Look, Booboo, no crisis

Naked Capitalism makes a very good point about the (relative) lack of a catastrophe in the ABCP market recently:

Many mortgage-related ABCP issuers have gone to a lender of last resort, namely the Federal Home Loan Banks, which have extended $163 billion of loans to them.

(The FHLBs are commonly thought of as equivalent to U.S. government risk, although they do not have an explicit government guarantee.) NC then references a Bloomberg article, quoting

Countrywide Financial Corp., Washington Mutual Inc., Hudson City Bancorp Inc. and hundreds of other lenders borrowed a record $163 billion from the 12 Federal Home Loan Banks in August and September as interest rates on asset-backed commercial paper rose as high as 5.6 percent. The government-sponsored companies were able to make loans at about 4.9 percent, saving the private banks about $1 billion in annual interest.

To meet the sudden demand, the institutions sold $143 billion of short-term debt in August and September, according to the FHLBs' Office of Finance. The sales pushed outstanding debt up 21 percent to a record $1.15 trillion, an amount that may become a burden to U.S. taxpayers because almost half comes due before 2009.

[...]

The home loan banks ``were the only game in town for a lot of borrowers,'' said Jim Vogel, head of agency debt research at FTN Financial a securities firm in Memphis, Tennessee. They are ``like an old watch your grandfather left you years ago, and you pull it out of the drawer and find it's the only timepiece you have.''

This is really scary. The FHLBs are lightly regulated (laws tightening the regulatory framework around them are currently stalled in the Senate), were one of them to fail or come close to failure, there would be enormous pressure for a government bail-out, yet their risk management practices are hardly likely to be in the same category as a Goldman Sachs. Avoiding a market crisis this way may just be storing up trouble for the future. Gangway, gangway, I want to get off this particular boat.

Black box trading and information fusion

According to Wikipedia, Information Fusion

refers to the field of study of techniques attempting to merge information from disparate sources despite differing conceptual, contextual and typographical representations

The convention is to keep the term data fusion for the situation where all information is quantitative, and use information fusion for the broader problem of integrating quantitative and qualitative data.
Another authority says that data fusion

takes isolated pieces of sensor output and turns them into a situation picture: a human-understandable representation of the objects in the world that created those sensor outputs.

Basically then, whenever you have diverse data which you have to try to turn into a coherent picture, you are performing data or information fusion.

Unsurprisingly much of the academic interest in this area occurs in limited problem domains: figuring out where the planes are from radar and visual data, for instance, or combining multiple different sonar sources to get a more complete picture of what's swimming around you. Many quantitative trading models of this class: they take feeds of market data and transactions and attempt to form a picture of where the market will go next. One simpler class of models, for instance, are basically trend followers. Often the idea of momentum is used: when markets are rising on increasing volume with low volatility, the models pile in, perhaps intensifying the rise. Decreasing volumes and/or rising volatility are sometimes used as triggers to reduce the size of the trade.

Many quantitative models, then, implicitly have a confidence estimate built in. When they strongly believe in their own predictions, they put a trade on. When they either don't believe in them, or they cannot make a prediction, the trade is taken off.

This feature is important: quantitative trading has been described as picking up pennies in from of a steam roller, and certainly many trading strategies act like short gamma positions, making a little money when they work, but losing a great deal when they are wrong. A false positive - a trade that you don't think will work and so don't make but in fact would have been profitable - is a lot less bad than a false negative - a trade you do think will work but turns out not to. The magnitude of this issue can be seen from Morgan Stanley's $480M one day quant trading loss.

For this reason, some quant traders use multiple models and only trade when all of them are giving the same signal. If the models are sufficiently different and do not share common assumptions this helps to reduce model risk.

It occurred to me recently that another approach might be to cast quantitative trading as an information fusion rather than a data fusion problem. That is, is there non-quantitative information that might be useful, in particular in avoiding false negatives by making the model more doubtful in situations where more care is needed? One of the anticedants here is the theory of prediction markets: when a large number of independent people have an opinion, a suitable weighting strategy can often lead to better predictions than any individual pundit. Note that I am not discussing analysts opinions here - there are clearly institutional biases at work there, and the history of collective analyst predictions is not that promising. Rather I am suggesting trying to use the commentariat, ideally as large a body of it as possible, as a signal akin to rising volatility. When enough blogs start to discuss a possible crash, that is a sell signal akin to rising volatility or rising market risk premiums. Such an information fusion based quantitative trading model would be of more use in global macro than in very short term applications like index arb, but the idea of using rising worry as a deleveraging signal could be interesting. Or it could just be a heap of potatoes.

