Daily Archives: 4 July 2010

Electoral Mythbusting 2: spotlight on Labour and boundary changes

The proposal to hold a referendum on changing the electoral system to the Alternative Vote is Labour’s policy, so you would have thought they’d be delighted that the coalition government is going ahead with it, wouldn’t you? The problem is, a) Labour’s commitment to the policy is at least partly tactical (designed to appeal to Lib Dem voters – and Lib Dem MPs in the event of a hung parliament. Ironically, the effect was to make a Lib-Tory coalition more likely) and b) the Tories are insisting on implementing the policy alongside their own reforms of reducing the number of MPs by 10% and “equalising” constituency boundaries in order to remove a perceived bias in favour of the Labour Party. Labour politicians are up in arms at this and are threatening to bring down the whole bill.

To those of us outside the big two parties, this debate is somewhat baffling. They are throwing claims and counterclaims at each other regarding “gerrymandering” with seemingly no self-awareness at the fact that the current system (and even AV) gives both parties a tremendous inbuilt advantage that no other party enjoys. The sense of entitlement on both camps is eye-watering. But, that aside, can we legitimately accuse the Tory proposal as “gerrymandering”?

First of all, if you support single member constituencies, then you support in principle the idea that constituency boundaries should be drawn up in such a way that give different parties an advantage over another party. That is gerrymandering by another name. The reason for this is basic mathematics and gets to the heart of why no system which uses single member constituencies exclusively can be called proportional. It is best illustrated by what is known as the “gerrymander wheel“. The wheel shows how you can dramatically change the expected seat share each party gets simply by drawing the boundaries slightly differently. It is a problem with any electoral system with constituency boundaries, but the problem is greatly reduced even with two member constituencies with multi-member constituencies it rapidly becomes difficult to gerrymander.

But there is another factor, and this is something that both the Tories and Labour have got completely wrong. The fundamental problem the Tories have under FPTP is not the way the boundaries are drawn up but where their votes are. Simply put, Labour’s supporter base is spread across the country while the Tories tends to be concentrated in specific areas. This means that no matter how much you redraw the boundaries, Labour will still do better than the Tories nationwide while the Tories will always tend to have a concentration of safe seats (all things being equal).

The result is, any attempt to redraw the boundaries is unlikely to change very much, as two seperate academic studies have shown. So why is Labour getting so het up about it? Well, a factor is almost certainly the opposition party playing opposition games, but they do have one point: with millions of people not on the electoral register, some constituencies contain many more people than the election results suggest. This tends to be a particular problem in urban areas, which are typically more Labour than Tory. It is a problem that Labour had 13 years to sort out and refused to, so it would be nice if we heard a little more humility about it, but that isn’t the fault of the people affected, and I would agree that this should be taken into account.

What this can’t be used as however, is an excuse to not hold a boundary review, or an argument against equalisation. It certainly wasn’t during the two boundary reviews conducted under Labour and we certainly should not assume that those “missing” voters would all vote Labour given half a chance, no matter how great Labour’s capacity for self-delusion might be. With a census due to take place next year, this is in fact a good time to conduct a boundary review taking this fresh data into consideration. The Electoral Commission are already in the process of studying how complete and accurate registers are (pdf), and so long as the Boundary Commission are required to take this into consideration (in a transparent way), I can see no reason not to proceed at this point. The Electoral Reform Society have suggested that it might even be slightly beneficial to Labour; so be it. I suspect these details will all get thrashed out in committee in any case.

But there are two other objections to this agenda which are also being bandied about. One is that the combined effect of “reduction and equalise” will be to weaken the constituency link by ending the practice of having constituencies reflect local communities. The other is that reducing the number of MPs is itself undemocratic and bad for Parliament.

Superficially, there seems to be some truth to the first argument, which does make a bit of a nonsense out of the Tories’ claim to be the great defenders of the single member constituency link. How can you argue for that in principle, while reducing the degree to which constituencies reflect communities? And of course, I should include my own disclaimer that as far as I am concerned, anything that weakens the single member constituency link and results in MPs doing their job as legislators instead of their phoney job as social workers, is an entirely good thing. Bring it on.

But let’s not fool ourselves that the current system does a good job at reflecting communities; it doesn’t. That is due to three reasons: there is no fixed size for a “community”, the average constituency size doesn’t come close to reflecting the typical community and the concept of community itself is more mutable than it was, say, 100 years ago.

Here, for example are all the constituencies I have ever lived in:

