Quangocracy - euthanasia or liberation?

Not for the first time, I have been scattering observations on an issue, in comments dotted around the place, rather than pulling the strands together for a piece on here. This time, the topic is the rising cost of the quangocracy. As a subject particularly close to my heart, I thought I'd pull the comments together before I forget where they are.

Dan Lewis, of the Economic Research Council (and some time of the Centre for Policy Studies), has produced an analysis of the increasing size and cost of our quangocracy. (It will be available on the ERC's site from Thursday 23rd August.) The Sunday Telegraph produced a piece on this analysis, in which the overall cost of the quangos ("nearly £170bn a year") was highlighted and compared with the budget of the Ministry of Defence (£32bn). Tim Worstall picked up on this article, and posted a brief comment arguing for the complete abolition of the quangos and the use of the savings to abolish income and corporation tax. Nothing wrong with the principle, but there is a bit of a problem with the calculations, as I commented on Tim's site (though rather too late to be helpful, as it turned out):

No one could be more opposed to quangos and determined to see most of them shut down than me. But a couple of things to bear in mind on the cost:

1. Around two-thirds of that total cost is attributable to quangos associated with one department of government - the Department of Health. Effectively, the NHS is a giant set of quangos, amongst which are the Primary Care Trusts and the NHS Trusts (which as a pair make up over £100bn of that sum).

2. Of the executive NDPBs (the busy-body, rather than funding quangos, which is what we usually have in mind when we rage against them), over half of the total cost of around £36.8bn lay within another department - the Department for Education and Skills. But the lion's share of this cost is also attributable to two quangos - The Learning and Skills Council (responsible for post-16 education and training) and the Higher Education Funding Council for England. It turns out, then, that although they calls these NDPBs "executive", most of their cost lies (like the NHS quangos) in the allocation of public funds.

This shouldn't diminish the attacks on the wastefulness of our quangocracy, but we should bear in mind that it is the sclerosis induced by bureaucracy that is their really harmful aspect. They cost and waste money, but only a fraction of that headline figure. We can probably save several billion pounds by scrapping most of them, and that's money worth saving, but we can't save anywhere near £167bn.

In the meantime, Tim had produced an article for yesterday's Times calling for the "euthanasia" of all the quangos, which he reproduced in another post on his blog. This prompted considerable commentary, most of it enthusiastic, on both Tim's site and at The Times, but one or two questioned the maths, so I felt the need to set the record straight again:

"You mean the whole NHS, or just the NHS quangos?" [from comment by Smidgeon on Tim's site]

Yes, "just" the quangos, but let's put that in perspective. Govt funding of NHS quangos: £88.5bn. Total gross expenditure of NHS quangos: £112.3bn. Total managed expenditure of the Department of Health in budgetary year 2006-7: £96bn. In funding terms, the NHS pretty much is the NHS quangos.

"Ambition, that’s what we need in politics today." [from Tim's article]

Yes, but even more importantly, we need careful and rational analysis. PT is bang on about the real reason to scrap most of the quangos, plus the removal of the sclerotic effect of a vast bureaucracy looking for ways to justify itself.

Sure, we can save some money, and whatever genuine savings we can make are worth having. But this is a giant bear-trap waiting for anyone who tries publicly to claim that they can save £167.5bn by scrapping the quangos. Public funding of the quangos comes to £123.8bn. Just to save that much, you've got to publicly espouse privatizing the NHS and a part of our education system, and doing away with all regulation (I don't disagree with that, but politically...). And even then, you haven't really "saved" the money, because a lot of the money you've put back in people's pockets from privatizing these services and reducing taxes, will have to come back out of their pockets to pay for the privatized services. It's still worth doing, in principle, for the various economic reasons that hopefully don't need stating here (though the Michael Moore objection on overheads is a counter-argument that needs addressing seriously), but politically it's cloud-cuckoo land, and it's certainly not a saving of anywhere near this scale in terms of the net impact on household budgets (which is what really matters to people).

And that's just to save the first £123.8bn. To save the remainder, you've got to show that all of those charges levied by the various quangos will vanish once the quangos are abolished. Yes, there are plenty of cases of quangos charging their "customers" for "services" that those customers don't want (anyone who has dealt with the Environment Agency and carried out yet another OPRA calculation will know that feeling, for starters). But what about health? That's where over half of those charges are levied. Do we think that prescription charges will be a large part of that? And do we think that prescription charges will be abolished as a result of the privatization of healthcare?

It's no good seizing uncritically on numbers that suit us. If it doesn't stand up to critical examination, it will damage the credibility of the case that it is being used to support, and of the person presenting the argument (Dan :-( ). There are more than enough good arguments for smaller government, without having to resort to bad arguments.

I was being a bit harsh there on Dan. This seems to have been Chinese whispers. Re-reading the Telegraph article, Dan doesn't seem to have made any outlandish claims of the extent of the potential savings.

Tim, to his credit, picked up on these comments, and pointed out the problem with his maths in comments on both his site and at The Times. And, as I said at the start, there's nothing wrong with his principle of getting rid of as many quangos as possible (all of them is probably a bit optimistic). We just need to be realistic.

