How bad can the business pages get?

As an exercise, I'd like to take a look at an article in today's Observer and deconstruct the text. The journalists responsible, Heather Stewart and Nick Mathiason, don't come out of it smelling of roses.

Let's begin near the start of their article with:

Britain's banking sector, once a bastion of financial rectitude overseen by cautious pin-striped branch managers, has been swept along in a frenzy of creative accounting, adapting strategies invented by Wall Street's 'masters of the universe' to feed the British public's voracious appetite for cheap loans.

This is just cheap. First we have the parody of 'pin-striped branch managers' - how does what a branch manager wears effect their job? - then emotive language like 'frenzy' and 'voracious' plus a cute little anti-American jibe with no evidentiary support. Perhaps we can tell already that this is not going to be a balanced piece of reporting.

Northern Rock [...] was certainly not the only UK bank to dive headfirst into the murky world of securitised loans, 'special purpose vehicles' and 'collateralised debt obligations' (CDOs).

Again, totally unjustified language: do the authors know that Northern Rock 'dived headlong' into CDOs rather than calmly and maturely took the business decision to use a perfectly reputable instrument? And just because something is 'murky' to what appear to be rather ill-educated journalists doesn't mean that it is murky to a professional banker, risk manager, or regulator.

The article then goes on to talk about off-balance sheet funding and the increasing gap between deposit funding and lending. It isn't long before we get another disingenuous piece of mud-throwing:

The regulators can hardly claim to be shocked about the blossoming of these byzantine arrangements.

Firstly regulators don't claim to be shocked by them, and neither are they byzantine. Just because you don't understand it, Heather and Nick, doesn't mean it is complicated. The article doesn't get any better:

In other words, instead of taking money in from depositors, and then lending it out again (how dull), they have become mere conduits, gathering in willing customers and selling their loans to investors keen to take a slice of the profits.

Deposit funded lending is not just dull, it also, crucially, has a low ROE. Banks have a duty to their shareholders, just like any other corporation. If a different method of doing business improves returns for acceptable risk, they are failing in their duty if they do not use it. And why is a conduit 'mere'? A central lesson of business over the last twenty years is stick to what you are good at and let others do everything else: it is hard to argue banks are necessarily both the best originators of credit risk and the best long term holders of it. You don't call Tesco a 'mere' grocery distributor: why criticise an originate-and-distribute bank?

Next we get something that is either a mis-understanding or a deliberate spin on the truth:

As the credit crunch hit last month, HBOS - the giant UK bank formed by the merger of Halifax and Bank of Scotland - was forced to announce that it would lend money to a so-called 'conduit fund' called Grampian, 'to repay maturing debt as market pricing was unacceptable'. This was code for a bailout: no other institution would lend the facility money.

In all likelihood, HBOS had written a standby letter of credit to the conduit and hence was obliged to provide liquidity since the ABCP market was disrupted. I rather suspect the authors have no evidence no one else would lend Grampian money because I rather suspect Grampian had no need to ask anyone else. Standby LOCs are commonplace for conduits and, while banks are probably not overjoyed about them being triggered, there is no evidence that they have thus far endangered the banking system. In particular, once a conduit's assets come back on balance sheet, the bank has to provide capital against them in the usual way, so the balance sheet relief only lasts as long as the liability belongs to the ABCP holder.

Of course, Heather and Nick can't resist throwing a little more mud:

No mention of Grampian is made in HBOS's 2006 annual report - an indication that the facility was held off-balance sheet. [...]

This complex financial merry-go-round has made a lot of people rich.

Good grief how terrible. HBOS followed accounting standards in not including an off balance sheet vehicle in its accounts and, even worse, some money was made. What a disaster for the financial system.

Heather and Nick continue with a deeply dodgy passage:

But for all their sophisticated risk management models, when defaults on sub-prime borrowing in the US shot up, and demand for CDOs and other asset-backed loans petered out, the banks realised they simply didn't know what their total liabilities were.

Umm, let's see. A liability is something on the balance sheet, like a bond you have issued. The subprime market has nothing to do with a bank's written debt, so they can't mean that, unless they are referring to covered bonds. They might mean the bank's assets because its easy to confuse an asset with a liability when you are writing polemic, but given the banks were mostly sellers rather than buyers of CDO tranches, that doesn't seem likely either. They might mean the value of the protection bought from the securitisation vehicle, but that isn't on balance sheet (although it is difficult to value at the moment) and it definitely isn't a liability so they can't mean that, can they? Perhaps they don't know what they mean.

Having fired off a string of alarmist propaganda, the authors conclude with:

[...] as it becomes clearer just how radically the banking sector has been transformed - and that the ratings agencies, accountants and just about everyone else in the City has been in on the act - it's hardly surprising if public confidence is shaky.

