Electing the whatever-we-call-the-Lords

Because I’m a sad geek who spends far too much time thinking about electoral systems, since last Wednesday I’ve been pondering the question of how to elect members of the Upper House of Parliament. So, assuming that the role of the Upper House is primarily one of review, scrutiny and amendment rather than a primary legislating chamber, here’s my suggestion.

First off, I like the idea that’s been suggested from several quarters of electing by thirds at five year intervals (probably at the same time as European Parliament elections) for a single fifteen-year term. That provides a chamber that’s not going to be as affected by short-term swings in electoral fortune, and single terms mean members will have a degree of independence because they can’t be concerned about their personal re-election.

As for size, I think around 300 members would be sufficient – the total workload of the Upper House would be less than the Commons and restricting the number of members should mean that they can concentrate on the tasks they have at hand, rather than having to make work for themselves. Given that members of the Upper House wouldn’t have the constituency commitments of an MP – especially as they don’t have to seek re-election – there’s even the possibility for it to be seen as a part-time commitment.

So, how to elect them? Well, given that we’re following the electoral cycle of the European Parliament, it seems simplest to use the same regional arrangements as those elections, with each of the 12 regions (Scotland, Wales, Northern Ireland and the 9 regions of England) electing an average of around 25 members each, meaning 100 Lords/Senators/Members Of The Upper House get elected each time. I’d also suggest that the allocation of seats to be regions be on a somewhat degressive, rather than strict, proportionality with the smallest regions having 18 members (electing 6 each time) and the largest 30 (10 each time) thus ensuring a balance of voices from throughout the UK.

When it comes to electing them, I would suggest – as anyone who knows me may have guessed by now – that a Single Transferable Vote system is used. However, there is a reason for this beyond my belief that it’s the best voting system, which is that it’s a voting system that encourages candidates to prove themselves as individuals not just as party functionaries, as well as allowing independent (cross-bencher) candidates to compete on a more even playing field. Candidates would actually have to work to get the voters to choose them from amongst their party’s candidates, not just hope to get a high enough place on a list, and voters would have the power to mix and match from amongst party and independent candidates. To go further, one suggestion I would make is that while candidates would be free to campaign as party representatives, party affiliation would not be listed on the ballot paper, making it near-impossible to hide behind the rosette.

Anyway, that’s just my thoughts, feel free to tear them apart in the comments.

5 thoughts on “Electing the whatever-we-call-the-Lords”

  1. I’m not going to rip it apart, it sounds entirely sensible to me.

    I hadn’t thought of using the European regions, even though they’re basically arbitrary, they are ready made constituencies and are a good size.

    I do wonder whether we even need that many peers, 100 would be fine as I think we actually have too many legislators at the moment.

    Of course, if you only have 100 it makes the smaller parties less likely to be represented which is probably a disadvantage from a representative democracy point of view.

  2. In Oztraya we already use several (minor) variants of STV, and combine that with proportional voting for the upper houses. The main thing to be aware of is that many people will be confused by it. It’s been used here for a while, and the donkey vote is still ~5% (people who put a 1 in the first box they see). I suspect that you will need the “above the line” option that makes this possible – essentially letting people just vote for their preferred party and the party controls their preferences. So you need to have some sort of regulation around party preference trading (even if just requiring parties to declare their preferences before the election takes place). One thing that STV does encourage is “fake” parties set up to collect votes from the naive, which at times leads to insanely large ballot sheets (over 150 parties at times). This pays off for the parties by splitting the vote – your mainstream party sets up a bunch of shells with attractive names and relies on people preferencing them out of ignorance.

    I still prefer this over simple proportional voting, but it does take a certain amount of intelligence and willingness to think about what you’re doing. Many people are not that interested… so Australia makes voting mandatory.

    One serious mistake Australia has is that there is no “intention of the voter” test, so there are a lot of “informal” (invalid) votes that are not counted because the voter fails to obsessively obey the instructions. Adding a smiley face to an otherwise valid ballot paper, for instance, makes your vote void. This is not very nice, IMO. If you have the intention test, then you open it up a lot. Sure, some people will stretch it by writing entertaining things, but you also open things up a lot for people who struggle a little with the details. And it also means that it’s possible to vote below the line on a tablecloth ballot (where a single error in 150+ numbers can render your vote void).

    Oh, and allowing people to exhaust their preferences is also very useful. Especially when there’s lots of parties. Just being able to number the parties you do care about 1..X, then after that your vote dies is great. It avoids the whole “do I number the ‘baby-eating monsters party’ higher than the ‘neofascist lackeys of Cthulu’?” question. That slightly reduces the chances of the Australian experience where surprising parties get up simply because preferences flow in unexpected ways (“the shooters party” have an MP despite getting ~1.5% primary vote in NSW, for instance. Not bad when it came through a block of 5 MPs in that district IIRC).

  3. Wouldn’t having constituencies electing 10 members at a time result in stupidly long ballot papers, match that with a lack of party symbols and slow readers like me would be in the booth for hours looking for the candidate we wanted to give our ninth preference to.

    Probably best to cap the maximum number of members to be elected to 6 at a time (that still gives a pretty proportional result right?) and split the regions, which outside of London and the North East make little sense to anybody in England anyway. Its not as if Boundary Commission (and geeks everywhere)wouldn’t enjoy the opportunity to draw lines on maps.

Comments are closed.