A step too far

Dealers have complained for some time that certain aspects of Europe's Markets in Financial Instruments Directive (Mifid) seem to have been designed with cash equities in mind. Nowhere is this clearer than in the requirement for best execution. It's tricky enough for equities, but, if price is a priority for a particular customer, a dealer can ensure it complies with the best-execution obligations by documenting quotes on all available execution venues, and taking exchange fees, commissions, taxes and clearing and settlement costs into account.

For derivatives, however, this is nigh on impossible. The European Commission seemed to acknowledge this in its Level 2 draft implementation guidelines, published in February. The paper concedes that the price of a derivative is based on a number of variables, such as the dealer's view on the credit risk of the counterparty, the amount of collateral held, and the volatility and liquidity of the underlying market. As such, best execution cannot be applied in the same way - although the Commission does state that dealers should act in the best interests of clients and, wherever possible, seek comparisons of market values used in the pricing process.

The UK's Financial Services Authority (FSA), however, seems to have taken the idea of best execution a step further. In its discussion paper on the implementation of Mifid's best-execution requirements, published in May, it moots the idea of using benchmarks in the fixed-income and derivatives markets. This has caused consternation in the banking industry - starkly illustrated by the joint response to the paper by the Bond Market Association, the International Capital Market Association and the International Swaps and Derivatives Association (see page 9).

They point out that benchmarking is incompatible with derivatives and structured products, that there is no 'right' price for a derivative, and that transaction terms are individually negotiated, incorporating a variety of parameters that differ from firm to firm.

The only way to obtain a benchmark would be to rely on internal models. However, different firms use different models, even for vanilla instruments - and those using the same models may still come up with different prices, depending on their view of credit risk and other subjective factors. Internal pricing models are seen as a competitive edge for banks - if forced to disclose details of these models, it could stymie innovation and persuade dealers to withdraw liquidity in certain markets, the associations say.

It's difficult to argue with these points, and it's difficult to understand how the FSA - which, on the whole, has a reputation for understanding how derivatives markets work - can have neglected to see the difficulties of using benchmarks for derivatives. It will now consider the responses to its discussion paper - let's hope it listens to the comments made by the industry.

Nick Sawyer, Editor.