Conservatives must not redraw the map (5 Oct 2009)
Eric Pickles has proposed reducing the number of MPs. But ‘equalising’ the size of constituencies is a deeply flawed policy
In his speech to the Conservative party conference today, Eric Pickles claimed to stand for “fair votes”. He did not mean an end to the first past the post system that gives all the power on the basis of under 40% of the vote and ignores votes cast outside the marginal seats. What he meant was a policy of cutting the number of MPs from 650 to 585 and a promise that “we will make all constituencies equal in voting size”. This would be accomplished in time for the general election after 2010.
Cutting the number of MPs is a bit of easy populism, made even easier by the expenses scandal. It might not necessarily be a bad idea; the number of MPs should be determined by the need of constituents for representation and the needs of parliament to function well as an institution. However, the Conservative rationale is cost-cutting. It is doubtful that cutting the number of MPs will really make much of a saving in terms of public spending – after all, the same amount of constituency casework will just end up being done by fewer MPs. There is also the possibility that unless the number of ministerial jobs is sharply reduced, there will be more executive dominance of parliament than we have already.
The principle that constituencies should be more or less the same size is generally accepted. The issue is how much tolerance for variation from the average constituency size one allows, and how frequently the boundaries are redrawn. Currently, the Boundary Commission allows around 10% either side of the ideal (ie 63,000 to 77,000 electors) with a bigger margin for geographically difficult mountainous or island areas. The Conservatives are talking in terms of a rigid rule not allowing more than 5% either side of the new ideal figure (77,000 after the number of MPs has been cut). To keep within this limit, boundary reviews would have to become more frequent and proceed faster than the current, admittedly ridiculous, system where the boundaries coming into force in 2010 are based on electorate figures from February 2000.
Pickles appears to believe that the major cause of the pro-Labour bias in the electoral system evident in the 2005 election was variation in constituency size. This is factually untrue. Constituency size was a small component of the bias, but most arose from other factors such as low turnout in safe Labour seats. Labour’s vote is efficiently distributed, partly because of tactical voting in 1992 and partly through New Labour’s successful electoral strategy. It is quite possible that less tactical voting, Conservative targeting of marginal seats with Ashcroft money and Cameron’s appeal to “Middle England” could cause a lot of bias to unwind anyway in 2010.
The Conservative boundaries policy would require a rapid boundary review during the next parliament – the shortest recent review (1991-95) took four years, so the Boundary Commissions would have to have extra resources to accelerate the task. The new boundaries could not be subject to the same scrutiny at public inquiries that makes the current process so lengthy – there are simply not enough assistant commissioners (usually barristers) available to run the inquiries.
Inquiries of whatever sort involve taking evidence from political parties, community groups and local councils, and in arguing for the boundaries that suit them, a party needs to be well-organised and professional. By having a quick review after the election, the Conservatives must be hoping to go into the process well-funded and prepared and facing a demoralised and impoverished Labour party before it has regrouped. The Conservatives would also be at their peak in local government, and could use council submissions to back pro-Conservative boundary schemes.
It is unclear to what extent the Conservative policy will address the handful of hard cases that are often used in calls for equalising boundaries – the Western Isles seat, Na h-Eileanan an Iar, has five times fewer electors than the Isle of Wight. It is quite possible that these anomalies would survive under a Tory plan – the number of seats involved is small, and the alternatives (a seat spanning the Solent, for instance, combining bits of Portsmouth with towns on the Isle of Wight?) are geographically absurd.
The effect in other areas is, however, only slightly less ridiculous. The policy is going to face a huge backlash when people realise what it means. Equalising the size of each constituency will mean crossing county and ward boundaries and ripping up what remains of the traditional map of community representation. A bit of Cornwall would end up in a seat based on West Devon or Plymouth, whatever its residents thought about the matter, and angry voices would ring out in every public inquiry. This would be repeated time and again, because there would be frequent changes to keep constituencies within the 5% threshold.
This freedom to cross ward and county boundaries also increases the ability of well-prepared parties to manipulate the process. It is no coincidence that the worst gerrymandering in the developed world is for US Congressional seats, where there is a rigid requirement of arithmetic equality within 1% of the ideal population size within each state. Legislators draw preposterously biased lines on the map which make no sense according to any administrative, social or physical geography, as long as the right number of people are corralled together.
Although apparently fair, “reduce and equalise” is a badly flawed policy. As a supporter of proper electoral reform, I sometimes mischievously think it should go ahead because it might hasten the end of first past the post. That mystical link between MP and single member constituency will be broken up because a large number of MPs will represent constituencies that correspond to no local community identity, and whose boundaries will shift around every few years.
The fundamental problem is that it is impossible to produce one-size-fits-all single member seats while keeping natural communities together. Communities, as any conservative should know, come in different sizes. To achieve numerical equity and community identity requires multi-member seats, as with the flexible single transferable vote (STV) system used in Ireland. STV would enable one to keep community boundaries intact while putting representation on a more equal arithmetical footing. It would be tragic, and ironic, if in the pursuit of arithmetic perfection and a chimerical public spending cut a Conservative government created electoral units that would make the Heath-Walker local government map look popular.