Could a phase-in save FRTB?

For regulators wondering how regional banks are getting on with their efforts to implement the Basel Committee's Fundamental review of the trading book (FRTB), Risk.net's inaugural survey on the subject should be an eye-opener.

Asked for their views on key elements of the framework, 11 regional banks from around the world answered – and their responses made for uniformly grim reading. Top of the list of concerns was securing approval for trading desks to use the internal model approach (IMA) under the regime. More than half of those surveyed feared 20% or more of their desks were likely to lose IMA approval; two said 80% or even 100% of their desks would be in the firing line.

Losing approval would be a blow: desks that fail to qualify for the IMA will be forced on to the more punitive standardised approach, which could produce a capital jump of between 2 and 6.2 times.

The biggest barrier to a desk gaining approval to use IMA will be Basel's fiendishly complex P&L attribution test, used to determine whether a dealer's risk models track the desk's actual performance sufficiently closely – something they can no longer rely on backtesting alone to prove.

In some quarters at least, regulators seem alive to the problem. An analysis prepared by the European Banking Authority (EBA) for the European Commission suggested approximately 40% of desks would likely fail the P&L attribution test. This matters, since it will be up to the EBA to pen the technical standards that will underpin Europe's version of the rules once the EC's legislative proposal has been agreed on and transposed into European law next year. A sympathetic hearing for dealers here would be very useful indeed.

Front and centre in the EC's proposal when it emerged in late November was an acknowledgement that implementing FRTB is likely to lead to a "steep increase" in banks' own fund requirements for market risks – something European legislators fear could lead to the continent's capital markets seizing up if all aspects of the rules enter into force simultaneously. That has led the EC to call for a three-year phase-in for the framework.

A longer lead-in time would help the market: almost half of Risk.net's survey respondents said they would need at least three years to fully implement FRTB. In addition, a desk wishing to use the IMA must have been running P&L calculations for at least a year before applying the attribution test, further eroding the time available to get the ducks in line.

Regional lenders, for which the barriers to implementation look more difficult still – either the markets in which they operate are too illiquid to apply the P&L attribution test successfully, or they have too few quants to cope with a shift in modelling approaches – will be hoping local market regulators take a leaf out of the EC's approach. The longer Europe seeks to delay full implementation of the regime, though, the greater the risk of global regulators' approaches diverging.