The failure of a monoline

One of the many threats to the ABS market at the moment is the perceived decline in the credit worthiness of the large bond insurers, aka the monolines. There are a small number of these firms, and most of them carry either a AAA or AA credit rating. They are vital to the functioning of the ABS market in that they wrap bonds, providing a guarantee that if the underlying collateral does not pay timely interest and ultimate principal, then the monoline will. This solves the information problem on less well understood ABS collateral, and lowers funding costs for pools with non-standard characteristics.

Unfortunately the monolines have wrapped a significant amount of subprime collateral. This is causing a drag on earnings with both AMBAC and MBIA reporting losses last week. Investors are concerned that this is only the beginning and have bid up default protection on the monolines. Five year CDS spreads have gone from a few tens of basis points to 300 for AMBAC and 200 for MBIA.

The monolines typically have very diverse pools of risk that they have wrapped, considerable claims paying ability, and fairly large capital bases. However they are also highly leveraged and, since much of their protection is legally insurance, they do not mark all of it to market, and even where they do have MTM instruments, these are often sufficiently illiquid that they are marked to model. They also engage in transactions other than public bond insurance, wrapping risk in private transactions and, in MBIA's case, managing a SIV. Given rising default and rising default correlations across ABS, then, investors are concerned that the monolines' capital models might not be robust and hence that their AAA status is questionable. Certainly were a monoline to fail the impact on the market would be very considerable. MBIA alone, for instance, from 2005 to mid 2007 insured $35B of bonds, their RMBS portfolio is over $45B, and their CDO book over $100B. The muni market would also suffer a massive case of illiquidity as investors scramble to understand bonds that previously had traded on the strength of their wraps.

Note finally that since the monolines are insurers, they are not regulated by the FED or the SEC. Any bail out would have to be coordinated between the insurance commissioners and the capital markets regulators. The threshold for 'default' is also higher, since a monoline cannot operate unless rated at least AA: a downgrade below that level would probably be taken by the market as more or less equivalent to failure. Meanwhile 300 over isn't a typical AA credit spread: the market evidently views a monoline failure as a real possibility.

Update. As at the 1st of November, according to the FT, the (AAA-rated) debt of Ambac and MBIA costs 345 bps and 310bps respectively to protect in the CDS market. XL Capital is trading at 445bp.

The limited role of transparency

Charlie McCreevy, EU internal market commissioner, has an interesting point about transparency in regulatory dealings. According to the FT, McCreevy will say that transparency has a useful role to play in dealing with opaque financial instruments but will add:

When the stability of a financial institution is at risk, the situation is best resolved behind closed doors. Unfortunately, in recent weeks, gold-plated transparency rules stood in the way of the quiet resolution of a problem before it became a crisis.

The result was that transparency rules that were intended to underpin investor confidence, when put to the test, actually promoted investor panic. Panic that culminated in a bank run – averted only by a central bank lifeboat which in turn spread moral hazard throughout the system.

I think the language here is a little too careless: it is not obvious that these situations are always best resolved in the dark. It was helpful, for instance, to know about the FED's Section 23A loosening. But it does seem reasonable that they sometimes are, and that supervisors and central banks should be able to act covertly if they think best, subject to later judicial and democratic scrutiny, of course. Alastair Darling certainly considers it potentially useful for the Bank of England to be able to act in secret.

This all boils down to market manipulation. Is it OK for the regulator to manipulate sentiment (and almost certainly conduct an off-market transaction) for the long term greater good? Personally I think it is, but these should be rare events. If we slide closer to Italian dirigisme, that would not be helpful.

Snow down on the MLEC

John Snow - Hank Paulson's predecessor as Treasury secretary - isn't that keen on the MLEC either. According to Reuters:

We've got all this paper out in the system, and my inclination is to say, let's accelerate the price discovery process on this paper [...]

