This should be of interest to you, comments on the BIS analysis of the crisis:
nakedcapitalism.com
“Basel III – the OK, the Unfinished and the Ugly”
By Richard Smith, who works for the London consultancy Cubematch, which specialises in risk, collateral and change management
The OK
The BIS analysis of the 2007-09 banking crisis floats my boat. Here is their headline list of causes: excessive on- and off-balance sheet leverage, diminutive and low quality capital bases, insufficient liquidity buffers at banks. That gave us a banking system unable to absorb the systemic trading and credit losses, and unable to cope with the reintermediation of large off-balance sheet exposures from the imploding shadow banking system. The crisis was intensified by panic deleveraging and by the interconnectedness of systemic institutions via OTC transactions. “During the most severe episode of the crisis, the market lost confidence in the solvency and liquidity of many banking institutions.”
In the tradition of a certain kind of official document, it gets more punchy as you get away from the executive summary; additional factors are either spelt out in the detail of the reforms, or implicit: OTC securities markets were mostly a bad idea; there was a complete failure to manage counterparty risk; opacity of accounting facilitated the gaming of capital rules, which concealed massive leverage; there were “basic lapses” in liquidity risk management; there was excessive reliance on ratings, which led to outsourcing of risk management functions properly kept in-house; initial margins were too low, and shot up during the crisis, intensifying the liquidity squeeze; there was “Measurement Error”; “model risk” was underestimated. It turned out that banks tested their credit risk models very badly indeed, and they were impressively defective. This last applies particularly to credit risk models: for instance, counterparties defaulted just when volatility was at its peak and exposure at its highest (think AIG, monolines); which appears to have come as a surprise. The Basel committee rejects more colourful coinages (“chocolate teapot risk”), and calls this “wrong way” risk.
That all sounds about right. It contrasts nicely with this year’s US regulatory reform programme, and it’s quite a refreshing change after a year listening to people trying to blame it all on the CRA, feckless homeowners, Fannie and Freddie, regulators and yada yada yada.
Anyhow, multiple market failures propagated across the globe at unprecedented speeds. There was a fierce contraction in global liquidity, in cross border credit availability and in international trade. The public sector had to step in, and taxpayers are on the hook for many years to come.
That apart, the talent earned its bonuses.
Finally (and perhaps I am putting a little more emphasis on this point than the Basel Committee do), the procyclicality of the existing regulations further amplified the procyclicality of in-crisis deleveraging; and there was regulatory arbitrage. Putting it another way, Basel I (1988) simply wasn’t relevant any more, and thoroughly gamed; Basel II (2004) didn’t help much either, and probably made things worse.
The Basel Committee reasonably concludes that “it is critical that all countries raise the resilience of their banking sectors to both internal and external shocks”, and that, since greater resilience at the individual bank level reduces the risk of system wide shocks, the needed reforms must have a dual focus: microprudential and macroprudential.
So the Basel Committee want to strengthen global capital and liquidity regulations. Excellent. Their auxiliary aims: improve banks’ risk management and governance; strengthen banks’ transparency and disclosures; strengthen the resolution of systemically significant cross-border banks. Quite some shopping list, on which they haven’t made uniform progress.
The capital rules first, which seem to be relatively cut and dried, and mostly not too bad, and more or less finished, plus or minus feedback and impact assessments.
First, with restated purposes, the Tiers are getting simpler, and a whole host of accounting dodges, sub-tiers and other wheezes are getting obliterated. Tier 1 is there to help a bank stay a going concern; it is to be common equity, or at least some kind of capital that absorbs losses in exactly the same way as common equity. Tier 2 is there to absorb losses once the firm is broke. Regulatory capital adjustments “must be harmonized internationally”; we’ll see how that works out. There are new disclosure requirements which might actually make the capital levels of different banks comparable, at least until the accountants figure out how to obfuscate the new disclosures. Tier 3 is swept away, along with concessions for goodwill, unrealised gains on investments, goodwill, deferred tax assets, investments in own shares, various cross holdings (other bank equities, not bonds as far as I can see, might be an important omission there), other investment stakes, pension fund assets, changes in the bank’s own credit risk. That list is not exhaustive, but you can see that a slew of favourite dodges appear to be doomed. “Yee-hah” is not classic Basel Committee language, but if this part of the document had a watermark, that would be it.
