Flickr/TiggerT. Some rights reserved.
The Electoral Reform Society (ERS), Britain’s leading campaign for electoral reform, has published a report, trailed on 30 May in the Observer, which argues that the first-past-the-post electoral system is not only producing increasingly disproportional but also more random results. In May this year, as they say, the wildly varying disproportional results of the general election were ‘splitting the union’, creating ‘artificial divides’, exaggerating political differences, perversely distorting the parties’ representation in Parliament and wasting millions of votes.
Very true. However, Katie Ghose, the Society’s CEO, takes the story further, stating in the ERS report, A voting system in crisis, that this is ‘the first study revealing what the results would have been under alternative voting systems’. This is not true. After the 1992 and 1997 general elections, a team from Democratic Audit at Essex, Birkbeck College and the LSE Public Policy Group, in which I played a part, carried out mock elections through large-scale opinion surveys asking people to vote again on the kind of ballot papers used for the three main alternative electoral systems as well as calculating three varieties of the List system. We published a major report, Making Votes Count: Replaying the 1990s General Elections under Alternative Electoral Systems (MVC), Democratic Audit Paper No.11, September 1997.
The ERS study also makes use of a large post-election poll to replay the 2015 election, but less stringently than we did, simply asking respondents to rank the parties in order of preference. But makes no reference to the 1990s report.
How to explain this strange amnesia about what was a prominent report (it sold out)? Is it just down to ignorance – among a new and inexperienced generation of research staff, say? Or is this a case of political calculation which is evident enough in other aspects of the ERS study? For that study is shot through with bias and is tellingly incomplete. The 1990s exercises ran mock elections under the Single Transferable Vote (STV, as used in Ireland and elections to the NI Assembly), the Alternative Vote (AV), (and its cousin, the Supplementary Vote), and the Additional Member System (AMS), similar to those used in Scotland and Wales for the parliament and assembly. Whereas, the ERS study does not include AMS in the alternative systems that are replayed and so it has no place in its comparative table. There is a perfunctory description of AMS and it merits only a slighting reference in a final summary which leads inexorably towards the verdict, ‘it’s now time for STV to take its place in Westminster’.
The ERS is a very odd institution. It was founded in 1884 with the aim of replacing FPTP with STV and it is bound under its constitution to campaign only for STV rather than electoral reform or proportional electoral systems in general. (Ironically, the Society did campaign for the Alternative Vote in the recent referendum, on the dubious grounds that the wildly disproportional AV is simply STV in a single constituency, albeit that it is shorn of all the benefits that the ERS values in STV.)
Various attempts have been made over the years to widen the ERS remit, but a priesthood of true believers, dug in behind amazingly complex constitutional defences, has repulsed all efforts to introduce a more pluralist approach to reform. Since AMS is STV’s closest rival for adoption in the UK, its omission from the alternative systems explored in the ERS report, and from serious consideration, is perhaps no surprise.
It may be also that a factional motive explains the amnesia surrounding the 1990s’ Make Votes Count report. A four-page leaflet summarising the main report shows that the results of the replayed 1992 STV ballot were mildly disproportional, but the 1997 replay produced a highly disproportional result with a significantly high deviation from proportionality score. By contrast, the AMS ballot, based on the classic German system, which is equally balanced between MPs elected in local constituencies and ‘top-up’ MPs, unlike the systems in use if Scotland and Wales, was consistently proportional in both 1992 and 1997.
The Electoral Reform Society is the lead organisation in the movement for electoral reform, in part because it is the richest of these bodies. It is in this sense ‘the voice’ of electoral reform in the UK. Its refusal to adopt a more pluralist policy is therefore disturbing enough. Moreover, while it may be prohibited from actually campaigning for any system other than STV, it is not prohibited from dispassionately discussing and comparing the arguments for STV and the rival alternative systems on an equal footing. Shoddy documents like A voting system in crisis do not only give the public incomplete and biased information, they reduce the chances of choosing the most suitable electoral system to replace FPTP.
STV and AMS are the leading proportional contenders to replace FPTP at Westminster. The aberration of the referendum campaign for AV has surely got that bizarre choice out of contention. List PR, the system used for elections to the EU Parliament, is too remote and distinct from British political tradition. STV suffers somewhat from its distance from recognisable constituencies people can identify with. Under STV, people vote in large multi-member constituencies - say, at a minimum for five candidates in constituencies five times larger than current constituencies. Voters can express as many preferences as there are candidates, ranking their choices both within and between parties. There is no party list and its chief merit, then, is that it gives voters a freer and wider choice and reduces the power of the political parties over that choice. However, it is not consistently proportional. Even the ERS’s own figures show that it was imperfectly proportional in 2015, over-representing the two main parties and under-representing UKIP and the Greens. (But the results were based on STV in smaller three to five member constituencies, as the ERS may have wished to play down the size of constituencies that the system requires.)
AMS comes much closer to recognisable British practice. The MVC replay assumed that seats would be twice the size of existing constituencies; thus retaining a closer constituency link than STV. We tested AMS as it operates in Germany, where half the MPs are elected locally, and half are regional “top-up” MPs. (AMS elections in Scotland and Wales produce less proportional results, as locally elected members take a higher proportion of seats in the legislatures).
The two main parties have resolutely clung onto FPTP as it has historically preserved their grasp on political power. It is probably too optimistic to think that the Labour Party will finally recognise that it comes off second best. Meanwhile, the electoral reform movement deliberately never came together to campaign for a particular system, fearing that any choice would split its ranks. It even adopted the rubric of ‘electoral reform’ rather than go for a proportional system, thus allowing AV to creep in disastrously. As things stand, the best course would be to go for AMS. It is consistently proportional, it is better known than STV and fits more closely into established practice. But the ERS, the most powerful reform organisation, would in its present disposition spoil the nation’s ballot papers.