Here’s a simple question: why does the UK have a single agency, HM Revenue and Customs, to collect most taxes?

This idea is so embedded in the UK political psyche that it is never, ever, questioned. But there are clear examples of alternatives that work perfectly well. In Denmark, for example, local government collects most national taxes. There are still national tax policies, but they are implemented locally.

If this question is taken seriously, it immediately raises a host of other issues. In Britain we have certain public services that have for many decades been seen as inevitably ‘national’ – tax collection is just the most obvious. For centuries this was through HM Inland Revenue and HM Customs and Excise, both of which were ‘non-ministerial’ departments in Whitehall.

There are other services that, almost by accident, have been ‘national’. The biggest of these has been the delivery of employment services and unemployment and associated benefits delivered through the now joint Job Centres Plus (merging the old Employment Service and Benefits Agency).

The other main ‘nationally’ organised service has been prisons – HM Prisons Service (but covering England and Wales only) and more recently the National Offender Management Systems (NOMS) which attempts to create an integrated prison and probation services (but again for England and Wales only).

In none of these cases – tax collection, unemployment management, or ‘corrections’ – is there any obvious or overwhelming reason why they should be delivered through national agencies, rather than local government. There are plenty of examples across the western world of where all of these services are delivered locally, even where the policies governing them are decided centrally.

Let me take one example – prisons. After the infamous ‘Derek Lewis’ affair, when the DG of the Prison Service was sacked in 1996, I was privileged to serve on a Review Committee of the organisation and management of HM Prison Service. This was an illuminating experience, supplemented by some others such as engagement with the Canadian federal corrections service.

What became obvious very quickly was that ‘corrections’ services (prison and probation) for all but the most extreme offenders were best organised locally. Most prisoners are short-term, and their recidivism rate is determined by the chances of meaningful personal relationships and job opportunities, both of which are lessened by them being sucked into an impersonal national corrections system. Currently, prisoners on short-term sentences may end up hundreds of miles away from where they live, undermining family connections.

What was also obvious was that there were two enormous blockages to developing a sensible corrections policy and service. First, no-one was willing to question why prisons needed to be organised ‘nationally’. This was especially ironic as we already had three prison systems in the UK: England and Wales; Scotland; and Northern Ireland. There were no obvious problems for the two smaller systems, yet questioning the ‘national’ nature of ‘HM Prison Service’ (for England and Wales) was absolutely off the agenda.

Second, there was clearly a ‘path-dependency’ problem that was embedded in the distribution of the prisons ‘estate’ (i.e. where prisons were located). Prisons had been built where they could be, not where they were needed (i.e. in the areas where most inmates came from). History means that a lot of prisons are simply in the wrong places, but the costs and long-term planning needed to correct this is a huge barrier to change.

I am pretty sure that similar critiques could be made of the ‘national’ organisations of tax collection and unemployment services.

The other two biggest areas of public services – health and education – have also become, in different ways, increasingly ‘national’. For education (in England) schools may have been given greater autonomy in various ways, but they have also been dragged away from local government and increasingly come under the dictat of the Department for Education in Whitehall. In health, the picture is slightly more complex but the one thing that hasn’t really happened is that local government has gained more control of local health services.

There have been all sorts of debates about ‘localism’ in England over recent years but what is most striking about them is their complete inability to challenge the status quo about the distribution of functions between central and local government. At the same time it is frequently pointed out that local government in England is much larger (in terms population served) than many of our European neighbours.

So here is a really radical thought: maybe we should make local government bigger and smaller.

By smaller I mean that local government should cover smaller population groups than it currently does, to provide greater opportunities for local engagement.

By bigger I mean massively increasing the number of functions local government performs, and thereby increasing their ability to really shape their areas. This could include devolving a lot of tax collection, unemployment and benefits services, and even prisons and probation to local government. It would also include re-establishing local government oversight of education and bringing most health services under democratic local government control.

None of these would prevent us having national policies on issues of tax, unemployment, incarceration, education or health, but it would create a huge national laboratory in which different local governments could try different ways of implementing national objectives and to learn from one another whilst doing so.

Local government would also improve through having much bigger financial flows at its disposal, attracting better staff and management and political leaders who could try and make a real difference to their areas. At the moment local government in England suffers from controlling only a tiny fraction of public services in their areas. ‘Joined up’ government would be massively enhanced in local areas if all (or most) public services were co-ordinated through local governments (as the ‘Total Place’ initiative tried to do, but in a very constrained way).

This would be a radical rebalancing of central-local relations and the fact that no-one even seems willing to question some of the currently entrenched ideas about ‘how we do things around here’ shows just how radical it would be.

There’s not enough space here to go into how radical rebalancing could also apply to the distribution of functions between UK central government, England, Scotland, Wales and Northern Ireland. But it is these fundamental questions, which apply also across England, which have been raised by the Scottish independence referendum.