EU Is Shocked That Banks Colluded on Libor

Once upon a time banks dreamed of cooperating to set an interbank lending rate. Then they did and it was great. Later, they cooperated a bit too much, and it was less great. This is their story.

A long time ago some big banks decided that it would be good to sell interest-rate derivatives. 1 To do that they needed an interest rate on which to sell derivatives. Various possibilities presented themselves -- Treasury rates or whatever -- but the interest rates that the banks themselves paid on short-term borrowing had an especially obvious appeal as an index. If you're a bank, that data is readily available to you, you don't have to worry about government-market idiosyncrasies, and it's easier to be hedged if your derivatives (and the floating-rate loans you write to clients) are indexed to your own borrowing costs.

But being like, "we'll exchange you a fixed rate of 7 percent for a floating rate of 3 percent over our cost of three-month borrowing" is kind of weird. For one thing, it makes the client nervous: What if the bank has an accident and its cost of borrowing goes way up? (What if the bank lies about its cost of borrowing?) For another thing, it makes interest-rate swaps less fungible and liquid: A swap of JPMorgan-plus-300-basis-points is not easily comparable to a swap of Citi-plus-325, so you can't really close out a position in one by selling the other.

So the banks got together and decided: Let's create a composite of our borrowing costs and all sell swaps against that composite. We're all talking to each other anyway as we go about borrowing from each other, so let's all just write down how much we're paying to borrow, send our costs to a trade association, take a trimmed average, call it the London interbank offered rate, and write all our swaps against Libor. We can even set up a different trade association to make sure that we all have the same documents for our swaps, so that all our swaps work the same and use the same rate that we all more or less agree on.

So they did that, and it was great. I mean, it was, for them. You can complain because Libor has fallen into some disrepute of late, and they can complain because "fallen into disrepute" really means "has racked up some enormous fines for Libor banks," but I don't want to hear it from any of you. U.S. banks -- U.S. banks alone -- made $2.8 billion just last quarter from trading interest rate derivatives. That decision to have a standardized thing that they all agreed on as the basis for those derivatives worked out just plain great for them.

It's a little awkward, I guess, in that all these banks that are supposed to be competing with each other are actually coming together to agree on things like contract terms and benchmark rates, but that is only worth worrying about in a very formal sense. The fact that the banks all get together to agree on this stuff makes the underlying derivative market more liquid and competitive and all that good stuff.

Later some people at those banks decided to lie about their borrowing costs in their Libor submissions, and to encourage their buddies at other banks to lie about their borrowing costs, in order to help their derivatives positions. As you may have heard. That was sort of unfortunate! But it was unfortunate for more or less the reasons that other "market manipulation" or "fraud" would be unfortunate: It made prices wrong, in a way that made money for the bad guys at the expense of innocent victims. That's bad.

But it's not really antitrust bad, is it? Nobody is competing to submit Libor quotes, so messing with someone else's Libor quote doesn't seem like interfering with competition. It's interfering with cooperation. If that cooperation is good, which it probably is, then interfering with it is bad, but not in an antitrust way.

Anyway, that seems to be the American conclusion, which is why the Justice Department, which has gone after Libor manipulators for manipulation and fraud, hasn't really pursued Libor antitrust stuff. It's also why a New York federal judge dismissed an antitrust suit over Libor in March. She said:

As plaintiffs rightly acknowledged at oral argument, the process of setting LIBOR was never intended to be competitive. Rather, it was a cooperative endeavor wherein otherwise-competing banks agreed to submit estimates of their borrowing costs to the BBA each day to facilitate the BBA's calculation of an interest rate index. Thus, even if we were to credit plaintiffs' allegations that defendants subverted this cooperative process by conspiring to submit artificial estimates instead of estimates made in good faith, it would not follow that plaintiffs have suffered antitrust injury. Plaintiffs' injury would have resulted from defendants' misrepresentation, not from harm to competition.

Europe disagrees! Today the European Commission fined eight banks €1.7 billion "for participating in cartels in the interest rate derivatives industry," which is a little silly. 2 Here:

Joaquín Almunia, Commission Vice-President in charge of competition policy, said: "What is shocking about the LIBOR and EURIBOR scandals is not only the manipulation of benchmarks, which is being tackled by financial regulators worldwide, but also the collusion between banks who are supposed to be competing with each other. Today's decision sends a clear message that the Commission is determined to fight and sanction these cartels in the financial sector. Healthy competition and transparency are crucial for financial markets to work properly, at the service of the real economy rather than the interests of a few."