We know that when you prop things up artificially -- Japan -- we know when you prop things up artificially -- the (savings and loans) in the United States -- you get bigger adverse consequences

He has a good point. Warren Buffett is not keen either:

I think there should be a requirement that before the securities are put into the new super-SIV, 10 per cent of the holdings should be sold into the market to people who are not associated [with the subprime problem]

Certainly the history of intervention to prop up prices in the markets is not encouraging, and suspicions remain that the MLEC will not use the right mark. Perhaps it falls foul of the Market Abuse Directive? Oh, and the ABX is on the way down again. Here are the 07-02 BBBs:

The Rock as a SIV

FT alphaville has a nice post on Northern Rock, gathered from the Bank of England's latest Financial Stability Review. They show this:

That's massive asset growth funded by issueing securitised paper. Hmm, that sounds like a SIV to me. After all, a SIV is just a simple bank with no deposit funding. That's NR these days.

Countrywide and Convexity

The news that Countrywide is modifying a huge swathe of mortgages brings out the issue of servicer related convexity again.

Countrywide is going to modify two types of borrower according to Naked Capitalism:

Borrowers who are current now but look unable to cope with an upcoming reset. That is budgeted for $4 billion of loans

Delinquent borrowers. These will be targeted with a "predetermined, preapproved" reduction. This program is targeted to $2.2 billion of mortgages.

For Countrywide the aim of this programme appears to be to keep as many borrowers current as possible and so earn their servicing fee. But remember a mod will typically decrease cashflow. If these mortgages have been securitised (as is highly likely) that cashflow reduction flows through the waterfall and hits the tranches. In other words, the servicer's proactive attempt to protect their fees is modifying the cashflows on securities they likely do not own. Which tranches are effected depends on the precise securitisation structure and the history of the security (losses and prepays so far). Nonetheless it is not obvious that Countrywide have the interests of all security holders at heart in implementing this mod programme.

Conduits and Consolidation

Bloomberg has an interesting article Citigroup SIV Accounting Looks Tough to Defend. The author, Jonathan Weil, makes the point that Citi is caught between the devil and the deep blue sea:

• If it supports its SIVs, accounting consolidation kicks in since the firm is deemed to be providing implicit support and under FASB Interpretation No. 46(R) that brings them back on balance sheet.
  
• If it doesn't support its SIVs, the reputational damage will be severe.

Weil argues this is one of the reasons for the MLEC plan: it would allow Citi to dig some of SIVs out of the mud without forcing it to consolidate. This is a good point and it might well be true. But there is another dimension, too: regulatory consolidation. It is a little known (if deeply, deeply boring) point that regulatory consolidation is not the same as accounting consolidation.

Regulatory consolidation very roughly works by identifying which parts of the group are financial, putting all of their risk on the top company balance sheet, subtracting the equity invested in non-financial subsidiaries, and calculating regulatory capital on the result. Accounting consolidation again very roughly seeks to identify whether an entity which a firm has a relationship with is controlled by, owned by or gives the results of its activity to the firm. If so, that entity is consolidated. In detail the topic is complex, particularly in the context of firms which have both banks, insurance companies and other activities in a single group - financial conglomerates - or where careful structuring has been used to try to dodge the consolidation provisions of IAS 27, SIC 12 or whatever for a given securitisation SPV, SIV or conduit. For now, though, note that in principle at least an entity can be consolidation for accounting purposes but not regulatory purposes and vice versa. And that's fairly odd. I can see that it makes sense to deconsolidate non-financial subsidaries of a financial holding company for regulatory capital purposes. I can even see how you might want to treat the same risk in different operating companies differently - for instance if a holding company owns both a bank and an insurance company. But the idea that an entity such as a conduit is consolidated for accounting purposes but not for regulatory capital seems deeply imprudent. And of course many conduits have been structured to ensure that provided no implicit support is given they will not consolidate for either purpose.

It is worth noting that for FAS accounters, FIN 46(R) on the Consolidation of Variable Interest Entities includes provisions for the recognition of an implicit variable interest, aka implicit support. This occurs amongst other things when a bank provides a conduit with credit or liquidity support that it is not contractually obliged to provide.

That brings us to the heart of the dilemma. Implicit support is poison. Regulators and auditors can only spot it after it has happened. But if conduit paper has been sold with a nod and a wink, so that the buyers expect implicit support and will exact reputational damage if it is not provided, then there is something rotten in the state of finance.

