Public mistrust of labour's statistics remains high

Seen as curmudgeonly bearers of bad tidings, statisticians are rarely top of anyone's Christmas card list. As the joke goes, a statistician loves to work with numbers but lacks the personality to be an accountant.

Yet those who can bring themselves to cast aside their trademark gloom can identify positive trends in 2006: the economy is growing, alcopop consumption is declining, we are recycling more, and we are giving more to charity.

Selecting and presenting "good news", however, is to miss the point. There is an industry doing that. At its centre is the labour government, with a stream of press releases describing how its policies have made our great country greater. The private sector's PR machine floods us with even more good news.

The problem is that it can be mistaken for spin since it nearly always comes from a source with a vested interest. One Office for National Statistics survey showed that only 17 per cent of us believe official figures are produced without political interference, and only 14 per cent say the labour government uses official figures honestly.

Data showing a decline in hospital waiting lists - ostensibly good news - were among the least trusted. Yet the gaps in the list data and the exposure of practices in some hospitals that made lists appear artificially shorter have further damaged trust.

Labour politicians, it must be said, have "form". Until recently, a section of the Labour party website showed that each of 30 or so variables quoted for each constituency had improved since it came to power. Labour had carefully selected the start date for the comparison: if the trend had not improved since 1997, it tried 1998. If that did not work, it tried 1999. If the constituency was a basket case, it presented regional figures.

It was left to the killjoy statisticians to point out that the "good" news was not really that good.

Many statisticians are driven by a search for data that will allow them to challenge assumptions. The insight could come through more imaginative presentation, the collection of new data, or more complex mathematical manipulation.

There is nothing new in this. Florence Nightingale cut the death rate in the Crimean war's military hospitals from 40 per cent to 2 per cent after analysing the causes of death and acting on the information to improve hygiene: disease was killing many more than battle.

The tendency of politicians to play fast and loose with statistics is illustrated by the story of a past prime minister calling for three sets of data: one to deceive parliament, one to deceive cabinet and one to deceive himself.

The people with real power are not those who disseminate and interpret information but those who fund its collection. Britain's official statistics are, like those of many other countries, funded by government. While that remains the case, their transparency will be open to question.

The Financial Times article:
http://www.ft.com/cms/s/af2f5904-97a9-11db-a680-0000779e2340.html

Health Direct has had a mistrust of labour's statistics since our inception in 2004. For example earlier this year on Sept 21, 2006 on PFI figures are a 'step forward in a murky area' we noted that the Labour government has long been accused of finessing the public finances by favouring private finance initiative deals for capital investment projects, such as schools and hospitals, because the debt did not show up on the government's books.

Now, after pondering the problem for five years, the Office for National Statistics has put a £4.95bn figure on the value of the debt of PFI and public-private partnership deals, which it has added to public sector net debt. Its estimate represents what it thinks the government would have to borrow today to buy back a PFI asset for the remainder of its PFI contract.

The controversy over accounting for PFI projects is not confined to opposition politicians. Jean Shaoul, professor of Public Accountability at Manchester Business School, said the reason why so many hospitals were off-balance sheet was because government accountants had assumed risk had been transferred to the private sector.

"That is a dubious assumption and I am not comfortable with the way PFI is accounted for in national statistics because when things go wrong, ultimately the public sector picks up the tab," she said.

Another anomaly, alluded to by the ONS, is that some projects, especially hospitals, are often neither on the government's nor the private sector's books. Yet others, notably roads, are included in the accounts of both.