The credit crisis has had a seismic impact on the global financial landscape. A central reason is an unforeseen interaction of fair value accounting rules and the illiquidity of structured credit products. Firms are required to mark to market their large portfolios of structured credit products, but for most of these products there is no longer any functioning market. Instead, as has happened on several occasions, we observe cumulative death spirals, in which the only prices that can be used for valuation are based on the fear that a Lehman or similar large firm will implode, and thus forced to sell a large structured credit portfolio. Mounting fears bring about large mark to market losses, increase counterparty margins, and bring about the very implosion that is feared in the first place.
Clearly recapitalisation of the banking sector is needed. But this malign interaction of accounting rules and illiquidity actually creates an opportunity, making it possible to substantially recapitalise the banking sector at little cost either to taxpayers or to bank shareholders. We believe this can be achieved through a direct solution to the trading paralysis, the setting up of a specialised exchange to allow a large number of traders to exchange information and views on the value of the underlying structured credit products and conduct trades at fundamental not fire sale prices. Prices established in such an exchange will provide more reliable valuations than those emerging from current bilateral over the counter trading and restore much of the banking net worth that has vanished in recent months.
Liquid markets are an efficient mechanism for assessing the average value across market participants. But market liquidity is a network good, without participation there is no longer effective aggregation and communication of information from a large number of sources and it is then no longer worthwhile for buyers and sellers to conduct trade: for this reason market liquidity is fragile and can completely disappear. How then should one infer value in the almost deserted market for structured credit products, where buying and selling is a relatively rare event?
When market participants do not trade, one can safely conclude that many potential sellers feel their own valuation is above current observed prices and thus refrain from trading. This is particularly true for illiquid markets (take the property markets in the UK and the US which have come to a standstill because many property owners wait for prices to recover). Those who do trade, typically at a low price, often do so because they have not much choice owing, for example, to bankruptcy. But there are many other potential sellers who stay out of the market because prices are well below what they regard as the intrinsic value of their holdings. Because prices in illiquid markets do not reflect the reservation price of those who do not trade, fair value accounting fails to reflect the valuation held by a large number of agents.
The solution is to engineer an improved flow of information between market participants regarding their internal assessment of the value of financial instrument, and provide the opportunity for trading based on these valuations, thus establishing higher prices that are less susceptible to temporary liquidity panic. Specifically we propose the establishment of a specialised trading venue, a central credit exchange with strict disclosure and transaction rules, in which participating banks will commit to place a small fraction of their structured credit portfolios and provide a range of value estimates corresponding to these assets. Trading can then be based on an auction mechanism that ensures honest disclosures and efficient prices. The resulting prices can be a much better basis for the fair valuation of traded structured credit instruments, balance sheets will then be repaired and financial stability restored.
Such a trading venue, while it will take some time to establish, has the potential to add about $300bn to the net worth of the global banking sector, without cost to either taxpayers or investors. This magnitude of this number may come as a surprise. It is based on the preliminary factual findings from our ongoing research on the accounting valuation of structured credit products.
(1) Structured credit securities are divided into good quality senior tranches and lower quality mezzanine and equity tranches. The policy of investment banks and commercial banks has been to retain most of the senior tranches (the best quality AAA and super-senior i.e. better than AAA). We estimate that banks are holding about $3-$3.5 trillion of these best quality structured credit assets (equivalent to about 25% of US GNP). [*]
(2) Prior to August 9th 2007, these securities traded regularly at prices that depended on their credit ratings and little else. All traded “par-par-par” and offered yields typically floating rate LIBOR + 30 basis points.
Since August 9th 2007, trading of all these better quality structured credit securities has effectively ceased, apart from a small amount of distressed trading at prices of between 10% and 30% below par. These low prices – often much below their intrinsic (long-term) value – are then the basis for recognition of losses under mark-to-market accounting in the balance sheets of other institutions holding similar investments. Those bank portfolios of these assets which are classified as “trading” or “available for sale” have had to be marked down accordingly, creating massive declines in bank capital, in the region of $300bn or more. [**]
These distressed prices have also been very volatile, creating an additional counterparty problem in repo trades. Specifically, the uncertainty in the valuation translated into both higher interest charges and increased ‘haircuts’ for loans secured against these senior/ super senior structured securities. This, in turn, further reduced liquidity in some cases making refinancing almost impossible.
