Were the Postal Votes really up by 19%?

The ‘Herald Scotland’ has reported concerns over Conservative links to an election count firm involved in postal voting.

Yet the irony seems to be that the problem in England and Wales is even worse.

As AC Grayling suggests:

Mere speculation, but: dodgy postal vote spike, dodgy Tory funding, Kuenssberg breaking electoral rules – what’s the chance of the election being voided, hung Parliament returning & this time getting it right: a government of national unity and a second-thoughts EU referendum?

It seems that the postal vote was 37%. In previous elections, the postal vote never got out of the teens… The postal vote was, in many areas, ‘managed’ by a private company that donated to the Conservative & Unionist Party. Private company Idox, who till 2018 had a Tory peer on its board, used a wholly owned subsidiary Halarose, to manage Postal votes in 12 Dec 2019 and yet Halarose is already listed on or before 19th December 2019 as being ‘dissolved on 24 Dec 2019’. This seems too convenient.

While even the Telegraph has reported that postal voting fraud is easy, yet surely 12 .2 million postal votes is absolutely remarkable – and yet the Conservatives remain still worried about the non existant actual crime of physical impersonation at the polling station.

In Scotland (where one might expect rather more difficult winter weather- easily justifying an enhanced postal vote) the postal vote went up by 1.2%, whereas in England the rise was, I understand, though I’m clear I cannot find a link for it, 19%. If that is true, something is not quite correct – never mind a lot of anecdotal, physical queues outside polling stations.

I’m not generally a conspiracy theorist but a 19% increase in postal voting against a 1.2% in Scotland is – shall we say – most unusual.

I shall certainly be writing to my MP, Ben Bradshaw, who at least has form on trying to get the ‘vote leave’ campaign properly and suitably promptly investigated to try to establish if there may be substance to this and more than meets the eye…

(Anyone with further evidence should contact the Electoral Commission here.)


  1. Kenmath -

    Serious questions were raised about tampering with postal votes in the Scottish Indyref of 2014: https://t.co/Vm7PUmBPL8

    There’s also a wealth articles online following Judge Richard Mawrey’s scathing 2014 critique of the UK Postal Voting system, so there’s a considerable weight of expert opinion that the system is vulnerable. It’s therefore worrying that management of it is entrusted to a commercial third-party. To carry out that management, the company must be given access to the metadata contained in the Voters’ Rolls maintained by local authorities. If this is done without surveillance and checking by the local authorities’ Electoral Registrars and/or the Electoral Commission, it’s effectively an open goal for interference.

    It’s not clear whether there has been tampering or not, but that suggests that the necessary surveillance/checking may not have been undertaken. Prevention of fraud is clearly preferable to suffering actual fraud.

    Another issue in 2014 Inyref was that Ruth Davidson stated on air that she had seen postal ballot papers in the days before the count and that what she saw had given her great confidence in the outcome. That similar statements were made by Laura Kuenssberg in the 2019 General Election suggests that control and security over postal votes pre-count is lax, quite aside from the breach of the law by Davidson & Kuenssberg (both, so far, unpunished).

    1. Peter May -

      Thanks – and the electoral commission say that postal ballot irregularities are a matter for the police, who won’t prosecute owing to lack of evidence -an admission on television doesn’t seem to count….

  2. Kim Sanders-Fisher -

    I have been in touch with the Electoral Commission multiple times since the vote asking probing questions. So far they are denying that they even have a master list of which constituencies outsource to which private vote management companies including Idoc.. That is gaping hole number one. Even if the did know there appears to be zero oversight.

    A bunch of us are trying to piece together the evidence and try to identify the worst anomalies in posts on a discussion forum on Craig Murray’s blog under Elections Aftermath. Please join us if you have any input.

  3. J -

    Hi Peter,

    Can you tell me how that figure of 1.2% increase in the Scottish postal vote was arrived at? I presume by comparing the postal ballot figures from the Ashcroft poll to GE2017?

    1. Peter May -

      As far as I’m aware it is the same Ashcroft poll that suggested the large increase in England. The Electoral Commission have said they do not yet have their own figures – goodness knows why….

  4. atmosphere -

    This 37%/38% postal vote is something of a red herring – they were apparently sourced from a Lord Ashcroft Exit Poll conducted on Election Day. https://lordashcroftpolls.com/2019/12/how-britain-voted-and-why-my-2019-general-election-post-vote-poll/

    The only data I’ve seen with regards to English postal votes is this article concerning Waltham Forest Council published Thu, 12 Dec 2019 15:19:06 +0000 – https://www.guardian-series.co.uk/news/18097920.waltham-forest-council-scrambles-hand-delayed-postal-votes-ahead-general-election/. They claim to have ‘dealt’ with 27,993 postal votes – whether this is in addition to or inclusive of the (very) last-minute deliveries of 1,470 postal ballots to voters, I’m not sure. If you compare those 28-odd thousand postal votes with the number of postal votes issued in 2017 for the constituencies of Chingford and Woodford Green (10,595), Leyton and Wanstead (10,203) and Walthamstow (9,016) then the levels of postal votes are going to be about the same between this election and the last (29,814).

    According to the specialist publication printweek – serving those undoubtedly involved in the production and delivery of postal ballot packs, poll cards and party literature – over 50% of their readers / market are Conservative voters, another 18% each were prepared to vote for the Brexit Party or Liberal Democrats, whilst only a mere 8% were potential Labour voters.

