The Paulson Asset Purchase Program is a done deal. It’s final form will likely be approved on Thursday, September 25, 2008, the final day of this session of Congress. Some details of the plan must be negotiated, and Congress is generating a lot of heat because their constituents are furious, but an acceptance of the core plan is an absolute certainty.
The events leading up to the crisis that made this program necessary are described in the post below this so will not be repeated here.
Whether this plan will work, or the degree to which it will work effectively, is still subject to question and will continue to be long after the law is passed.
The plan will certainly avert a near-term continuation of the grave credit liquidity crisis that occurred last week, and that by itself is a strong positive. That crisis was as serious as is being described to Congress yesterday and today in hearings on this bill. A destructive global financial collapse was imminent on Wednesday of last week..
As the details of the plan are being debated in Congress, one serious tradeoff with both political and economic dimensions has emerged and is driving the debate on the final configuration of the plan.
The two parties in the tradeoff are the banks and other financial institutions who now own the bad debt versus taxpayers. These financial institutions will hereafter be called “banks,” but keep in mind that commercial banks in the United States hold only about $2 trillion worth of the approximately $10 trillions of U.S. mortgages outstanding. The rest is held by Fannie Mae and Freddie Mac, investment banks, foreign banks, and other players who are not commercial banks.
As will be explained below the outcome of the tradeoff will be set by a single, simple array of variables – the prices set for the assets that will be purchased by the government because of this plan. That same issue, pricing, will also determine the long-run effectiveness of this program and, to some extent, the probability of the crisis revisiting in another form in the future.
The Political Dimensions of this Plan
The timing of the crisis and hence the articulation of this plan is unfortunate. This has emerged just six weeks before a presidential election and only a few days before Congress is scheduled to adjourn (they don’t convene again until January, 2009). The Congressmen who are shaping this plan in cooperation with Federal Reserve Chairman Ben Bernanke and Treasury Secretary Henry Paulson are gravely concerned about the outrage that they are hearing from their constituencies. All members of The House are standing for election as are one-third of the Senators. They know that if the final version of the plan is perceived as an expensive taxpayer bailout of recklessly irresponsible banks, including foreign banks, the political cost will be high.
This political reality is clearly influencing the emerging details of this plan. Various attachments are being added to the plan, such as a vaguely defined component that promises to “help” homeowners with these troubled mortgages and a punitive provision that limits severance packages (“golden parachutes”) for fired bank executives.
Congress will almost certainly insist that the purchase plan be rolled out in installments over time, providing more interim oversight.
These secondary issues will not be addressed in this post.
More important, as be seen yesterday and today on the televised hearings on this bill, this political reality is firmly impacting the allowable methods for pricing the troubled assets that are purchased by the new federal agency.
As will be shown below, whether this turns out to be an expensive taxpayer-funded bailout of this banks will be determined almost entirely by the prices set for the purchased assets.
The Essential Features of the Asset Purchase Program
The new law will authorize the creation of a chartered federal (government-owned) corporation similar to the Resolution Trust Corporation created during the Savings and Loan Bailout. For the sake of reference this corporation will be called the Asset Purchase Corporation (APC).
The APC will be given the right to directly or indirectly (through the U.S. Treasury) issue up to $700 billion of debt instruments, which will almost certainly be a mix of bills, notes, and bonds similar in maturity to those issued by the U.S. Treasury presently. It is possible that the U.S. Treasury will sell a mix of conventional Treasury bills, notes and bonds and release the proceeds to the APC.
The APC will then use the proceeds to buy the Collaterized Mortgage Obligations (CMOs) from the banks who now own them, on a voluntary basis. The CMOs will thereafter be owned by the APC (i.e. the U.S. Government) who will be entitled to all cash principal and interest payments from the active mortgages in the pool, but will also absorb the default loss from the same pool of mortgages. The banks will be permanently rid of these troubled assets and will gain some cash to improve their liquidity. More important, they won’t have to continue the quarterly “mark to market” required by the Sarbanes-Oxley Act of these troubling assets nor will they have to keep drawing heavily from the interbank lending market to continue their operations (the two provisions that provoked this crisis).
From these details is should be apparent that the final cost to taxpayers will be nowhere near the $700 billion funding authorization. The money will be used to acquire assets which have value and promise considerable future cashflow to the APC. The cost to taxpayers (in the form of debt service and debt redemption of course) will largely be determined by (1) the interest costs of the securities issued by the APC and (2) the spread between the cost of buying these assets and the present discounted value of the principal and interest payments of the performing mortgages in the CMO portfolios.