Second, the Committee is jacking up the capital requirements for counterparty credit risk exposures arising from derivatives, repos, and securities financing activities. This is meant to cope with “wrong-way” risk and mark-to market hits from highly correlated assets all heading south at the same time in heavily coupled markets. This bit isn’t all good and I’ll come back to it.
Third, they want to supplement the risk-based capital requirement with a leverage ratio. What’s not to like about that? It should “help contain the build up of excessive leverage in the banking system, introduce additional safeguards against attempts to game the risk based requirements, and help address model risk”. Reading between the lines, only a little bit, the Committee thinks its new Basel III measures may be just as porous as the old Basel II ones, and is slapping this on as ‘belt and braces’. Highlights: no netting for repos, no allowance for purchased CDS protection, written CDS included at notional value. This is where off-balance-sheet items get their comeuppance, too, though again, there are accountants in the mix, so who knows how heavy a blow will actually land on VIEs? The implementation of FAS167 in January will be a straw in the wind: my gloomy bet is that rather less than expected will actually revert to bank balance sheets. But that’s not the regulators’ responsibility, yet.
Next up, the global liquidity standards. The intention here is not so much to prevent fiascos like Northern Rock or HBOS, (both of which would have been lossmaking, rather than hopelessly illiquid, if they had had sane funding models), but rather, to enable the authorities to get warning of an approaching liquidity problem, or at the very least, to give them a known timeframe in which to sort out the problem once it’s hit. So there are three prongs: first, a 30-day liquidity coverage ratio requirement, so that the authorities have that long to sort out new capital from somewhere, even in the face of a bit of a run. Despite the BoE’s legendary slow response to Northern Rock, 30 days really ought to be long enough to sort out new capital for an institution of that size. For the likes of JPM or Citi, probably still enough, though there might be pressure from a more severe run than the base scenario in the document. Second, there is a longer-term structural liquidity ratio to which banks must adhere; common sense, really. Third is a common set of monitoring metrics to assist supervisors in identifying and analysing liquidity risk trends at both the bank and system wide level; also common sense.
The Unfinished
Impact assessments…rather a lot of impact assessments. One would really love to see how those changed capital and leverage requirements affect European and US banks. Will it be sharp intakes of breath, stunned silence, and perhaps some shouting (good sign), or ritual whingeing and (in private) big shrugs all round (bad sign)? Wait and see.
Then there’s the consultation, a.k.a. watering down process.
And they really need to put a whole lot more meat on the “reducing procyclicality” idea, which is a rag bag of ideas to mitigate the massive positive feedback loop in asset prices in a crisis (for which mark-to-market accounting get some of the blame). The Basel Committee are in a bind with mark-to-market. They kind of knew that mark-to-market was procyclical, but crossed their fingers and went ahead. That didn’t work out so well; now it’s back to the drawing board.
The next idea is to mandate countercyclical capital buffers that can be drawn upon in periods of stress. As opposed to paid out in bonuses, for instance. Me, I think the UK’s interim 50% tax on bonuses is a great stopgap. In the spirit of offering predictions to suit all tastes, I suggest you look for that tax to vanish mysteriously in the next few weeks, last until the next election, or persist until Basel III is adopted by the UK regulators; and maybe get copied by Spain.
The next aspiration, not really even an idea yet, is to address systemic risk and interconnectedness. The Basel Committee are looking at charging extra for systemically interconnected banks. If I remember correctly, those are the ones with the really powerful lobbies. We’ll see how it goes.
