Uncategorized

Electoral Systems and Welsh Elections, 1: An Introduction

An important feature of elections in Wales as elsewhere, but one often overlooked, is the Electoral System. This post is the first of a series which will examine the electoral systems we currently use, as well as some that might potentially be used, for elections in Wales. This initial post will concentrate simply on describing the main systems currently used or prominently proposed for use.

First, though, what do we mean by an Electoral System? There are numerous rules that govern any election, such as laws on who may legally vote, who may become a candidate, on what day(s) the election will be held etc etc. Most of these rules, though undoubtedly of importance to the conduct of reasonably free and fair elections, are not part of what is meant by an electoral system. The latter concerns something more fundamental: as Prof David Farrell concisely puts it, “Electoral systems determine the means by which votes are translated into seats in the process of electing politicians into office” (Farrell, Electoral Systems: a Comparative Introduction, 2001, p.4).

Electoral systems differ on two main dimensions. The first is District Magnitude: this refers not to geographical size but to the number of representatives elected for any given area. As explained below, District Magnitude can vary from 1 to a much larger number. The second dimension is Ballot Structure. This concerns the type of vote people cast: the main distinctions here are between Categorical and Ordinal, and Candidate-based and Party-based votes. (Analysts sometimes include a third dimension, the Electoral Formula, but for most purposes this is, I think, rather less fundamental).

What then, are the main electoral systems currently used in Wales?

Single-Member District Plurality (SMDP): This is probably the most familiar system to most people. It has long been used for UK general elections (albeit its full application came later than many people realise, with some two-member seats existing until 1950). This system, sometimes popularly termed ‘First Past the Post’, is how we elect Wales’ 40 Members of the House of Commons: by dividing Wales into 40 geographical districts (known in Britain as constituencies), which each elect one representative (that’s the single-member bit). The person elected is the one who wins the most votes (i.e. a plurality).

The District Magnitude under this system is, therefore, 1. The ballot structure is categorical – each voter has one vote, which can be cast for one candidate only. (If you try to vote for two candidates, that ‘spoils’ your ballot paper and your vote is not counted.) As well as being used in UK general elections, SMDP was used from 1979-1994 to elect Wales’ representatives in the European Parliament, with Wales being divided into four constituencies in 1979, 1984 and 1989, and then five in 1994. SMDP is also used for elections in some other countries – for most elections in the USA, in Canada and in India, to give a few prominent examples.

Multi-Member District Plurality (MMDP): This minor variant on SMDP is used for many local council elections in Wales. Many local council wards elect more than one councillor. Voters are then given a number of votes equal to the number of representatives to be elected. Each vote is categorical; those X candidates with the most votes (where X equals the number of representatives to be elected in that ward) are deemed elected.

Party List: This type of system has been used to elect Wales’ Members of the European Parliament since 1999. It is also, of course, used in national elections in many countries. It is a form of proportional representation – where the number of representatives a party wins is explicitly linked to their share of the vote.

Under such systems, District Magnitude is larger than 1 – often much larger. Voters cast a (categorical) vote for a party, not a candidate. Those elected are normally the first X people on a party list (where X is the number of seats that a party wins).

These systems are sometimes applied to a single, nationwide party list (such as in Dutch general elections). Often, however, a country is divided into large regions, with separate contests among party lists in each region. For European Parliament elections in the UK since 1999, Wales has been one of eleven regions in mainland Britain. In 1999, five MEPs from Wales were elected; since 2004 (because of adjustments to national seat allocations in the European Parliament to make room for new members states of the EU) Wales has had four MEPs.

An interesting side-note – well, interesting to a few election freaks like me – is that seats are allocated between mainland British regions using one mathematical formula, Sainte-Lague; but the allocation of seats between the parties within each region is done by another formula, d’Hondt. The d’Hondt formula helps larger parties (i.e. those with the larger vote shares) more than Sainte-Lague; and it was, of course, large parties who crafted the relevant legislation in the UK parliament. Were d’Hondt used to allocate seats between regions, the Electoral Commission has calculated that Wales would have one MEP fewer, with that seat being allocated to London. (http://www.electoralcommission.org.uk/__data/assets/pdf_file/0017/156230/Distribution-of-UK-MEPs-among-electoral-regions.pdf).

A final important element to note is the nature of the list system used. For EP elections in Britain, and the List element of AMS used in devolved elections, the List is ‘Closed’: i.e. voters cast a categorical vote for a party list only, and have no ability to favour individual candidates within that list. Some List systems operate differently, using an Open or Ordered list where voters can favour individual candidates.

Additional Member System (AMS): This system has been used for elections to the National Assembly for Wales (as well as for the Scottish Parliament) since 1999. Essentially, it combines SMDP and Party List systems. Each voter is given two (categorical) votes. The first vote is for a constituency member, chosen by SMDP. For National Assembly elections in Wales, there are currently 40 constituency members elected, from the same constituencies as used in Westminster elections. (In Scotland, the constituencies used for Westminster and devolved elections now differ; in formal language, Scotland no longer has co-terminosity of constituency boundaries). The second vote is for a party list in one of five larger regions (based around the former European Parliament constituencies). After the constituency seats are elected, the party list votes are counted and the list seats allocated proportionally (in Scotland and Wales via the d’Hondt formula) having made allowance for the number of constituency seats a party has won. Party lists are ‘closed’ rather than ‘open’.

AMS combines the local representation of SMDP with the proportionality of a party list system. This ‘best of both worlds’ quality helps explain why it is increasingly popular around the world. But it also has notable disadvantages. Having two votes, one for an individual in a constituency and another for a party list in a broader region, rather complicates the election for the voters. It also produces two different classes of elected representative, with those chosen through the constituencies sometimes viewing list members as ‘second-class’ representatives.

Single Transferable Vote (STV): This system is not yet used in Wales, but it is worth mentioning because its use has been proposed for both local and National Assembly elections; it is also used extensively in Ireland (both north and south). Under STV, instead of voting categorically for their most favoured party or candidate, voters give an Ordinal series of numbered preferences (1, 2, 3, 4…) from their most to their least favoured candidates. Normally 3-6 representatives are elected within each, fairly large, constituency; and parties often stand more than one candidate in a constituency. Counting of votes proceeds through multiple rounds, with unsuccessful candidates successively eliminated and their votes re-allocated on the basis of voters’ preferences.

Advocates of STV argue that it maximises voter choice. Instead of requiring an ‘all-or-nothing’ choice, STV allows voters to cast a more subtle verdict. They can also choose between both individuals and parties: thus avoiding the dilemma, often faced by voters under SMDP or Closed List systems, of what to do if they like a party but dislike its particular candidate. STV also produces fairly proportional results for the parties. But it is undoubtedly a somewhat complex electoral system, both for the voters and the parties (not to mention those doing the counting!)

 

These various electoral systems, and the differences between them, can often seem very arcane. But electoral systems matter deeply. In the short-term they can determine who gets elected, favouring some over others. Over the longer-term they can potentially influence a political system much more broadly. In following posts I will consider some of these systems, and their relationship to the sort of politics we have had, do have, and might have here in Wales.

Comments

  • David Costa

    In practice STV is not necessarily proportional. Parties in both parts of Ireland have developed sophisticated systems to “cheat” STV by advocating different candidate orderings in different parts of the constituency. In the North this helps to explain the rise of harder line, more disciplined parties on both sides – Sinn Fein and the DUP. This suggests that contrary to the expectations of many supporters of STV, it can increase the rewards for effective political machines.

Leave a Reply

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