Saturday, 16 May 2015

Non-Geographical Constituencies: Not as boring as it sounds...

Imagine, for a moment, that the Queen, her family, everyone in Westminster and every aspiring politician simultaneously gets tired of it all. A calm, peaceful and economically developed Britain suddenly finds its government packing up and going home, with nothing to replace it. What should we put in its place?

Putting together a political system is one of the more difficult, confusing and time-consuming jobs around, partly because it doesn't happen very often. It might take a long time to learn how to put together your first house, what with all the bits and pieces that have to fit together, but once you've built twenty of the things, you're probably going to be OK. Countries do, occasionally, need new constitutions: the Arab Spring produced a few, the world's newest country, South Sudan, needed one and there is no doubt that Scotland would've had to get writing in September last year if it had voted for independence. There are people who study these things very closely, even people who fly to new nations to help them put together a country, but the fact is that there are so many ways that a nation can be designed, and so few countries to test these methods in, that we have barely scratched the surface of possible political systems.

Imagine if we randomly balloted each year for a government, like conscription or jury duty: it would be fair, equal and probably quite efficient. What if we hired pollsters to see if the public agreed with certain policies and pass them if they did? It would be a lot cheaper than referenda and much quicker than having everyone actually discuss it. When we think about political systems we are, inevitably, constrained in both our options and our imagination by the things that have come before. When we talk of powers, executives, legislatures and the like, we are only rearranging the ideas that went before into new and interesting combinations. Every system has an unimaginable number of possible alterations and the costs of trying out something unsuccessful are huge, but that doesn't mean we shouldn't try imagination. There is no doubt that every political system is far from perfect and that most are ludicrously biased towards the status quo -  think Britain voting down Alternative Voting, or the U.S. continuing to use the electoral college. There are always ways that we can improve the democratic process: sometimes these ideas come from looking at other countries, but sometimes they are untried, untested and grounded entirely in the politics of ideals.

In this post, I intend to outline what I believe is an interesting and important way of re-imagining representative democracy. To start, I want to return to the world in which Britain is suddenly without a government. The first questions would be about decisions and power: how do we make decisions as a society? Who should we give power to? How should we monitor and control that power? Someone will no doubt suggest that we begin with constituencies, arguing something like this:

"There's no space or time for us all to make decisions together. If we split up into constituencies and elect people who we think will represent us well, we can save a lot of time while getting decisions that are just as good (if not better). Every few years they can come back to us and seek re-election, at which point we can either vote them back in or kick them out, depending on how well they did. This means that, not only can we get rid of bad representatives, every representative will have an incentive to do their best."

"Sounds good!", says someone else, "So these representatives go on to decide not only what the law of the land is, but who forms government, who is given power and how laws are implemented. Why do there have to be constituencies for that though? Couldn't we just all put a cross next to the party that best represents us then give them seats proportionally to their votes?"

"Hmm, that is a good point, but people often have specific views and grievances that are localised. Special interests matter and everywhere is different, I want to be able to have someone who is accountable to me and people like me, as well as a say in who forms the government in general."

"Great! Well given that I've now given up on Proportional Representation, probably so I can discuss it in another post some time, I guess we should get going with constituencies. I suppose if you wanted them to be more proportional you could have two or three members per area, instead of one. So first things first, let's make Oxford a constituency and see how many MPs we want it to have..."

At some point, without anyone really noticing, we all decided that the only legitimate way of dividing up a country for political representation was physical geography. Of course it makes sense to have MPs from the north of England! People there have very different concerns, beliefs and values from the south. If we just had PR, their voice would be drowned out by all the Oxbridge/Westminster elite. Surely if you're going to have nationalist parties, like the SNP, you're going to want them to be able to stand specifically in Scottish seats (or whatever area they represent). Nationalism is a legitimate political desire, and if we have a system that doesn't adequately express it, then the system isn't doing very well. As it happens, the system is doing an incredibly good job of representing geographically concentrated concerns, not just in Britain but pretty much everywhere in the world. Rural parties, regional parties and nationalist parties all make strong showings in developed democracies, and the reason is this: we divide our constituencies on the basis of geography.