And lastly, the timetable:
“A comprehensive impact assessment of the proposed capital and liquidity standards, to be carried out in the first half of 2010”. So the final regs will depend a lot on where the lines are drawn in this impact assessment. Which banks are to be deemed representative? The “average” ones?
“Review the regulatory minimum level of capital in the second half of 2010, taking into account the reforms proposed in this document to arrive at an appropriately calibrated total level and quality of capital.” Groovy!
“A fully calibrated set of standards by the end of 2010 to be phased in as financial conditions improve and the economic recovery is assured.” Whenever that is. It makes you wonder how the banking lobbyists will argue the case a) that we’ve recovered and b) we haven’t recovered enough for the new rules to be phased in. I’m sure they can thread that needle just fine.
Aiming at implementation by end-2012; with “appropriate transition and grandfathering arrangements”. So: (with thanks to the Basel Committee for their economic forecast) a bit of work for the IBs, nicely spread out, as banks rejig their capital. It’s an ill wind…
The Ugly
An intended side effect of the increased counterparty capital requirements is to increase incentives to move OTC derivative exposures to central counterparties and exchanges, which, it is hoped, will reduce coupling. This sounds fine for the more vanilla end of the market – interest rate swaps and the like, where genuine hedges of various kinds exist. It’s less convincing with Credit Default Swaps, which essentially start out as a mechanism for pretending that you can short corporate bonds, when you can’t, and spiral off into full-blown psychosis from there. To point to the anti case, I’ll just recycle my link from my recent post about the FSA and Basel III. Net effect: FSA has parted company from the Basel Committee on this point already. It looks as if there will be a shot at getting CDS traded via a central counterparty anyway, so we’ll get a chance to see who was right, eventually.
Rating agencies reform isn’t pretty. They are as firmly embedded in the new Basel rules as they were in the old. Once again the Committee is in a cleft stick. They rightly mistrust banks’ internal ratings as an alternative, but are reluctant to admit the possibility that the rating agencies aren’t really all that independent from banks. Perhaps they are just being polite to the US.
As mentioned, banks are given a general whack on the muzzle for not testing their credit risk models properly, and are given a whole slew of new specifications of what their models should be like. We are now to have stressed EPE (an attempt to tackle wrong-way risk), stressed VaR (actually this was specified back in July), extra back testing requirements, extra stress-testing requirements, extra model validation requirements. This will be a geeks’ bonanza.
The jury is out on how well one can expect to manage wrong-way risk. I can’t see how any numerology can possibly do it, since it doesn’t have any insight into the deteriorating fundamentals that drive a company’s creditworthiness.
But at least there’s another chance to take a kick at the corpse of VaR, now tricked out as Stressed VaR, but just as dead as it was before the makeover. I am indebted to David Murphy of the blog Deus Ex Macchiato for some nice references, quotes and data points here. Stressed VaR has had a wee bit of testing (in here, section 5). On average, using stressed VaR increased market risk capital requirements by 280%. Not on average, it increased market risk capital requirements by anything from 7% to 700%, depending on the bank, a level of variability in the results that makes the average pretty meaningless.
Not as meaningless, however, as Stressed VaR itself. From Murphy: “One need only compare JPMorgan’s capital allocation for market risk — $9.5B at Y/E 2007 — with its VAR — $107M — to see the problem with trading book capital based on VAR alone.”
OK, now multiply that $107M by 700%. I think JPM still won’t believe their VaR numbers….
In another twist, the Committee promotes the use of CDS to hedge against credit valuation adjustments (CVA – this is where the value of a position is adjusted by changes in the counterparty’s creditworthiness – it led to huge mark-to-market losses in the Crisis). So the wrong-way risk engine had better be a good one, otherwise the CVA hedges will blow. Not sure that the new world is much different from the old, here. Anyone (apart from banks, that is) want to be long credit risk on a group of banks in a crisis?
The End
That’s about it. Wait for the impact assessments and the final rules. Don’t hold your breath. |
