An article in the FT last week (Forex trading speeds up moves to computer platform) made the astonishing claim that a kind of fraud fatigue arising from continuing revelations about rogue traders is prompting banks to accelerate the move from voice to electronic trading. Further, the piece argues that this is futile since electronic trading is itself also run by bank staff who, though differently skilled, are no less likely to be corrupt. The article was not penned by a blogger on the lunatic fringe but the authoritative chief Investment Banking correspondent of the Financial Times: the claim is astonishing but it may nonetheless be true.

In reality, it doesn’t matter whether or not fraud fatigue is driving banks to go electronic: they haven’t got any choice. Clients have voted with their trades to transact more electronically and this has been motivated by convenience rather than suspicion of fraud. The effect of suspicion isn’t to increase the e-ratio but to change the rules under which e-trading is conducted. Regulators understandably want to minimise the opportunities for dealers to collude amongst themselves, use client trades to their own advantage or treat clients inconsistently and unfairly. In prescribing how e-trading must work, regulation is increasingly shaping Fixed Income, Currencies and Commodities (FICC) markets to an Equities template. It’s far from obvious that this is the best outcome for anyone but the fraud factor introduces an urgency that now guillotines more considered analysis.

Regarding the FT article, a friend from a large trading firm lamented the “incredible amount of misunderstanding and myths of black magic” around e-trading. The sense amongst observers, including regulators, that FICC market-making counts amongst the dark arts is leading them to demand simplicity and transparency.

Unfortunately, for electronic market-making to work transparency at trade-time has to be suspended. To simplify a little, the job of the dealer (or, more accurately, the dealer’s market-making computer) is to trade with the client at the tightest price consistent with the expectation that the other side of the trade can be found with another client before an adverse market move leads to a loss on the position. This is a fair service in which the client gets a trade executed without any risk of price slippage and the dealer gets paid through spread capture. Dealers compete, to the benefit of the client, to win business through tightness and consistency of spread.

Weighing against the dealer are a multitude of (non-bank) traders who seek to aggregate numerous small profits from very short-term price fluctuations. To them, knowledge of the price pressure arising from a dealer working a risk position from client trades is monetisable to the dealer’s disadvantage.

For the dealer’s role to be viable a number of practices must be permitted. For example, the dealer must be able to show a skew – a better price on one side of the trade than the other – to clients who may thus be attracted to place risk-reducing trades, while hiding skews from counterparties who might use the skew as a “signal”. The dealer must be able to assess continually the likelihood of short-term price moves and to show tighter spreads in quiet markets and wider spreads at more volatile, riskier times – and a better service is possible when the dealer can balance price consistency against spread tightness differently for different clients. Further, the dealer must be able to show tighter spreads and more liquidity to clients whose trades are not routinely associated with adverse market moves at around trade time.

These and other practices add opacity to pricing in order to improve it. As Profit and Loss’s Colin Lambert says, best execution and transparency don’t go together. Nevertheless, regulation is propelling FICC towards an Equities model in which the role of the market-maker is significantly diminished and the burden of reducing “slippage” is transferred to the client, who is partially compensated with an enriched market in DIY execution tools. Equity exchanges, which do have rules about trade-time transparency, are highly congenial to high-frequency electronic quantitative traders, whose activity dominates trading volumes. They are, predictably, the only markets yet to have experienced a flash crash.

What does this have to do with eCo? If banks are to argue for the retention of FICC-style electronic market-making – and it may be too late – the suspicion of dark practise needs to be exorcised. This requires a different kind of transparency. One remedy would be for a major dealer to make an electronic market-making stack available on eCo. This would recognise that the franchise value enjoyed by the top banks springs from their client relationships, their credit lines, their global distribution and the completeness of their platforms. The techniques used for implementing e-trading may use PhD mathematics and buzzword technology but that is not where their franchise value lies and the aura of rocket-sciencery does nothing to build public confidence. Opening up the apparatus of electronic market-making offers the best shot at securing  a place for it in the post-crisis landscape.

The other rationale for an eCo code listing is to promote excellence: code sharing is quality-increasing. No matter how smart they are, a limited number of experts working for one bank are more at risk of oversights than a wide community of buyer-users whose scrutiny is not clouded by pride of authorship. And in e-trading the golden rule is Make no errors.

I have no expectation that the Front Office software that implements what is often referred to as the “secret sauce” will be posted on eCo – not, at least, by the leading Investment Banks. It will be an opportunity missed. This will be another area from which Investment Banks retreat in response to the immense pressure to focus squarely on making loans to parties who may not repay them.


Next week and going forward at the suggestion of Bill Stockton at KLever this column will come out on Tuesday rather than Monday.