgage defaults, as consumers price the option of default very efficiently. Falling property prices and foreclosures will lead to less consumer spending, resulting in a potentially very severe recessionary cycle, as rising unemployment and depressed property prices are likely to lead to a second wave of delinquencies and defaults in late 2009-2010.
So far, the anticipated effects of these losses have been mainly felt via mark-to-market of securities and derivatives based on mortgages, which are held by trading entities or in the available-for-sale or trading books of commercial banks, where they must be marked-to-market, or in the absence of real prices, to model which is meant to approximate market. We have not yet seen - for the most part - the effects of losses on the raw underlying loans, which are generally held in accrual books and not marked-to-market. The capital support and pricing read-through from WaMu and Wachovia indicate that these losses are substantial, and they will be widespread as almost every US bank has exposure to mortgages.
Most institutions are therefore either holding mortgage-related assets at accrual price, or marking to model at unrealistically-high prices. How do we know they are too high? Banks and broker/dealers have an immense incentive to get these assets off their books, and if they could sell at the current (written-down) book values they would surely do so to eliminate investor and counterparty concerns.
There certainly exist investors who would buy these assets at some price and who have cash. As no one is selling, we can conclude that the prices are in general higher than true market. Anecdotal evidence, such as a view of the bids for Lehman assets in bankruptcy, supports this view.
Therefore, we can conclude that most financial institutions have undisclosed uncertain (and not easily calculable with public data) losses û hence the uncertainly of investors and counterparties. All we know is that the published balance sheets are wrong.
Why Tarp is not a solution
The Tarp (Troubled Asset Relief Program) will of course help to some extent, but it is inefficient and does not target the underlying problems of the economy and credit markets.
First, it is oriented, at least in its original conception, towards purchase of mortgage securities and mortgage derivatives. This will not help the underlying mortgage borrowers, and it is their failures which will contribute to the severity of recession, as well as causing the cascade of losses from raw loans up to securities and their derivatives. So, it might rescue some institutions, but they will be primarily the highly-levered trading ones rather than direct credit-extending ones. If the objective is to restart the flow of credit in the economy, bailing-out traders rather than lenders is less than optimal.
Secondly, the discussion around the price at which assets will be purchased has been highly disingenuous.
If the Tarp buys assets at true market value, the selling institutions will incur additional losses as these are very likely to be below their book value. The institutions most in need of assistance are the ones least able to absorb such losses ù or they would have already sold. If the Tarp buys at market value from healthy institutions, it will create reference transactions which will have to be used instead of mark-to-model at other institutions holding the same or similar assets, and by forcing the realisation of losses probably reveal insolvencies elsewhere.
Likewise, if the Tarp buys assets at book values, it will most reward those who have been least honest, and it will be buying assets for more than their worth to prop up distressed institutions. This will as above create false marks which will be used elsewhere, and thus perpetuate the uncertainty of true balance sheet values. If the desire is to effectively give away money to failing banks, it should be done openly and without creating a false market and its associated inefficiencies.
The meretricious idea that there is some magical ôheld-to-maturity valueö of the assets being purchased by the Tarp, which value is not evident to the market (including PIMCO, pension funds, endowments, insurance companies, and others with essentially permanent capital and long-term investment horizons), but which will be discoverable by Secretary Paulson or his designees, and which is enough above true current market value to prevent institutions which sell at such a price from incurring lethal losses, is ostensibly laughable.
What more can be done?
The financial sector has two separate but related problems stemming from the credit crisis: a solvency issue and a confidence issue. The steps thus far taken are intended to address the latter, but have not yet made any impact on the former, and thus in isolation are bound to fail. Four proposed steps which would collectively restore some order to the financial system:
Step 1: Stem financial sector panic
Interbank rates are essentially notional at this point, as no banks are lending to each other; instead, the Fed is a single counterparty to the entire industry. To combat this, the Fed should on a temporary basis explicitly guarantee all interbank lending with maturity of less than one year, as well as bank and bank holding company commercial paper. This would remove fear without much incremental cost, as the Fed is clearly not prepared to let any bank fail if it would take down other insured institutions as well.
