Following an investigative piece in the Wall Street Journal by Mike Bird, ASI Executive Director Sam Bowman wrote to the Cabinet Office asking for a review of current procedures around pre-releases of ONS data to government ministers and civil servants.
Dear Mr Gummer,
I am writing to you in your capacity as Minister for the Cabinet Office and Paymaster General. I wish to draw your attention to new evidence published in today’s Wall Street Journal showing dramatic moves in financial markets in the 24 hours prior to the publication of surprisingly strong or weak market-sensitive CPI and other market data. (https://www.wsj.com/articles/market-drift-suggests-some-investors-may-be-trading-on-u-k-economic-data-ahead-of-release-1489397403)
These moves, revealed by the WSJ’s Mike Bird and Prof Alexander Kurov, a professor of finance at West Virginia University, are large and suggest that official, confidential data may be being leaked to certain traders for the purpose of insider trading, before the data is made public. These leaks may be coming from inside the government and as such it is vital that they are investigated.
The UK is unusual in giving advance notice of this sort of sensitive data to a large number of people before it is revealed to the public – 118 different politicians and civil servants are given advance access to labour market data, for example, including ministers and their special advisors at the Treasury, Department for Work and Pensions, and other ministerial departments, and at least fourteen different government press officers. This creates a much greater risk of data leakages, enabling traders to profit illegally from secret information undermining trust in the markets.
In light of this new evidence, will the Cabinet Office review the people it gives pre-release access to, to determine whether these leaks are taking place and, if so, who is responsible? Furthermore, will the Cabinet Office re-investigate the rejected proposal from the UK Statistical Authority (made in 2010) that the maximum pre-release period ought to be cut to a maximum of three hours?
In the US official data is typically released less than 24 hours after it’s completed and physically delivered by courier to the White House, where only the President and the chairman of his Council of Economic Advisors have access to it. Prof Kurov found that evidence of pre-release drift was much less likely in the US. We ask that the Cabinet Office consider US-style security measures to prevent future leaks and ensure that leakers cannot enrich themselves or others through insider trading.
In the long run, the solution may be to move towards greater openness and transparency at the ONS so that leaks of this kind are not possible. There is little reason for the ONS not to make its data collection and generation open to the public throughout the process, so that we can see in real time what the statistics suggest about the health of our economy and ensure that no individuals can profit from corrupt leakages of government data.
Adam Smith Institute