The Berwyn experiment

Berwyn Prison near Wrexham, which began operations in February this year, will be the biggest prison in the UK by some margin, with very tight staffing. It is in several ways something of an experiment.

The privately operated Oakwood and Northumberland prisons – also Cat C male, so these are all the same type – have been criticised by opponents of the private sector both for being too large, and having high staff/prisoner ratios, thus causing unsafe conditions.

But Berwyn will be much bigger - but not quite so tightly staffed (1):

                                                                             capacity                                operational staff per prisoner (2)

Oakwood                                                         1605                                                      1:5.9

Northumberland                                        1318                                                       1:6.8

Berwyn                                                             2106                                                      1:5.5

It will be interesting to see what happens, as Berwyn builds up numbers – it is currently only at about a third of capacity .

If it manages to maintain a stable, safe regime, that may put in question the assumption that staffing cuts are the main cause of the worst prison crisis in a generation. Might it also suggest that MoJ has found the 'sweet spot' where staffing is lean, but still safe? No: since prisons with similar or lower ratios are in deep trouble. Even so, MoJ's 9 'new for old' prison programme assume still further staffing cuts, to generate net savings in operating costs. But might it not also suggest that MoJ as customer has driven staffing ratios too high in private sector prison contracts - such as, indeed,  Oakwood and Northumberland? (3)

If, on the other hand, Berwyn turns out to be chaotic and unsafe, it will validate the hypotheses about low staffing - and call in question the judgement of those who planned it. And - here's the difference relative to the privately run prisons, which are held to contact - staff numbers may have to increease.

There are differing views about mega-prisons. I haven't any expertise in the matter. They offer economies of scale, so are cheaper per place.  Opponents say they will be factory-like and impede the relationships essential to good prisons. Advocates say that 3 or 4 units for different types of prisoner within a mega jail, each with their own identity, management and staff, would in fact be a series of smaller prisons, not one big one, and would have benefits for prisoners eg better specialist healthcare on site. Maybe: but I suspect that despite that, it will still feel like a mega jail, and something will be lost.  Their sheeer size means that prisoners cannot be held as close to home as would the case if we built a network of small  community prisons instead, though that would cost more.

Berwyn marks a major change in accommodation policy. Hitherto, we had always aimed at one man to a cell: that was how overcrowding was measured - how many were held two or even three to a cell.  In Berwyn, many(I think the majoirty) will be housed two to a cell, albeit one designed for double occupancy, whether they like it or not.  No credible reason is advanced for this, other than packing more in, and driving down construction and operating cost per place. I am surprised how little comment there has been on this historic change. It should be independently evaluated.

I haven't mentioned cost. The headline figures are:

                                                                              annual cost per place, £

Berwyn (est., when when full) (1)                     14, 000
Oakwood (4)                                                                 14, 348
Northumberland (4)                                                 14, 548

So why are privately run prisons more costly, even though they have leaner staffing? Because they aren't. To compare like with like, you'd need to make several additions to the Berwyn figure:

1. the private sector has to insure against damage, riot etc. and being sued; the public sector doesn't, because the Treasury pays all such bills.
2. the public sector prisons consume various support serves supplied and paid for nationally, which the private sector does not.
3. the private sector funds its pensions, the public sector does not. There is no fund. Pensions paid out to today's civil service pensions are netted off against the contribution of today's staff to their future pensions, and if there is a shortfall – for example, if there are far fewer civil servants in the future than now – the future taxpayer will just stump up the difference (but won't, in fact, be able to trace the shortfall back to any one group of staff or period of time) No wonder the PCSPS has been called a Ponzi scheme! (5)

We can estimate these missing figures:

Past experience of trying to ensure a level playing field in competition suggests an addition of around 2% for item 1 (6).

MoJ accounts show the on cost for item 2 is 25% for the private sector but 60% for the public sector, thus we should allow 15% differential on cost to the public sector (7).

3 is unknowable - but pension experts think that even under the reformed PSC scheme, the combined employer and employee contributions are likely to be 5% short of actual costs even on current assumptions, and it could be much more. But we'll never know – that's the beauty of the thing. Uncosted and uncostable (8). As so often nowadays - pensions, nuclear power, climate change, the NHS, social care, the ageing housing stock - when long term costs  look too big to ever be affordable, we just ignore them.

That suggests the true comparison requires an additional 22 % for the public sector, thus:

                                                                                           annual cost per place, £

Berwyn (when full)                                                            17, 500
Oakwood                                                                                14, 348
Northumberland                                                                14, 548

About 5-7% of the higher cost in the public sector is having more operational staff; the rest is a a mix of higher (though hidden) pension costs, more non operational staff, higher pay, longer holidays, more sick absence, shorter working weeks, less flexibility, or more costly supplies and service contracts including central services (6). Of course, if we were looking at operational cost, rather than cost to the public purse, the gap between sectors will be even larger, because of the profit margin.

The significance of these figures is considerable. It means that in stating to Parliament that public and private prisons are equally efficient, Government has told an untruth (9); indeed that Government doesn't actually know what the true comparison is (10); and that in terminating competition on the bogus basis that there is no longer significant cost difference between sectors, MoJ has shown precisely the kind of 'producer self interest' which might be expected, given that the public sector has, until recently, both awarded contracts for prisons and has also been the biggest competitor in those competitions. (11)

It also implies that on a true cost comparison, the private operators could provide a better service than the public sector for the same money – but is not allowed to, because its costs are being mis-represented, relative to the public sectors.

No wonder the companies I worked for were so wary about competing against public sector prisons! (12). The match is rigged.



NOTES

1. Data from FoI applications
   
   
2. Public sector = prison officers + OSGs, private sector PCOs
3. http://www.julianlevay.com/articles/cuts-and-the-crisis-in-prisons-the-impact-on-the-private-sector
4. 'Costs per prison place and cost per prisoner by individual prison establishment 2016 to 2017 tables'
5. Chapter 9 in my book explains the way the PSCPS works. For enthusiasts only!
6. See chapter 11 of my book
7. Costs per place and costs per prisoner by individual prison HM Prison & Probation Service Annual Report and Accounts 2016-17 Management Information Addendum
8. See also Robert Peston, http://www.bbc.co.uk/news/16419885
9. Reference at note 3: I asked Treasury to justify the statement: they cannot
10. The last proper cost comparisons were done in 2000, see note 6; FoI applications reveal that none have been done since.
11. See the end of chapter 15 of my book
12. In the case of MITIE, the Government not only aborted the competition but then adopted the company's key idea.