Step 2: Quantify the losses
Uncertainty is the killer of financial markets, much more so than losses. Given that there are massive undisclosed losses within the financial system yet and that hardly anyoneÆs balance sheet is what it appears, it is perfectly rational for market participants to fear lending to each other. We must identify where the losses are in order to address them, resolve their owners if necessary, and remove suspicion from those who remain healthy. Banks must be forced or strongly encouraged to sell bad assets in the market in arms-length transactions, and to recognise the losses ù whatever their magnitude.
There is ample precedent for this from the S&L crisis, where banks were encouraged to sell their mortgage portfolios at market, even if they subsequently bought back similar assets in the market. I have no doubt but that the same might happen in this instance.
There will undoubtedly be resistance to selling at ôfire-sale pricesö, but unfortunately it is almost axiomatic that the bottom of the market will be where and when these holders sell û holding longer in hopes of a market upturn will only prolong the agony. This is validated not only by previous experience in the US, but by our own Asian crisis experiences in Thailand, Indonesia, Korea, Japan, and so forth. The requirement to sell at the bottom is also mitigated by the ability for banks to get out there in the market and buy the troubled assets of others as and when they see compelling value.
Step 3: Provide capital support
If institutions are going to be forced to potentially reveal their insolvency, the government must be prepared to correct it in some way, or no one will comply, as it would be corporate suicide. This can take the form of regulatory forbearance, as during the S&L crisis, where the government agrees to let banks take realised losses for tax purposes (and thus claim substantial refunds from prior year profits) but amortise the losses for regulatory capital purposes over 10 or 15 years.
In the current environment, forbearance alone is not likely to be sufficient, and so capital injections will be required as well. If this is done via preferred shares with potential conversion options or warrants (as was the case in Japan), or via a combination of common and preferred equity as the UK authorities have this week proposed, it should be sufficient.
Step 4: Address the underlying problem
All the while, these weak mortgages are still out there, and still defaulting. Propping up property prices is difficult (some wacky proposals include buying foreclosed houses en-masse and bulldozing them), and risks re-inflating the bubble. Buying raw loans at discounted prices from banks does not directly help the problem either.
As unpalatable as it is, the best solution is to recast mortgages to keep people in their houses. It is a distorting subsidy, but at least one which does not prop up property prices in general, and it would be a broad subsidy from taxpayers to taxpayers. With T-Bill rates at very low levels, the government could in effect pass along its cost of borrowing (via re-finance of mortgages via the GSEs or subsidy to existing mortgage holders) to borrowers on owner-occupied, first mortgages originated between, say, 1Q04 and 1Q08, for the next 10 years.
These rate-subsidised mortgages would not be transferable or assumable, so they would not stimulate anyone to go out and buy new houses. They do not cut the principal amount due, although that should also be considered when it is the best response, as encouraged by the Frank-Dodd housing bill last summer. Over time, as people move or refinance, all the loans will disappear or reset to market rates.
Even with these actions, the US will continue to see a high level of foreclosures and will most likely have to endure a stiff recession. Consumers will have to cut back, begin to save once more, and repair their own personal finances, even as banks do likewise. Trust in the financial system and its major actors will not fully return for a long time, if ever. However, if comprehensive action is quickly taken, perhaps we will be able to say that this is at least the end of the beginning, rather than the beginning of the end.
Here he presents his manifesto for resolution of the credit crisis.
Letter to America: Action this day
Winston Churchill, upon assuming office as British Prime Minister at the beginning of World War II, would prod a slow-moving government corpus with demands for ôAction This Dayö, the order that was attached to his most urgent missives.
Observing the collective market reaction to the policies of the US government in its attempts to resolve the current credit crisis, it would seem that a box of ChurchillÆs stickers, to be dispensed liberally, is in order.
The G7 and the IMF are anxiously petitioning America for further aggressive action to stem the progress of the credit crisis which began in the US, and is now affecting the entire world.
As we in Asia are not only impacted by the crisis, but know a thing or two about banking crises ourselves, perhaps it is an appropriate time to return the favours of 1997-98 and advise the US on appropriate steps to take.
First, let us review the underlying problem, and the root causes of the crisis:
Credit crisis: Root causes
American consumers are broadly over-leveraged relative to income. This borrowing has been used to finance consumption, and is one of the reasons for the preternaturally-resilient American economy over the past ten years. The US economy is now dependent on consumer spending well above the normal trend as a percentage of GDP.