With journalism like that it is hardly surprising that public confidence is shaky. But the fact remains that securitisation has improved bank profitability, allowed banks to focus more on their customers, given investors access to a much wider range of assets, and spread risk in a fashion that has, so far at least, caused distress but no actual bank failures. How bad might the sub-prime crisis have been if risk had not been securitised? I can't suggest that the regulatory, accounting or ratings systems are perfect - far from it in some cases. But no one's interests, apart perhaps for the authors', are served by articles written in the same vein as Heather and Nick's. Mind you, informed balanced reporting of the credit crunch in the mainstream UK press is about as rare as Giraffes in Canary Wharf.

Angela's still arsey

Entirely predictably, the British Bankers Association has called for an overhaul of the banking regulatory system in the wake of the Northern Rock crisis while attacking Mervyn King. The BBA is acting here as an industry lobbying body so we should not be surprised to find some special pleading, and indeed we find the usual charge of 'you should have done more'.

One can have some sympathy with the claim that the regulatory system needs review: the deposit protection scheme is sub-optimal (although you can bet the BBA will be among the first to scream if the Bank suggests an insurance-based replacement with banks paying premiums based on risk), and the combination of the market abuse directive and the takeover rules may have prevented a Bank-organised takeover of Northern Rock. Still, none of this is entirely the Bank's fault, nor FSA's. Given the rules as currently written, the Bank has done a good job acting as lender of last resort, rather than, say, lender during a mild panic. It hasn't protected the foolish, only the system, as its remit requires. FSA may have had concerns with Northern Rock's originate-and-securitise model but, contrary to much press comment, this wouldn't have seemed an extraordinarily risky strategy six months ago. It would have been inappropriate to intervene much earlier than the Bank did.

Quote of the day

From Bloomberg:

``Moody's does not structure, create, design or market securitization products,'' Kanef [group managing director, asset finance group, Moody's Financial Services] said. ``We do not have the expertise to recommend one proposed structure over another, and we do not do so.''

Update. After that confidence building statement, we have the FT reporting that a senior U.S. official is considering splitting up the advisory and ratings functions of the agencies to reduce their conflicts of interest. It will be interesting to see if there is a knock-on effect to Basel 2 here: after all, ratings are at the heart of the new Accord, and now it is clear how untrustworthy they were in some cases, the supervisors might have a change of heart.

Lending decision based on counterparty shock

Over at the FT there is a slightly sensational article Libor’s value is called into question. Apparently it surprises the writer than an average rate like Libor is not everyone's cost of funds. Surprise, some banks are funding significantly sub-Libor at the moment, and some way, way over.

The FT raises the concern that some players may have written derivatives contracts referencing Libor and now discover that this index does not reflect their cost of funds. Most derivatives traders worked out that arb sometime in the late 80s: Treasury pays L-10 but the pricing system assumes you fund at Libor flat, so if you can raise money at L-5 you book a profit vs. the model and treasury eats the loss (or at least it did before transfer pricing caught up with that little game). Anyone who didn't understand that writing a contract based on an average interbank rate introduces funding basis risk shouldn't be allowed to write something that will be with the holder for many years.

Model risk in economics and finance

Over at Overcoming Bias, there is a discussion of model checking in the economics literature:

One thing that bugs me is that there seems to be so little model checking done in statistics. Data-based model checking is a powerful tool for overcoming bias, and it's frustrating to see this tool used so rarely. [...]

But, why, if this is such a good idea, do people not do it?

I don't buy the cynical answer that people don't want to falsify their own models. My preferred explanation might be called sociological and goes as follows: We're often told to check model fit. But suppose we fit a model, write a paper, and check the model fit with a graph. If the fit is ok, then why bother with the graph: the model is OK, right? If the fit shows problems (which, realistically, it should, if you think hard enough about how to make your model-checking graph), then you better not include the graph in the paper, or the reviewers will reject, saying that you should fix your model. And once you've fit the better model, no need for the graph.

The result is: (a) a bloodless view of statistics in which only the good models appear, leaving readers in the dark about all the steps needed to get there; or, worse, (b) statisticians (and, in general, researchers) not checking the fit of their model in the first place, so that neither the original researchers nor the readers of the journal learn about the problems with the model.

Sadly this rings all too much of a bell in Finance too. If a firm has one pricing model and no really incisive model verification, it is often happy. If it has more than one model or a good model verification unit, the scope of possible error, uncertainty and failure of assumptions becomes visible. People dislike uncertainty so this kind of mature attitude can be rare. I don't go quite as far as this (from Equity Private):

Large unwieldy models are almost universally produced by financial "professionals" who have no clue whatsoever about their predictive value (hint: it is vanishingly small) and therefore the size of the model is, in my view, inversely proportional to the technical competence of the employee.

But I would like to to see something like this:

We are doing our best with this product given the resources available but the model might be wrong. By stressing parameter inputs, using other models, and reviewing assumptions we have quantified our likely error as $28M.