Wile E. Remembered

The head of the IMF, Rodrigo Rato, is worried about a Wile E. Coyote moment for the dollar too. Speaking at a meeting in Washington he said:

There are risks that an abrupt fall in the dollar could either be triggered by, or itself trigger, a loss of confidence in dollar assets.

The uncertainty ... comes from downside risks that are much higher than they were six months ago. The turbulence in the credit markets is a warning that we cannot take the benign (global) economic environment of recent years for granted

When the party ends, someone is often found passed out in the loo. It looks like it might well be good old USD's turn for the thumping hangover this time.

Sliding SIeVes

In sharp contrast to some mild improvements in other parts of the credit markets, FT alpahville presents this enlightening chart, courtesy of Fitch. It shows how the average NAV of the SIV's assets has continued to slide.

The better players have been doing fine, but the worse ones are catching a severe cold, and the average is still declining. At this rate more triggers start to be hit, and forced liquidations are putting further pressure on prices. And that of course leads to more write downs. The rumor is that that $5B at Merrill is only the start for instance. We'll see...

Update. It's another $2.5B at Merrill, apparently. And that's not the end of it, obviously, as the markets have been so illiquid that it is likely that all the big players still have a significant part of their position. Unless the recent ABX kickup is the start of a trend, there will be more writedowns like this.

Section 23A of the Federal Reserve Act

Wow, that's an appetising title. This is not even by the lamentably low standards of this blog a sexy post, but the news is interesting. RGE via Naked Capitalism reports that the FED is permitting large banks to borrow money versus collateral and to lend it on to their affiliates. Section 23A refers to the intergroup lending: normally banks can't lend on large amounts of money to non-bank affiliates. This stops a group holding company using a bank within the group to access the FED window or raise deposits which are then lent on to a broker/dealer. As IRA put it:

Section 23A is one of the most important parts of the Federal Reserve Act. It prohibits "covered transactions" with any one affiliate of a Fed member bank in excess of 10% of the bank's capital and surplus, and up to 20% in aggregate for all bank affiliates. The purpose of the section is to protect the capital of the bank, even if that means allowing non-bank units or the parent holding company to be decapitalized or even fail in a "market resolution."

So far we know that the FED has granted Section 23A exemptions to Citi, JPMorgan, BofA, Barclays, RBS and Deutsche. This is potentially significant: it means that the FED is sufficiently worried about the consequences of the broker/dealers having to dump assets because they can't fund that they are willing to suspend a key regulation. At best it creates serious moral hazard, allowing firms to avoid the losses on dumping assets in the current market. At worst it means that if the problem gets really serious, the broker/dealers will bring the banks down.

The brotherhood of the MLEC

Strangely enough, some bankers are not happy with the idea of Citi et al. manipulating the ABS market by waving assets into the MLEC at the wrong mark. From the FT:

A committee of top international bankers on Sunday warned that the proposed $75bn mortgage securities superfund must be transparent in its pricing of assets if it is to help restore market confidence.

The statement by the Institute of International Finance reflected concern that the superfund, which is backed by Citigroup, JPMorgan Chase and Bank of America, is proposing to buy assets from troubled investment vehicles at higher than true market prices. Josef Ackermann, chief executive of Deutsche Bank and chairman of the IIF, stressed the importance of “proper valuations” and the need for financial institutions to “take the hits”.

One does wonder if the 'one conduit to rule them all' plan is going anywhere...

Keep bailing

Rich Bookstaber has a proposal:

[...] what if the government maintained a pool of capital on the ready to buy up assets of firms that are failing, much as Citadel did for Amaranth and Sowood? Of course, if a private entity is willing to step up to the plate, all the better. But as a last resort, what if the government took on the role that Citadel did in these instances. There would be no moral hazard problems, since the firm still fails. But the collateral damage would be contained; the market would be kept from going into crisis, the dominos would be kept from falling. And the taxpayer would have good odds of pocketing some profits.

That is all very well if the failing firm's assets were really purchased at a market clearing level. But determining that level is very difficult, and the value of a portfolio in a distressed firm (as doubtless Citadel knew) is rather lower than in a going concern. There would be enormous pressure on such a government fund to overbid giving a better outcome for the firm's shareholders but a worse one for the taxpayer. Also where is the government going to find the right kind of people to run its vulture fund? Great hedge fund traders tend not to be the type who want to be government employees. It's a superficially attractive idea but I suspect it isn't very practical. Mind you, the current asset dry dock MLEC doesn't look that practical either...