    Lord Ashcroft’s Poll isn’t totally uninteresting, however. If you download the linked data-set https://lordashcroftpolls.com/wp-content/uploads/2019/12/Lord-Ashcroft-Polls-GE-2019-post-vote-poll-Full-tables.xlsx. They don’t appear to have captured non-voters for this election but they did survey 562 people who were non-voters in the 2017 election and 630 non-voters in the 2016 EU Referendum.

    You can see that the rates of postal voting are roughly similar between people who voted for any party in 2019 and for any party in 2017. Regardless you can compare the level of postal voting captured in this poll with the official rates given by the Electoral Commission for 2017.

    Could do with a statistician / mathematician to go through this – as it doesn’t count the spoilt / defective ballots – but if you reduce the percentage points for each party in increments of five then run those percentages off against the number of voters for each respective party in 2017 – then you can determine how many percentage points you should reduce the 2019 postal vote rates by.

    So if you reduce the percentages across the board by 15 pts then you end up with roughly 7,468,269.85 valid postal votes, if by 17 pts then 6,824,186.17, if by 20 pts then 5,858,060.65 valid postal votes – the middle figure comparing favorably with 6,986,581 valid postal votes returned (7,155,315 postal ballots returned in total). Reducing the 2019 percentages by 17 pts yields a return of roughly 6,791,216.53 valid postal votes – mirroring a reduced voting pool despite the increased electorate.

    If you return to the non-voting columns for 2017 and 2016 then presumably percentages of 30% and 31% respectively are for voters registered to vote by post be declined to do so. So then we go to the non-voting but eligible population of 14,558,161 – remembering that most voters wouldn’t know whether their ballot had been ruled invalid or not – if we try to arrive at the number of postal voters who didn’t vote by running the percentage quoted in Lord Ashcroft’s Poll and reducing it in increments then we end up with the following figures:

    30% | 4,367,448.3
    20% | 2,911,632.2
    15% | 2,183,724.15
    13% | 1,892,560.93
    10% | 1,455,816.1

    According to official figures 1,256,745 postal voters didn’t return their ballot before the close of poll, and a further 29,760 ballots were received within 25 days after closing.

    If we subtract the that figure of 1,256,745 from the above numbers we end up with the following excess.

    30% | 4,367,448 | 3,110,703.3
    20% | 2,911,632 | 1,654,887.2
    15% | 2,183,724 | 926,979.15
    13% | 1,892,560 | 635,815.93
    10% | 1,455,816 | 199,071.1

    So theoretically between 600,000 and 900,000 votes were cast by post in the names of those registered to do so but were unaware of. The Conservatives increased their share this time around by a little under 330,000 votes.

    In fact official figures predict out of a non-voting population of 14,558,161 only 8.63% would have been registered to vote by post.

    There’s some discrepancies in the Electoral Commission’s data for 2017 as I don’t know why these two sets of figures aren’t equal

    4. Total number of postal ballot papers issued (1 to 3)


    14. Total numbers 6 to 13 (this should be the same as that in 4 above)


    6. Number of covering envelopes received by the Returning Officer or at a polling station before the close of poll (excluding any undelivered or returned under regulation 77(1) (spoilt), regulation 78(1) (lost) and regulation 86A (cancelled ballot papers))


    15. Number of covering envelopes set aside for the verification of personal identifiers on postal voting statements


    1. Peter May -

      Thanks for that analysis. On that basis it looks rather as though the queues we saw to vote were mostly in the wrong (for Labour) places…

  5. Brendan O'Brien -

    Could you please provide a source for your claim that postal votes wen up to 37% in the UK general election?

    There is in fact no evidence to support this social media claim but by all means tell me if you have some.

    Possibly you are, like many others, misinterpreting the exit poll of Lord Ashcroft, which was most definitely not predicting that 38% (or even 37%) of the votes would be postal votes. His report merely said that of those interviewed on the day, 38% had already voted by post. This is unsurprising because the interviews were mostly in the morning. He reported almost the same percentage in 2017 and when the Electoral Commission reported the actual figures, it was a normal figure (~ 20%).

    The Electoral Commission will report the full figures in a few months. Meanwhile, a number of people on social media have arranged to collate the responses to Freedom of Information requests in order to find this figure out right now. One such exercise has very diligently obtained 191 responses and found an average increase of the postal voting rate of a paltry 0.8%age points. Another, of about 30 constituencies, has found that the rate went slightly down.

    However, I repeat. If you have evidence of an increase of 37%, I invite you to provide it (though wonder why you did not provide it in your article).

    1. Peter May -

      The source is the same Ashcroft poll, which, now six weeks later, it transpires was probably incorrect. I find it worrying nonetheless that the BBC and Raab both mentioned the decisions of postal voters, which they should have had no way of knowing, and yet we still have no figure of how many there were and in particular how many were not cast or invalid.

      1. Brendan O'Brien -

        LK and Raab were foolish to report rumours of PVs. By doing so, they opened the flood gates to questions about how PVs are managed. Your article is an example of this.

        I assume that by now you know the two ways these rumours always come about, in every election. The explanations have been all over the place so I won’t repeat them.l here.

        This blog, by the way, is well worth reading (not by me).


Comments are closed.