Advertisement

Jim Armitage: We must act to stop our official statistics being leaked

Jim Armitage: Andrew Tyrie’s decision to quit as an MP leaves a huge hole in Parliament’s banking expertise
Jim Armitage: Andrew Tyrie’s decision to quit as an MP leaves a huge hole in Parliament’s banking expertise

Andrew Tyrie’s decision to quit as an MP leaves a huge hole in Parliament’s banking expertise. Though the former Treasury Select Committee chairman is best-known for grilling Barclays’ Bob Diamond and Bank of England officials over Libor, he was also an informed and feared voice who improved much of the City after the financial crisis.

Finding a replacement as capable of spotting the sharks from the red herrings will be tough.

One Great White he recently started chasing offers a fine example: the suspicious regularity with which traders correctly predict surprise UK official economic statistics.

Unlike most developed countries, dozens of British politicians and mandarins are allowed sneak previews of official data. They say it helps formulate government responses. Maybe, but the added risk of leaks is obvious.

Today, academics at West Virginia University released the most thorough analysis yet into suspicious trading around Office for National Statistics releases.

From January 2011 to March 2017, they show that, in the hour before surprise data, traders in the pound unerringly guessed the reaction against the dollar in advance.

From the hour before to the hour after 9.30am data releases, sterling moved up or down an average of 0.127%. And of that, 0.065% — half the overall move — happened in the hour before.

Small movements, perhaps, but when multiplied by millions of pounds wagered on every basis point, day in, day out, the accurate “guesswork” is making someone big money.

Now compare the UK with Sweden, where nobody outside the stats office gets a glimpse in advance (as the Wall Street Journal puts it: “not even the Prime Minister”). Overall, the krona reacts to surprise data more than the pound, moving 0.163% by an hour after the release. But in the hour before, it barely ever moves in the right direction: a statistically insignificant 0.003%.

It could be that UK stats are just easier to predict. But given the British economy’s bigger size and complexity, I’m sceptical.

Far more likely is that Britain’s treatment of official statistics is lax and being abused. Just like our setting of Libor before it was humiliatingly stripped from us to be run properly by the Americans.

The statistics regulator has long pushed for an end to “pre-release access” but has consistently been knocked back by Westminster.

Tyrie appreciated what a vast issue this is and last month called for a full investigation. Whoever takes his post must keep up the pressure. After all why, in this financial capital of the world, is convenience for Downing Street spin doctors deemed more important than keeping cheats out of the markets?