Does a steeper skew presage a crash?

The S&P skew has steepened significantly of late. Does that mean anything, other than more buyers of downside options? In particular is there any evidence that a steeper smile is associated with large moves down? In the other direction it works - if there is a big fall then vols rise and the skew steepens. But I know of no studies that suggest skew steepening is a useful predictor of falls. It would be interesting to know though... after all, we are close to the anniversary. Certainly it seems likely that with all the press coverage, the big 2 0 is spooking investors. But of course that doesn't mean we won't have a crash.

Not MBS? Wave it in...

From the FT:

Investors scrambled to buy a multimillion-dollar catastrophe bond on Wednesday as the market for natural disaster debt continued to boom.

“Investors like these bonds because of the high premiums and the fact there are very few pay-outs. It is also a great diversification play. These investments are not linked in any way to subprime. That also explains why the market has continued to boom despite the credit problems over the summer.”

This is a BB+ cat bond paying Libor plus 275, and people are waving it in because it is not a credit instrument and specifically because it is not a mortgage backed security. Gosh, that let's avoid the place the lightening struck last week and go and stand under a different tall tree during a storm strategy always works, doesn't it? The bond may or may not be a good deal but it has to be a cause for concern if it is being bought by people who are more interested in what it isn't than in what it is.

The illustration shows some underwriters' desks at Lloyds of London. Perhaps they have left the building thanks to the willingness of the capital markets to take their risk?

Two misconceptions about the credit crunch

Reading Naked Capitalism's discussion of an otherwise reasonably good article by Angel Ubide, it occurs to me to clear up two mistakes Angel makes.

Central banks have added liquidity to a situation of already "excess liquidity" to tackle an apparent liquidity crunch, and yet nothing has got better. Perhaps it was not about liquidity, after all.

This assumes all liquidity is the same, and it isn't. The reason the extra liquidity the central banks injected didn't help much was it was the wrong sort of liquidity: it did little to ease the difficulty of funding Libor-based assets as it didn't address soaring liquidity premiums. We did have a liquidity crisis, but it was one in the Libor markets. Central banks may wish to think hard about how to address this characteristically modern market stress.

What is the right response from a risk management standpoint to a sudden increase in balance sheet risk, volatility and uncertainty? Reduce positions dramatically [...]

This just turns an unrealised and uncertain loss into a certain realised one. It is exactly the wrong thing to do. Rather banks should (and one suspects from their results that Goldman did) have enough headroom in their risk appetite and enough spare funding so that they can buy cheap assets in this situation from the forced sellers. Risk management isn't always about reducing risk: it's about taking the right risks at the right price.

Update. The 'having funding' part above is important. It is estimated that banks have taken over $280B of assets on balance sheets from conduits and SIVs since the credit crisis began. For some banks that means they have to find a lot of cash before even thinking about funding new asset purchases. Maybe those undrawn lines were worth paying for after all.

What's in it for the Treasury?

The U.S. Treasury Department has been heavily involved in talks over the MLEC according to Bloomberg. Why, I wonder? It seems strange for a right wing Republican administration to interfere in free markets unless there is a very real systemic risk. Perhaps they know something about Citi we don't that is worrying them. In any event the more we find out about this attempt to bolster the system, the stranger it seems.

Update. Greenspan has been sticking the boot in again. According to the FT he opined:

What creates strong markets is a belief in the investment community that everybody has been scared out of the market, pressed prices too low and there are wildly attractive bargaining prices out there. [...] if you intervene in the system, the vultures stay away. The vultures are sometimes very useful.

This seems pretty reasonable. Providing an artificial mark via a non-open-market purchase of these assets into the MLEC, if that is indeed what is intended, will not be good for confidence.

Where's the mark, Chuck?

The importance of my earlier question about how the MLEC will price the assets it is going to purchase is highlighted by an article in the FT today:

[It has emerged that] Axon Financial, a SIV linked with the US hedge fund TPG-Axon, had taken losses of $110m on sales of $3bn of its investments.