The major increase in borrowing has been via secured lending with residential property as collateral. This has been encouraged by government promotion of home ownership, regulatory capital calculations which assume only nominal loss on all first-lien mortgage lending, and ratings agency somnolence. It has also been embraced by borrowers themselves, accustomed by steadily-rising property prices to being able to refinance their way out of any problems servicing debt, and thus looking to maximise their gains by taking on as much property as they could possibly finance.
With little recent history of loss and the acquiescence of bank regulators, underwriting standards slipped greatly. At the peak of the property bubble it was possible for borrowers not only to make no down-payment when buying a new home, but actually to take money out immediately from the first mortgage, over and above the cost of the house. This new loan might not only have a low ôteaserö interest rate, but also negative-amortisation features which would cause the principal balance to actually rise each month for the first several years. By the time interest rates reset and the loans began amortising, both borrowers and lenders assumed that the collateral would be worth much more.
The chain of mortgage losses
As these badly originated mortgages season, they are defaulting at a high rate. In addition, negative-amortisation and/or low-teaser-rate loans will be resetting in large numbers through 2010, and are much more likely at that point to also begin defaulting. The initial collateral values backing these loans either were never accurate, or have since fallen substantially. Thus, mortgage-holders will incur substantial loss, even before accounting for the large costs of foreclosure and resale.
Sales or overhang of foreclosed property will further depress property prices. This effect by itself will lead to additional mortgage defaults, as consumers price the option of default very efficiently. Falling property prices and foreclosures will lead to less consumer spending, resulting in a potentially very severe recessionary cycle, as rising unemployment and depressed property prices are likely to lead to a second wave of delinquencies and defaults in late 2009-2010.
So far, the anticipated effects of these losses have been mainly felt via mark-to-market of securities and derivatives based on mortgages, which are held by trading entities or in the available-for-sale or trading books of commercial banks, where they must be marked-to-market, or in the absence of real prices, to model which is meant to approximate market. We have not yet seen - for the most part - the effects of losses on the raw underlying loans, which are generally held in accrual books and not marked-to-market. The capital support and pricing read-through from WaMu and Wachovia indicate that these losses are substantial, and they will be widespread as almost every US bank has exposure to mortgages.
Most institutions are therefore either holding mortgage-related assets at accrual price, or marking to model at unrealistically-high prices. How do we know they are too high? Banks and broker/dealers have an immense incentive to get these assets off their books, and if they could sell at the current (written-down) book values they would surely do so to eliminate investor and counterparty concerns.
There certainly exist investors who would buy these assets at some price and who have cash. As no one is selling, we can conclude that the prices are in general higher than true market. Anecdotal evidence, such as a view of the bids for Lehman assets in bankruptcy, supports this view.
Therefore, we can conclude that most financial institutions have undisclosed uncertain (and not easily calculable with public data) losses û hence the uncertainly of investors and counterparties. All we know is that the published balance sheets are wrong.
Why Tarp is not a solution
The Tarp (Troubled Asset Relief Program) will of course help to some extent, but it is inefficient and does not target the underlying problems of the economy and credit markets.
First, it is oriented, at least in its original conception, towards purchase of mortgage securities and mortgage derivatives. This will not help the underlying mortgage borrowers, and it is their failures which will contribute to the severity of recession, as well as causing the cascade of losses from raw loans up to securities and their derivatives. So, it might rescue some institutions, but they will be primarily the highly-levered trading ones rather than direct credit-extending ones. If the objective is to restart the flow of credit in the economy, bailing-out traders rather than lenders is less than optimal.
Secondly, the discussion around the price at which assets will be purchased has been highly disingenuous.
If the Tarp buys assets at true market value, the selling institutions will incur additional losses as these are very likely to be below their book value. The institutions most in need of assistance are the ones least able to absorb such losses ù or they would have already sold. If the Tarp buys at market value from healthy institutions, it will create reference transactions which will have to be used instead of mark-to-model at other institutions holding the same or similar assets, and by forcing the realisation of losses probably reveal insolvencies elsewhere.
Likewise, if the Tarp buys assets at book values, it will most reward those who have been least honest, and it will be buying assets for more than their worth to prop up distressed institutions. This will as above create false marks which will be used elsewhere, and thus perpetuate the uncertainty of true balance sheet values. If the desire is to effectively give away money to failing banks, it should be done openly and without creating a false market and its associated inefficiencies.
