Electoral Reform in Ireland – Slight Return

This really is the last episode of my exciting series on electoral reform in Ireland. I was talking about how there is something to be said for introducing a mixed-member electoral system in Ireland, with my own eccentric suggestion being electing half the TDs by STV in large multi-member constituencies, the other half by closed national lists. One thing I did not really talk about is how mixed-member system operate in practice. I do not mean so much in terms of whether they are associated with "good" or "bad" political outcomes, more how the system itself tends to operate. This is not something I have read too much on*, but I understand that one common feature of mixed-member systems is that list MPs tend to cultivate particular localities with a view to becoming constituency MPs for that area. In this context, it is maybe interesting to note that Germany's chancellor is not head of the national CDU list, but is rather a constituency MP for somewhere in Mecklenburg-Vorprommen.

I am not entirely sure why the list MPs are so keen to become constituency MPs. Maybe the ones who are individually elected are seen as having greater legitimacy. Or it could be that being a constituency MP is seen as being more secure – a list candidate needs to keep in with the party hierarchy to ensure that they are being placed high enough up the list to get elected, while a local MP just needs to keep in with the local party grandees to ensure they get reselected, if they are in a safe seat. Or maybe some other factor is at play.

If a mixed-member system were rolled out in Ireland, it would not therefore be too surprising if it failed to vanish the genie of localism from Irish politics. We might end up with list TDs who neglect national issues and instead focus on local issues and look to become constituency TDs.

I will for the moment leave the topic of electoral reform, but not without throwing out a question. Talk on this issue is driven by the idea that a localist orientation in politics is bad. But is it really so dreadful? In many countries, people complain about politicians who are remote from the people who elect them. Maybe we should be glad that ours are always available to address our petty concerns.


*in this respect I am like a great many people who advocate profound institutional change.

Electoral Reform in Ireland – Part 4: More mixed-member action

FINAL EPISODE! Last time I suggested that a mixed-member system – where some MPs are elected in constituencies and some by national lists – would be worth introducing in Ireland. Then I mentioned research undertaken for an Oireachtas committee, which found that if the German system (plurality voting in constituencies for half the members, closed national lists for the rest) were introduced here we would end up with Fianna Fáil sweeping the constituency seats and the other parties winning all the list seats. This skewed result did not appeal and the Oireachtas committee recommended against moving to a mixed-member system. So are mixed-member systems that rubbish? Well, no.

There was maybe a certain failure of imagination on the part of the people doing the research for the Oireachtas committee, in that they only seem to have considered an exact replica of the German system*. What would produce less extreme results would be a mixed-member system where half the TDs were elected by STV in multi-member constituencies, the other half by list. The multi-member constituencies would of necessity be very large, but we could reasonably expect that one party would not win all or even most of them. The likelihood is that candidates would still compete for these seats on something like the localist manner we are used to now. The other seats – the party list seats – would allow at least for the possibility that parties could win seats by campaigning on national issues. If there was some kind of hitherto unfulfilled drive for national politics then this would provide some chance that it could find expression.

There are a couple possible issues to this kind of system. One of them is that the STV constituencies would be so big that they would often have to include several counties. The likelihood would be that people from some of the less populous counties would have no constituency TD from their county. This is only a serious problem if you think that every county, no matter how underpopulated, needs to have at least one TD.

Another issue would be that if you have TDs elected in two different ways then one set of TDs might see themselves as having more legitimacy than the others. The constituency TDs, in particular, might scoff at the list TDs on the basis that they, as individuals, had never been chosen by the voters. In practice, though, politicians throw all kinds of brickbats at each other, and another set of insults should be easy enough to shrug off.

The downsides of mixed-member systems are manageable. The advantages are that the constituency elections allows people who like voting for individual TDs to keep doing so, while the list election makes it easier for a less locally oriented politics to emerge. Electing the constituency TDs by STV in multi-member constituencies prevents the unpleasantly skewed outcome a straight import of the German system could produce.

