Saturday 13 February 2010

Edmund Burke must be turning in his grave

John Rentoul has a significant and informative post with more on the Tory plans to change the rules on constituency boundaries if they are elected to government.

The main thing that can be said about this approach is that, even if designed to favour the Conservative Party's electoral interests, it is a fundamentally unconservative approach to redrawing the political map of Britain.

The main rule is simply this:


(3) The electorate of any constituency—

(a) shall be as near the electoral quota as is practicable, and all other special geographical considerations, including in particular the size, shape and accessibility of a constituency, shall be subordinate to achieving this aim,

(b) and shall in any case be no greater than 103.5 per cent. and no less than 96.5 per cent. of the electoral quota, except where this makes it impossible to meet the provisions of clause 2 of this schedule.

(Clause 2 says that constituencies may not cross national borders: "A constituency shall be located wholly within one of England, Wales, Scotland or Northern Ireland.")


It is a "one size fits all" cookie-cutter approach which will surely do much violence to the natural boundaries of communities, historic constituencies, local government districts and other factors. It will produce ever more Parliamentary constituencies which have no natural resonance at all - and which would make the 1974 local government boundaries seem like a model of local sensitivity.

Any conservative with a nodding acquaintance with the work of Oakeshott or Burke would be flabbergasted by any issue of special local community or geographical relevance being always "subordinate" to this tight utilitarian quota. It is quite out of step with Steve Hilton's localist mantras and buzzwords too. (And the contrast with the indicative STV map produced by Denis Mollison to respect local boundaries is very striking)

To take the most obvious examples: the Isle of Wight is currently 50% larger than the average English constituency. It would not, under these rules, be possible to maintain that, nor to have two MPs for the island. So there will have to be one MP for perhaps three-quarters of it, and a substantial chunk of Isle of Wight voters to be a minority of constituents for a mainland MP. The impact on the Western Isles would be more dramatic still: it will have to be absorbed into a constituency which no mortal could be expected to properly represent.

One of the supposed great virtues of first-past-the-post, the MP-constituency link, is inevitably weakened by such an approach.

It will be interesting to see what current and former Tory MPs, and their local papers, make of the plan when the implications of the splicing and dicing of every constituency becomes apparent. One trusts that any true localist - Simon Jenkins comes to mind - will go ballistic.

And for what purpose?

Of course, this does not achieve the 'equal value of votes' aim which Eric Pickles and David Cameron have argued is its aim. There will be safe seats which are never competitive, and marginal seats which matter, as now.

As Left Foot Forward reported, the non-partisan academic expert advice is that this is not a route which can successfully address the sources of the sharply shifting electoral bias under first-past-the-post, of which constituency size is only one relatively minor variable.

We have an authoritative advice on that because David Butler convened an expert group of academics - all holding different views on normative questions, including FPTP and PR supporters - which agreed on many of the technical questions about the workings of the current system. The group concluded that:


“The principal sources of disproportionality have nothing to do with boundary-drawing or the detailed statutory rules which the Boundary Commissioners have to apply. Changes in these rules would do very little to make results more proportional


Some estimates suggest that constituency size amounts for perhaps 15% of the current electoral bias. Other factors include differential turnout and tactical voting. (One in ten voters say they do not vote for their preferred candidate or party). And any redistricting which does not take into account differentil voter registration, as this one does not, is also open to the charge of a partisan gerrymander.

While Labour's proposal is to put an electoral reform to a public referendum, which will require a majority of those voting to proceed. The Conservatives are intending to simply legislate this extremely flawed plan, presumably on a whipped vote against other parties.

In the end, it is very difficult to see how first-past-the-post can be reliably fixed.

Its ability to give the largest party an overall majority depends on a happy accident of electoral geography. In any close election, this has become a matter of Russian roulette, as I argued in a Fabian Review essay looking at this fatal flaw for FPTP's majoritarian ambitions.

That does not say anything about the argument between the principle of proportional or majoritarian systems. Peter Kellner presents strong evidence that the Alternative Vote is a more reliable majoritarian system (including in his own submission to Jenkins). If they do not wish to advocate that, then Conservatives and others who wish to defend the principle of giving the electorate a clear choice of majority governments would do much better to advocate a system designed to achieve it: the direct election of the Premier or governing party being the most obvious route.

No comments: