Monday, 6 May 2013

Bulk Ward: Counting errors and swinging seats (updated & corrected)

Original story posted 6 May 2013
Updated 7 May 2013 with corrections and LCC comment:

Counting errors at the Lancaster City Council by-election count on Friday came close to costing the Green Party a council seat for Bulk Ward, we can reveal.

When the votes were counted in, the result was provisionally provided (ie not announced as I previously wrote)  to the candidates and their agents as being  Green 676 and Labour 895, indicating at that point a triumph for Bob Clarke, the Labour candidate, at which the Labour contingent present began to celebrate, as well they might.

Councillor Tim Hamilton-Cox of the Green Party has developed a reputation for having an eye for detail. It is thanks to his assiduous accountancy, we hear,  that we owe the holding of the council's nerve over Council Tax support levels. It's one thing to complain about injustice. It's another to plough through masses of paperwork and accounts and budgets and work out realistically sustainable ways to make things fair.

Councillor Cox had been keeping an eye on the counting tables and knew from watching the paper move that the score between Labour and Green was too close to call. When the provisional result came out as a substantial shift to Labour he realised it didn't fit with what he'd been seeing.

It could feel embarrassing, to start demanding a recount when your side looks to have lost. It's a bit sour grapes. Everyone's been working hard, and now they have to start again. In this case (I stand corrected), the words 'kitchen' and 'heat' apply. Most people have few opportunities to make their wishes count.  Protesters are regularly informed in courtrooms that that the appropriate time to express their views is at elections. So, to keep that faith,  the election count must be transparently honest and reliable. If there are doubts, they must be disproven, so everyone can go home sure it came out fair. This is Lancaster, not Florida. And so, at the point when the provisional score is provided, candidates and their agents are asked if they are satisfied and offered the opportunity to ask for a recount. Usually the potential loser does so when the scores look very close.  In this case the problem was that they didn't, so further recounts were requested.

The votes for each candidate are counted into bundles of 50. These are labelled with the candidate's name and stacked in boxes. The recount found that 3 bundles had been miscounted. We understand that two Green bundles and one Conservative bundle had been mistakenly labelled for Labour. And within one Green bundle were found 8 Labour votes. Almost 10% of the by-election ballot had gone astray.

After the recounts the final result was agreed and declared,  Green Party candidate Caroline Jackson had polled 769 votes. Labour's Bob Clarke 753 and Conservative Kevan Stuart Walton 142. Meaning that the Greens held this Bulk Ward seat by just 16 votes.

Correction: I previously wrote that some of the local results for the County Council elections might be re-examined in the light of the errors in the by-election. However, I am informed by the City Council that the final opportunity to challenge the result is at the point when the provisional result is provided. Once the final result is declared, as these were on the day, then the election process is closed and the Councillors elected are mandated to get on with their jobs. (See their comment below)

Had the error been smaller, say just one bundle astray, it might have been missed and Labour might still be celebrating.

As indeed might Centros, the company managing the planning application for the proposed Canal Corridor development, the site of which lies in Bulk Ward. Originally a safe Labour ward, its three seats landslid into the hands of the Green Party back in 2007 after a series of increasingly large and frustrated public meetings about the development proposals which its Labour councillors did not find time to attend.

Prior to the election the Labour campaign became increasingly concerned about the reception they were receiving on previously friendly doorsteps. But they could still count on one resourceful and well-funded ally. The final week of the campaign was marked by a £16,000 telephone 'survey' of local shopping habits commissioned by Centros and timed for the Labour Party to use on their final election leaflets.  However a phone call to ICM, the company commissioned by Centros to carry out the poll found that the interpretation of their findings had been surprisingly creative (see report). Even when expensively spoonfed by Centros, however, the Labour leafleteers made the mistake of spinning together their favourite bits from two different sets of ICM stats - which then added up to a facepalming 107% on their election leaflets.

