Annals of market manipulation, FX edition

By Felix Salmon
June 12, 2013
Bloomberg has found anecdotal evidence of market manipulation in the FX world as well.


" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

First it was rates, now it’s FX: in the wake of the Libor scandal, Bloomberg has found anecdotal evidence of market manipulation in the FX world as well.

The problem here comes as little surprise. There are huge investors out there — largely ETFs and index funds — which invest internationally and which need to engage in a lot of large FX trades, especially at the end of the month. Their substantial FX trading notwithstanding, those funds don’t tend to employ dedicated FX traders. Instead, they do a deal with a big wirehouse — probably Deutsche, Citi, UBS, or Barclays — which promises them execution at whatever the spot FX rate is at the end of the day.

So at 3:30pm or so at the end of the month, the index fund tells the bank that it will need to (say) exchange 1 billion dollars into Swiss francs; the bank promises to do so, at whatever the official market rate is at 4pm. The official market rate is determined by WM/Reuters, which looks at all the trades taking place between 3:59:30pm and 4:00:30pm, and then takes the median rate of all those trades.

The bank’s trader now has both upside and downside risk. As Matt Levine explains, if the bank can’t manage to buy those Swiss francs at the WM/Reuters level, then it will lose money; if it can get a better price than the WM/Reuters official rate, then it will make money. When a trader buys a large amount of FX in a short amount of time, that can work in his favor or against him — if such a thing is done tactically, it’s called a concentration move. Here’s Izabella Kaminska:

“Concentration” tactics are normal practice for the industry. It’s the equivalent of creating economies of scale and then choosing the moment to transact so that the depth of the market, and its likely impact on the price, is most beneficial to you. It’s called skillful execution.

If you’re an investor rather than a trader, you generally hate the concentration issue. A big trade will move the market in and of itself, and you’ll end up getting a worse price than a small investor would. On the other hand, if you’re a sell-side trader rather than an investor, and you’re gifted with half an hour and a billion dollars of firepower, then you’re likely to be able to make quite a lot of money by being tactical about exactly how and where you do the transaction.

Essentially, the trader’s job is to procure a billion dollars’ worth of Swiss francs as cheaply as possible. The amount that the client will pay for the Swiss francs is set, and the trader will pocket the difference.

But here’s where the market manipulation comes in: what happens if you “bang the close”, and drive up the price of Swiss francs in the crucial 60 seconds between 3:59:30pm and 4:00:30pm? Then you’re not just trading to get the best execution for your own end of the deal; you’re also trading to get the worst possible price for your client.

Does this happen? Undoubtedly, yes. But as Kaminska says, it probably happens less now than it used to in the past, thanks to what she calls “a tech-related platform revolution”. And remember that no one is putting a gun to the index funds’ heads and forcing them to negotiate deals based on the WM/Reuters market close. If they want to just go into the market and buy the FX themselves, they’re more than welcome to do so. While the funds are surely concerned about traders trying to manipulate closing prices in the FX market — just as traders also try to manipulate closing prices in the stock market on days when options expire — they will have made an internal determination that it’s cheaper to live with that, compared to the cost of hiring their own traders. Especially since implicit FX costs only turn up in infinitesimally smaller returns, while explicit payroll costs show up directly in all-important fund-management fees.

The bigger picture here is that trades move markets, that traders know that trades move markets, and that traders are always going to try to profit from their ability to move markets. As Kaminska says, that’s execution, more than it is manipulation. If traders are really manipulating the close in order to make easy money from index funds, that’s a bad thing: those index funds are meant to be their clients. But when you have an unregulated, Wild West market like FX — which, by its nature, is almost impossible for any national regulator to oversee — it’s simply naive to believe that such behavior is never going to happen. My advice to the index funds is that they renegotiate their contracts, and try to put in place a system which aligns incentives more. Because if the incentives are opposed, as they are here, then the client is never going to end up the winner in the deal.

Comments
2 comments so far

felix –
I think an important point is that “civilians” tend to get upset about both the “frontrunning” aspect, and the “banging the close” aspect… as a commenter on Levine’s post noted (I quoted this commenter in my own post), you can’t have it both ways…

thanks for the heads up on Izzy’s post. i had missed it.

Posted by KidDynamite | Report as abusive

It seems that quite a few very problem on to kelly felix, Attempting transform i would say some sort of football landscape designs and many types of, And i also stated the Philadelphia masses isn’t the most sufferer most typically associated with contingents. But when he considers heat comes to an end from now on, Choose he provides the actual their environment while you are their own better player sets. To make neo error in judgment, This item some kind of so if perhaps, Probably and never which, McCoy comes in contact with damage.

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/