[The following is an edited summary of my remarks to the Political Reform conference, hosted by University College Cork, March 26 2010.]

It is good to see that political reform is now on the agenda of some of our political leaders. More and more people are calling for large-scale political reform, with an ever-growing list of targets for reform. Electoral reform is frequently mentioned as one of those targets. I certainly agree that it is worth exploring the question of electoral reform, but whether it is one of the political institutions that actually should be reformed is a moot point. I’ll develop this argument in four main points.

1.) Don’t do this in isolation
First, if electoral reform is to happen then at the very least it should certainly not be done in isolation. Let’s start by thinking about what it is about Irish politics that electoral reform is supposed to fix. The charge is that TDs are acting as little more than constituency messenger boys and girls, running aimlessly around their constituency patches chasing funeral hearses and helping sort out welfare and pension payments – all of this at a cost to their role as legislators in the Dáil. What comparative evidence we have does, indeed, support the contention that the constituency focus is greater here than in comparable countries. The accusation is that TDs are in intense competition with each other for personal votes, and the blame for this is laid firmly on the STV electoral system.

But to what extent is it STV that is really to blame for this? We know from comparative work (e.g. a recent survey of MPs in Malta also elected by STV – see http://bit.ly/9UH78d) that politicians in other countries that use electoral systems like ours do a lot less constituency work. Changing our electoral system is unlikely to be enough to fix this problem. Arguably, rather than focusing our reforms on the supply-side (on the service that TDs supply for citizens) reform would be better addressing the demand-side (on the service that citizens demand of TDs) on such things as the strengthening of local government and on improving the interface between public service departments and citizens. In short, whatever reform we go for must be joined-up, large scale and effective.

2.) Will Irish voters support electoral reform?
My second point is that electoral reform is not worth pursuing without fair chance of it succeeding. We need to take account of the particular circumstances of the Irish case, namely the fact that the STV electoral system is enshrined in the Constitution ad can only be changed by referendum (something that has been unsuccessfully attempted on two occasions). If we’re not careful we could waste a lot of time and money designing a spanking new electoral system only for it to be defeated in a referendum.

One of the things that make the Fine Gael ‘New Politics’ document (http://bit.ly/auiSCA) so attractive is the proposal to establish a citizen assembly to look into the question of designing a new electoral system. The point is that there is no such thing as a bespoke electoral system. Any proposed new electoral system would have to be designed from scratch to take account of the particular circumstances of the Irish case. The will place the PROCESS of electoral system design into sharp relief. Any whiff of a notion that the new electoral system is being designed to serve the interests of one political party or another and it will be roundly defeated in a referendum. In short, any process of electoral reform must engage with and actively involve citizens; it must have the stamp of popular legitimacy (for more, see Ken Benoit’s piece in the Irish Times – http://bit.ly/awwzgb).

3.) Do we really need to change our electoral system?
My third point is that, actually, it might well be that on reflection we won’t want to change the electoral system. Let’s look at the alternatives, of which there are three. The first of these is the British first-past-the-post system (or, perhaps the Australian alternative vote), a non-proportional electoral system (which was what Fianna Fáil proposed in the 1959 and 1968 referendums). One of the best-established social scientific laws – ‘Duverger’s Law – states that non-PR electoral systems produce two-party systems. In other words, this electoral system would produce a Dáil with two large parties (FF and FG) and at best a smattering of seats going to a tiny rump of a Labour party (much like the British House of Commons) – hardly a terribly attractive prospect for a modern, inclusive democracy. The second alternative is the PR-list electoral system common across much of Europe. The question then is which form of list system would we opt for. We could adopt the version used in Finland (‘open list’) or Switzerland (panachage) in which voters declare their vote for individual candidates rather than closed lists, but how would this change anything? We would be swapping one electoral system (STV) in which voters vote for candidates with another electoral system (open list) in which voters still vote for candidates. Do we really think that this would change anything? An alternative option would be to go for a closed electoral system such as used in Spain, in which voters vote for a party list and the determination of which politicians win parliamentary seats is decided by the parties who rank-order their candidates on the party lists. The problem with this option is that you may end up doing little more than replacing a constituency emphasis on the ELECTORATE with an emphasis instead on the SELECTORATE: instead of running around chasing support from voters, TDs may still end up running around only this time chasing support from party members.

This leaves is with our third option, the mixed-member proportional (MMP) system used in Germany, and the system that is often referred to as a likely alternative here (e.g. by Noel Dempsey; and also by Enda Kenny in an earlier version of his New Politics document, which talked of a variant of MMP for Ireland). The first thing to note about this electoral system is that it has only been adopted by a small number of countries – many of them barely even democracies – and so its properties are still not fully understood. To date, it has been adopted by: Albania, Bolivia, Lesotho, New Zealand and Venezuela. And before we rush to look at the New Zealand case, it would be well to note that next year they will be holding a fresh referendum with a view to possibly abandoning this system.

One reason why MMP is seen as so attractive is because of the way it mixed two electoral systems – the PR list system and the British non-PR first-past-the-post system. It is seen as a fair compromise between the two extremes, as ‘the best of both worlds’ – for here we can have a mix of constituency politicians (elected in single-seat constituencies) with party list politicians elected to produce a fair proportionality in the election result. But, actually rather than see this mix in terms of a mid-point between PR and non-PR it is more accurate to thing of it in terms of an analogy to a bowl of soup – in which two ingredients have been mixed. The question now is whether the ingredients have been mixed in the right way. Recent academic research into MMP systems has found clear evidence of ‘contamination effects’, in which one part of the system effects the operation of the other half. A good example is in the constituency emphasis of the politicians. There is evidence of the list MPs operating constituency offices and paying close attention to the constituency needs of voters in much the same was as their constituency-MP counterparts: the role of the list MPs is being affected by (contaminated by) the role of the constituency MPs.

In short, a move to a new electoral system may not necessarily improve things.

4.) Perhaps fixing STV is better than replacing it?
My final point is that rather than full-scale electoral reform it might be preferable to introduce some fixes to some of the details of how STV operates. This argument has been rehearsed elsewhere (http://bit.ly/9UTThu) so I will just summarise the gist in bullet points. Of the following items, only the final one would require constitutional reform.
• We could increase constituency size (we used to have constituencies with up to nine TDs in each), which would increase the proportionality of the electoral system, bring in more small parties and reduce the hold over the political system of the larger parties (thus increasing the influence of the Dáil).
• We could change the way candidates are listed on the ballot paper: e.g. to the Malta version in which candidates are listed under their party lists. This might help to reduce some of the constituency emphasis because voters would be encouraged to think more in party terms than candidate terms when deciding on their vote.
• We could make a change in the STV counting rules regarding how surplus votes are transferred. This can quickly get into a pretty arcane world (for an illustration, see http://bit.ly/cVkFn9) but the gist is that by moving to what’s known as the ‘Gregory method’, or the ‘Senate rules’, we would help to reduce the element of chance that currently prevails in Irish elections that can result in the wrong candidate winning the final seat in a close election count.
• One final change to consider is replacing by-elections with other means of filling vacant seats. Ireland is one of the very few proportional electoral systems to still use by-elections, which is an anathema because, by definition, a by-election (an election to fill one seat) is a non-PR election.