Basically every other photo for foreign exchange is just screens of numbers. So, this.

Photographer: Philippe Lopez/AFP/Getty Images

Banks Will Charge Extra for Not Manipulating FX

Matt Levine is a Bloomberg View columnist. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz and a clerk for the U.S. Court of Appeals for the Third Circuit.
Read More.
( Updated
)
a | A

Finance is about managing and allocating risk, and one of the most important risks in the corporate world is the risk of looking slightly stupid. Not because it's the worst thing that can happen to you, but just because modern life presents so many opportunities to look slightly stupid. If you can cut back on those opportunities, life is so much more pleasant.

For instance, if you're a corporate treasurer or mutual fund manager, you might need to convert 10 million euros into dollars today. But what time today? Ugh, what an awful question. If you bought dollars when you got to work today at 8 a.m. in New York, when the euro would buy around $1.128, then by 3 p.m., with the euro around $1.1325, you'd look, you know, slightly stupid. Not catastrophically stupid. Not like the guy who decided to stop hedging Swiss francs the week before the Swiss franc exploded. But you bought $11.28 million for your euros, and you could have bought $11.325 million if you'd waited a few hours. You lost $45,000. You look a little dumb.

But what could you have done? Predicting the intraday movement of currency markets is hard, and there's no reason to think that you, in your corporate treasury department, would be great at it. You're probably not even really trying: You're just buying dollars between doing other things, without giving any thought to picking the perfect minute at which to do it. Odds are that you'll be below-average, and so you'll do worse than the average price during the day. You'll go to your boss and tell her that you got done at $1.128, and she'll see that the price is now $1.1325, and she'll quietly judge you for not getting as many dollars as you could have. And you and your fellow corporate treasurers and mutual fund managers and so forth will come away from the foreign exchange markets with a feeling of sadness. 

That is not good advertising for the foreign exchange market! That market -- I mean, the dealers who make money in that market -- want you to be happy, so you come back and trade more foreign exchange. So they came up with a way to do that. It goes like this:

  1. Establish some price -- say, the price at 4 p.m. in London -- as the day's benchmark price.
  2. Guarantee you the benchmark price.
  3. With no fees.

All of those parts matter: Guaranteeing you the 4 p.m. price does you no economic good if the euro moves up after 4 p.m. You still, abstractly, look a little dumb for missing out on an opportunity. But if you can convince your boss that the 4 p.m. price is the only one she has to look at, and you deliver her that price, then you have entirely de-risked the process. (For you.) You have eliminated an opportunity to look slightly stupid. And you've done it without paying a fee! The bank that has given you this deal is your hero.

Of course this is no skin off the bank's nose. If the bank knows that it has to sell you 10 million euros worth of dollars at 4 p.m., then it can go figure out how to make money with that information. Simplistically, if the bank knows that the euro will be going up between 3:59 and 4:01 p.m., then it can sell you dollars at 4:00 and buy them back cheaper at 4:01. If it knows that the euro will be going down over that period, then it can buy dollars at 3:59 and sell them to you dearer at 4:00. 

Why would the bank's traders know what will happen? Well, first of all, it is their job. Corporate treasurers don't spend all day monitoring the fluctuation of foreign exchange markets, but bank FX traders do, so they really ought to be better than average at predicting it. Second, they have order-flow information: If all their clients want to buy dollars at 4 p.m., then that's good reason to think that the dollar will go up, and they can position themselves accordingly.

Third, they can illegally manipulate the fixing price, and they seem to have done quite a ton of that

The financial industry is full of similar trades, aimed in large part at making agents look good in front of their principals -- corporate treasurers in front of chief financial officers, mutual fund traders in front of portfolio managers, public company executives in front of shareholders.  If you can establish a benchmark for the agent to be measured against, and then guarantee the agent execution at the benchmark (or better), then you've got yourself a business. And then customers ask for execution at the benchmark, which creates optionality and information for the bank, which the bank turns into money. Banks are good at turning optionality and information into money; corporate treasurers usually aren't. And the banks use all that money to pay themselves bonuses, yes, but also to subsidize the customers' trading costs. So they can guarantee the benchmark with no fee. 

