Saturday 18 December 2010

More on the lunacy of the Basel Accords

I was looking at the preferred asset classes under the Basel Accords in my previous post on why central banks are so determined to stave off a government default, and realised that every single asset class that is given less than a 100 percent credit risk weighting is now tainted by widespread default, scandals or bailouts.

The credit risk weightings mean that instead of reserving the standard 8 percent of capital in respect of a debt, the bank can cut that by the weighting applied to the asset class. Effectively, the reduction in credit risk weighting operates as a powerful subsidy to the borrowers and equally powerful incentive to over-leveraging the lenders.

As a baseline, all financial, consumer and corporate debt must be reserved at a credit rating of 100 percent of 8 percent, unless explicitly discounted. A weighting of 50 percent, for example, means that instead of holding $8 reserves on a loan of $100, the bank only needs to hold $4 of reserves. A zero weighting means they lend $100, but hold no reserves at all.

Mortgages get a credit risk weighting of 50 percent, and we all know how well the mortgage market is performing. Mortgages and mortgage backed securities became the largest asset classes globally in a matter of years thanks to the credit weighting subsidy and securitisation. If I recall correctly, our present long crisis started with the collapse of the sub-prime market and now all categories of US mortgages are impaired by the ongoing mess with MERS and fraudulent or missing documentation. Borrowing short to lend long brought down Northern Rock in the UK and many other over-leveraged mortgage banks.

Interbank debt gets a credit risk weighting of 20 percent. We've seen from the collapse of interbank lending that banks do not trust each other. At the same time, inter-bank exposures and credit derivatives mean that financial institutions are massively dependent on each other, such that bailouts are justified as essential to prevent systemic collapse. If Too-Big-To-Fail is predicated on the systemic impact of a bank's failure on other banks, it would seem that the 20 percent inter-bank risk weighting was and is unsound.

Government agency debt gets a risk weighting of 10 percent. Looking at Fannie and Freddie, and the serial scandals and bailouts they have occasioned over the past decade, it is hard to see how such a subsidy can be justified.

Finally, sovereign debt of Zone A states is zero weighted - no reserves required at all. Zone A includes any country in the EEA, full members of the OECD, or states that have concluded special lending arrangements with the IMF except that any state that reschedules its debt is excluded from Zone A status for five years.

So the current financial crisis started with bad mortgage debt, spread to bad bank debt, carried over into bad agency debt, and now encompasses bad sovereign debt. Each of these categories was given preferential capital weighting under the Basel Accords, and now all are open sores on the financial system and the stability of excessively indebted governments.

Not only did the Basel weightings encourage poor risk assessment, they directly contributed to the inadequate capitalisation of banks for the risks they assumed.

And yet, have you heard any regulator take responsibility for regulatory failures yet? I haven't. In fact, I've seen no historic analysis of capital requirements, deregulation of credit markets, securitisation, derivatives, demutualisation, or any of the other regulatory policy innovations which should reasonably be matters for review in assessing the causes of the current crisis.

As the bankers and regulators do not seem keen to be reflective about their own policies and conduct, it's hard to imagine that they can craft constructive reforms to make the system safer or more efficient in future.

I've downloaded the draft Basel Accord III, released this week, for leisure reading. Sadly, I'm trending to the view that all harmonised regulation is likely to end in disaster as it precludes independent judgement and sensible challenge to orthodoxy. Once something has been agreed by a big enough committee, it becomes impossible to question whether it makes sense. Ultimately the unintended consequences of incentives and distortions mean it won't make sense, but by then it's far too late to change course and break from the herd.

Monday 13 December 2010

Basel Faulty: Sovereign Defaults and Bank Capital*

A foolish consistency is the hobgoblin of little minds, adored by little statesmen and philosophers and divines. - Ralph Waldo Emerson
When historians of the future look objectively at the era preceeding this long financial crisis, they might well conclude that failure of the globalised capital system is traceable to the Basel Accords.** The unreasonable assumptions and myriad distortions introduced in this one-size-fits-all paradigm of bank capital adequacy fatally undermined the practice of independent judgement in assessing credit risk and prudential supervision of banks.

Bankers of the past had to assess the creditworthiness of a debtor or counterparty based on balance sheet, revenue potential and management reputation for competence. They husbanded their scarce capital, aware that each dollar lent remained at risk until repaid, with cash reserves proportional to the bank's assets usually 8 to 10 percent.

A primary fallacy of the Basel Accord is that OECD government debt is risk free and requires no bank reserves. Better yet, the banks can count the government debt they hold as Tier 1 capital, reserving against other debt assets. The Basel Accords assume all OECD government debt is a cash proxy, being liquid in all market conditions.

Walter Wriston's 1970s dictum that "sovereigns can't default" was disproved in the Third World Debt Crisis of the 1980s, but somehow the BIS Committee on Bank Supervision still embraced it when applied to OECD state debts.

