What are we to learn from the charges that came out against interdealer broker ICAP and its staff over the manipulation of Libor rates? In three separate regulatory complaints, from the UK Financial Conduct Authority, US Commodity Futures Trading Commission and Department of Justice, once again we read emails and chat logs outlining intent to collude or deceive in setting Libor. We learned of a new nickname, ‘Lord Libor’, the nom-de-guerre of ICAP cash broker Clive Goodman, named along with two colleagues in the DOJ’s criminal complaint.

The Libor investigations began in 2012 and have already led to hefty settlements at Barclays, RBS and UBS, while Deutsche Bank, Rabobank and Citigroup are yet to come. I would characterise ICAP as a half-way point.

Looking backwards to what we already knew, the complaints further clarify the central role of former UBS and Citigroup trader Tom Hayes, known as ‘Rain Man’. Hayes was named in a DOJ complaint last December and was charged by a London court in June.

The complaints also emphasise the importance of interdealer brokers in facilitating Libor manipulation. Because of the way Libor is set by averaging contributions from panel banks and excluding outliers, a single bank can only have a limited impact on the daily fixing.

In its 2012 complaint against Hayes, the DOJ identified examples where UBS singlehandedly moved Japanese Yen Libor by one-eighth of a basis point, generating a few hundred thousand dollars profit for Hayes on a trade. However, to earn the $260 million in profit he made for UBS between 2006 and 2009, Hayes needed other banks to push Libor in the direction he wanted.

Traders at some banks, such as RBS, helped Hayes do this at his request, but brokers were more useful because they could influence other banks without them twigging what was happening. The way it worked was that brokers such as Goodman provided so-called ‘run-throughs’, or daily emailed assessments of Libor levels.

These run-throughs were supposed to be an independent view of the money market, but in reality Goodman was controlled by Hayes via the ICAP yen derivatives desk whom Hayes rewarded with ‘bro’ or brokerage fees. The quasi-comic aspect of the ICAP complaints is Goodman’s increasing impatience over time, as he grew tired of free curry and champagne and asked for cash payments instead.

Goodman’s activity highlights once again the tail-wagging-the-dog role of over-the-counter derivatives compared to cash markets. Libor was originally developed as a benchmark for London’s offshore dollar deposit market. By 2007, unsecured interbank lending dried up in the financial crisis, while derivatives referencing the benchmark continued to grow.

As the UBS and RBS complaint documents already show, the big banks effectively created figures like Hayes by giving their derivatives traders control of Libor submissions. ICAP did no more than mirror what some of its largest clients were doing for profit.

Given that there are an average of 2,000 new interest rate derivative trades per day according to data from the Depository Trust & Clearing Corporation, compared with a handful of interbank Libor loans according to BNP Paribas ( See Note 1), the incentives to do this are not going away.

The charges are likely to raise questions about the role of brokers such as ICAP as independent price reference points for the over-the-counter derivatives market. Bank risk departments often use third-party broker quotes to check the accuracy of their traders’ marks, but JPMorgan’s failure to properly do this was highlighted in the recent London Whale scandal. If these third-party quotes are tainted by the dependence of brokers on dealer fees, then the functioning of OTC derivatives is called into question.

This is particularly relevant to ISDAfix, a benchmark controlled by ICAP and used for pricing numerous interest rate derivatives such as constant-maturity products. The FCA and CFTC are currently trawling through millions of emails and messages relating to ISDAFix. If evidence of wrongdoing at ICAP is found here (and none has so far) then regulators will be under pressure to curtail the market even more aggressively.

Indeed, we may never know how much manipulation really takes place under the current system for trading OTC derivatives. Comically, Hayes and his co-conspirators use instant messages to warn each other not to be too obvious in documenting requests for manipulation. What about traders who were more circumspect?

In the absence of self-incriminating messages, regulators are going to have to depend on statistical analysis to prove that market manipulation or collusion took place. Dubbed ‘forensic economics’ by Justin Wolfers, this is a burgeoning and fashionable area of study, and has only just started to be applied to Libor.

In a 2012 working paper, Connan Snider and Thomas Youle deploy an economic model of Libor rigging which they test against historical data. They find evidence of manipulation in the form of bunching together of Yen Libor submissions, but their research doesn’t single out UBS and doesn’t take brokers into account

Contradicting what ICAP and the banks have said so far, Snider and Youle also say that their research implies that manipulation is systematic and not a question of a few ‘bad apples’ at particular institutions. This is relevant to Hayes, who in his forthcoming trial is likely to point to senior bankers who condoned his activities.

The dog that hasn’t barked so far is the question of who was a victim of market rigging. After all, if Hayes made $260 million for UBS partly by manipulating Yen Libor, does that mean UBS counterparties who traded with him are victims? Without full disclosure of Hayes’ trading records, it’s hard to say.

A suit brought by the City of Baltimore against Libor-setting banks was largely dismissed by a US court in March. In the UK, Barclays faces an important civil test case from Guardian Care Homes, which has succeeded in including Libor manipulation as part of an interest-rate swap mis-selling complaint. But how much of Guardian’s losses were the result of Libor rigging, as opposed to moves in sterling interest rates?

These are the questions we have to look forward to as the Libor scandal passes its half-way point.

(1)  “We always kept our fixed income trading business separate from the treasury, so the treasury which was quoting Libor had no interest in the bonus pool of the traders. Today, we have a problem because there are far fewer transactions which makes it difficult to quote. Sometimes, for USD 3m Libor we only have four transactions per week, something like that. It’s difficult to give a quote every day”. – Philip Bordenave, chief operating officer of BNP Paribas, Bloomberg Risk interview, May 2013