That, then, is my proposed new electoral system for Ireland. There is no great prospect of it being adopted, as I have never heard it suggested by anyone else as an alternative electoral system for Ireland. Even if my wonderful proposal was somehow adopted, we should be realistic about the likelihood of it actually effecting any great change to how politics works. I am highly sceptical of the power of institutional setups alone to transform politics, and experience suggests that the localist impetus in Irish politics is sufficiently strong that under any electoral system it will still dominate. Still, electing some TDs by party at a national level might just concentrate some electors' minds on the fact that elections are about picking people who will form a government.


*maybe I should look at their report and see if this is actually the case.

Electoral Reform in Ireland – Part 3: The magic of mixed-member systems

In the last post I made a case for electoral reform in Ireland, but then argued that a number of electoral systems people sometimes talk about moving to are a bit problematic. Astute readers may have noticed that I omitted any discussion of mixed-member systems, the subject of today's post.

Mixed-member systems are so called because they mix up how parliamentarians are elected, typically electing some of the MPs in local constituencies and some nationally by list. Germany is the great mixed-member poster child, with its adoption of the system in the Federal Republic's Basic Law often seen as one of the things that embedded democracy in (West) Germany after the Second World War. Germany elects half its MPs in constituencies, using Westminster-style plurality voting. The other half is elected from closed national lists. The list seats are allocated so as to ensure the overall proportionality of the Bundestag. There is also the 5% threshold – if a party wins less than 5% of the list vote then it wins no list seats.

The German Federal Republic has proved to be a rather successful country, especially given the travails it experienced under previous regimes. This means that people are always talking about borrowing aspects of its institutional setup. Advocates of electoral reform often talk about introducing mixed-member systems in their country. The advantages of the system are seen as being that it allows people to keep voting for a local representative while ensuring a proportional overall result. In Ireland's case, the list side of the election offers the possibility of voters' minds being focussed on national issues, while letting them continue to vote for individual politicians.

As it happens, a move to a mixed-member system here was considered not too long ago. Following a 1996 report by the Constitution Review Group that suggested mixed-member systems were worth looking at, an Oireachtas Committee commissioned further research on the issue, bringing forth a report in 2002. The research was not particularly favourable. Looking at how an exact replica of the German system in Ireland would operate, the report found that it would produce a very skewed allocation of seats between list and constituency members – basically, Fianna Fáil would win all (or almost all) of the constituency seats, with almost all of the list seats then going to the other parties*. The committee recommended against adopting a mixed-member system.

So is that it for mixed-member systems? Come back next time for the FINAL EPISODE and see.


*This was based on the relative levels of party support then applying. Fianna Fáil's support is currently much lower, and if a general election were to be conducted right now using the German system then it might not produce such a skewed result.

Electoral Reform in Ireland – Part 2

In part 1, I discussed proposals periodically made to change Ireland's electoral system. The proposers of such change hope that by doing so they can orient Ireland's politics away from localism. The potential for intra-party competition in the current system is typically seen as causing our localist politics, so advocates of change typically propose electoral systems where politicians will not have to compete with their own party colleagues. I suggested that a lot of this thinking is a bit woolly, in particular claiming that these people overstate the extent to which electoral systems drive politics.

For all that, a case can still be made for electoral reform. Even if our electoral system does not cause localist politics, it could be said to assist it; STV provides a fertile ground for localism and does not encourage politicians to take a more national view. A new electoral system would not conjure a more agreeable politics into being, but it could provide space for it to emerge. This assumes there is a latent drive towards "good" politics currently being blocked by the electoral system.

When thinking about what kind of electoral system to move to, you need to first think about what is wrong with the one we have at present. People typically see the opportunities it offers for intra-party competition as driving clientelism in Irish politics. I think this is over-stated. One feature of the current electoral system that is, I think, more relevant is that we vote for people in constituencies – if you elect people by locality then it should not be too surprising if they spend a lot of their time trying to look after the locality. If you have an electoral system that elects people at a national level then there is more scope for a nationally based politics.

