The government has come under renewed pressure to review its policy of granting ministers and civil servants early access to national statistics, as new research suggests the information is being leaked to traders ahead of official publication.
Early access to statistics across Whitehall has come under fire – Photo credit: Fotolia
The Office for National Statistics grants pre-release access to statistical datasets to around 100 people, including the prime minister, ministers, and key civil servants, such as policy and press officers, 24 hours before they are made available to the wider public.
The idea is to allow teams time to prepare statements, briefings and reports based on the data, but many commentators have warned that it increases the risk of data being leaked to others outside the group, especially when it is market-sensitive data.
More weight has now been added to these concerns following a new analysis of trading data that has been revealed this week by the Wall Street Journal.
Related content
New ‘data campus’ planned by UK Statistics Authority
Office for National Statistics told to pick an Innovation Champion
Bringing government data to life
The research, which was carried out by Alexander Kurov, an associate professor of finance at West Virginia University, shows that the markets “often move sharply in the 24 hours” before reports are released.
This indicates that some investors may be trading with knowledge of the statistics before they are made public, the WSJ said.
The research looked at 207 releases of economic data – including industrial production and labour market statistics – between April 2011 and December 2016.
Of these, 172 came in above or below the consensus estimate of analysts, meaning they were generally considered by the market to be surprises. However, in 59.7% of these cases there were movements in the markets that anticipated the outcomes before the official publication of the datasets.
‘Clearly open to abuse’
Responding to the research, Hetan Shah, the executive director of the Royal Statistical Society, said: “By definition the more pre-release access to statistics, the more you increase the chances of leakage. In the case of market-sensitive economic data, this could lead to insider trading which is both illegal and risks affecting confidence in the City.”
He told PublicTechnology that the new research “should make the government realise that it is right both in principle but also in practical terms to end the practice of pre-release access, which is clearly open to abuse”.
The society has been calling on the government to put a stop to the pre-release access to statistics for some time, writing in its 2015 Data Manifesto that ending the practice would help build public trust.
“In order to improve trust, the public needs to be reassured that statistics are not being misused to suit political or media agendas,” the RSS said at the time. “Decisive action against pre-release access is one of the actions needed to respect the impartial nature of official statistics, and that all should have equal access to them.”
Meanwhile, Sam Bowman, executive director of the economics think tank the Adam Smith Institute, said that the report should be “taken extremely seriously” by the government and the ONS.
“If flaws in ONS procedures are enabling market-moving leaks to take place, allowing some individuals to profit off secret information, then something has gone badly wrong. The Cabinet Office should review ONS procedures and consider substantial reforms in light of this new evidence,” Bowman said.
The Wall Street Journal acknowledged that there were other possible reasons for the market shifts, such as traders using privately collected economic data to estimate consumer-price inflation themselves. But it noted that research showed other countries do not anticipate economic data from government departments “with the same precision as it happens in the UK”.