It might appear to be a mess, but it does properly characterise (to use what is becoming word of the week) the epistemology of the situation.

Sub Libor Silliness

If I had seen this at the time, it would have been the strongest sell signal imaginable.

The rules of trade

Stolen more or less wholesale from Naked Capitalism:

Dani Rodrik has [...] set forth the conditions that have to be in place for trade liberalization to enhance economic performance (short answer, a lot); in another [post], he reviewed the analyses that claimed that our current trading system produced large economic gains and found the logic to be badly flawed.

Rodrik in turn refers to Deconstructing the Argument for Free Trade, an excellent paper by Robert Driskill. In particular Driskill begins by asking what the metric is:

What does it mean for a change in economic circumstances to be "good for the nation as a whole"?

He then goes on to review various possible metrics, and discuss their advantages and disadvantages. Encouragingly, he ends not with a conclusion but a methodological recommendation:

Trade economists should [...] be forthright about the epistemological basis of their policy advocacy of free trade.

In other words, be clear about why you claim something is a good idea, not just that it is one.

Losing my religion

Two comment articles from recent days focus on faith schools. First on Thursday Zoe Williams argued against the state funding of faith schools:

We all [...] have the odd qualm here and there about Islamic schools, and whether they invest proper rigour in the propagation of gender equality, but Christians, we think ... now they're different. They provide a sound education, they don't discriminate on the basis of class, they're not exclusive, they've been doing this for years. They can have as much taxpayer money as they want.

It's balderdash. For a start, they are cherrypicking middle-class children (the Institute of Education at London University just produced this finding, after the most extensive research yet undertaken) and, much more important, in many cases they are prosecuting an agenda that is repugnant. Are we really happy to sit back and pay for this?

Her point is reasonable. If the state gives any organisation money, it is implicitly affirming its values and efficacy. Do faith schools really do what we want? In particular, is it good for children to be exposed to the diet of difference that is faith? Believers think they are somehow privileged, after all (ignoring for a moment a few Buddhists and perhaps the Quakers). In its currently fragmented state, do we really need yet another thing dividing people and encouraging prejudice being taught in schools?

Moreover, all faiths are not equal: atheists and agnostics are not allowed to play at the same table, as this story indicates:

A headteacher who tried to reduce the influence of religion inside the classroom by creating the country's first secular state school had his plans blocked by senior government officials who called it a 'political impossibility'.

Personally I would be perfectly happy to let my taxes go towards faith schools if there was good evidence they lead to better outcomes for children and society. But as they seem to be just another piece of Mr. Tony dogma unsupported by the facts, I tend to view them with some suspicion.

Truth Hurts

This is my favourite quote of the year so far. It's in the running for quote of the decade.

Man is basically lazy. Innovative and complex incentive and disincentive structures must be continually created and refined to compel any desirable behavior (including the absence [of] self-destructive behavior). Excessive gaming of the system will be employed at every opportunity to avoid doing anything resembling work.

It is from a deeply entertaining selection of posts at Going Private and it is true: my only quibble is whether 'excessive' is a reasonable term for what is basically playing by the rules (albeit subverted to reflect your own selfish needs).

The Dollar and the fans of Wile E. Coyote

Paul Krugman uses the term 'Wile E. Coyote' moment for when traders find a currency level is unsupported by fundamentals and it drops precipitously. Certainly some currencies display very fat tails: they tend to have long periods of stability, followed by one or more greater than five s.d. moves. (The interested reader may at this point wish to fit the Generalised Pareto Distribution to twenty or more years worth of dollar/yen returns, and compare the GPD VAR with the normal one at 99.99%: typically it's very roughly four times higher.)

Krugman further suggests that a Wile E. Coyote moment may be approaching for the dollar. This is more than just suggesting that the dollar will fall: the fall has to be steep to qualify. Tanta over on Calculated Risk has some supporting evidence (which is a little glib but bear with me):

Bear in mind that the principal channel through which Fed policy affects domestic demand is via housing. If a burst housing bubble is part of the economic problem, the Fed’s leverage over the economy will be greatly reduced, and even a zero Fed funds rate might have only modest stimulative effect.

The problem is too many people are talking about this possibility: many currency strategists expect dollar weakening, so existing dollar shorts will tend to make large falls much less likely. The macroeconomic picture is not encouraging for the dollar, it's true. As Long or Short Capital put it, albeit amusingly bluntly:

I challenge you to find one measure of wealth OTHER THAN THE DOLLAR which shows the US economy as worth more now than in 2001. If I wanted to buy our country it would cost me 30% fewer euros today than it did in 2001, it would cost me less bars of gold, less barrels of oil, less ounces of copper, less btu’s of natural gas, less cubic feet of lumber, less of almost anything that has intrinsic value. Yet you keep reporting GDP growth, why? Because your quick fix is to effectively print more money so that in dollar units everything is getting more “valuable”. But guess what, to the 95% of the world that doesn’t use dollars the true value of the US economy has been shrinking, rapidly.

