At its heart, Libor is supposed to be a measure of the borrowing costs of banks.
Libor is actually not just one interest rate. It is the name for rates calculated in 15 currencies for loans of 10 different maturities, ranging from overnight to 12-months. (See the latest Libor and Treasury/Euro-Dollar credit spreads.)
The setting of Libor begins each morning between 11:00am and 11:10am (London time) when someone in one of the designated Libor panel banks enters a number into a piece of Thomson Reuters software that asks the question: “At what rate could you borrow funds, were you to do so by asking for and then accepting inter-bank offers in a reasonable market size just prior to 11am?”
Over at a Thomson Reuters office somewhere in London—the exact location is a closely guarded secret—clerks look over the submissions for possible error. The clerks will call a bank to confirm a number that seems off for some reason. A number that greatly varies from the submission of the prior day might trigger a back-check. A number highly divergent from the submissions of other banks can also trigger a double check. Obvious typos are surprisingly frequent. The clerks are careful, however, not to reveal any other banks' submissions prior to the official publication.
What the clerks do not do is challenge a number confirmed by the bank.
“Our goal in the checking is to make sure we use the number the bank intended to submit,” a person familiar with the matter said.
The numbers are then entered into a “calculation engine” that ranks them in descending order and trims off the highest and lowest 25 percent of submissions. The middle 50 percent are then averaged, producing a figure that is published as that day’s Libor at 11:30am. In addition, Thomson Reuters publishes the separate submission of each panel bank.
This methodology is set by Foreign Exchange and Money Markets Committee of the British Bankers’ Association, which also determines make-up of the Libor panel—the group of banks whose rates are used in the calculation—for each of the 10 currencies. The banks are supposed to be the biggest, best and most reliable in the world, selected according to their credit rating, their scale and their expertise in the relevant.
The BBA is mainly a trade-group that lobbies on behalf of the financial sector. Until this week, its chairman was Marcus Agius, who resigned from his BBA chairmanship on the same day he resigned his chairmanship of Barclays.
The Libor committee is formally independent, governed by its own board and its own chairman. But, as The New York Times has detailed, that board has four senior BBA officials, including its former chief executive. What’s more, the secretary of the committee, John Ewan, is said to do much of the Libor oversight work—and he is also a senior BBA official.
One of the things the committee does is receive a report each week from Thomson Reuters about the details of every inquiry and back-check it made with the banks. It can then tweak the process Thomson Reuters uses to collect and check the submissions.
The biggest change to this process came in 1998. Prior to then, the banks were asked a slightly different question: “At what rate do you think interbank term deposits will be offered by one prime bank to another prime bank for a reasonable market size today at 11am?”
This was thought to offer too much wiggle room for banks to potentially fudge rates. The current formulation of the question requires banks to estimate their own borrowing costs rather than that of a hypothetical bank. Thomson Reuters says this provides “accountability” to the process.
The question is still a bit vague. Each bank can decide for itself what a “reasonable market size” borrowing might be. What’s more, the question is not about actual transactions. Because not every panel bank will have to borrow in the relevant currency on a given day, they are asked about a hypothetical borrow. It’s not “what rate are you paying to borrow?” It’s at what rate you think you “could” borrow “if you were to do so.”
The answers are not supposed to be wild guesses—much less the self-serving rigged numbers Barclays has admitted to submitting. The bank should know its own credit and liquidity risk and understand prevailing market conditions well enough to produce a somewhat accurate model.
Whether this works in practice, however, is a subject of much debate. Even prior to the current Barclays scandal, numerous journalistic investigations and academic studies have questioned whether the Libor submissions are reliable.
Now we know that at least one bank—Barclays—provided unreliable data to the Libor system. The question that remains open is what other banks were also taking the low road with Libor.
Follow John on Twitter. (Market and financial news, adventures in New York City, plus whatever is on his mind.) You can email him at email@example.com.
We also have two NetNet Twitter feeds. Follow CNBCnetnet for the best of the days posts, including breaking news. Follow NetNetDigest for a feed of every single post each day.
You can also be our friend on Facebook. Or subscribe to John's Facebook page.
We're on Google Plus too! Click here for John's Google+ page.
Questions? Comments? Tips? Email us atNetNet@cnbc.comor send a text message to: 917-740-8477.
Call us at 201-735-4638.