Please let me keep three screens.
Today’s post is from Adrian Blundell-Wignall, Special Advisor to the OECD Secretary-General on Financial Markets. The view expressed here is his own and does not necessarily reflect that of any OECD government.
This week, the German Parliament’s Finance Committee invited Paul Atkinson and me to comment on a draft bank-separation law. The draft is strongly influenced by the 2012 Final Report of the High-level Expert Group on reforming the structure of the EU banking sector chaired by Erkki Liikanen.
Liikanen proposes assigning trading and available for sale securities above a threshold of 15-25% and all activities related to market making to a separate, well-capitalized, subsidiary. This would maintain the advantages of the universal bank model in a holding company structure, but insured deposits could not be used to subsidize the trading activities. The OECD has long proposed a non-operating holding company structure for separation, by ring fencing and separately capitalizing the different activities without restricting a bank from offering a complete range of services to customers. In general terms Liikanen is similar to this. It has the advantage of also of promoting a level playing field for bank competition with stand-alone securities trading firms.
However, the proposal involving a €100 bn “separation trigger” for total assets held for trading and available for sale is not sensible. No bank under €400bn total assets with any amount of derivatives would ever be considered as long as it kept no more than 24.9% of trading assets? If a bank with around €100bn trading assets was subject to a plus or minus 10% volatility cycle in asset values, it could separated and reunited as prices rose and fell.
The Liikanen report urges the Basel Committee and the EU to deal with the shortcomings of model-based risk weighting approach in the capital rules, so that the trading subsidiary (in particular) is well capitalized and not subject to error. The Expert Group also stresses that the directives on resolution and bail-in are an essential complement to its separation proposal. Clarification and pre-notification of instruments that are not guaranteed and will be subject to bail-in should be transparent to promote trust. As far as possible, such instruments should be marketed to non-banks.
Strengthening boards, promoting risk management and disclosure, tackling incentive schemes and sanctions to ensure compliance were also recommended by the Liikanen group.
While the OECD has supported much of this over the years, we disagree with Liikanen on two important points, concerning separation and minimum capital standards.
The first major problem is that, while idea of a threshold for separation is good, the Liikanen group has not chosen the right variable on which to base the threshold. Recent OECD work has sought empirically to explore the factors that make a bank more or less risky: i.e. that take it towards or away from the default point. This research was necessary, because policy after the crisis had to be made ‘on the run’ without enough detailed empirical evidence on which to base reform. Most of the research referred to in the Liikanen report pre-dates the crisis, or is related to recent policy developments, but none of it contains research relating business model features to banks’ distance to default. Yet it is crucial to know which mechanisms are and are not supported by the data.
With respect to the business model features of bank risk, the OECD study shows that liquid trading assets, properly separated from the gross market value of (mainly over the counter illiquid) derivatives, helps to increase the bank’s distance from default and make it less (not more) risky. On the other hand derivatives are overwhelmingly the business model feature that gives rise to interconnectedness risk and default paths arising from illiquidity in crisis conditions (for example the massive margin calls in 2008-2009).
This makes intuitive sense too. Most derivatives are not standardized and trade over-the-counter, i.e. directly between the two parties without being supervised by an exchange. An institution can find itself in a position where it cannot operate because it doesn’t have the liquidity to meet immediate calls for payments on derivatives markets. Dexia is a recent example of such a failure, and if AIG’s derivative commitments had not been met from official sources, bank collapses through interdependence channels would have been difficult to contain. Liquid trading securities, on the other hand, can be sold precisely to meet margin and collateral calls —a very good thing.
This is a fairly major problem for the Liikanen report—they are not looking at the right threshold variable. The idea of a threshold makes sense, but it must be based on the key variable if banks are to be safer. In the OECD view this is derivatives: any bank with a gross market value of derivatives above 10%-15% should be considered for separation.
Putting to one side the empirical evidence for a moment, consider intuitively the case of Wells Fargo (appropriately converted to International Financial Reporting Standards – IFRS – derivatives concepts) and Deutsche Bank. Wells Fargo offers most essential services to its customers, has very low leverage, had no issues in the crisis and is one of the most profitable banks in the world. Wells Fargo received no payments from the US government in settlement of the AIG counterparty positions. Yet Wells Fargo would be considered for separation under Liikanen ‘percent-of-assets’ threshold test. Its trading and available for sale assets were around 21% in mid 2012, but its derivatives were only 6.5% of its adjusted balance sheet—a safe business model for interconnectedness risk according to the OECD research.
Deutsche Bank, on the other hand, with 40% of its balance sheet in derivatives and only 14% of liquid trading and available for sale securities would not be considered for separation on this rule. Deutsche Bank received a payment from the US government in settlement of its AIG positions equal to 37% of equity less goodwill. Enough said.
The second main issue after separation, minimum capital requirements, is dealt with extensively in the Liikanen Report, with calls for “…more robust risk weights…more consistent treatment of risk in internal models [and that] the treatment of real estate lending…should be reconsidered,…”
The Expert Group should have had the courage of its convictions. The core problem is the risk weighting system as proposed in the so-called Basel III international regulatory framework for banks. This introduces an illusory “risk sensitivity” that relates minimum capital requirements to “risk-weighted assets (RWA)”, instead of to actual balance sheets. This has evolved into a system of extreme complexity that invites institutions to look for regulatory ways to reduce RWA relative to total assets (including negotiating with supervisory authorities) rather than ensuring they really have enough capital, defeating the entire purpose of capital adequacy rules.
So long as capital requirements are based on RWA, whose relationship to the actual balance sheet is effectively a management tool, many banks (and the system as a whole) are likely to be under-capitalized. The best solution would be to scrap the risk-weight system at both global and European levels in favor of something vastly simpler and more effective. Failing that, the equivalent can be achieved by strengthening the role of the (non-risk-weighted) leverage ratio to the point where it overrides the risk-weight system.
Useful links
OECD work on sovereign debt and financial stability
