2nd opinion: Deprived of improvement

We've grown accustomed to seeing the top of league tables as the place we want to be, whether they are the Audit Commission's inspection rankings or the Premiership. Some charts, though, you don't want to head – for instance, the roll of dishonour published last week by the government, the official indices of multiple deprivation. These are rankings of council areas showing where unemployment, crime, income, school results and life expectancy are worst.

The chart for 2004 is not strictly comparable with the last one, which was compiled in 2000. Crime and environmental conditions have been added to existing measures of deprivation, along with a new summary of local data called "barriers to housing and other services". But the new table looks remarkably similar to the old.

The geography of deprivation has hardly changed at all. The worst-off areas remain Knowsley in Merseyside, Easington in County Durham, Nottingham, Hull, Liverpool, Manchester and the three London boroughs of Tower Hamlets, Hackney and Islington.

The supporting cast is equally recognisable: Southwark and Haringey, Newham, Lambeth and Barking show up between positions 10 and 20. And the London map of deprivation is still most densely coloured in the east, with great extensions northwards to Enfield and south through Brixton into Thornton Heath.

We mustn't forget Westminster, either. The borough may be solidly Tory but this year it is ranked, unlike in 2000, as one of the 88 most deprived councils. That's a significant number because it is used for inclusion in the Neighbourhood Renewal Unit's funding catchment.

Outside London, the ranks of the next most deprived areas include Stoke, Birmingham, Sandwell, Blackburn, Salford and Hartlepool with Blackpool and Middlesbrough pressing their claim.

It's a fair bet that these are areas with large numbers of social tenants, physically poor housing conditions, arrears and so on. The map of housing renewal areas (outside London, that is) ought to overlay this map. These also ought to be the areas that get most social housing grant.

One problem, however, is that official boundaries for, say, council or health service districts include the better-off households who live alongside poor neighbours, even in the poorest parts of the country. The money may arrive in the town hall or primary care trust, but it will not necessarily get through to this estate or that housing enclave – let alone to individual households.

The ODPM has tried to address this, at least in data collection. It has abandoned local authority wards as the unit for assessing social need and replaced them with "super output areas" of between 1000 and 3000 people. These statisticians' constructs are supposed to get closer to neighbourhoods and that ever-elusive thing called "community". There's also a new acronym, of course, SOA, which might become the basis for new policy.

But success in all this will hinge on Whitehall being prepared to set need alongside public spending, and to calculate much more accurately than it has ever done just where the money goes.

Even though these indices of multiple deprivation will be used in computing support grants for councils, for distributing health service money and other funds, we don't actually know whether these poor districts do receive the most public money – which they ought to, at least under a government concerned with equality.