Glen Newey · Libor
In his Treasury Select Committee evidence yesterday, Paul Tucker, deputy governor of the Bank of England, denied nudging or winking to Barclays over the Libor rate. Whatever the semiotics, warning signs over Libor were already plain long before Tucker’s supposed nictation in October 2008. Minutes from November 2007 of the Bank’s money markets liaison committee meeting show its members were already exercised, eleven months before the exchanges between Tucker and Bob Diamond at Barclays, over lowballing Libor prices: several expressed concern that ‘Libor fixings had been lower than actual traded interbank rates’. The oversight regime resembles a dead chameleon, stuck on orange, while the system goes critical.
Libor affects the real economy via the cost of commercial loans, and indirectly through interest rate swaps, floating rate notes, derivatives and so on. Libor rates are a truncated mean, averaged from the eight median prices submitted. If a bank is keen to nudge the rate up or down, it should pitch as close as possible to one or other extreme in that median range. That can be used to help its own trading book: between 2005 and 2009 Barclays derivatives traders made no fewer than 257 rate-rigging requests. Libor submitters are meant to submit the rate at which they ‘could borrow funds’, on the basis of inter-bank offers ‘in a reasonable market size’ at a given time. In acting as their own Standard and Poor’s, banks go not on real transactions (which may be well-nigh non-existent for, say, the 11-month dollar rate) but their impressions. That leaves ample room for submitters’ ‘judgment’ rather than just reporting facts.
Barclays has been berated for lowballing estimates to Libor, the implication being that the bank wanted to underplay its liquidity problems following Lehman Brothers' sub-prime crash earlier in 2008. Diamond knew the government was threatening to recapitalise – i.e. part-nationalise – Barclays, which had been weakened by exposure to bad loans. Certainly, graphs show the bank’s Libor submissions taking an abrupt plunge in late October 2008, from around 4 per cent to around 1.5 per cent by mid-December. But other submitters like J.P. Morgan, and indeed Libor itself, moved in the same direction over this period. Though the submissions from Barclays remained high compared to the others, the gap between the overall Libor rates in October and December was much bigger than that between Barclays and the other submitters ever was.
There’s no particular reason to believe that Barclay’s earlier highballs reflected honest reporting of market rates: if players know that the system’s bogus, Libor bids no longer reflect institutions’ credit rating. It’s perfectly possible that Barclays, not worrying before Lehman collapsed that highballs might signal a lack of creditworthiness, simply hoped to suck up liquidity from the money markets when its competitor institutions were clustering round the lower Bank-approved rate. For its part, the Bank wanted to keep the taps open, and worried, as Tucker said yesterday, that that would be scuppered by a rogue institution bidding high.
So during that quarter, lowballs became the order of the day among Libor panellists. Barclays insinuated before Bob Diamond’s appearance last week at the TSC that Tucker had given Barclays the nod to lowball its Libor submissions, a claim Tucker denied yesterday. It’s odd that Diamond knew nothing of the rate manipulation until last month, despite what on his own testimony was a strong nudge that way by the Bank and the fact that Jerry del Missier, co-head of Barclays investment bank and a close colleague of Diamond, had told Barclays’ Libor submitter to do just that after talking to Diamond. At any rate, all converged in lowering their estimates of their lending costs after the over-leveraged Lehmans went belly-up. The banks duly swallowed the extra liquidity later released by the Brown government to beef up their asset base, which helps explain why quantitative easing failed to arrest the downturn.
The system, or in Tucker’s word, ‘cesspit’, worked admirably in easing convergence, as even the gnome-friendly Economistaccepts. You don’t need to be a game theorist to see that when key actors’ interests favour convergence, their behaviour will converge. Libor panellists’ prices are published within an hour of submission. So if Barclays had been pitching higher than most of the other submitters before Tucker contacted Diamond, its submitted rate would have been in the top range excluded from the Libor calculation (and hence above the final average). All had an interest in misreporting their creditworthiness, as the Bank did in conniving at it. The chameleon always was dead.
Now we know that Tucker didn’t nudge Diamond; Diamond didn’t wink to del Missier; but del Missier got his Libor man to do the necessary. Meanwhile, as Tucker stressed, the Brown government wasn’t nodding at anyone. The great thing about tipping the wink is its deniability. But then, when interests converge, co-ordination is often unnecessary. As Adam Smith wrote, ‘we rarely hear’ about ‘the combinations of masters’ but ‘whoever imagines... that masters rarely combine, is as ignorant of the world as of the subject. Masters are always and everywhere in a sort of tacit, but constant and uniform combination.’