Moreover, central banks, notably asian central banks, are not buying enough dollars to provide a floor. As Brad Setser says:

The world’s key central banks have concluded that they have more reserves than they need, and are rapidly losing interest in adding to their dollar reserves. China’s central bank has made it known that it thinks it has enough reserves. Some in China think the PBoC already has far more reserves than it needs. Korea’s central bank has indicated -- at various points in time -- that it has more than enough salted away. The ADB agrees.

Still, currencies are often a long way from macro-economic equilibrium, and the U.S. has historically shown an astonishing ability to grow out of difficulties. Despite the fundamentals then instinct suggests that while one might not want to be short dollars yet. That just leaves buying the potential to benefit from a Wile E. moment via the options market. Perhaps selling short term downside gamma (in the belief that Wile E. will take a while to arrive at the cliff) generating cash to pay for a longer-dated further from the money position might be interesting.

I'm sure a salesperson will soon christen this 'the Coyote trade': look for a range of North American mammal structured notes at your friendly investment bank shortly.

Our one is better than your one

Melvyn might not have convinced everyone in his testimony today, but at least people aren't publishing stuff like this about him.



Funny, though, no? (From Long or short capital.)

Mourinho and the Myth of the Man Manager

(With apologies to Fred Brooks.) So Jose has gone. That's hardly a surprise, although the timing is interesting. English football will be better off without his swaggering arrogance. He typifies the myth that management is the business of heroes, that the individual is all important. The boring reality is that it is teamwork that is most often the deciding influence in football or business: Arsenal's players are worth a good deal less than Chelsea's, but because they play as a team, they are four places above Chelsea in the table. Sunderland under Roy Keane are another example: they punch above their weight (one hopes given Keane's reputation not literally so) because they have team spirit.

Mourinho's Chelsea were too often a collection of prima donnas, selfishly strutting around the pitch and not quite delivering. Let's hope we see a return to sportsmanship in sport, with the team seen as more important than the individual and team building as more than just shopping.

Stepping back from Basel

It will be very interesting to see how bank supervisors react to the current credit crunch in the longer term. We are less than a year into the new Basel 2 Capital Accord, and already it looks flawed in a number of areas. This is hinted at in the FT today. At very least in the UK we will see a revision of the deposit protection regime. But (admittedly in the heat of the battle and without mature reflection) Basel 2 looks vulnerable too, especially in the areas of credit risk mitigation and the treatment of off balance sheet vehicles.

A related issue is the procyclicality of Basel. This has been around for a while in the context of VAR models (asset prices go down, vols go up, so VAR goes up, pushing banks over their risk limits and hence causing asset sales which further depress prices), but Basel 2 extends this to the banking book via bank's PD estimates in IRB models. The problem is, risk sensitivity goes hand in hand with procyclicality. It will be interesting to see if the supervisors can bring themselves to make Basel 2 less risk sensitive.

Here's a view towards Canary Wharf from the City of London.

The Bank between a Rock and a Hard Place

Today the Bank of England said it will lend £10B to commercial banks in an emergency three-month auction and widen the range of securities it accepts as collateral to include mortgage backed securities. That should bring the 3 month swap spread in. The question is, why have they caved in to the banks after holding the line admirably until now? Some possibilities, the first three none too savoury:

• The political clamour over Northern Rock could not be ignored and the Bank was worried that someone like Alliance and Leicester or HBOS might be next.
  
• The Bank knows that another player is in serious trouble.
  
• The knock-on effects of the high cost of Libor-referenced funds in the real economy are becoming too large.
  
• The rate the Bank is charging for this extra liquidity is sufficiently high that it provides appropriate liquidity yet makes anybody imprudent enough to need it pay through the nose. This wouldn't be too bad but one wonders why at least some of this £10B was not provided earlier.
  

I agree with Nils Pratley that this is not a resigning matter for Mervyn King - the Bank has played a reasonable hand. But we do need understand what their thinking is. The performance today will be interesting.

Update. King is blaming the Market Abuse Directive for preventing him lending to Northern Rock covertly, and the takeover code for stopping him organising a quick-but-effective sale. This is fascinating if it's true.

What's in a Mark Part 157 (not 39)

Be careful what you embrace, particularly in the area of valuation policy. According to Bloomberg:

Absolute Capital Management Holdings Ltd. will stop clients from pulling money from eight hedge funds with $2.1 billion of assets after co-founder Florian Homm quit.

Investors will be asked not to remove cash for a year as the firm restructures the funds, Absolute Capital said in a statement today. Seven of the pools invested in over-the-counter U.S. stocks that can't be sold at the prices at which the firm had valued them, affecting as much as $530 million of assets.

If an asset is marked at price different from where you can sell it, how exactly is that mark to market?

