Mark Pack

[FOI #18547 email]

26 October 2009

Dear Mark Pack

You raise the matter of the possible use, for the fourth time, of the electronic counting of votes at the Mayoral and Assembly elections in 2012.

To be clear at the outset, I set the budget but the actual running of GLA elections has to be in non-political hands, and so all aspects of their management is delegated to the Greater London Returning Officer (GLRO), supported by GLA staff and using GLA codes of conduct and procurement.

GLA elections have only ever used electronic counting (e-counting). This is not new technology and it has been used successfully around the world for many years. The first GLA elections in 2000 were run by the Government Office for London and the complexity - multiple ballot papers with each elector having up to 4 votes under 3 different electoral systems - was deemed to justify the use of e-counting for accuracy as well as speed.

The GLRO maintained that practice in 2004 (when it was used for the European Parliament elections as well) and 2008. Both of those projects came in within budget and the results were accepted by all the candidates without a single challenge.

The recent publicity about the GLRO's decision in principle to use e-counting has given only part of the story.

The Electoral Commission and others published reports last year on the May 2008 elections and two clear recommendations were, first, that there should be a national framework setting performance standards for e-counting, and secondly that the GLRO should undertake a cost benefit analysis on the alternatives of manual or e-counting.

As it happens, the GLRO had already initiated such an assessment before the Commission reported - the first time that such a study had been undertaken in the UK, and it took a number of months, involving discussion with every local authority in London about the way they would tackle such a complex manual count.

The assessment had to be based on the information available at the time, namely the actual costs of e-counting in 2008 and the best estimates of the boroughs of the costs they would incur if they had to count manually. An electronic count can be completed in one day, whereas a manual count is reckoned to take three days, possibly four. As to the accuracy of e-counting systems, extensive testing before the 2008 elections showed the e-count system to be more accurate than manual counting.

The cost difference was, as has been rightly reported, of the order of £1.5 millions, although since then the Ministry of Justice has introduced a new funding formula for national elections which, with the possibility that London Councils will review their fees for local elections next year, means that the cost benefit analysis will have to be re-calculated at some point in 2010. Further, the assessment allowed that if tenders were invited for more than the one set of elections (i.e. 2016 as well as 2012, with appropriate provisions to ensure the second event benefited from the technology available at that time) the unit cost might be substantially lower. It was with that in mind that the GLRO took his decision in June.

Leo Boland

Chief Executive

Greater London Returning Officer

The Office of the Greater London Returning Officer

London Elects

City Hall

The Queen's Walk

More London

London SE1 2AA

Tel 020 7983 4123

Fax 020 7983 4191

www.londonelects.org.uk