Very crudely scaling from $110m on $3b to Citi's $100b in conduit assets gives us a loss of $3b. So we know a real mark to liquidation would really really hurt. The market is so illiquid at the moment that it's hard to be sure until you try to sell of course. So it should be no surprise to learn that many banks have not yet marked their conduit assets down:

One banker said last week: “The banks have varied enormously in terms of how much they have marked down their books – of course there are some that want to avoid the big write-downs.”

The MLEC then looks like an attempt at creating a new vehicle the banks can claim is arms length and which they can use to justify valuations which might not really be liquidation levels. That helps in turn helps the banks other conduits and the value of their on-balance sheet ABS. If you tell people it's worth par often enough, loudly enough, maybe it will be...

Incentive structures in capital estimation

A capital model creates an incentive structure: if a firm's estimate of the capital required to support its business rises, then that implies it is taking more risk. At some point increasing risk becomes unacceptable given the firm's desired soundness standard, and so positions are cut.

Recently there have been some discussion by Gillian Tett in the FT (quoted by Naked Capitalism) of this effect with regard to VAR models. The basic problem in VAR is that risk estimates can increase either because the portfolio has changed or because market volatilities and correlations have increased. Thus with a regularly updated VAR model the same portfolio in a crisis produces a higher capital charge and hence banks are incentivised to cut at the worst moment. Similarly risk estimates are lower in calm markets, encouraging banks to over-leverage.

The effect can be significant. As Ms. Tett points out

[The Bank of England] estimated that a typical bank’s VAR might theoretically double, with the same assets, if volatility increased.

A similar problem occurs in Basel 2 IRB models - in an economic downturn, banks' estimates of PD and LGD rise, increasing capital, and so discouraging lending. This may intensify the intensity and duration of the downturn.

The phenomenon is known as pro-cyclicality, and it is clearly undesireable. One problem here is that regulators have confused two different kinds of risk sensitivity. Clearly at any point in time having a larger capital estimate for a riskier portfolio than for a less risky one is a good thing: let us call this portfolio risk sensitivity. (Basel 2 doesn't completely satisfy this either, but we will ignore that for the moment.) Then there is temporal risk sensitivity: here the risk estimate of the same portfolio changes over time as market factors used as inputs to the capital model change. It is much less clear that complete temporal risk sensitivity is a good thing. Using long term average inputs to VAR or IRB models might produce better incentives than short term current market estimates. Such models would have the helpful (in a crisis) property of failing to respond quickly to changes in market conditions.

It might be argued that this means that banks are under-capitalised during tough markets. That would be a reasonable argument if VAR produced capital estimates which reflect possible losses in these markets - but it doesn't (and it was never designed to). One needs only examine Morgan Stanley's latest 10-Q to see the phenomenon: their VAR was very roughly $100M yet they suffered a one day loss of $390M. This does not mean that their VAR is broken: VAR is not intended to give an account of how big losses might potentially be. But it does illustrate that modern trading activities can generate losses far in excess of VAR capital estimates, and hence that other risk measures such as stress tests are important too. This relegates VAR to its proper place, as one risk measure amongst a number. In this setting market risk capital would not be based on the VAR alone, and shows that there is no need to have overly temporally risk sensitive capital estimates.

They have some unusual animals in Regent's Park at the moment...

...and some of them are even more rare than buyers in the ABCP market. That however, may be about to change.



The proposal apparently (the details are sketchy) is for a group of banks to set up a mega-SIV. This new vehicle will acquire the mortgage assets now held by some existing SIVs and conduits. The range of participants is unclear - Citi, with over $100B in conduits, is apparently leading the deal structuring, with JPMorgan interested in selling the new paper. The basic idea is that having a vehicle with a more diverse range of assets will avert the need for many banks to sell their existing conduit assets causing a crash in the RMBS market. Things are not exactly going well there as it is, as you can see from the recent price action on the ABX Home Equity BBBs (this graph is for the 07-2s):


(Graph from Markit via Calculated Risk. This is pretty good on the ABX if you need some background.)

The credit enhancement for the new vehicle hasn't been made public thus far: one structure might be for all the contributing banks be jointly and severally liable for some bottom tranche, followed by a layer of seller-specific credit enhancement. The proposal is apparently to be dubbed M-LEC, for master liquidity enhancement conduit.