The meretricious idea that there is some magical ôheld-to-maturity valueö of the assets being purchased by the Tarp, which value is not evident to the market (including PIMCO, pension funds, endowments, insurance companies, and others with essentially permanent capital and long-term investment horizons), but which will be discoverable by Secretary Paulson or his designees, and which is enough above true current market value to prevent institutions which sell at such a price from incurring lethal losses, is ostensibly laughable.
What more can be done?
The financial sector has two separate but related problems stemming from the credit crisis: a solvency issue and a confidence issue. The steps thus far taken are intended to address the latter, but have not yet made any impact on the former, and thus in isolation are bound to fail. Four proposed steps which would collectively restore some order to the financial system:
Step 1: Stem financial sector panic
Interbank rates are essentially notional at this point, as no banks are lending to each other; instead, the Fed is a single counterparty to the entire industry. To combat this, the Fed should on a temporary basis explicitly guarantee all interbank lending with maturity of less than one year, as well as bank and bank holding company commercial paper. This would remove fear without much incremental cost, as the Fed is clearly not prepared to let any bank fail if it would take down other insured institutions as well.
Step 2: Quantify the losses
Uncertainty is the killer of financial markets, much more so than losses. Given that there are massive undisclosed losses within the financial system yet and that hardly anyoneÆs balance sheet is what it appears, it is perfectly rational for market participants to fear lending to each other. We must identify where the losses are in order to address them, resolve their owners if necessary, and remove suspicion from those who remain healthy. Banks must be forced or strongly encouraged to sell bad assets in the market in arms-length transactions, and to recognise the losses ù whatever their magnitude.
There is ample precedent for this from the S&L crisis, where banks were encouraged to sell their mortgage portfolios at market, even if they subsequently bought back similar assets in the market. I have no doubt but that the same might happen in this instance.
There will undoubtedly be resistance to selling at ôfire-sale pricesö, but unfortunately it is almost axiomatic that the bottom of the market will be where and when these holders sell û holding longer in hopes of a market upturn will only prolong the agony. This is validated not only by previous experience in the US, but by our own Asian crisis experiences in Thailand, Indonesia, Korea, Japan, and so forth. The requirement to sell at the bottom is also mitigated by the ability for banks to get out there in the market and buy the troubled assets of others as and when they see compelling value.
Step 3: Provide capital support
If institutions are going to be forced to potentially reveal their insolvency, the government must be prepared to correct it in some way, or no one will comply, as it would be corporate suicide. This can take the form of regulatory forbearance, as during the S&L crisis, where the government agrees to let banks take realised losses for tax purposes (and thus claim substantial refunds from prior year profits) but amortise the losses for regulatory capital purposes over 10 or 15 years.
In the current environment, forbearance alone is not likely to be sufficient, and so capital injections will be required as well. If this is done via preferred shares with potential conversion options or warrants (as was the case in Japan), or via a combination of common and preferred equity as the UK authorities have this week proposed, it should be sufficient.
Step 4: Address the underlying problem
All the while, these weak mortgages are still out there, and still defaulting. Propping up property prices is difficult (some wacky proposals include buying foreclosed houses en-masse and bulldozing them), and risks re-inflating the bubble. Buying raw loans at discounted prices from banks does not directly help the problem either.
As unpalatable as it is, the best solution is to recast mortgages to keep people in their houses. It is a distorting subsidy, but at least one which does not prop up property prices in general, and it would be a broad subsidy from taxpayers to taxpayers. With T-Bill rates at very low levels, the government could in effect pass along its cost of borrowing (via re-finance of mortgages via the GSEs or subsidy to existing mortgage holders) to borrowers on owner-occupied, first mortgages originated between, say, 1Q04 and 1Q08, for the next 10 years.
These rate-subsidised mortgages would not be transferable or assumable, so they would not stimulate anyone to go out and buy new houses. They do not cut the principal amount due, although that should also be considered when it is the best response, as encouraged by the Frank-Dodd housing bill last summer. Over time, as people move or refinance, all the loans will disappear or reset to market rates.
Even with these actions, the US will continue to see a high level of foreclosures and will most likely have to endure a stiff recession. Consumers will have to cut back, begin to save once more, and repair their own personal finances, even as banks do likewise. Trust in the financial system and its major actors will not fully return for a long time, if ever. However, if comprehensive action is quickly taken, perhaps we will be able to say that this is at least the end of the beginning, rather than the beginning of the end.
¬ Haymarket Media Limited. All rights reserved.