When is a region not a Region?

As part of an ongoing effort to remove the regional government framework, Eric Pickles has announced that the Department for Communities and Local Government will no longer be publishing regional level statistics.

The arguments for and against this course of action are well-rehearsed and I do not intend to repeat them here (although you can read Shelter’s response to the consultation on this subject).

However this decision does raise a few issues.

The Department will include statistics for Local Enterprise Partnerships (LEPs) and ‘upper tier’ authorities (i.e. County Councils, Metropolitan Districts, London Boroughs and Unitary Authorities) in releases ‘where it would be appropriate and meaningful to do so.’

For all their failings, the nine English regions provided a manageable way to break the country up; using ‘upper-tier’ local authorities would mean splitting the country into 122 different areas – a rather more unwieldy number. And as for the 39 LEPs, they appear to be even less consistent and meaningful than the old regions.

A quick look at this map reveals some of the reasons why. Not only are there swathes of the country which are not covered, but there are other areas which are covered by more than one LEP.

As the DCLG points out, people may be able to arrive at totals for the regions merely by adding up the relevant local level data (indeed we will be continuing to do so on our Databank).

But this will not be possible with the English Housing Survey (EHS) – one of the key sources of housing statistics. One of the big problems with the EHS has always been that the sample size is not big enough to allow meaningful analysis at a local level. Removing regional statistical breakdowns in the EHS could mean that we are left with a major survey which supplies us with no published sub-national figures.

For organisations like Shelter, this would be annoying, but not the end of the world; we have access to the software and expertise necessary to run our own analysis of the raw data. However, for this ‘army of armchair auditors’, and indeed for smaller voluntary organisations, this will be nearly impossible.

Both the Ministry of Justice and the Department for Work and Pensions publish statistical data at a regional as well as local level. Neither has expressed any intention to end this practice – maybe because they find the framework, for all its supposed failings, to be a convenient level at which to analyse data.

We could be facing a situation where – particularly if DCLG pursue the LEP route – government departments are publishing data using incompatible geographies. It is difficult to see how this will help anyone. It appears that the DCLG is alone in not seeing that a region is a useful and convenient way of presenting statistical information, even though Regions are no longer a formal part of governmental framework.