In this context it occurs to me to wonder why holding assets off balance sheet in conduits have such a preferential capital treatment compared with on balance sheet credit enhancement. To see this, consider the following two trades: 1. A bank sells a diverse $500M portfolio of assets into a conduit, retains a $10M seller's interest, funds by issueing three month ABCP, and writes a back-up liquidity line; 2. A bank holds the assets on balance sheet and buys a three month credit derivative on losses in excess of $10M, rolling it as it expires.

In both cases the bank is exposed to losses between 0 and $10M but not above. In 1., it has to fund the assets if the ABCP market is disrupted, whereas in 2. it knows that it has to fund the assets in all conditions and hence can lock in term funding. Therefore arguably 2. is less risky than 1., and certainly not riskier. But you can guess which situation has the higher regulatory capital, can't you?

Update. The plan is now out. Or, at least, the intention to come up with a plan that may lead to the MLEC is out. I wonder how they are going to value the assets the MLEC will wave in...

If not exactly early, then at least not too late

Another sign we are on the upswing: the market for commercial paper in US is expanding again. Buy illiquid structured securities. And LBO loans.

Long jumbos, short agency PTs

Obvious really.

Stressing Sants

The FSA's Hector Sants has been getting a grilling at the Treasury select committee today. One of the more troubling features of this was Sants' revelation that FSA's stress tests did not include turmoil in the credit markets.

According to the Guardian:

"We did not perform to my satisfaction with the stress-testing scenarios," said Mr Sants. "There are lessons to be learned."

[If that really was Sants' grammar, he might consider a remedial English course as part of the lessons to be learnt.]

Mr Sants said that the FSA had performed its last full assessment of Northern Rock in 2006, and was not due to conduct another until 2009. He explained that while the bank had been identified as a "high-impact organisation", the regulator had concluded there was only a low risk of it getting into difficulty.

It would be easy to conclude that this is shocking dereliction on FSA's part, and certainly a 1998-style credit crisis and liquidity meltdown would be an obvious stress scenario. But was the assessment of Northern Rock as 'low risk' really so obviously wrong? Low risk doesn't mean no risk, and low PD firms can still default. Originate and distribute was a perfectly respectable business model until mid 2007, and even if the securitisation market dried up, FSA might have concluded that Northern Rock could have borrowed in the interbank market. The fact that they couldn't was difficult to foresee. I don't think FSA comes out of this smelling of roses, and certainly the UK regulatory system did not operate perfectly, but we should not fall into the assumption that just because a bank had to access the lender of last resort FSA necessarily did something wrong. It might have done, but so far the case seems not to be proven.

Diving in the deep end - the valuation of infrastructure assets

News from Bloomberg that a JPMorgan-led group is buying Southern Water from RBS leads to the question How can market players disagree over the value of infrastructure assets? After all, these are amongst the easiest assets to value: they have stable cashflows in industries with very high barriers to entry and predictable demand. Surely valueing a business like this is little more than an MBA exercise -- so how can RBS and JPM both think they have a good deal?

One of the answers is funding. Whole business securitisation has transformed the market for utility assets as it allows much higher leverage than the traditional public equity/senior debt funding model. If JPM and friends believe that they can squeeze more leverage out of Southern then their equity stake becomes smaller and hence they may be able to get a higher ROE even after paying higher interest expenses.

Another perhaps less remarked upon issue is the assessment of the equity required to support the business. Suppose you want Southern to have a 1% PD. Then you calculate the economic capital required to support the business at a 99% confidence interval. Since cashflow volatility is low - the regulation of the water industry helps matters here - this capital estimate is low and hence your model suggests that a private equity model using double leverage of infrastructure assets is safe. But suppose you want more safety. At a higher confidence interval considerations such as possible alterations in the regulatory environment become significant. The business risk here is that changes in regulation will dramatically change the likely future cashflows of the business: suppose for instance the regulator decides that Southern has to spend more to renovate its pipelines. This risk makes the tails of the return distribution for utility businesses very fat, and hence the capital at a higher confidence interval is much larger. A more cautious owner, then, would assign much more equity and hence need to earn a rather higher return. Thus for instance a cautious RBS and a more aggressive JPM could agree on the all in cost of debt and return distribution yet disagree on the value of the asset. If this were true, RBS could be a seller and JPM a buyer.

Mark my kangaroo down, pa

Surely banks' robust mark verification processes and extensive financial controls would not let this happen would they?

"If you're a smart CEO, you're going to write off everything and then some, maybe even to below-market prices, because you're going to be hidden in the woodshed with everybody else," says Daniel Genter, chief executive and chief investment officer of RNC Genter Capital Management [...]

"They'll make it look a lot worse than it is, but that's the smart move, because you've got little to lose and you might get some of it back in a quarter or two."

Of course they wouldn't do that. You just stand there and tell the tide not to come in.

Monetising prejudice premia

Over at Calculated Risk, Tanta takes issue with this report from JPMorganChase (via Creditflux):

In a new research report entitled "Leveraging CLO illiquidity premia", JP Morgan says that the combination of historically wide CLO liability spreads and near-zero default rates makes this an optimal time for buy-and-hold investors to consider investing in CLOs-squared. It says this is a way to efficiently monetise the current illiquidity created in the "spread rout of 2007".

The report concludes that CLOs-squared offer reasonably low risk relative to the underlying CLOs. Junior tranches in particular offer higher spreads than triple B and double B tranches of regular CLOs with similar or lower risk.

Normally I have a high regard for Calculated Risk, but Tanta's sarcasm is rather disheartening. She says:

[...] because this whole problem, you see, was just a matter of the underlying collateral and had nothing whatsoever to do with levering up complex derivatives or having to unwind some goofy structured deal.

Leverage and complexity aren't good or evil. Particular deals might be, or perhaps even more specifically particular deals sold a certain way to particular investors might be, but that's it. You can't conclude that all structured trades are necessarily bad. After all, buying stocks on margin has been common for U.S. retail investors for many years, but no one is suggesting that this form of leverage doesn't meet some legitimate investment objectives. So JP's investment idea should not be dispensed with by waving a magic wand of disapproval at it: you actually have to analyse it. Still with people talking this way, it might be a good way to go long some cheap complexity.

See Wile E. run

From the Telegraph:

The Saigon Times said this morning that the State Bank of Vietnam was abandoning the attempt to hold down the Vietnamese currency through heavy purchases of dollars.

Separately, the gas-rich Gulf state of Qatar announced that it had cut the dollar holdings of its $50bn sovereign wealth fund from 99pc to 40pc, switching into investments in China, Japan, and emerging Asia.

This action is certainly adding pressure to the dollar and further falls seem likely. But will we see a rout (aka a Wile E. Coyote moment)? Put on your gamma, take a seat, watch the action.

Building a new system of financial regulation

There has been some discussion in the wake of the Northern Rock panic of the structure of UK banking supervision. Some commentators have suggested that bank supervision should go back to the Bank of England from FSA. Surely this is just the wrong conclusion: organisationally it makes sense for banks to be supervised by the same body as broker/dealers, insurers and others. No, rather the issue is why the Bank is responsible for financial stability. Surely that belongs with regulation at FSA. If the Treasury wanted to offer the Bank a sop, it could allow them to continue to run the window, but give FSA control in the event of a crisis over what collateral was eligible and how much liquidity to offer to the market.

Humourous quiz of the day from dealbreaker.com:

First, we want to offer a sincere congratulations to Alan Greenspan, for proving that in spite of his retirement and status as an octogenarian, the value of his opinions haven’t yet been placed somewhere between those of a house plant and Jim Cramer (in descending order).

[...]

So why, then, does the guy have to be sort of a prick toward his successor? It’s not enough that he insists on remaining in the spotlight instead of going to Florida to die like all good retirees do. Greenspan’s made it his “post-work, I need something to do all day besides putter around the house” hobby to undermine the Beard of Irrelevancy (that's what AG calls him) and the Federal Reserve (and central banks in general), saying in Amsterdam that “The Fed, ECB, BOE, and BOJ are all losing their ability to influence longer-term rates,” and suggesting that Ben Bernanke be fitted for dentures and get an Rx for Cialis. It’s almost as though he wants to see them fail. But why?

a. So he can be called in at the 11th hour to save the day.

b. His thinly-veiled hatred of Bernanke masks a deep-seated bitterness over never having been able to grow facial hair.

c. He’s an asshole

d. He’s bored

Personally I'd welcome an e. All of the above, but quizzes are often like that. Maybe Ben should dip in the funds a little more to send Alan on a nice long cruise?