Rejoice! Only a fifth of government legislation is nonsense! (last year it was a quarter)
August 27, 2014Great news today from the Regulatory Policy Committee, the independent scrutineers of Impact Assessments. There has been an improvement in the number of Impact Assessments marked as “fit for purpose”, from 75-77% to 80%.
Yes, that’s right. Instead of a quarter of the government’s legislation having an evidence base which is not fit for purpose, it’s now only one in five – rejoice!
Let’s take a step back and look at what this means, shall we? First of all, what is an Impact Assessment? It’s a document that sets out the reasoning behind the government making a piece of legislation, particularly what the costs and benefits will be, and whether there are any other impacts on, for example, equality, small businesses, carbon emissions and, coming soon (if we are to believe David Cameron) impacts on families. They are supposedly a key part of policy development, making sure that the only legislation which sees the light of day is based on robust evidence. I say “supposedly” because in my experience they are also on occasions produced at the last minute, between the policy being finalised and the announcement seeing the light of day, solely to justify the actions being taken rather than as part of a judicious consideration of alternatives. At their heart, though, Impact Assessments should show you that there’s a good reason for the government to take the action that it’s taking.
So why would a panel of independent experts find that the ones you were publishing were “not fit for purpose”? Well, let’s look at a few examples, shall we? How about the BIS attempt to change the Trades Unions’ register of members regulations, where they managed not to know what they were requiring unions to do, not to give a long enough consultation for the unions to talk to them about it, and not to work out how much it would all cost to implement. No?
Well then, how about the Cabinet Office trying to consult on the proposal to introduce a register of lobbyists, where they managed to forget to explain why they were proposing the change in the first place, what options were available, whether there were any benefits from what they were proposing, oh, and to base their costs on a register of dental professionals that the RPC thought was “unclear how relevant”!
I’m sorry, but as a former Impact Assessment professional you have to allow me my moment of schadenfreude here. There is a serious point, however, which is that by the time an Impact Assessment goes to the RPC for its opinion, the responsible Minister will have physically signed the form (Jo Swinson in the case of the TU register) if it’s a final IA, or will have approved the documents for issue if it’s a consultation (Oliver Letwin and Mark Harper for the Lobbyists consultation) so the IA is the place where “the rubber meets the road” – the place where the responsible Minister has to rely on his Civil Service to give him the facts. You don’t expect him or her personally to investigate whether the numbers should be 42 or 43, but you do expect them to be able to be confident that when they sign a piece of paper saying it’s 42 they’re damned certain there’s an infrastructure in place that gives them assurance they’ve got the right figures in their hands.
It’s embarrassing to the Minister, then, to be found wanting by the RPC. It’s embarrassing to the government to have its expertise found wanting by a panel it appointed to give it independent scrutiny.
And then sometimes they go ahead and just plain do it anyway.
So that’s all right, then. The impact assessment shows you the rationale and evidence for a piece of regulatory legislation. Around a fifth of them aren’t fit for purpose. But then the government can go ahead and do what it likes anyway, regardless of whether there’s any evidence underpinning what it wants to do.
But what of tax changes, I hear you ask?
Hmmm… well, for tax changes the impact assessment is contained in the TIIN, the Tax Information and Impact Note. How do I know? Because David Gauke told Parliament it was, so by definition it must be true. But, oddly enough, the New Approach to Tax Policy Making somehow forgot to include external scrutiny of the evidence base for tax changes, so the TIINs don’t go to the RPC. I’m a lot keener on the idea that they should now that I no longer work on them, of course. Practical experience tells me it’s enough of a nightmare to get the book of TIINs out of the door in time for the Budget and the autumn statement, let alone having to wrangle them past an external scrutiny panel first.
It would be difficult to do, and inconvenient for the civil servants who have to do it, and expensive for the government (because they’d need shed-loads more people on the RPC and so they’d need to resource them better, not to mention they’d need a big spike in analyst resource to get the TIINs produced in time to get them to the RPC in time to get the Budget out of the door on time so they’d probably have to buy in some resource there, too…) But those are project management problems, not issues of principle.
It would be difficult in terms of Budget secrecy, too – increase the number of people who know about a package of measures by the number of people needed to give them independent scrutiny and you of course increase the number of opportunities for things to go astray. Again, though, a practical rather than a principle issue.
Is there a principle behind the lofty insistence that tax is different and special?
No, I don’t have an answer: it’s a genuine question. Is there?
tiintax 
I think tax law is genuinely unusual in that it is totally artificial – it relies entirely on the monetary values which get ascribed to real world transactions, hence papers like this one: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1107662
But that alone probably doesn’t justify a different approach to IA. What might justify differentiating tax IAs is the fact that it’s not just about government spending money on its policies, but about taking money away from taxpayers – sometimes on policy grounds, sometimes just to raise the revenue to pay for other stuff.
So in fact, I’d say yes, tax law is different and special – because it’s not just about spending money for our benefit, it’s actually about (potentially) impoverishing people, or even unduly benefiting lobbyists. So it should be subject to more scrutiny, not less – and because of the odd way tax law works, you might need different qualities of the reviewers, who need to assess not just the intended impacts, but also the likely unintended outcomes (which may be facilitation of avoidance, or just undesirable), as well as their interaction with other legal constructs like company law.
Shabana Mahmood MP described the UK approach as tax policy “done on the cheap” at a recent event, and there are too many examples of MPs having voted for tax measures which they subsequently protest about, presumably on the basis that they didn’t understand the full ramifications when first proposed. You’re right, it would cost a lot to get tax measures properly scrutinised & assessed. But what’s the cost of not having a sound evidence base for tax policy?