Thus, a key point which many commentators failed to appreciate, it is the combination of credit losses on low quality structured credits and market to market ‘liquidity losses’ on senior structured credits that has been pushing firms to the wall. The liquidity losses are only temporary, if firms avoid bankruptcy then the value of senior securities will come back to par or very close to par, either when they mature or when the world economy begins to come out of recession. Easier therefore to address the liquidity losses, not the credit losses.
Sub-prime mortgage lending is only about one-third of the overall problem. If there is a substantial global downturn then there will be further underlying credit losses and liquidity losses on leveraged loans, commercial mortgages, CDOs built on corporate bonds and ‘synthetic’ CDOs. The situation will deteriorate further if the lack of liquidity is not addressed.
The rules of the trading venue require elaboration and discussion, but will be something as follows:
All participating banks agree to put forward AAA and super senior securities regularly from their structured credit portfolios on an agreed scale (e.g. $10million per week, enough to hurt, not enough to be strategically central) for sale on the London central credit exchange (such sales might start on a weekly basis, moving later to a daily basis).
In addition to placing a security, the placing bank will provide own estimate of the intrinsic value (or more likely a valuation range since point estimates are less reliable), possibly with some independent checking to ensure it is not set too high
The exchange will initially operate as an auction. The lower end of the intrinsic valuation range made by the placer of the security will be a reservation price, no sales will take place at lower than that value.
All bank members will be obliged to either submit bids, or record that their valuations are below the reservation price. Some rounds of bidding may be needed to establish a final selling price.
Governments and long term investment funds, including sovereign wealth funds will be able to bid on an optional basis. So government can use this auction to buy and keep prices strong and non-banks will have an opportunity to get hold of undervalued senior credit assets for their own portfolios.
There will be detailed disclosure of both the underlying collateral performance and the auction bids for these securities. All this information will be in the public domain (unlike the current private OTC markets). Thus even a small auction can provide key information to banks worldwide to help them more accurately value their portfolios. The auctions will start initially with those securities where underlying collateral performance can be most easily summarized.
For such an arrangement to have an impact, the key issue is not the quantity of trading but the dissemination of information. Banks worldwide should be able to value their own portfolio by referring to prices on this central credit exchange.
The main objection to this proposal is that structured credits may be too heterogeneous for successful centralized exchange trading of this kind. As a result large information asymmetries and price discounts may persist. To address this heterogeneity a large analytical effort is required, to standardize the disclosures of underlying collateral performance and to allow one security to be compared with another. Our early research findings, especially the evidence of a widespread view amongst structured credit professionals that the senior securities are now grossly undervalued, helps us believe that these information problems are not so large as to prevent exchange trading establishing sensible pricing. In any case it is clear that the effort is well worth making.
Our proposed central credit exchange is a complement, not a replacement, for the Paulson plan now before Congress, for a $700 trillion fund that will purchase sub-prime mortgage related assets from US banks and for other proposed measures to recapitalize the banking system. The focus of the Paulson proposal remains rather unclear. Initial press reports suggested that it would purchase the lowest quality assets, hence compensating banks for their underlying credit losses. However chairman Bernanke’s testimony to congress of 23rd September, 2008, suggests that the fund will be buying better quality structured credit assets, at a values close to par, and should make a profit when these are held to maturity i.e. in Bernanke’s view the fund will tackle the same illiquidity problem as we seek to address. However the fund will only tackle mortgage related debt, not the severe and growing problem of corporate structured credits. The Paulson fund also faces the problem of establishing appropriate purchase prices. We envisage that the Paulson fund will be a major buyer of credit related assets on our proposed central credit exchange and our exchange will help the fund establish appropriate purchase prices.
Gilad Livne is associate professor of accounting and Alistair Milne associate professor of banking and finance at Cass Business School, City University, London.