The turnout massacre: what went wrong with the Lib Dem internal elections?

My first reaction on seeing the Liberal Democrat federal committee election results was one of puzzlement. Surely I couldn’t be reading the results correctly, because those numbers looked much too small to be people’s first preference totals? But no, they were indeed.

What had happened was that the number of votes cast in the elections fell off a cliff.

Take the number of valid votes for Federal Policy Committee:

2004: 1,158
2006: 935
2008: 1,208
2010: 1,669
2012: 1,204
2014: 656

Two things stand out from these numbers. First, how exceptional the 2010 number of votes was – driven most likely by the party being newly in government and so the importance of committee elections suddenly seeming that much greater.

Second, just how low the 2014 figure is. Unfortunately, the party has never published official turnout figures, but comparing 2012 and 2014 remember that the party’s membership has gone up since then (31 December 2012: 42,501, 30 September 2014: 44,526). It’s worth emphasising the “up” bit as one myth-in-the-making is that the lower vote totals reflect a drop in party membership.

UPDATE: In the comments thread, Martin Tod has added turnout figures circulated within the party previously:
Lib Dem committee turnout

The electorate is the voting reps elected by each local party, which uses a sliding scale based on membership, so although membership totals don’t reveal exact electorate, the two go roughly in step. Moreover, in Spring 2014 that sliding scale was changed to make it significantly more generous (e.g. a local party with a membership of 30-50 went from 2 to 8 reps).

One sort-of electorate number was published in October during the party debates over moving to OMOV: “Of our total membership, just 5% are appointed as voting representatives”. 5% of 44,526 is 2,226 which would imply a turnout of around 29%.

But even that figure is probably too high as that 2,226 is the number recorded in the party’s records, not the theoretical maximum if every local party had got a full slate of conference reps recorded. This latter point will be overtaken by events given that Glasgow conference voted for the principle of introducing OMOV. Yet that shouldn’t blind us to how big the fall was in turnout in 2014.

Something else went wrong.

Most likely, it was the further move to online voting. This time those for whom the party doesn’t have a working email address didn’t get a posted mailing with a ballot paper. They got a letter telling them how to vote online and with an option to get in touch to make alternative arrangements if online voting wasn’t viable for them.

It’s highly plausible that sending a ballot mailing without a ballot paper hit turnout, exacerbated by the problems with the despatch of both the ballot mailing and the online voting emails.

Not only did they go out late, requiring the closing date to be put back, but also there are numerous reports from activists on social media of problems not receiving their mailing, even after multiple requests. One member even said it took them five times of asking before they were able to vote and stories of two or three attempts before getting the chance to vote are common.

Will the move to OMOV see a similar turnout massacre?

Delving into the evidence in more detail than I have done to get to the truth of the matter needs to be done, else the party’s move to OMOV risks being done in a way that also massacres turnout.

The new Federal Executive and Party President need to find out the answers – and then solutions, especially if the explanations involving overworked staff short of time – to:

  1. What were the actual turnout figures this time and in previous years – and why was there such a big fall it appears this time?
  2. Who decided the change in the process for posted ballots, and why wasn’t the FE involved? (FE members I have spoke to can’t recall this change being reported to them let alone being put to them for decision.)
  3. What caused the huge logistical problems that forced the delays to party elections?*
  4. How good is the data in the party’s new Salesforce-based membership records? (Faulty data could have depressed the ‘registered electorate’, i.e. number of recorded voting reps, and/or messed up the distribution of the mailings.)
  5. How efficiently were complaints about missing ballots handled?
  6. Why were candidates not kept informed about the delays, the revised closing date or the revised results day?
  7. In public elections Liberal Democrats rightly value transparency in election counts, with candidates and the press both having different rights of access, in order to give everyone confidence in the system. With the changing dates and lack of information, what transparency was there this time? (This isn’t a hypothetical question; the stories I can tell of party election counts of the past are for another day but show that transparency helps everyone as it guards against innocent mistakes as well.)

Presidential candidates have differing fates

Two Presidential candidates – one now withdrawn – featured in the results. Pauline Pearce, who pulled out over the summer, was comfortably elected to both the Federal Conference Committee and the Federal Executive. Given her vocal comments about the party’s need to do better on diversity, the latter is particularly good news.

Less good I suspect for Liz Lynne is her re-election to Federal Conference Committee. It was a decent solid performance by her, but given all the extra publicity she’s got through the President contest, I suspect not doing better in the FCC elections is a sign that she won’t shine in the Presidential ballot.

Churn ahoy!

Meanwhile, it’s worth noting the high level of churn in the committee results.

  • Of the 12 FCC members elected in 2012,  none were defeated, 5 did not stand again (Blundell, Brinton, Gaszczak, Harris, Rendel) and 7 re-elected (Epps, Gidley, Lynne, Maines, McGuiness, Tilsley and Wiseman) – a churn rate of 42%
  • Of the 15 FE members elected in 2012, 4 were defeated (Afzal, Gallagher, Rendel, Williams), 4 did not stand again (Bagshaw, Cooper, Shaw, Vernon-Jackson) and 7 re-elected (Dewan, Doughty, Gurling, House, Lindsay, Lishman, Tod) – a churn rate of 53%
  • Of the 15 FPC members elected in 2012, none were defeated, 8 did not stand again (Bastone, Brinton, Care, Church, Greaves, Rennard, Wallace and Willis) and 7 re-elected (Blundell, Brack, Buch, Epps, Harris, Pack, Smith) – a churn rate of 53%

I’ve not done the churn calculations for previous years and I believe no-one else has, but I think these are in line with previous years. Certainly the rate of churn I’ve noticed in previous results has always been much higher than the ‘same old faces always get elected’ trope that you often hear in the party. Courtesy of retirements much more so than defeat, they don’t.

The more rigorous data this year confirms that it’s a myth – and a dangerous one if it puts people off standing.

* There were two problems with delays. First, the ballots went out late. Second, on the day of the Presidential election count the verification stage took much longer than expected, delaying the result by several hours. This was due to a discrepancy of 67 votes thanks to different weight paper having been used to print different ballot papers, and hence weighing them to count them did not give accurate results.

Leave a Reply

Your email address will not be published. Required fields are marked *

All comments and data you submit with them will be handled in line with the privacy and moderation policies.