Academics wrangle over fuel poverty and winter deaths

The Independent reports today that deaths due to fuel poverty are three times higher than government estimates, reaching 7,800 in an average winter.

Its headline “Winter deaths three times greater than first thought” is misleading. There is no change to the estimates of excess winter deaths (25,700 is the provisional figure for 2010-11, and most recent winters show comparable numbers) which are collected by ONS from death registrations.

Nobody is arguing about this total. What is at issue is the number attributable to fuel poverty, defined as households which spend more than 10 per cent of their income on fuel in order to keep warm. The Independent’s source appears to be Transform, which on Monday launched a campaign calling on the Government to use the money it collects from carbon tax revenue to make UK homes more energy-efficient.

 In a report backing the campaign, Transform says: “According to the World Health Organisation, between 30 and 50 per cent of excess winter deaths can be attributable to cold indoor temperatures. This means that on average at last 65 people a day die in the UK in winter as a result of illnesses due to cold homes.”

This contrasts with the figure reached by Professor John Hills of the London School of Economics in his review of fuel poverty for the Government. His final report is not yet published, but his interim report which came out in October 2011 concludes that 2,700 deaths a year are related to fuel poverty.

This estimate is described as “peculiar” in The Independent’s article by Professor Christine Liddell of the University of Ulster. “I see no justification for it” she is quoted as saying. “I believe the figure of 7,800 is much more realistic as it is based on WHO’s most recent estimates.”

She is referring to a review for the European Region of WHO which includes a chapter by Janet Rudge, an architect and academic at London Metropolitan University, as well as a member of the steering group of the National Right to Fuel Campaign. She cites a range of studies that lead her to conclude that 30-50 per cent of excess winter deaths are attributable to housing.

The lower of these estimates, applied to ONS’s five-year average of 26,000, produces Professor Liddle’s 7,800 claim. (I actually make the five-year average a bit higher than this, but the difference is small.)

How do we square these two widely-differing estimates? The first point to make is that fuel poverty is not the same as cold housing. Some houses are cold because their occupants cannot afford to heat them; but others are cold because their occupants, though quite rich enough to heat them, choose not to.

Professor Hills suggests that only half of cold homes are a consequence of fuel poverty, and then calculates his deaths in a different way. Citing work by the Marmot Review Team, he attributes about a fifth of the total number of excess winter deaths (21.5 per cent) to the coldest 25 per cent of houses. That would be 21.5 per cent of around 25,000 (it’s not clear what year’s data he is using) which is 5,375. If only half of these deaths are due to fuel poverty, that gives Professor Hills’ total of 2,700, rounded to the nearest hundred.

This is an odd way of making the calculation, though I’m not sure I’d call it peculiar. What he is saying is that 2,700 deaths occurring in the coldest 25 per cent of homes are attributable to fuel poverty.

What Professor Liddle is saying, by contrast, is that 7,800 deaths are attributable to cold homes, which she then assumes are cold because of fuel poverty. If she agreed with Professor Hills that only half are, her estimate would come down to 3,900 – which, given the huge uncertainties involved, is not that far distant.

Professor Hills may have modified his figure by the time his final report emerges. His interim report went out to consultation and attracted over 60 responses. Considering the number of UK academics with a strong interest in this field, I’d be surprised if some of them didn’t query his calculation.