What other option do we have, though? Answering this question is easier if you think why we're so happy about local communities having representatives in the first place. The people of Wales do have significant political differences from those around them and, in many systems, those differences could easily be drowned out. Who else will stick up for a community stricken by flooding? Who will fight to protect an industry that is keeping a town alive? What happens when farmers and country-dwellers are ignored by every party? Special interests and under-represented communities matter and giving them a voice is one of the most important things politics can do. With that in mind, I'm sure anyone can think of hundreds of important political divisions where a group of people might not be heard: age, gender, class, race, sexuality, industry, disability. When you do get parties that represent people along boundaries other than geography, they gain their strength when certain groups happen to be concentrated in different areas. Northern cities have always been a stronghold for Labour because they are traditionally working class areas. People who like low taxes, dislike gay marriage and enjoy hunting foxes also happen to cluster together in the leafy green hills of the rural South-East. What if, instead, we could provide an option to vote along those divisions directly? How would that look?

Non-geographical constituencies

Here we reach the point where the thrust of my argument should be relatively clear, but understanding the technicalities of the system is more complex. To keep things simple, we can add in one new type of constituency: one based on age. People turning up to vote on election day are handed two ballot papers: one for the area they live in, one based on how old they are. To start with, imagine two age brackets across the whole of the UK: above 40 and below 40. There will be, say, 10 seats set aside for age-based voting, which are distributed to the different constituencies based on how many voters are in each. If 60% of people are below 40, then 6 seats will be allocated to the Young constituency and 4 seats will be allocated to the Old constituency. In the ballot box, people will put a cross next to their preferred MP, just like now (or rank them like in Australia), and then rank their preferred candidates for their age-based MPs. The votes will be shipped somewhere in the middle of the UK and counted up, before the votes are announced. Seats will then be distributed to different parties and candidates based on the votes (through whatever system you like) and now those 10 new MPs will take their seats in Parliament, bringing the number up to 660.

It is important not to get confused by the numbers or exact systems I used, all of which were selected to illustrate the more general point. It could be that there is only one seat per age-based constituency. Age-based constituencies could be sub-divided by region (the MP for South-East 20-25 year olds). The MPs could be elected by any form of voting within their constituency. The total number of MPs can be whatever you like it to be and how they balance with the pre-existing geographical MPs is totally up to you. The real core concept is that, when you go into the polling station (having registered beforehand) you now get two ballot papers based on two pieces of information about you: your residence and your age.

If you've understood everything so far, it should be clear that the number of possible constituency-types (and by extension ballot papers) is pretty unlimited, and can come in any combination. You could, for example, stand to be the MP for LGBTQ Welsh 25-30 year olds, but that is, sadly, unlikely to ever be the case. Exactly how a system like this might work is left for another post, this instead serves as an explanation of concept and clarifying some justification and implications.

Doesn't this benefit some people more? Will it still be equal?

No and yes. As in definite answer to each respectively, not an ambiguous answer to both. One of the most important principles in any democracy is the idea that, while we might have different powers and responsibilities at other times in our life, everyone's vote has the same weight. This does not change that. Everyone will be registered in a geographical constituency (like they are now) and everyone will have an age (like they do now), so will obviously be registered to some age-constituency. Everyone, then, has two votes, but you have exactly the same influence on the overall make-up of Parliament as before. One way of thinking about it is that where you once had one vote, exclusively in one constituency, you now have two half-votes which you put into two different constituencies. This would be like if students were half registered at university and half registered at home. They'd have no more of a vote than anyone else but it would be a better reflection of where their concerns lie.

What other categories could it be used for?

As I mentioned above, the list is pretty much endless, and every time you come up with a new category, it can be combined with other categories to make yet another constituency. That said, the more serious and practical ones (in my opinion) are gender, ethnicity and (perhaps) socio-economic background in some way. These are things that can change over the course of a life: racial and gender identity are fluid and people can change their economic circumstances, but they are nonetheless important and relatively consistent parts of people's identities and, by extension, their political beliefs and values. A politician saying "I will stand up for Durham" is just as legitimate as "I will stand up for people of colour", but the political pay-offs for those statements are very different. Incorporating important forms of identity as and when it is appropriate would help to correct that imbalance.