  • Ravensbourne (Bromley), which incorportated the council wards of: Biggin Hill, Bromley Common & Keston, Darwin, Hayes, Martins Hill and Town, West Wickham North and West Wickham South. As a West Wickham resident, I considered my “area” to be West Wickham, Pickhurst, Hayes and Bromley. Biggin Hill might as well have been on the other side of the planet. I couldn’t even tell you where Martins Hill is.
  • Manchester Gorton (Manchester), which incorporated the council wards of: Fallowfield, Gorton North, Gorton South, Levenshulme, Longsight and Rusholme. As a student, I identified with the Oxford Road corridor, which incorporated much of Manchester Central. Much of Rusholme was, in fact, in Moss Side ward (Manchester Central). Much of Fallowfield was, in fact, in Withington (Manchester Withington). I very occasionally saw people in Levenshulme. Gorton was a completely different place, both ethnically and in terms of student population (I also lived in Central and Withington at various times, the same basic pattern applied).
  • Leeds Central (Leeds), which contained various wards in central Leeds, most of which had little in common other than that they were in Leeds itself. Leeds North West, where I was agent in 2001, was even more disparate. Shaped like an ice cream cone, it included the student-heavy Headingly at one end and the rural villages of Otley and Wharfedale at the other.
  • Warwick and Leamington (Warwickshire): To the extent that this constituencies contained two distinct communities, I suppose it counts. But even then, it wasn’t entirely cut and dried, as at the time it also included half of Kenilworth.
  • Hendon (Barnet), which currently includes Burnt Oak, Colindale, Edgware, Hale, Hendon, Mill Hill and West Hendon. Again, most of these places might as well not exist as far as I’m concerned. I live in Mill Hill and own a flat in Colindale. I’ve been to Edgware once in my life and Hendon not much more frequently. Finchley and Golders Green, where I lived shortly prior to now, was also two extremely distinct communities (if not more).

Looking at all these constituencies, a pattern quickly forms. The size of the constituency is such that as far as local identification is concerned they are neither fish nor fowl. You DO get identifiable communities at a council ward level, you can even make a case for a community at local authority level (although in both cases there will always be issues around boundaries), but constituencies are typically at such a size that they should be regarded, at best, as collections of multiple communties. Indeed, within London the boundaries have got even stranger since this election, with numberous constituencies crossing local authority boundaries (I am technically a member of Lewisham and Beckenham North Liberal Democrats for example, and Hampstead and Kilburn is an aggregate of Brent and Camden wards).

Reduction and equalisation won’t change that. The tighter equalisation rules might, around the edges, cause a few more odd boundaries through the middle of towns and villages, but for the vast majority of constituents, their constituency will be the same impersonal lump it was before the change. Equally, there will no doubt be some areas that become more coherent as a result of the boundary changes. One of the advantages of STV is that by creating larger multi-member constituencies, each one would conceivably represent a more meaningful piece of geography such as a county or a borough, but that is another matter.

Of course there is also the fact that people’s sense of place differs wildly depending on their lifestyle. As a public transport user for example, my bit of North London is effectively Mill Hill, Finchley and, to a lesser extent, Golders Green – i.e. the bits which I go to frequently because of my daily commute. I can’t even get to Hendon directly by bus or tube. If I used a car, I would no doubt have a different perspective. “My” Manchester involved both sides of Oxford Road, from the centre out to Fallowfield – but that was because I was a student. As we all become more mobile and more culturally diverse, talk of constituencies needing to represent distinct communities becomes increasingly bunk. So to get precious about the constituency sizes we have now is frankly silly.

The final objection is that reducing the number of MPs would be bad for democracy, yet the House of Commons is unusually large by international standards. ERS have included a comparative table here. The conclusion they invite the reader to draw is that the UK doesn’t have a particularly oversized Parliament after all, but I’m not convinced. After all, the statistics do indeed show that the UK House of Commons is large by global standards.

For starters, the assertion that only countries with federal systems should have smaller Parliaments is a little dubious. Certainly, countries with legislative chambers at a sub-national level have fewer things for their legislatures to do, but it doesn’t follow that you therefore need more bodies to do it. MPs all have to vote on the same number of laws, no matter how many MPs there happen to be. And while, conceivably, more MPs means more people who can share the load in terms of scrutiny, in practice it doesn’t work that way.

For example, the Commons Select Committees have just been reduced in size from 18 members down to 11. Far from being about reducing the amount of scrutiny, this is actually about ensuring there is more. In the past, each select committee effectively consisted of a hardcore and a group of malleable part timers who would contribute very little and were more susceptable to influence from whips. Smaller committees are generally regarded as better in terms of building a consensus and doing the hard work.

I don’t have statistics, and would love to see them, but I would guess that public bill committees tend to be dominated by a bunch of usual suspects. Similarly, you either have an MP who reads things like papers on statutory instruments, or you don’t. It isn’t the number of MPs, or even the number of laws particularly that is the issue here, but the culture in Parliament that seems to reward citizens advice over and above legislating.

Either way, a reduction of MP numbers by 10% is unlikely to have much impact. A bigger reduction might do, for the simple fact that we have such a large payroll vote with our current system of government. But 10% is unlikely to have that much of an impact, and we should be reducing the payroll vote (if not seperating the legislature from the executive altogether) in any case. Another useful thing would be to increase the amount of research staff each party is entitled to employ, which would arguably do a far better job at ensuring there is more scrutiny than a handful of extra MPs at £100,000+ a throw.

Ultimately then, neither the “reduction” or the “equalise” part of these reforms are likely to make much of a difference, either to the political breakdown in the House of Commons or the nature of MP’s roles. Reforming the voting system to AV may be a modest reform, but compared to either of these tiny steps it is revolutionary. They are certainly a price I have no problem paying in order to keep the Tories happy (although it looks as if some backbenchers are determined to scupper the referendum bill in any case). What I find baffling is why Labour are claiming that some kind of massive point of principle is under threat here, when for the most part they are just totemic changes. Watching both parties scrap in this debate looks remarkably similar to two bald men fighting over a comb.