Update. I like this fragment from the Independent so much I want to quote it.

Sandy Chen of Panmure Gordon [Absolute Capital's Broker] said the apparent valuation shortfall "could prove fatal for the equity funds" and highlighted "the risk of investor lawsuits as an unquantifiable threat to earnings". He slashed his price target from 750p to 150p...

Is that target change great? I just wish more analysts were willing to move their forecasts with such speed and robustness.

TechnoDemo

A few things have made me think about democracy recently.

• First the Olympic Delivery Authority are sticking to the idea that the new Olympic stadium should only be used for athletics - despite a bid from West Ham to use it as a football stadium and athletics use requiring a downsizing of the building from 88,000 to 25,000. This is my money you are spending, people. I am sure most Londoners would prefer a new full size football stadium and a billion quid to spending more money on an athletics venue. So why is the ODA allowed to flaut the wishes of the people who are paying for their project and their salaries?
  
• The West Lothian Question is in the news again. Most politicians seem to think differently from most voters on this one.
  
• Finally I came across a really good article in the LRB (behind a firewall I'm afraid) that points out that political arrangements evolve and that democracy in its current form might be dating fast.
  

The problem is that democracy has little to do with the exercise of political power by the people. The people vote for a party, then the winner does what they want regardless of the people's wishes. In extreme cases (Blair, for instance) the party leader even rides roughshod over the wishes of his own party. The only decision the people get is who to vote for once every five years or so.

Proportional representation makes the problem a little better in that coalitions are likely so governments have to be more moderate, mostly. But there is still the problem that it is politicians making most of the decisions, not the people. The model of having professional government might have made sense in the 18th century when figuring out what the people wanted was difficult and time-consuming but it is not clearly appropriate today. And professional politics encourages decisions that look good or send the right political message rather than reflect the people's wishes - like replacing Trident.

There are alternatives. We could use referendums much more. We could elect different parties for different things - a Lib Dem could be responsible for law and order, for instance; a Green, the environment; a Tory, Defence; and Labour, the economy. There would be coordination issues of course but it isn't obviously necessary to have the same party running every ministry. Or we could have a pre-emption mechanism so that any policy enough people disagreed with was put to a popular vote. If something is not done then respect for the political process will continue to decline. People are sick of their vote not counting. The good news is that it is fixable, especially given the potential to use technology for fast, easy voting. The bad news is that fixing it requires the turkeys to vote for Christmas.

All Hail The King

Here is the edifice that houses the Lutine Bell in Lloyds of London. Traditionally the bell was sounded to signal bad (or good) news in the insurance market.

Talking of bad news, two Bloomberg reporters seem to have it in for Mervyn King:

Two days after King, 59, told lawmakers on Sept. 12 that central banks should avoid giving the impression they will help lenders that made bad decisions, the Bank of England provided emergency funds to Northern Rock Plc in the biggest bailout of a British bank in three decades.

Subsequently the Bank has been forced to guarantee all of Northern Rock's depositors (as opposed to 100% of the first £2000 then 90% of the rest up to a maximum of £33,000* under the UK's Financial Services Compensation Scheme).

``It's a crisis of confidence, and the bank is confused,'' said Patrick Minford, an economics professor at Cardiff University who advised former Prime Minister Margaret Thatcher. ``They want to be hands-off, but in this situation they can't be. I don't think this has done King any good.''

Minford was Thatcher's economist, a recognised hardliner and currently out of political favour. Perhaps he isn't the most disinterested observer of the situation?

Bloomberg continues:

King's credibility is in question for his refusal to emulate other central banks and take early action to help cash-strapped lenders. With Northern Rock's failure, he is finding himself subject to the same charge of excessive caution being leveled at U.S. Federal Reserve Chairman Ben S. Bernanke, whose office adjoined King's at the Massachusetts Institute of Technology in the 1980s.

Horror of horrors, two intellectuals. Your money cannot possibly be safe with anyone who went to M.I.T.

King held back until markets forced his hand. Last week he said that too much help ``encourages excessive risk-taking, and sows the seeds of a future financial crisis.''

And he was right to do so. The only criticism of King I would offer is that he failed to engineer a takeover of Northern Rock by Lloyds or HSBC earlier in the week, something that would have represented the typical dirigiste approach of the Bank to a crisis (cf. ING's purchase of Barings). But someone will come along sooner or later with an open pocket, and meanwhile King is rightly signalling to the market that institutions must bear the consequences of their own liquidity risk management decisions.

*One might argue having only a 90% guarantee introduces an incentive which encourages bank runs: no one wants to lose 10% of a chunky amount of money. Perhaps now we are actually seeing bank runs again HM Treasury, the Bank and FSA might like to redesign the scheme?