That collusion is the only thing that Almunia has jurisdiction over, 3 so he has to go around acting shocked to find banks colluding when they "are supposed to be competing with each other," but of course they're not supposed to be competing with each other over Libor. The goal of Libor isn't to get banks to submit the lowest, or highest I guess, possible interest rate. The goal of Libor is to get all the banks to agree on a standardized generic interest rate that they can all use. Competition doesn't enter into it.

Anyway, here is a very boring document from the commission that lists all the fines but is light on the fun details that make a Libor settlement a Libor settlement. Instead, there is a statement that, "In setting the level of fines, the Commission took into account the banks' value of sales for the products concerned within the EEA, the very serious nature of the infringements, their geographic scope and respective durations," and tables like this: 4

Source: European Commission press release

Those "infringements" are not naughty e-mails; they are what seem to be sort of arbitrary periods in which each bank was participating in a yen interest-rate derivative "cartel," that is, sending naughty e-mails to other banks and interdealer brokers. Presumably in the intervening periods they were still sending naughty e-mails within their banks. Also presumably when they were participating in a "cartel," the banks weren't colluding on yen Libor every day during that 5 or 8 or 10-month period. 5 (If they were, that would be news, and you'd think the commission would mention it.) They were just sometimes (often?) calling in favors at other banks to manipulate the market.

Which, again: bad. But notice what the commission did not take into account in setting the fines: how much money each bank made by manipulating Libor, or how much damage it caused, or even apparently how culpable it was and how frequently it manipulated Libor during those periods. And, in particular, it couldn't take into account all the intramural, non-cartel manipulation in which banks just wrote down the wrong rate in their own Libor submissions. The main problem was banks writing down the wrong rate, but the commission has to focus only on banks writing down the wrong rate because another bank asked them to.

Deutsche Bank "said the fine 'reflects, in particular, the high market share held by Deutsche Bank in the markets investigated.' " That's a sensible way to punish a cartel -- the bigger you are in the cartel, the more you pay -- but a strange way to punish market manipulation. I suppose punishing the Libor manipulating banks is better than not punishing them, but this does feel like punishing them for the wrong thing.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
  1. This is fairy tale, not history, but it's true enough.

  2. Possibly sillier than the commission's immediately prior antitrust enforcement action, "Commission fines four North Sea shrimps traders €28 million for price fixing cartel," though I am in no position to judge, I suppose North Sea shrimps trading is no laughing matter.

  3. In fact this seems to be an antitrust case for regulatory-jurisdiction reasons, not substantive antitrust ones. Here is DealBook:

    Unlike its American and British counterparts, the European Union has limited enforcement powers over financial firms, which are primarily regulated in their home markets or where they conduct the bulk of their business.

    As a result, European Union antitrust authorities had to build a case based collusive conduct among a group of financial firms, rather than improper behavior by a single entity or group of traders at one bank.
  4. I guess we should talk about that "UBS received full immunity for revealing the existence of the cartels and thereby avoided a fine of around €2.5 billion for its participation in five of the seven infringements." You can speculate on the appropriate reward for UBS's compliance officers who reported this. Fifty percent of revenue for compensation is a rough rule of thumb for front-office jobs, but how do you count the revenue here? After all, if they had just stopped their traders from manipulating Libor the whole €2.5 billion thing wouldn't ever have come up.

  5. I guess you can match up the times to get who was colluding with whom in yen Libor. So UBS colluded with Deutsche for 10 months, RBS for 8 months and again for 5 months, JPMorgan for 1 month, and Citigroup for 1 month (and with RP Martin, an interdealer broker, for one of those two one-month collusions with JPM or Citi). RBS colluded with UBS twice and with Citi once. Deutsche with UBS for 10 months and with Citi for 2 months. The euro Libor thing seems to have been more of a mush.

To contact the author on this story:
Matthew S Levine at

Before it's here, it's on the Bloomberg Terminal.