Mark Wadsworth, a fellow flat-taxer, liked the comment enough to post a note on this site in another thread (much appreciated), and to do some research on a different angle to get at the potential savings. Mark's approach is to try to estimate the number of extra jobs created within the quangocracy, the assumption being that the savings from their abolition would be roughly equal to the savings in wages and on-costs. On that basis, Mark reckons that about half of that £167bn cost could be saved. It's a clever idea, but again we need to be careful with the numbers:

That's a very useful reference to employment statistics on your blog. It tells us a number of things:

  • Between summer 1999 and winter 2005/6, 1.5m jobs were added to the economy. That included 1.6m jobs in "Public admin, education and health" (PAEH). In other words, the number of jobs in the rest of the economy actually declined in the period, while the PAEH sector grew like topsy. I hesitate to say that the productive part of the economy has been shrinking at the expense of the unproductive part, because teachers, doctors and nurses are no less intrinsically productive than laywers or accountants, but one could say that the part of the economy that produces profits and saleable goods on which taxes are paid has been shrinking at the expense of the part of the economy that is tax-funded and free at the point of use. This is obviously unsustainable.
  • Mind you, Major wasn't much better. The big break from the past occurred between Spring 1991 and Spring 1992, when the proportion of the workforce employed in PAEH jumped from around the 21% mark that it had occupied during the Thatcher years to around the 24% level, from where it has crept upwards under Labour. A clear indication that public-service employment is being used to disguise economic woes, as the obvious trigger suggested by those dates is White Wednesday.
  • The jump in public-sector employment under Major is mirrored by an abrupt drop in manufacturing employment. Thatcher is often blamed for the decline in manufacturing. The proportion of the workforce employed in that sector was indeed going down steadily under her, but the absolute numbers barely declined at all (by fewer than 12,000 jobs, or less than 0.2% of the total in the sector). The declining proportion simply reflected the fact that total employment was increasing strongly, largely through the creation of jobs in the service sector. The real decline began under Major, with over a million jobs being lost between Spring 1991 and Spring 1994, which looks like a load of businesses giving up the fight after the economic shocks of the previous couple of years. But it stabilized at this lower level for the rest of the Major years (for which much of the credit should go to Ken Clarke - one of the best Chancellors of the twentieth century, to whom Tony and Gordon owe a great deal of thanks). Under Labour, there has been a renewal of decline, with another 1.2 million manufacturing jobs having been lost by Winter 2005/6.
  • The dependence on PAEH jobs to disguise decline can be contrasted with the Thatcher years. Between Spring 1984 and Spring 1990, two and three-quarter million jobs were created, of which 650 thousand were in the PAEH sector. This also gives the lie to the suggestion that public services were starved of funds - 650,000 new jobs in the public sector is not insignificant, but it is more proportionate to the growth of the overall economy than the employment changes under Major and Blair.
  • We need to be a bit careful ascribing the increase of 1.7 million jobs in the PAEH sector under Blair purely to the quangocracy or the public sector. The spreadsheet gives figures for the public sector as a whole. There is a discrepancy of about 1.2 million between the number of new jobs in the PAEH sector and the number of new jobs in the public sector, i.e. PAEH was growing much faster than the public sector as a whole. A number of factors may be at play. It is possible that cuts in other parts of the public sector were balancing against new jobs in the PAEH sector, but it's hard to think where these other parts and cuts might have been. More likely causes include (a) the outsourcing of services in the PAEH sector (e.g. cleaning in hospitals), so they no longer count officially as public sector, though they remain part of the Greater Bureaucracy, and (b) an increase in private health and education provision. You probably don't want to be attacking the latter, and the former (if it is things like cleaning and catering) isn't generally dispensible.

All the same, the generality of your comment on your site is still true in the broadest sense. Whatever the proportion of employment in the private and public sectors and the specific numbers in the quangos, the overall levels are up dramatically, which means we ought either to expect significantly improved levels of service, or to be able to find quite a bit of flesh to cut. I doubt, for the reasons above, that it will be anywhere near 1.7 million jobs that can be cut. Perhaps 500,000 - the increase in the number of public-sector jobs - would be closer, though even there, a significant number of those jobs will be new doctors, nurses and teachers. But as increasing private provision should be balanced by decreasing public provision, that 500,000 disguises a larger increase in public-sector flabbiness, so the potential savings would be greater than just the difference between that increase and the increase in the number of front-line employees. As a very round and rough stab in the dark, how do you fancy calling it around £25bn that could be saved?

Of course, the net saving will be less if most of the 500,000 go on the dole. The benefit to society will depend on how many of them are usefully employable in the private sector, and the number of those that choose to take private-sector jobs. The incentives under our tax and benefits system will be crucial in determining whether public-sector job cuts turn into a burden on society in the form of increased unemployment, or a benefit to society in the form of converting unproductive work into productive work.

Others who ran with the story included Tim's colleagues at the Adam Smith Institute, and one of the Samizdata authors (who spotted the NHS gotcha). The document (PDF) on which Dan's research is based can be downloaded from here. It is an annual document that was last produced annually about a year ago, so another reservation would be that this research may soon be out-of-date.