Stop-gap rules on banks' trading books may add perilous complexity.

The nnew year hangover throbbed agonisingly for investment bankers this year. Blame Basel 2.5, a new set of international rules which charges banks higher capital for the risks they run in their trading books (as opposed to their banking books, where they keep assets that they intend to hold to maturity). Those charges were too low before. And heaping higher costs on banks should please politicians and Joe Public. But they add another layer of complexity to banks' risk management.

Basel 2.5 came into force on December 31st in most European and major world financial jurisdictions. Switzerland applied the rules a year early, and the costs are substantial. Third-quarter figures for Credit Suisse show a 28% increase in risk-weighted assets, and hence capital charges, for its investment-banking activities purely because of Basel 2.5.

The most notable laggard is America. US financial regulators do not oppose Basel 2.5, but it clashes with the Dodd-Frank act, America's big wet blanket of a financial reform. Basel 2.5 uses credit ratings from recognised agencies such as Moody's and Standard & Poor's to calibrate capital charges. Dodd-Frank expressly forbids the use of such ratings agencies, whose poor judgments are held partly responsible for the crisis. Instead American regulators are working on their own cocktail of credit-risk calibrations for Basel 2.5, using market data and country-risk ratings from the OECD. Their solution is still months away from application (though not as distant as implementation by the Russians or Argentines).

Basel 2.5 for the first time charges banks extra capital for the credit risk of what they hold in their trading portfolio (because the crisis showed that markets are not always liquid enough to be able to offload assets). That includes a charge for the risk that a counterparty goes bust. It also imposes heavy charges on securitised bundles of assets unless the credit risk of each piece of the bundle has an identifiable market price. Banks that have portfolios of trading positions which they reckon offset each other have to convince regulators that their risk models work or face being charged at a cruder, standardised rate.

The problem with Basel 2.5, recognised by regulators and bankers alike, is its complexity. The risk of a trading portfolio must now be broken down into five “buckets”—value at risk (VaR), a measure of how much could be lost in an average trading day; stressed VaR (how much could be lost in extreme conditions); plus three types of credit risk ranging from the risk of single credits to those of securitised loans. Traders are understandably confused. For some banks, developing risk models and getting them approved is just too expensive: more complex businesses will be shut down. That will please those who want banks to be more boring.

But unintended consequences will doubtless follow. Useful products may become less tradable. Trading of riskier products could migrate to unregulated entities. Banks may be tempted into new forms of regulatory arbitrage, by juggling assets between their trading book and their banking book. Worst of all, perhaps, is the increased risk of back-office bungling because of the extra complexity.

Regulators recognise this risk. The Basel Committee on Banking Supervision, which drew up the rules and is also responsible for the full Basel 3 regime that will come into force in 2019, is still conducting what it calls a “fundamental review” of capital rules for banks' trading books. Publication is not expected before March. Those sore heads will not soon clear.