That's pretty much the name of the bank, yes. Photographer: Simon Dawson/Bloomberg
That's pretty much the name of the bank, yes. Photographer: Simon Dawson/Bloomberg

Here is the story of how a high-ranking trader at Lloyds Banking Group allegedly led a shadowy cabal of global currency dealers in a sinister scheme to rip off Lloyds Banking Group. Wait hang on. What?

The Lloyds’s dealer, Martin Chantree, alerted the other trader on Jan. 31, 2013, that his desk had received instructions from the bank’s treasury department to swap more than 300 million pounds ($499 million) for dollars and that they would continue selling regardless of price movements, said two of the people, who asked not to be identified amid a probe into alleged rigging of the currency market. The recipient of the tip worked for oil company BP Plc, the other two people said.

In the seven minutes between the communication at 10:53 a.m. and the time Lloyds began executing the order, the pound fell 16 basis points against the dollar, or 0.16 percentage point, according to data compiled by Bloomberg. As the U.K. currency tumbled, costing Lloyds an estimated $750,000, Chantree told colleagues that maybe he shouldn’t have shared the information, two of the people said.

Maybe! I read that last sentence in my best George Costanza voice, and Chantree has since been suspended, so signs do point to "shouldn't have shared." But who knows, right? Here is BP's response:

“BP did not sell sterling until after it was widely known in the market that Lloyds was an active seller,” the London-based company said in a statement. “We strongly refute any suggestion that any BP FX traders engaged in inappropriate trading activity.”

That's a paragraph worth lingering on. There's no central repository of FX trades identified by seller; Lloyds makes no public disclosure of its daily currency trades. Certainly not within seven minutes. The standard that BP seems to hold itself to is not "don't trade when you have undisclosed information," or whatever. It's "don't trade when you have information that isn't widely known in the market." Where "the market" is that shadowy cabal etc. The information only gets out to the market through informal methods: basically, traders talking to each other. About their clients' orders. Even when their clients are their own banks.

I've long assumed that some significant part of the foreign-exchange rigging investigation would turn out to be entirely explainable hedging: If a client puts in an order at 3:30 to sell at the 4 p.m. fix, it's just good risk management to do some selling before the fix to hedge your exposure. In hindsight, it might look like manipulation or front-running, but the fix for that is to disclose it more openly to the client, not to stop doing it.

But this does not seem to be that. If your client gives you a market order to just sell a lot of pounds, you have no price risk. The only reason to leak this order is to be helpful to your buddies.1 Which apparently is a good reason?

In the clubby, close-knit world of foreign exchange, traders interviewed by Bloomberg News on condition they not be named said they would share information with each other and favored clients to secure future business or get details about transactions being done elsewhere. Findings by Andre Spicer, a professor at the Cass Business School in London who’s researching the behavior of traders, support that.

“There’s a very tight network between foreign-exchange traders,” Spicer said. “If they want to make a career in the market, they need to keep sweet with each other rather than with their employer.”

Will that change? Oh let's find out,2 though I guess my money would be on "sure." A lot of people have been suspended, anyway, which might cause them to re-evaluate their sweetness priorities.

But were those priorities bad to begin with? A simple model of the job of a trader is that the more information about the market you get, and the less information you share, the better you will be. (And the better you are, the more profits you can share with your clients, and the more business you will get.) Those goals are adverse to each other, since the easiest way to get information is often to sit around gossiping with your trader buddies at other firms who have similar information-obtaining incentives. Everyone game-theories out how much information they should share and what they'll get in return, and the ones who are good at that survive and end up doing a better job for themselves, and their clients.

That's like a hopelessly abstract and idealistic view of how trading works, and obviously clients are thrown to the wolves all the time. But the assertion that one of those betrayed clients was this trader's own bank is a useful data point. It's easy to say, "I was just doing my job and being part of the information flow" if you're paying for that information with your customers' money. It means a bit more if you're paying for it with your own money. Well, your employer's money anyway.

1 Well, not the only reason. Really there are plenty of reasons you might want to get your friends to front-run your own bank's treasury department. For instance, your treasury department might be a more annoying client than an external client: You might get to charge the home team less, or have to do more boring unprofitable facilitation trades, because you all work for the same company. Or they might just be annoying personally, and you spend more time with them than you do with other annoying clients, and sometimes see them in the cafeteria or whatever. So you might mess up their trades out of resentment.

And since traders are humans, or at least mammals, this happens sometimes. The UBS Libor settlement documents feature one UBS trader manipulating Euroyen Tibor in a way that hurt UBS, because the trader whose position was hurt and the trader doing the manipulating "were rivals at UBS and had feelings of animosity towards one another." Markets: Not always rational.

2 Like if Chantree, at home during his suspension, is profitably day-trading FX based on tips from his still-employed buddies, then you'd have to say that Spicer is still correct. If not ... ?

To contact the writer of this article: Matt Levine at mlevine51@bloomberg.net.

To contact the editor responsible for this article: Tobin Harshaw at tharshaw@bloomberg.net.