Of course the way that the banks turn optionality and information into money is -- I mean, look, money is money. The bank is trading against the customer. That's not a statement of values, just of fact: If you're buying dollars from a bank, that bank is selling you dollars. It's on the other side of the trade. It wants to get as many euros as possible for its dollars; you want the reverse. It uses the information and optionality that you give it against you. If it's thrilled to sell dollars to you at the benchmark price, with no fees, that can only be because selling to you at the benchmark price is better, in expectation, for the bank than selling to you at some other price would be. Which means that the benchmark price is worse, for you, in expectation, than the other price would be.

But that is incredibly abstract. Yes, "you," the corporate entity, would in expectation be better off trading intelligently on your own, instead of accepting the bank's guarantee of the benchmark price. But then "you," the corporate treasurer, would have to actually go and trade intelligently, developing an expertise at timing the intraday currency markets. That's hard, and exposes you to the risk of looking dumb. You're willing to pay the bank in information and optionality to avoid that risk. Especially since what you pay the bank is invisible to your boss: She sees only no-fee execution at the benchmark price, not the alternative reality in which you could have done better than the benchmark.

Anyway:

Barclays Plc, Deutsche Bank AG, and JPMorgan Chase & Co. have told customers in recent weeks they may start charging fees for trades executed at the WM/Reuters rates, including the 4 p.m. London fix, said the people who asked not to be identified because the discussions are private. Citigroup Inc. is weighing similar plans, one of the people said.

Credit Suisse is also apparently planning to charge for benchmark execution. The impetus for this, of course, is that last year Citibank, HSBC, JPMorgan, Royal Bank of Scotland and UBS AG got caught manipulating currency fixings, and fined a lot of money. After that, regulators recommended some changes to the currency markets, including that banks should start charging these fees.

The fees are related to two other changes to the market. First, the 4 p.m. fix is going to change from being an average of prices over one minute to five minutes -- a change that will reduce the banks' optionality from the fix.  And, second, after all those fines, banks probably can't go around sharing client information and manipulating the fix and doing all the other bad things that they were previously doing to make money. It's probably even harder to do the legitimate things that banks were doing to make money: Regulators raised their eyebrows at the "potential conflict of interest" even when a bank is "legitimately managing the risk arising from its net client orders at the fix rate." Because that legitimate risk management is pretty hard to distinguish from front-running.

So the banks can't make as much money as they used to from information and optionality. And the banks are not charities. So they'll have to make money the boring way, from fees: Instead of paying the benchmark price, plus optionality and information, now all the corporate treasurers will pay the benchmark price plus a fee. "This will likely diminish the attractiveness of this fix as a standardized FX rate for commercial transactions and reference," says a guy. The perfect mechanism -- set a benchmark, guarantee the benchmark and make it free -- is broken, and the whole thing falls apart.

Which is a little sad! The old system basically made everyone happy. Customers paid no fees, avoided the risk of looking stupid and got guaranteed execution at the benchmark price. And they had no reason to look past that fact: It was of little interest to the clients, or at least the treasurers, if the benchmark was "good" or "bad." Meanwhile banks made money, had happy customers and had frequent opportunities to look clever, the pleasing flip side of taking on their customers' risk of looking dumb. The problem is that the banks couldn't resist the temptation to get too clever, and so ruined the whole system for everyone.

(Updates paragraphs 13 and 14 of article published Feb. 9 to name fined banks and reference regulatory recommendations.)

  1. Especially after fees or bid-ask spreads.

  2. I mean, look, I know the foreign exchange market doesn't really need to advertise itself. But each dealer needs to get clients, and providing a happy experience helps.

  3. A classic benchmark is volume-weighted average price in equity markets. There is a whole industry of ways for banks to beat the VWAP, and to pass on some of the savings to you. 

  4. Intuitively, a 5-minute average is more likely to reflect the "actual" average price around that time than a 1-minute average. It's easier to match the longer average, but harder to beat it.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

To contact the author on this story:
Matt Levine at mlevine51@bloomberg.net

To contact the editor on this story:
Zara Kessler at zkessler@bloomberg.net