Update. Willem Buiter and Anne Sibert writing in the FT are wrong about liquidity premiums:

The Bank is not blameless in the Northern Rock debacle, however. A bail-out might not have been needed if the Bank had a more sensible collateral policy for its open-market operations and discount-window borrowing. The ECB accepts private securities rated at least A-; the Bank should too.

This risks the Japanese problem of excess liquidity and a lack of liquidity premiums causing falling returns on illiquid assets, deflation, and a reluctance to invest. The Bank absolutely should not accept dodgy collateral at the window in ordinary conditions. It should widen the definition of eligible collateral when necessary to manage the three month swap spread but not otherwise.

The Centre of Power

Depending on exactly how you calculate it, the geographical centre of the UK is somewhere round Accrington in Lancashire. On the other hand, the political centre is London. As I was coming back into the capital on Sunday night, the idiocy of this came home to me. The entire South East is crowded, its infrastructure is straining, and it attracts a disproportionate amount of attention and resources. So let's move parliament and the major government departments to somewhere more representative. Accrington, say. I'm sure a nice big brownfield site could be found to re-develop, and the boost to the local economy (currently suffering from the decline of manufacturing) would be huge. Think of the opportunities for journalists, restaurants, pubs, and of course high class courtesans.

Meanwhile down South valuable land would be freed up, the exodus of people would depress housing prices, relieve some of the pressure on the South East's straining transport system, and most importantly of all refocus the government on the needs of the whole of the UK rather than just the parochial concerns of the South East. Admittedly administrative capitals that are not a major population centres tend to be pretty boring places -- think of Canberra, Ottawa, or Bonn pre 1999 -- but imagine the burst in artistic creativity once all those policy wonks, lobbyists and other government types aren't dragging down London's atmosphere. Blackburn and Burnley haven't got much of a scene to ruin either. We could keep the House of Commons as a tourist venue: it would make a fantastic museum. Surely the opportunity to knock down the Ministry of Defence, a building of positively Stalinist ugliness, is enough to wave the decision through by itself?

Liquidity Risk is real

The BBC doth protest too much. First:

Shares in one of the UK's largest mortgage lenders, Northern Rock, have fallen 32% after it had to ask the Bank of England for emergency funding.

The next paragraph however:

But experts say it does not mean Northern Rock, which has £113bn in assets, is in danger of going bust.

Liquidity risk has nothing to do with solvency. You can go bust with £113B in assets and £1 in liabilities if you can't find the cash to pay the liability and it is due today. If you have to go to the Bank of England as Lender of Last Resort because, well, it is a last resort, then you are in danger. And it is no surprise that an institution whose business model was based on originating mortgages then securitising them has a funding crisis. What is more surprising is the number of Corporal Joneses going around crying 'don't panic'. Panic seems entirely appropriate. Unless you are naked long CDS protection on mortgage banks, of course, or long gamma on the stock...

Remember, if it looks like a duck, smells like a duck and quacks like a duck, it is probably a duck. Similarly if it looks like a liquidity crisis, smells like a liquidity crisis and the central bank acts like it is a liquidity crisis, then it's probably a liquidity crisis.

How much?

From today's FT:

Continued high overnight interest rates forced the Bank of England to offer £4.4bn additional cash to commercial banks on Thursday morning, in an effort to normalise the money markets.

A decent sized conduit or SIV is £5B so the bank's extra liquidity is less than one conduit's worth. Is anyone else surprised how relatively small the bank's offer is? Personally I think they are doing an excellent job in not rescuing the imprudent, but I wonder if £4B is significantly different from zero.

Why I hate HP

I sometimes don't mind a stylish, considerate rip-off. My shaving foils are like that. They are eye-wateringly expensive for what they are, but I put up with them because they do actually shave reasonably well, and the product doesn't go out of its way to point out how much I am being gouged.

My HP printer on the other hand, is an irritating rip-off. The print quality isn't great. The printer looks cheaply made. But what really gets my goat is the great cartridge racket. Probably the most expensive part of the printer is the system designed to ensure you can only use HP cartridges. Every time it starts up it checks to see that it has the 'right' cartridges and that they are full enough. After all, HP wouldn't want to miss an opportunity to sell you another over-priced piece of plastic, would they? And it's all too obviously about selling more ink. Apparently IT industry slang for an inkjet printer is 'cartridge vending machine'. So now I'm going to throw my vending machine away and buy a black and white laser printer which I can buy cheap toner for. And it won't be an HP. Nothing I ever buy again will be an HP.

What money market?

The term 'money market' is pretty unhelpful in current conditions. There isn't one. There is a short term Libor market, which is decidedly interesting at the moment, and a short term govy market, which is awash with liquidity at least in some currencies. These two markets have decoupled.

The FT, reporting on a speech by Axel Weber, president of the Bundesbank, says:

[...] the tools that modern central banks possess to address liquidity problems can only directly address such runs inside the traditional banking sector, and do not directly touch the non-bank financial sector, which has been hardest hit by the current credit crisis.

