Other ways to improve democracy by picking your politicians at random

Obviously I’m a big fan of randomisation, and I think it would be interesting to use it in voting systems. The big manifestation of this is my 80%-serious endorsement of using random ballot for electing the house of commons.

I thought it might be interesting (well it’s interesting to me which means you, the reader, get to come along for the ride) to point out two and a half other interesting ways you can use randomization to improve your democracy. I think these are all both less obviously a major improvement and maybe more obviously good/politically tenable ideas than more perfect democracy proposal.

Sortition for the house of Lords

I really like having the house of Lords as a concept for keeping the commons in check. I’m not convinced it always works very well, but I have to admit my level of attentiveness is not high enough that I’m entirely sure that I would.

Know what I’m not a fan of? Its selection process. It’s a body whose election is mostly immune from the democratic process (there is sometimes voting involved, just not by us. Members are sometimes previously elected members of the commons, but we have no say in their appointment to the lords, etc). We’ve mostly removed the role of hereditary peerages from it, but we still have Archbishops of the Church of England getting seats. It’d all be charmingly quaint if it weren’t a major part of our governance.

You know what would be nice? If the body responsible for keeping the commons in check and preventing them from screwing us over were actually representative of our citizens instead of our aristocracy, a bunch of religious leaders and the buddies of the very people we want kept in check. This may be a wild idea, but I think it it might have legs.

And you know what a great way to make sure a large body of people is representative of the population is?


What I would propose in these circumstances is very much a true sortition. When what you’re interested in is a body of people whose primary purpose is oversight and review, I think the benefits of a hybrid system like random ballot are vastly outweighed by the benefits of having a fully representative sample of the population.

In all honesty, I can see very few downsides to doing it this way. One objection I have heard to this notion is that it is undemocratic – there is no popular oversight over the lords. Personally I think that this is a very narrow view of what democracy looks like, but fortunately it’s easy to fix so that we’re all happy: Candidates are elected by sortition, but you have semi-regular votes to evict and replace them with a new randomly selected candidate. I don’t think this is actually an improvement (it introduces more room for biasing the pool by e.g. being more likely to evict candidates who are not white), but I think the scope for causing problems is still much smaller than that of a normally elected system.

Random elections

So 2015 will be the first of our fixed-term general elections since the change in 2011 (which, full disclosure, I had either missed or completely forgotten about). Fixed terms are all very well – the previous system where the incumbent basically got to pick the timing of the general election was always a bit weird to me. Not necessarily wrong, it just feels a bit like stacking the deck.

I’m not totally sure I’m a fan of fixed terms either. It’s all a bit “ho hum we can do what we want EEK PANIC TIME DO POPULAR STUFF TO GET PEOPLE ON BOARD SO THEY RE-ELECT US”. This doesn’t seem like a healthy cycle to me.

It would be interesting if every year were potentially an election year. It would keep politicians on their toes and it would keep the rest of us more interested in politics.

So here is what I propose. Every year in a grand ceremony, parliament convenes and rolls the dice (metaphorically. You’d use a commitment scheme in practice). With one chance in four, a general election is triggered.

The result is that the expected term length remains the same as we currently have (5 years – you get one gratis and then are basically running a geometric distribution with expectation 4), but you can have unusually short terms or unusually long ones depending on the roll of the dice.

I think this idea has legs. The ceremony would be a nice spectator sport and the constant concern that they might have to stand for re-election keeps the government a bit in check. It possibly over-emphasises short-term populism, but I can’t say forcing people to make decisions that the population actually approves of sounds like a bad idea right now.

…with staggered elections

So I don’t know if they’d work with the house of commons (I think the idea has merit, but it also has problems), but I really like staggered elections like occur in e.g. the US Senate. Essentially the idea is that rather than have a giant general election in which every single MP is up for re-election, you space them out and re-elect a proportion of them over shorter periods. In the US Senate this is a third of them every other year. You could equally do a fifth of them every year in order to maintain the 5 year cycle.

The interesting thing that would happen in the UK parliament is that because the prime minister is whomever has the support of the majority of the house of commons, re-electing only a portion of the MPs could still trigger a change of government. I am undecided if this is a feature or a bug.

But setting aside the question of whether this is a good idea in the UK, it’s an excellent opportunity for randomization! We essentially implement the previous idea of having a big “Hey you’re up for re-election!” ceremony, only instead of doing it at the full blown national level we do it per MP.

The result is that each year about one MP in five will come up for re-election. Which? We’ve no idea! That’s the beauty of it! Every single MP is thus kept a bit more on their toes and is more susceptible to public scrutiny as a result. An expenses scandal in the year before your number comes up is a lot more stress inducing than one early in the general election cycle for example.

In Conclusion

Basically, I generally think there’s a lot you can do to use randomization to improve decision making, and I think voting is one of the most important instances of formalised decision making we regularly engage in. As such, when you elect me supreme dictator for life, I will definitely do my best to make your voting experience more random.

Or something like that.

This entry was posted in voting on by .

2 thoughts on “Other ways to improve democracy by picking your politicians at random

  1. JJ

    Hi, I’ve been reading your posts on voting which are really interesting and unusual. Thanks!
    A couple of questions/thoughts – and I mean them as honest contributions, not Flaws in Your Plan.

    Regarding the annual round of elections, I wonder if you’ve been thinking about the problem they have in the US with ‘perpetual campaigning’ – there’s a major election that’s going to CHANGE EVERYTHING almost every year in a lot of states, so politicians seem to spend all their time campaigning and very little actually governing – in particular, the selection process for the House starts almost as soon as the elections are over, so a new Representative is straight away having to negotiate the Party Faithful (who are even more insane there than here, but ours are bad enough for all mortal purposes).

    And whilst I’m commenting on things electoral, what do you think the effect of the system you propose in ‘Towards a More Perfect Democracy’ (which is a great idea) would be on the party system? I’d guess (and I am a psephology geek, but I see no further through a brick wall than the next human) you’d get a tremendous upswing in voting for minority parties and especially local independents, compared to any other system, because you’ve got the every-vote-counts thing as in PR, but also a very strong constituency link. In 2010 there were, if I recall rightly, about 30-40 seats with a serious Independent candidate polling somewhere up there with the main parties; the vast amount were standing either against an incumbent in a safe seat, or on a single local issue. In a system where those sorts of candidates have a decent chance of getting in, I’d expect there to be a lot more. Does the likelihood of getting a lot of local independents negate the ‘nationwide proportionality’ of the system? Not all single-issue independents are alike.

    There would also be really interesting consequences for the whole notion of political parties in such a system – no safe seats for the leadership, seriously limited ability to whip someone against their local interest for the good of the nation (so you could for example forget wind farms and HS2), and I don’t even begin to have an idea what would happen to the selection process of such parties as kept existing – the “Purity Now Purity Forever Down With The Traitors To The One True Way” ideologues would have a field day for a few years, but it’d be a great laboratory to find out whether (or rather, when and why) core-vote strategy really, truly works. I will link some people who are much more knowledgeable on the nuts and bolts than I to your posts…

    Of course, it’s never going to happen, because the reformist interests (who are going to get a tremendous filip in 2015 when someone winds up in Downing St on <35 and quite possibly <30% of the popular vote) already know what they want, and they probably can't get anything at all anyway, but it's great to speculate about, and maybe one day you will be supreme dictator (temporary) and able to try it for real.

  2. Pingback: Bashkimi Kombëtar, Plani B | Ekonomisti

Comments are closed.