The beginning of the end for Basel?
The Basel Committee has agreed to delay a decision on the most controversial aspect of its new capital rules. It promises to catch up ‘in the near future’. But two years into the process, a conclusion looks no closer than at the start.
On Tuesday, the Basel Committee pushed out a short, enigmatic press release, announcing that “more time is needed to finalise some work”, and that therefore next week’s big reveal — supposed to be the end of the post-crisis regulatory drive — was postponed.
The Committee has had plenty of work on its plate for the past seven years, so a short delay need not suggest a disaster.
But the Committee has not left just a tidying-up exercise until last but rather the centrepiece of the new regulatory architecture. If the Committee can’t agree on how to calibrate capital floors, it might as well pack up.
The journey to this point started in 2014, when, fresh from passing Basel III rules — a hasty post-crisis safety net — the Committee started looking at how to optimise them.
A prime target in the new regime was bank internal models for capital calculations.
Bank bashers hated the idea that banks were “setting their own rules” by using internal models, while US institutions and US regulators saw such models as a sop to undercapitalised European banks — a vehicle for sweetheart deals that cut capital costs in some European jurisdictions.
In December 2014, the Committee published its proposed solution.
Banks would have a limit on how much they could cut capital using their own models. Internal model-driven capital would be between 60% and 90% of the capital mandated by the standardised approach (a far more prescriptive method where the Committee sets out exactly how to calculate the capital allocated to each individual asset).
For banks with large books of low-risk assets, this spelled trouble, in the shape of sharply higher capital requirements. Nordic banks were expected to see their corporate loan costs jump sharply, as were Dutch and German mortgage banks, and UK buy-to-let lenders. Universal banks with big legacy asset backlogs were also expected to face fierce capital hikes.
The intellectual problem with removing banks’ ability to model the risk of their own assets was obvious. The Basel Committee had to treat all corporate loans, mortgage loans or any other asset as equivalent. The Committee could not very well pass a rule that treated Irish mortgages as riskier than German mortgages, even though banks, and the wider market, clearly do.
Bankers also pointed out the Committee had already passed a rule, the leverage ratio, designed to stop banks manipulating their capital requirements too much. Cut risk-based capital using internal models overly, and the leverage ratio, rather than risk-based capital, would become the main constraint.
To reassure the market (and soften the banker bashing tone), members of the Basel Committee took to repeating two assertions.
The first was that this was just the completion of Basel III, not the beginning of a whole new set of capital requirements called Basel IV.
The second was that the rules would not increase capital requirements — at least, not in the aggregate — though requirements for individual “outliers” would indeed rise.
Bankers, though, struggled to square these commitments with the clearly stated consultation document. Plainly, the Committee was planning a new capital rule, which would indeed increase capital requirements — otherwise there would be little purpose in introducing the rules.
Then the authorities got involved, with the internal disagreements on the Committee spilling out publicly. Late last year, Andreas Dombret of the Bundesbank seemingly threatened to pull out of the project.
“In theory, it acts as a way of putting a stop to the frivolous calculation exercises associated with using internal models,” Dombret said. “In practice, however, it works against the focus on risk. This is something we’re not prepared to accept."
It’s in this context that the Basel Committee announced Tuesday’s delay.
Since the publication of the consultation in 2014, the basic battle lines have been obvious. Since planned revisions to the standardised approach were published in December 2015, followed by limits on when models can be applied in March 2016, there haven’t even been any major technical changes to pick over. The decisions left to fight about are picking a number between 60% and 90%, and deciding how quickly Committee members should implement the rules.
Chewing over these issues for another couple of weeks, however, seems unlikely to clarify anything. The Committee faces the same choices it faced in the summer —armed with the same information — and the same set of interests on either side.
It can either pass a rule that the Europeans, led by the Germans, will accept — something sufficiently soft or with a long enough phase-in that it might as well never happen — or it can pass a US-friendly proposal and risk European authorities ditching the process entirely.
A delay might bring a changed approach from the incoming Trump administration in the US, but the only change that would be useful is dropping the US demand for a higher capital floor of 75% or more, which is just another way of gutting the proposals.
The decision on which way to jump is basically a political one, and a consensus-based technocratic talking shop, used to passing near-unanimous and non-binding resolutions, is not set up to settle such questions.
But if it can’t come to a conclusion on the fundamental questions of how much capital banks should have, perhaps the Basel project can’t do much at all. This delay could be a sign that Basel is finally broken.