Brazen bank bandits busted
To paraphrase Gwyneth Paltrow, life is harder for bigshot currency traders than for your average nine-to-five smartass.
They've just taken the sleepyheads at Big Index Fund for megabucks with a transaction so clever only a coke-fuelled, three-nippled, obsessive-compulsive PhD astronomer could think of it, and they can't tell anyone. Because if they did, everyone would know how to do it and they wouldn't be able to make millions any more. So they have to bottle up their bragging rights.
No hints, not even a little piece of origami left on BIF's front desk. Must be tough, like Clark Kent having to hide his superpowers. Still, it is no wonder alleged foreign exchange market manipulation can stay under the radar so long - the only people likely to know about it were those filling their boots.
Who would blow the whistle on their own bonus?
It also helps explain how the deeply flawed market practices that allow manipulation can be left undisturbed for years. There are two - well, at least two - big faults in the market well known to those in the trade. Combined, they created perfect conditions to nurture the behaviour that could lead traders to collude.
Given the allegations so far by customers of Bank of America, Barclays, BNP Paribas, Citigroup, Credit Suisse, Deutsche Bank, Goldman Sachs, HSBC, JPMorgan, Morgan Stanley, RBS and UBS, Chalkie will be intrigued to see what comes out of the Commerce Commission probe in New Zealand. The implication of the regulator's statement so far is that there was a cartel and that a New Zealand member has owned up to avoid being clobbered by enormous penalties, which can be up to 10 per cent of turnover. How the NZ squealer did its thing remains to be seen, but from reporting by Bloomberg and the lawsuit filed in New York we already know a fair bit about what some traders are alleged to have done.
For all its size, the currency market is almost entirely conducted in private.
Transactions happen between two parties without the intermediation of an exchange. Typically, traders at the big banks transact with each other to execute the wishes of clients, who may be companies or investment funds, for example.
They make money by taking a small slice of each trade in the form of a spread between buy and sell prices. The size of the spread depends on the level of competition among traders and the ease of finding a counterparty to the trade.
As well as transactions for clients, banks will trade on their own behalf as they try to profit from their knowledge of the market.
In this situation, information about what transactions are about to happen becomes valuable.
If a trader knows client A has a big currency order on the way which will affect the price, they can use that knowledge to take positions beforehand that will profit from the price move. This is known as front running.
In practice, however, the currency market is so big that it can be risky for traders to rely on knowledge of their own bank's order flow - who is to say Rival Bank does not have its own client transactions in the pipeline that will affect their plans?
The problem can be particularly severe around 4pm, London time, when currency rates are sampled for benchmark exchange rates. Since 1994, the WM Company has compiled a standard set of currency rates to make it easier to compare equity portfolio valuations.
One of the most important spot rate benchmarks has become the WM/Reuters London close, which uses the median of trades taking place for 30 seconds before and after 4pm. The data are captured from electronic dealing systems run by Reuters, Currenex and EBS. Crucially, the calculation does not involve any weighting of the trades, so a small trade counts just as much as a big trade. This is big flaw number one.
The London close may have started out as a mere convenience but it has turned into a key number for large transactions ordered by index funds - passive investment funds designed to track indexes such as MSCI World.
If the MSCI index uses the London close, so must the tracker fund.
This creates the potential for large currency transactions to flood the market during a single minute as funds change their holdings to match an index at its quarterly rebalance.
Big flaw number two.
What happens here is that traders may be required by index fund clients to transact big orders at the London fix, with the result that huge volumes jam into a short period. Simply getting a deal done may require breaking up a client's order into small chunks and transacting over a longer period, which opens the bank up to the risk the overall deal will not meet the 4pm price.
Traders, whose bonuses rely on profitable trades, would rather not take that risk. According to the lawsuit, since 2003 the banks conspired to manipulate the London close by exchanging their client order flow information via chatrooms and text messages, eliminating their risk and allowing them to make super profits through advance knowledge of the market. It alleges their top traders "brazenly named their chat rooms The Cartel, The Bandits' Club, The Mafia, and One Team, One Dream."
As the Financial Times has commented, conversations in the chatrooms were "peppered with allusions to drinks, drugs and women".
Chalkie would have been more surprised if City traders chatted about seasonal vegetables and the novels of Evelyn Waugh but allowing them to talk to each other in private chatrooms when they had such a strong motivation to share information was asking for trouble. According to the lawsuit: "A transcript provided by RBS to the UK [Financial Conduct Authority] revealed that JPMorgan's Richard Usher wrote ‘messages to traders at other firms [that] included details of his trading positions'. Defendants' traders confirmed that ‘chatroom discussions between rival traders . . . allowed them to share information about pricing and order books'."
The techniques used to benefit from information sharing included front-running, "banging the close" and "painting the screen", it was alleged.
Banging the close involved splitting big orders into numerous smaller trades within the 4pm minute and using them to drive the fix up or down - a method made possible by the unweighted median calculation of WM/Reuters.
Painting the screen involved creating fake transactions with another trader "to create the illusion of trading activity in a given direction in order to move rates prior to the fixing window".
Since Bloomberg first published the forex price-rigging story last June, 29 senior traders - all men, strangely enough - have been named as involved and most have been placed on leave, suspended or fired.
Chalkie has heard it said that the artificial movement of forex rates at the 4pm fix can be thought of as the price to pay for having so much liquidity available at a specific minute in the day.
Perhaps that is so but it does not make colluding with competitors any less wrong.
The remarkable thing about this affair is how easy it would be to prevent.
If indexes used an average price for a whole day rather than a 4pm fix, for example, the problem would disappear.
So why didn't this happen years ago? Chalkie reckons it comes down to the financial industry's traditional view that it is fine to screw customers as long as they don't realise they are being screwed. Silence is golden.
Chalkie is written by Fairfax business bureau deputy editor Tim Hunter.