©John Walton/PA Archive/Press Association Images
It is often argued that one of the desirable features of a democratic election is that every vote should be of equal value. One of the ways in which that principle can be realised is by ensuring that the ratio of representatives to voters is more or less the same everywhere.
In the case of Britain’s single member plurality system, that means ensuring that the number of voters in each constituency is more or less the same. If that is not the case, the votes of those in constituencies that contain fewer registered voters may be thought to have more weight and influence than the votes of those in seats with more voters. And if one party does especially well in constituencies with fewer voters, it is likely to secure what might be thought to be an unfair advantage in terms of its overall tally of seats.
At the last general election, there were some big discrepancies between constituencies in the number of registered voters that they contained. Leaving aside the island constituencies, where special geographical considerations may be thought to apply, there were on average just over 48,000 people to vote the ten smallest constituencies in the UK, but over 90,000 in the ten largest.
There are two main reasons why these discrepancies exist. First, until now, the rules for creating parliamentary constituencies have lacked any mechanism for equalising the sizes of constituencies across the four parts of the UK. Apart from a once and for all increase in Northern Ireland’s representation in 1983 and cut in Scotland’s in 2005, the number of constituencies in each part of the UK has simply been determined on the basis of how many seats it contained last time, even though over the long-term the electorate in England has grown more rapidly than elsewhere.
Consequently, whereas in May 2015 constituencies in England on average contained just over 72,500 registered voters, those in Scotland had just under 69,500, seats in Northern Ireland 68,700 voters, while those in Wales (which have never been aligned with those elsewhere) typically only contained just over 57,000. Meanwhile, whatever justification there might have been before the advent of devolution for the over-representation of Scotland and Wales certainly does not apply now that both of them have their own law-making bodies with responsibility for substantial areas of domestic policy.
"The average urban seat in provincial England gained less than 1,800 voters between 2005 and 2015; the average more rural one gained nearly 3,200"
The second main reason for the discrepancies is that the current constituencies were drawn up on the basis of electorates that are now well over a decade old. In England the current constituency boundaries are based on the electorate in 2000, in Scotland and Wales in 2002, and in Northern Ireland in 2003. Outside of London at least, much of the population movement in the intervening period has been out of more urban areas into more suburban ones. The average urban seat in provincial England gained less than 1,800 voters between 2005 and 2015, whereas the average more rural one acquired nearly 3,200 more.
Given that Labour performs better in Wales and in more urban constituencies, these patterns mean that at the time of the election on average seats won by Labour in 2015 contained just under 4,000 fewer registered voters. Even within England alone the discrepancy is almost 3,000 voters.
So, it is bound to be the case that any redrawing of parliamentary boundaries will be relatively disadvantageous to Labour. In truth, because of the long-term movement of population out of many of our cities, this has been a feature of every one of the last four boundary reviews. The one added twist of the latest review, the initial recommendations of which are due to be released later this month, is that new rules for drawing up constituencies introduced by the Conservative-Liberal Democrat coalition and now being implemented for the first time require that constituencies should be of the same size throughout the UK. Thus Wales in particular will lose out.
But does this mean that the latest review is entirely uncontroversial? Certainly not. First not only are constituencies being made more equal in size, but the number is being reduced from 650 to 600—a hangover of promises made by both the Conservatives and the Liberal Democrats in the wake of the MPs expenses scandal. This means that the review will be much more disruptive—for Conservative MPs as well as those on the opposition benches—than would otherwise have been the case. It will not just be Labour MPs who find that their seat has disappeared—so also will some Conservatives.
Second, the success of any endeavour to equalise constituency sizes depends on the accuracy of the electoral register, and that is in some doubt. The new review is being conducted on the basis of the registered electorate as of 1st December 2015. Controversially the Conservative government decided that this register should be compiled wholly on the basis of a new process of individual, rather than household, registration, a change that has been being phased in gradually for some time. The upshot of that decision was that the names of 770,000 people whose right to be on the electoral register had not been verified were deleted a year earlier than previously planned. Although the Electoral Commission has concluded that most of those names were probably correctly deleted, the decision seems to have exacerbated the pre-existing high levels of under-registration among younger people and private tenants. Certainly the difference between the average electorate in seats held by Labour and in those won by the Conservatives is now some 400 greater than the 4,000 difference already in evidence at the time of last year’s general election, meaning that the review will hit Labour even harder than already seemed likely back then.
Indeed, the incompleteness of the register upon which the current review is taking place has been dramatically demonstrated by the marked increase in registration in the run up to the EU referendum. An extra 1,800,000 names were added to the register during that period. If we really want the system for drawing up boundaries to be seen to be fair to everyone, we are going to have to do much better than we have been at ensuring that people are actually on the electoral register in the first place.