Preventing things like "tactical identification" without denying people their strongly held identities is obviously a difficult task, but it's not impossible. People, if they really wanted, could move to constituencies where their vote counted for more, but it's definitely not something we worry about too much. Any major constitutional change brings with it technicalities, but I do not believe they are enough to sink this idea.

Won't we have worse politicians?

A common thing that people bring up when I mention this idea is the notion that we will have single-issue candidates with no idea about "serious" issues like defence, budgets and the like. While there is a lot to say in this (which I will say another time), it is worth making the point that having people who explicitly represent areas doesn't seem to prevent them from adequately caring about the whole country. People expect a good representative to do their job for the country well, since voters care about the world beyond their constituency. There's no reason to believe that this wouldn't be the case for different kinds of constituencies. Young voters would care about issues specific to them, but also about fiscal policy, foreign policy and responsible government. Black voters would want someone to address racial inequality, but also someone who could do their job competently and fight for foreign aid and medicare.

So it's not necessarily worse, but why bother?

Maybe, in practice, non-geographical constituencies would make next to no difference to the political process. Maybe it would result in a proliferation of single-issue parties or maybe the new representatives would have had no idea what they were doing. That said, making sure that political differences across communities are heard should be central to the design of our political system. When people's influence over the electoral and deliberative process of government is highly imbalanced, favouring nationalist and local concerns over those of more disparate communities, we live in a less democratic society. In the modern world, communities are far more complex and far less stationary than they used to be. The internet, travel and the modern media make running a campaign across the UK (or any large area) far more feasible than it used to be. How can we adequately represent young people? How can we represent people who don't stay in one place?

To return to the original problem (building a constitution from the ground up), I hope what I have made clear is that there is no reason that physical space should be the only we way we think about politics. It is important, yes, probably more important than any other political division, but the only one worth incorporating into the rules of society? Probably not. The idea is utopian, unlikely and quite possibly totally unfounded (as I hope readers will point out), but the first step along the road to building a better society is improving the way we make change. That means a more democratic world, where we actively seek out different voices and ask people to speak out. However we accomplish that goal, the sooner we do it, the better.


  1. NZ have a "Maori Constituency" and that works pretty well. People can move in and out of it every few years, but that is no longer a technical requirement so much as a legacy.

    I'd be interested to see how that worked with multiple constituencies per person and a fairly long lead time for changes to prevent strategic flipping. Viz, I can choose to be part of the "ecofascist", "guaranteed basic income" and "pro-life" constituencies, and can only change once per three years with a one year delay after I elect to change. That might cause a problem with short careers as the politicians on the margins of their constituency list pop in and out of parliament. Possibly synchronise things, everyone gets to choose every 3 years but at the same time. Election day :)

  2. That idea, BTW, depends on proportional representation. Otherwise building multiple parties within each constituency becomes difficult. But with PR you just get seats in proportion to your voters... with a single vote per voter it's effectively single electorate pure PR as used in Israel (but hopefully with more than 30% of the population allowed to vote). It'd be amusing watching the coalition building that resulted from that - I can imagine a plethora of single MP constituencies making negotiation very difficult, especially for the former "major parties" who are used to being able to whip their members to achieve antidemocratic outcomes.

    Allowing multiple votes and possibly requiring them to be spread opens things up. Spreading is tricky, it encourages "liberal party too" type gaming of the system. But it's probably not required, I suspect most people would split vote (they do in NZ with MMP, where almost all Australians party vote rather than referencing)

    I'm in Australia where the problem is a constitution specifically designed to make change difficult, but as compensation we have an effective, independent electoral authority. The UK very much needs something like that purely to address the gerrymandering. Unfortunately I think that's as likely as getting the Australians to vote for a major change to the electoral system (referendum, requiring a majority in federal parliament(s), a majority of voters and a majority of voters in a majority of states... the "triple majority")