Roughly, the risk weights of the main asset classes under Basel I were:
- zero for Zone A (EEA and OECD) government debt of all maturities and Zone B (non-OECD) government debt of less than one year;
- 20 percent for Zone A inter-bank obligations and public sector entity debt (e.g. Fannie Mae, Freddie Mac, et al.);
- 50 percent for fully secured mortgage debt;
- 100 percent for all corporate debt.

The post-Basel bank supervisors applied their prudential supervision models unthinkingly, to rubberstamp the bankers' leveraging of their balance sheets toward ever greater excess. No one bothered to ask whether Basel Accord assumptions made sense. They were the harmonised norm for prudential supervision and too deeply embedded in the fabric of international finance to adjust, except to allow more and greater leverage in Basel II through liberal recognition of derivatives. Basel II allowed the banks to offset even more risk exposure with even less capital through collateral, securitisation, credit default swaps, and recognition of the validity of internal asset valuation models.

Global harmonisation of prudential supervision around the Basel Accord meant that the hobgoblin of excess leverage became systemically entrenched in all markets, in all nations. The foolish consistency of harmonised capital adequacy was adored by little minds of global bankers and central bankers worldwide.

The zero weight for OECD government debt must have appeared a harmless subsidy to OECD governments in 1988, promoting liquid government debt markets and enhancing the competitive positioning of OECD-based global banks who stood to gain most from the harmonisation of global bank regulation and capital rules.

Leveraging their balance sheets to work every dollar of capital harder became the obsessive preoccupation of two generations of bank executives once the Basel Accords were adopted. Risk management departments were less about controlling exposure to adverse credit events than about identifying deal structures which would minimise the amount of regulatory capital allocated to any exposure.

Fannie Mae, Freddie Mac, and their ilk were an early mechanism to reduce the reserves required from 50 percent on an individual mortgage to twenty percent or even zero, allowing the banks to write more and more mortgages with less and less capital. When these entities proved inadequate in the go-go 1990s, asset back securities allowed banks to get sub-prime mortgages - and thereby the capital requirement - off their books entirely, passing the risks to yield-hungry investors. With Basel II they could reduce capital even further by writing each other a daisy chain of credit default swaps for all categories of exposure. Who could have known that it would end badly?

OECD government debt is zero risk weighted and accounts for a disproportionate bulk of Tier 1 capital of major banks. A default by any EEA or OECD government will force banks and central banks to recognise that government debt has inherent risk like all other debt. This would force recognition of a positive risk weighting, and bring into question the assumption that government debt can be counted as a cash-proxy in Tier 1 reserves. The illiquidity of impaired or defaulted government debt would undermine its role as a Tier 1 reserve asset in bank capital models.

At this writing the OECD governments at risk of default are Greece, Portugal, Spain and Ireland, with other states queuing up in the wings. Already the ECB is the only buyer in the market for much of the impaired government debt.

If any OECD state were to default there would be very serious implications:
- The Basel Accord zero risk weight of government debt would be proved fanciful;
- The assumption of government debt as a liquid asset suitable for bank Tier 1 reserves to meet unanticipated and sudden cash demands will become unsustainable;
- Banks would be forced to recapitalise at much higher levels, forcing even sharper deleveraging and contraction of lending;
- Governments would lose the captive, uncritical investor base they have relied on to finance excess public expenditure for the past 30 years;
- Central banks could be forced to suddenly monetise even more government debt if required to meet the cash demands of a run on their undercapitalised banks.

Looked at this way, you should be able to understand why the ECB keeps repeating that there can be no Eurozone sovereign default, and why the UK and US are staunchly behind them in preserving the illusion of state solvency for all Eurozone states.

It will likely prove impossible to reform the bankers and central bankers dependent on the Basel Accords for their business models and careers. Harmonisation of global standards was supposed to make the world safer. A foolish consistency on bad policy and bad practice led instead to a world on the edge of financial implosion.

This hobgoblin haunts us all.
______________________________

* Hat tip to Tracy Alloway at FTAlphaville for suggesting a revision to the title for this post.
** Several commenters questioned my spelling of Basle Accords in the original post. I prefer the British/French Basle which was standard in my youth on Basle Accord documents, to the German/American Basel more common and current today. Nonetheless, I yield to the customs and usage current today in changing the spelling to Basel.

Thursday 9 December 2010

The Return of London Banker?

I am back in the City that I love, at my pleasantly cluttered desk, after an adventure of nearly two years. I hope I did some good.

Looking back at what I published here and on Roubini.com in 2007 and 2008, I am proud of the passion and fluency I wrote with as we shared the experience of the financial crisis in its early stages. Some things I got right, some I got wrong, but that is to be expected when observing events honestly and with imperfect information as they unfold. I have not written with such passion since. I am not sure how or whether to start again.

Does anyone remember that I used to write? Does anyone care if I write again?

Let me know in the comments below.