The value of the latter, however, cannot now be determined with any confidence. That value will largely be determined by future mortgage default rates and no one is sure of what those will be, Statistically it is a random variable with a very wide and changing dispersion. In fact, its dispersion may become a function of the outcome of this bill.
Given this, the cost to taxpayers is indeterminate right now. It could be zero or even slightly profitable (not likely). But no present estimate, unless it has a very wide range, is now credible.
The outcome will also depend on how these CMOs are priced when they are purchased.
Pricing
This crisis emerged largely because banks are now required under the Sarbanes-Oxley Act to declare any reassessment of the value of their assets in their quartely earnings statements. In the jargon of the financial media, they are required to “mark to market” the value of their assets. This writeoff reduces shareholder equity and in the case of commercial banks lowers their capital reserves.
Because default rates are high and rising on the mortgages in the low-quality CMOs owned by these banks, these banks have been required to recalculate again and again the value of their CMO portfolios relative to their original notional value and then make those revaluations public. Not only does this depress stock market values and reduce capital reserves for banks, it also makes it harder for banks to borrow in the interbank lending market and other sources for short-term funds.
When Lehman Brothers failed, part of that failure was due to the fact that they had marked to market their CMOs to only 39% of their value. In effect, that would mean that a CMO originally valued at $10 billion would thereafter be perceived to have a value of less than $4 billion.
Because these markets are now illiquid these marked-to-market values are the closest approximations to the true value of these portfolios. They are what economists call “shadow prices.” But they may be biased prices and they may not accurately reflect the true long-run value of these CMO portfolios.
From this discussion it should be clear that the “price” of any CMO being referred to here is a percentage of the original notional value of the CMO.
To say that the dispute is over pricing of these CMOs, the reference is the percent of original notional value that the APC will pay for the purchase of these assets. Should they pay 40% or 60% or 52%?
Each CMO will have a different price because each CMO has a different risk profile and default history.
That begs the question about how the price should be set. Generally, the lower the price the less the benefit to the bank but the higher the price the higher the likely cost to taxpayers.
If the CMOs are priced at a level close to the level recently implied by marking them to market, the purchases will not help banks very much, if at all. For one thing they would effectively converted a accounting loss not yet realized (and a loss from which there may be some recovery in the future) into a permanent realized loss. If the bank is crippled at present mark-to-market values, it will remain crippled after the sale if the price is close to that same value.
But if the current mark-to-market value is a reasonably good estimate of the true discounted cashflow value of the CMO, any price above that will entail taxpayer cost, and the higher the price, the higher the cost.
Clearly, the pricing mechanism is a critical component of this asset purchase plan.
Proposals for Auction
Because of the pricing issue, in his testimony before Congress on Tuesday, FRS Chairman Ben Bernanke suggested that some of the assets can be acquired through a reverse auction.
In a reverse auction the selling bank would begin the process by tendering an “offer” to that market. Effectively the bank might say, “We are willing to sell this at a price of 80%.” Ideally that would trigger a bid from an interested buyer at a lower price, like 60%. In a liquid and orderly market offer and bid should converge until a price is settled and the purchase would proceed at that price.
Selling through a Dutch Auction is also a possibly (a procedure too complicated to describe here).
But neither procedure would work for a project of this size undertaken so quickly. Such markets work only if there are a lot of players and there is confidence in the market. Since most banks are precluded from the bid side (if they could buy these they don’t need help) and they would normally be a large part of this market, buyers with adequate capital to participate are restricted to sovereign investment funds and large hedge funds and a handful of large private and corporate buyes. They might dabble but would never shoulder the entire risk. Far more important, the program implies that only the U.S. government would be on the bid side.
Auctions don’t work between two players.
Because of that Bernanke is actually proposing a “trial” or sample market, where this would be done (offered to the private markets) for a small subset of these CMOs, resulting in a shadow price that can be applied to the remainder of the CMOs.
That won’t work either. The liquidity issue still exists for even a small sub-set of auctions. Far more important, if the small auction results in a shadow price that would be applied to later purchases of CMOs, there is a huge incentive for the private players impacted by the pricing to game that auction, even if results in a CMO purchase that is anticipated to produce a loss. For example, if a $1 billion auction is meant to establish a price for a $100 billion CMO purchase, far better that the price be 76% than 52%, even if it produces a $200 million loss on auctioned CMO.
The price issue is therefore completely unresolved.
[To be continued, either by amending this post or in new posts].
Posts
No comments:
Post a Comment