The current election campaign also coincided with Centros and British Land, the new owners of the redevelopment site, delivering their new promotional leaflets to every local household in the last fortnight. Their consultation stand will be in Lancaster's Market Square on Tuesday 14 May from 10am - 7pm. They also welcome your comments at

7/5/13: Lancaster City Council have asked us to add the following clarifications:

"At every count it is standard practice to share a provisional result with candidates and agents for them to consider and provide the opportunity for them to request a recount.  It is misleading to describe this part of the process as announcing the result.  

"This procedure is in place so that potential inconsistencies can be examined and rectified. This was precisely the case at Friday’s count. Following the provisional result for the Bulk ward by-election a counting agent of the Green Party requested a recount and the returning officer agreed to the request. 

"It is far from embarassing to request a recount; it is a routine event.  One of the reasons candidates and agents watch the votes being counted is to identify potential inconsistencies and therefore request a recount.
It is not unsual, particularly in district council elections where the number of votes being counted is relatively small, for recounts to be requested. 

"Candidates and agents in the county council election counts were also provided with the opportunity to request a recount after they were informed of the provisional results. Once a result is formally declared, it is final and cannot be amended. "


Anonymous said...

Thanks for setting this out. Curious that this distinctly dodgy sounding affair has not been reported elsewhere... I guess the other media in this town don't have the enquiring minds nor the balls to do so. Labour, if they had any grace, should ask for a recount in the other close run ward, just to prove that there was no skullduggery.

Chris Satori said...

I don't think anyone is suggesting that there has been skullduggery. I don't see how it could be proven, and may quite simply have been due to an unfortunate lapse of concentration on one table while counting thousands of bits of paper. It is simply that an error margin as large as 10% is unusual and possibly not something people would have taken into consideration at the earlier counts. Perhaps the lesson candidates will take away is to always request a thorough recount when you lose. Even when the majority looks decisive.

Karen Leytham said...

I think someone is up to mischief because this years council rent freeze has absolutely nothing to do with Tim's alleged 'assiduous accountancy' skills. Within my portfolio I was the one responsible for recommending the rent freeze and at the same time bringing forward the building of new 1 bed council properties.

Chris Satori said...

@ Cllr Leytham: I am sorry if I have not given you credit where it is due. I imagine it is awkward for all parties when this happens and I have removed that part of the article pending further consideration.

We may be able to clear it up for any future reference if you will help me. The rent increase of 7.8% last year was delivered with an emphasis from yourself on the need to keep in line with government calculations. Further projected rent increases of 4.6% a year for the next two years were included in this decision.

However it was not unanimously supported even within the Labour Party membership. Since then the direction of the council's rent policy has undergone a sea-change. If or when you have the opportunity, you may be able to set the record straight on how this shift came about.

With regard to your reference to 'mischief': One problem that we face here at VL is the imbalance in the amount of news we receive from each local party. Labour are well in the lead on this score, and were we to print everything we receive, as we receive it, it might almost seem as if there were only Labour members on the council. This is far from being a criticism. We appreciate the healthy transparency and consideration. The Conservatives have also, over time, publicly highlighted a spectrum of issues of particular concern to them and this helps towards a better understanding of their concerns and abilities. However the newer political groups represented on the council are less frequent with their press releases, which more usually focus on planning matters. Furthermore, where an issue of note is under consideration, councillors are often restricted from disclosure and one cannot solicit them but must spend hours researching local and national websites and fora, council PR and minutes, FOI etc or phoning stakeholders or companies like ICM for a more rounded picture, when other commitments allow (Volunteers, please!). Trying to arrive at an understanding of how local democracy works is thus not always an efficient process and a story may be updated or amended as further clarifications and developments emerge. Sometimes this is the only way to bring an issue of real public interest to light at all.

This particular story is about how a 10% error margin in a count is, perhaps, unusual and unexpected, and about the human and statutory processes that brought it to light.

It is a pity that this may be lost in unfortunate side issues (and again, mea culpa). In this case it appeared to indicate a decisive majority for one party when in fact another had polled more votes. Errors happen, I can relate to that. And the system worked, in that it caught a noticeably large provisional error.

My thought is that once candidates and potential future candidates of all hues recognise the possibility of so generous an error margin in provisional results, they may be inclined to increase the frequency with which recounts are requested.