Mr Weber’s analysis highlights the dilemma facing central banks, which cannot channel funds directly to the non-bank financial sector, and may therefore have to resort to easing monetary policy instead.

Once the Libor market disassociates from the govy market, there isn't much that a central bank can do. They control financial operations in the govy market, and they can inject extra liquidity there via accepting a broader range of collateral at the window, cutting rates, or whatever. But they have no power over the Libor market. They can hope that if the differential between the two markets becomes large enough banks will step in, borrow from the central bank, and lend into the Libor market, but they can't force that to happen.

Weber claims that the current situation is like a run on a bank, but one effecting conduits and SIVs rather than banks per se. In that these vehicles are suffering a liquidity squeeze that they are vulnerable to due to mismatched funding, I'd agree. But Weber says this "is a total over-reaction". I'm not so sure. If you can get Libor plus 100 for lending a AAA Libor-based funder cash (because cash is really scarce and you have it) why wouldn't you?

Yes, it is still liquidity

Today's FT discusses the current problems for SIVs and conduits and identifies two main ones:

They have been hurt as funding in short-term debt markets has seized up.

Simultaneously, the values in the kinds of assets they hold have fallen as investors deserted all asset-backed bonds in repsonse to fears of contagion from the US subprime mortgage markets.

Now the killer punch:

Analysts at Moody’s said during an investor call on Tuesday that the funding problem was by far the most serious for most vehicles.

“The ongoing liquidity crisis has deepened and broadened since . . . July,” said Paul Kerlogue, senior credit officer for SIVs at the agency.

“Vehicles that fund [by means of] the issuance of commercial paper have found financing either impossible or achievable only at exorbitant levels.”

There are three ways out of this. Default; pay up for CP at current levels and hope you don't run out of cash to pay spread before the ABCP crisis ends; or deleverage. Your asset quality effects which of these alternatives are available to you but it doesn't change your exposure: even if your assets and capital structure mean that your debt should still be AAA, you nevertheless still have a problem. And it is going to get worse, at least for a while. There are tens of billions of dollars of ABCP due to expire in the next ten days, and even more falls off next month. Some of that will be financed in the short term via backup CP lines with banks, but that will just force the 3m swap spread out further and the ABCP market is likely to become even more arid. The key to understanding the current crisis is the liquidity market.

Is alignment of interests a panacea?

Probably not, but it surely helps. Martin Wolf in the FT says:

Good reasons can still be advanced for shifting exposure from the balance sheets of thinly capitalised banks to those of better capitalised outside investors. The theory was that risk would thus be shifted on to those best able to bear it. The practice seems to have been that it was shifted on to those least able to understand it.

The supply of such fools has, if only temporarily, dried up. In the short run, securitised debt is likely to contract, as existing debt is paid down or written off. In the longer term, intermediaries will have to find a way to make their products more transparent to the buyers. Unfortunately, the ratings agencies, which once served this purpose, have lost their credibility.

The reason many of these assets have ended up with fools is that if the equity was a good tranche, it was retained, and if it sucked, it was sold. For many deals the mezz was sold - that should tell you all you need to know about how good that tranche is going to turn out to be. One quick fix would be for regulators simply to cap the benefit of securitisation at 50% for any tranche. That would give banks no benefit from selling more than half of each tranche, and hence encourage them to look after the interest of all the securitisation buyers as a side effect of looking after themselves. It would not stop cross subsidy between tranches, of course, but it would be a reasonable start.

That's not fair Mummy

Or perhaps it is. I've just got around to looking at FASB Statement 157 on Fair Value from the tail end of last year. It is remarkably sensible. The definition of fair value is more or less the same as before

The price that would be received to sell an asset or paid to transfer a liability in an orderly transaction between market participants at the measurement date.

Note though that the definition is based on an exit price (for an asset, the price at which it would be sold) so the intention is clearly to mark at bid/offer rather than mid or acquisition price. This presumably means everyone will show a loss when the buy assets (since they will buy at the offer, mark at the bid), although there is some mealy mouthed language that the standard "does not preclude" the use of mid-market prices or other "conventions as a practical expedient".

The innovative part of 157 is the hierarchy of valuation principles:

Level 1. Valuation using quoted market prices for identical assets or liabilities in active markets.

This is the traditional idea of fair value: I have 2000 shares of IBM, and I mark them at the market price.

Level 2 — Valuation using observable market-based inputs, other than Level 1 quoted prices (or unobservable inputs that are corroborated by market data)

This is what firms mostly do in OTC derivatives: we see traded implied vols for some plain vanilla options (or perhaps simpler exotics in some markets) and we use those implieds (together with rates, dividend rate estimates, underlying prices and so on) as inputs to a valuation model to mark the entire book. Here we are saying 'this 1478 strike 70 week OTC call on the SPX I own is worth $50,000 because brokers are quoting a 1450 strike 78 week call at $57,400 and Black Scholes calibrated to that known price give me $50K for my asset'.

