Inconvenient truths about Libor
We may well see more heads roll at major banks before the scandal surrounding Libor is over. But some senior members of the international financial community are increasingly wondering about the future of Libor itself.
As one senior international regulator put it to me: "The benchmark is broken. It needs to be fixed. Or perhaps it will just go the way of the dodo. The world has changed."
That's because the scandal has shed light on an inconvenient truth about these interbank rates which are used to determine the price of so many hundreds of trillions of dollars worth of global financial contracts.
That inconvenient truth is that even when London Interbank Offered Rates are not "fixed", they may still not bear very much relation to reality - because banks are not actually offering much unsecured money to each other at all.
This has been an open secret among bankers and regulators since the start of the credit crunch in 2007. What any outsider would find surprising is that until now, neither the Financial Services Authority (FSA) nor the Bank of England have really made it their business either to replace Libor - or make it more accurately reflect reality.
Interestingly, this is what the US regulator, the Commodities Futures Trading Commission (CFTC), has tried to do in its settlement with Barclays. If and when it ends its investigations of other banks, it will presumably try to do the same with them.
We know that the British bank has been fined £290m ($450m) by the Financial Services Authority and the CFTC for making false claims about the rate at which it was borrowing from other banks - either to make profit or to protect Barclays' reputation at a time when it was seen to be paying more than others to borrow.
There has been less focus on the "undertakings" which Barclays was also forced to make to the CFTC, about how it would calculate and report its interbank borrowing rates in future.
These come into force 14 days from the signing of its agreement with the US regulator and will require the people reporting the bank's cost of borrowing for the BBA to base that report on actual transactions that Barclays has been involved in - in other words, on actual loans it has obtained or made on the wholesale market.
If Barclays hasn't, in fact, obtained or offered such loans, its new rules say it must calculate the rate on the basis of other secured loans that the bank has been involved with (i.e. loans backed by collateral). If there haven't been any of those either, they have to refer to similar transactions that other banks have been involved in which Barclays folk have themselves observed in the market - and so on and so on.
The point is that somewhere along the line, there has to be an actual transaction involved - and the system for deriving the Barclays interbank borrowing rate from those transactions has to be entirely transparent.
You might wonder why that is worth an official "undertaking" in a legal settlement with the US regulator. After all, isn't that what the Libor rate is supposed to reflect - the rate at which banks are able to borrow and lend from each other?
The answer to that question is yes, that is what it is supposed to be. The trouble is that it does not - cannot - mean that in an environment in which banks are finding it very difficult to get unsecured loans from anyone for any length of time.
As we keep hearing, that was the situation for large parts of 2007, 2008 and 2009, when some of the "fixing" and attempted manipulation of Libor at Barclays was taking place. But, as Robert Peston often reminds us, it is also true of many banks right now - especially across the Channel.
Many banks on the continent - in so called "core" economies like France as well as Spain, Portugal etc - cannot really borrow on the interbank market at all at the moment.
In the City, the borrowing by banks for more than one week is almost entirely happening on a secured basis. Collateral is king.
Many banks in the eurozone are almost entirely dependent on the European Central Bank (ECB) for day-to-day liquidity. And since the ECB started its long-term refinancing operation (LTRO) programme last November, they have been getting quite a lot of longer term funding from that source as well.
And yet, throughout this period, the Euribor - the Euro Interbank Offered Rate, also caught up in the Barclays manipulation scandal - has been faithfully reported, day in day out, by more than 50 European banks. You have to wonder what those rates actually mean.
True, included in that list of 50 are some major global banks who are unlikely to have much trouble borrowing on the wholesale market: Deutsche Bank, for example, and JP Morgan. But you also have banks like the National Bank of Greece, Allied Irish Bank, La Caixa Barcelona and Societe Generale.
We're not just talking overnight borrowing here. The Euribor rates that are reported include the cost of borrowing for anything from one week, to an entire year. If you ask banking industry insiders how many of those 50 banks are actually getting unsecured loans for a year - or even one month - on the wholesale market they are quite likely to laugh in your face.
So, maybe the Euribor and Libor rates correspond to something, but we know for a fact that they do not always, or even usually, correspond to an actual transaction. Nor, if you look at the guidelines for setting Libor on the British Bankers' Association (BBA) website, are they required to. Banks merely have to report the rate at which they "could" borrow funds before 11am, were they to decide to do so. How, exactly, each bank determines the rate is largely up to them.
The CFTC settlement says Barclays had "no internal procedures and controls" determining how the Libor rates were calculated. The BBA does not seem to have had any problem with that.
As it happens, the strangeness of this situation was captured very well by Sir Mervyn King in testimony to the Treasury Select Committee in late November 2008, when he had this to say about Libor.
"It is in many ways the rate at which banks do not lend to each other, and it is not clear that it either should or does have significant operational content. I think it is convenient, very often, for people to justify what they do for other reasons, in terms of Libor, but it is not a rate at which anyone is actually borrowing. It is hard to see how it can actually have much of an impact."
And yet, despite their inherent fuzziness and lack of "significant operational content", despite the lack of formal checks on banks' internal procedures for coming up with these rates, Euribor and Libor are the benchmark for pricing transactions worth trillions of dollars. US dollar Libor, for example, is the basis for the settlement of the three-month Eurodollar futures contract, which had a traded volume in 2011 with a notional value of $564 trillion, according to the CFTC.
Many of you will find all of that pretty odd - and pretty shocking. I know most economists would.
We are all understandably interested in what Bank of England deputy governor Paul Tucker and other senior officials may or may not have said about the Libor to Bob Diamond or other bankers, at the height of the credit crunch in 2008.
But, perhaps we should also be asking why those senior officials and regulators continued to allow Libor to play such an iconic role in global financial contracts - when even the governor of the Bank of England knew quite well that they did not paint a remotely accurate picture of reality. Even when everyone was playing by the rules.