No to AV

Tomorrow sees the first referendum in the UK for over 35 years and the decision at hand is whether to change the voting system for Westminster elections from the traditional first past the post system to the alternative vote (AV). I’ve written this post as I wanted to share my reasoning for deciding to vote no in the poll.

Firstly let me stress that this referendum is a massively important decision, one of the most important decisions that is to be made about the UK political system for over a generation so please don’t underestimate the significance of casting your vote! Whether you agree with my points or not, you should be sure make your voice heard on May 5th – set aside a time of day to go to the polls and cast your ballot.

I was first introduced to the alternative vote back in 2003 when I started my studies at the University of York and decided to run for election to the Halifax College Students Association (as it is now known). As I became engrossed in campaigning and the fervour of the election trail I came to realise that AV has many flaws, interestingly some of which are not being actively mentioned by the current No to AV campaign.

Initially I noted that the primary flaw seemed to be the time taken to count. In first past the post each ballot paper is required to be reviewed only once, whereas with AV you review some ballot papers a number of times; if the top candidate has less than half of the votes then you review all the papers for the candidate with the lowest number of first place votes to decide if to discard the paper or add the vote to one of the remaining candidates. This could, in theory, continue until almost half of the ballots have been reviewed at least twice. Despite what the yes campaign says there is no way that this can possibly counted as fast as first past the post with standard resource allocation. If they wanted to compete on speed then they’d have to have more counters or resort to electronic voting. For reasons of cost and reliability neither of these two options seem palatable, even to many in the yes campaign. The final nail on this one of course is that if we just accept it will take longer, keeping the count manual and the number of counters the same, our “morning after” results that we currently enjoy (and the excitement of election night) may well be a thing of the past. As someone who has enjoyed election night even since before I was able to vote this is not something I’m willing to let go easily.

It’s probably possible to infer what my next gripe was based on the above, the increased likelihood of counting mistakes. While I’m sure all counts are and would be conducted to a high standard, the vote re-distribution factor in AV along with the increased number of ballot paper options increases the probability that a good counter will make accidental mistakes. Given the counting time factor, if the vote is close, called into question and a recount requested (as is sometimes the case for certain marginal seats) this process could take longer. Don’t get me wrong, I’m a quality over quantity person but I also subscribe to the Keep It Simple Stupid (KISS) principle; if something can be done simply with less chance of error than that is usually the best approach unless there are compelling reasons not to. This of course applies equally to the voter making their choices on the ballot paper although given how long polls are open for during the day any slight time increase in voting time is unlikely to cause any problems.

My final and certainly biggest issue with AV (indeed my decision really rides on this issue) is the political spectrum conundrum. Many of the worked examples you’ll find online for how AV works show a scenario in which there are several similar candidates politically with small but defining differences between them and one politically different candidate. In this situation AV outperforms first past the post in fairness as those placing the several similar candidates in ranked order will ensure one of them gets elected over the politically differing candidate thereby ensuring the political wishes of the majority are catered for. This however assumes a certain polarity of political spectrum in each constituency and in the UK we simply do not have that. Most common is the 3 party ballot in which most votes are shared between them. In AV, the temptation would be to rank amongst these three but politically this makes little sense as if you drew a venn diagram of how policies overlapped you’d find little in common. The result of this is that voters are tempted to mark on the ballot against more than one candidate under the perception of increased choice when in fact doing so is increasing the chances that a party that does not sit well with their true political opinions being elected. This also increases the chances to engage in tactical voting. At present tactical voting goes on but based on past results and opinion polls as you can only vote for one candidate. With AV you can deliberately set out to disadvantage one candidate over others by simply ranking all the others and leaving the last candidate off the ballot. Clearly not everyone will do this but something that even has the potential to encourage people to focus on voting out a candidate rather than voting one in somewhat defeats the objective of polling political opinion which is, after all, what a general election is really about.

Having said all of that, ultimately everyone must make their own decision and I’ve no doubt that many will disagree with my views and my rationale behind them. Whatever you believe though be sure to do one thing; as I said at the start, go out and vote. To stay at home is to let others decide for you and as always when it comes to the decisions others make on your behalf, you might not like what they choose.

   

3 Comments

  1. Colonel Jessop's Boss Said,

    May 5, 2011 @ 2:17 pm

    Blimey!! How long did it take you to write this last night?? No wonder you complain of paucity of sleep :P

    In the light of the above write up, I think the fact that voters might fail to express themselves effectively if they do not understand how AV works is worth underscoring- eg : some voters may think expressing a second preference will jeopradise their favourite candidate’s chance of winning etc- but this is wrong lower preferences are not counted as long as the first preference candidate remains in the race. Instead one can argue about voters wasting their vote on a candidate who cannot win. Having said that AV is designed to elect a candidate with widest support. Eg: One right wing candidate and 2 left wing candidates, under FPTP, the right wing candidate will win even though majority of the voters are left- because of the a,,b,c vote divide in 3 candidates. In AV, the voters would flock for their favourite left candidate.
    Lastly, touching upon the point of political manifestos (from last night)- AV could result in bipolar multipartism- 2 stable blocks and therefore stable coalitions with less conflicting agendas.

  2. Joe Hames Said,

    May 10, 2011 @ 12:45 pm

    “Yes” was my second choice.

  3. Norman Said,

    May 12, 2011 @ 12:40 pm

    Your arguments are strong, although I think the counting error and counting time ones are fairly trivial as the error rate in counting the votes is normally very low, bordering on statistically insignificant. Unless there were 5 or 6 candidates, all with similar shares of the votes the cumulative counting error would remain to low to be significant. In marginal seats where recounts are common, those responsible for counts tend to be more aware of the likelihood and can accommodate more staff accordingly. Most returning officers in marginal seats know they are in marginal seats and manage the situation very well.

    Your third point is, as you say, more significant. However, let us not be naive enough to think that under the current system people vote candidates in. It is a matter of course that in many seats voters actively vote for a candidate they don’t want in order to prevent a candidate they want less out.

    I actually voted yes in the referendum based on two points. First, the actual impact of introducing AV would be limited to around 25% of seats that are marginal. The majority of the parliamentary seats would remain safe and would probably only ever need one count. This means that some of the risk of your third point is mitigated. The second reason was the fact that the No campaign didn’t use any real arguments. With such a close decision there needed to be a really compelling reason either way to make up your mind. The Yes campaign was ever so slightly more persuasive (or rather less made up of lies and innuendo).

RSS feed for comments on this post