Level 3 — Unobservable inputs (that are not corroborated by observable market data)

This is where we mark a mountain range option to (spread to) a historical correlation, for instance, or a private equity position based on projected cash cashflows.

Firms are required to use the lowest level possible, so the most market based marking must be used. Best of all, firms are required to disclose the split between the levels, and to discuss the basis of unrealised level 3 gains and losses. This should be a really useful disclosure for the readers of the financial statements issued by large financial institutions.

Now, consider Jos Ackermann's recent call for banks to reveal their losses in the current subprime/ credit crisis. He's right of course: banks should have an idea of what their exposure is, and they have a duty to reveal that to their investors. But at the moment a lot of securities and derivatives that would have been valued at level 1 last year would currently have to be valued at level 3 - there isn't a market for many credit products right now. This shows the importance of FASB's hierarchy: knowing that a bank has definitely lost $1B is very different from knowing that on the basis of non-market based estimates a bank thinks it might have lost $1B.

Update. Bloomberg backs me up here:

Dealers stopped providing prices on subprime bonds when trading dried up during July and August [...] the firms are reluctant to give low quotes that suggest clients have lost money and are even more hesitant to give high estimates that they would then have to honor as market-makers by purchasing the securities.

``The dealers are finding themselves away from their desks a whole lot more, [...] No one is willing to put out a quote.''

It was liquid, now it's not, and you should be disclosing that fact.

It's not just rates, stoopid

There were some interesting remarks from James Hamilton at the FED Jackson Hole meeting. My favourite part is:

The instrument of monetary policy that we tend to think of first is the time path of short-term interest rates. It's natural to start there, because it's easy to quantify exactly what the Fed is doing.

But another instrument of monetary policy that I think needs to be discussed involves regulation and supervision of the financial system.

That certainly fits with the theme of this blog - that the rules of the game determine the dynamics you see, and you need to engineer the rule set carefully to get the right type of behaviour (and stop the wrong type, whatever right and wrong mean).

Naked Capitalism (where I picked up the report) talks about this in the context of the large complex financial institutions - the Bank of England's term for the largest most systemically important firms (of which it identifies 16).

This graph from the Bank's Financial Stability Review is quoted there:



What we see in the large is rising assets and (assuming asset quality remains roughly constant, which might well not be true) decreasing credit quality. Certainly leverage is increasing. Basel II is likely to make this worse if the results of the quantitative impact studies are to be believed, and the increasing prevalence of SIVs and conduits to get assets off balance sheet is not helping either. Stepping away from the details of risk sensitivity, the level playing field and so on, is the regulation of large complex financial institutions heading in the right direction if they are getting larger and their leverage is increasing?

The 16 LFCIs include Barclays, ABN AMRO and RBS, so if ABN is taken over by either of its two current suitors, the 16 will become 15. Are a smaller number of larger, more complex firms more or less systemically risky than a larger number of smaller, less well capitalised ones?

Modifications and spread compression

For a CDS on a corporate loan or bond, restructuring is often a credit event so in a CDO of corporate debt, adverse modification of the loan causes a loss which flows up the waterfall (albeit sometimes a minor one if the restructuring is purely a technical one).

This is not true for a CMO of retail mortgages. If a retail mortgagee is in trouble, the CMO docs often permit a modification of the loan at the discretion of the servicer. That's huge. The servicer might well have no economic interest in the cashflows of the mortgage yet they often have the discretion to modify them to the probable detriment of the CMO tranche holders.

And the FED wants them to. This is from Interagency Statement on Loss Mitigation Strategies for Servicers of Residential Mortgages via Calculated Risk:

Servicers of securitized mortgages should review the governing documents for the securitization trusts to determine the full extent of their authority to restructure loans that are delinquent or in default or are in imminent risk of default. The governing documents may allow servicers to proactively contact borrowers at risk of default, assess whether default is reasonably foreseeable, and, if so, apply loss mitigation strategies designed to achieve sustainable mortgage obligations. The Securities and Exchange Commission (SEC) has provided clarification that entering into loan restructurings or modifications when default is reasonably foreseeable does not preclude an institution from continuing to treat serviced mortgages as off-balance sheet exposures. Also, the federal financial agencies and CSBS understand that the Department of Treasury has indicated that servicers of loans in qualifying securitization vehicles may modify the terms of the loans before an actual delinquency or default when default is reasonably foreseeable, consistent with Real Estate Mortgage Investment Conduit tax rules.

The last sentence is particularly telling:

Servicers are encouraged to use the authority that they have under the governing securitization documents to take appropriate steps when an increased risk of default is identified

Now this is presumably in the interests of the home owners. I can see why the FED is doing it. But did the tranche buyers know this could happen? Caveat emptor.

A letter you might not want to get...

Or perhaps you might. It is funny.