So, what electoral system might encourage a more nationally oriented politics? Other systems based on geographical constituencies are not going to break the link to localism; that stops me from advocating anything like the alternative vote, plurality voting (the crazy Westminster system), that funny two-round voting thing they have in France, and so on. Then there are the various types of list system that are used for proportional representation in other countries. If you want to challenge the localist orientation of Irish politics then you will want national rather than regional lists.

Even so, list systems remain problematic. You can have closed lists (where the order in which people are elected from a list is determined by the party) or open lists (where people on a list are elected on the basis of which individuals on it have the most votes*). I suspect that a move to closed lists for Dáil elections would be unacceptable in Ireland – people are too rooted to the idea of voting for an individual candidate rather than a party. The problem with open lists is that by retaining the element of competition between individuals, they allow for politicians to continue differentiating themselves on local issues. This might not happen in a country like Finland, where open lists are used for parliamentary elections, but in a country like Ireland with a strong local tradition politicians might well continue to look for votes from people a particular geographical area.

That leaves us in a bit of a pickle – I have suggested that a change in electoral systems would be desirable, but have then raised problems with any of the electoral systems we could consider moving to. Come back next time as I attempt to resolve this conundrum.


*I am somewhat simplifying how open lists work, or how they can work – there are open list systems that give the voters astonishing abilities to reorder, split, and combine lists.

Electoral Reform in Ireland – Part 1

Ireland uses an unusual electoral system – the Single Transferable Vote in multimember constituencies (STV). Only two other countries use STV for national elections, and of those one is very small and the other only uses it for the less important of its two parliamentary chambers.

Every so often someone proposes a move from STV to some other electoral system*. Whenever this proposal is mooted, it is argued that STV is a major cause of the political woes afflicting Ireland, so getting rid of it is necessary to improve the political climate. The argument basically works like this. As is, Irish TDs** spend most of their time on local issues – either directly servicing the needs of constituents or bringing home pork for their locality. This is seen as being because STV allows for competition for seats between politicians of the same party. To differentiate themselves from each other, they compete on their ability to service their constituency. Thus, the electoral system leads to the Dáil being full of locally oriented politicians who neglect national issues. Moving to some other electoral system would lead to a situation where parliamentarians are more engaged with national issues; the hope is then that the likes of the current economic crisis would never arise again.

A lot of this thinking is a bit woolly. The idea that it is intra-party competition that drives politicians’ localist orientation is somewhat problematic. It ignores the historical record, with it apparently being the case that politicians engaged in considerable amounts of constituency work before the foundation of the state, when a completely different electoral system was used. It also misses that TDs from parties that only field one candidate in their constituency still engage in plenty of constituency work. It does appear that there is something embedded in Irish political culture that drives politicians towards pork-barrelling and to work as direct service providers for their constituents.

International comparisons are also instructive. People here think of localist politicians as an exclusively Irish phenomenon, but looking further afield suggests differently. MPs in the UK, Canada, and France devote considerable energy to constituency work; none of these countries have electoral systems that not encourage intra-party competition. There are even examples of countries that use closed national lists to elect MPs seeing parliamentarians doing constituency work – despite not having to compete against party colleagues for votes and not even having constituents.

People also tend to forget that there are other electoral systems where candidates compete against members of their own party for the electorate’s favour. I am thinking here of the commonly used open list PR elections, where voters pick one candidate from a party’s list and the party’s seats are then allocated on the basis of which candidates have the most individual votes. In some of these cases, you see politicians engaging in a lot of constituency work, and in some you do not. My suspicion, therefore, is that constituency work is not solely driven by electoral systems, and so moving to another electoral system would not banish localism from Irish politics.

For all that, I think there is a case to be made for electoral reform in Ireland. Join me in part two for a discussion of what direction that reform should take.


*two examples, in articles that are also about other things: How inertia became the iron law of Irish politics & Opposition parties must tell electorate hard truths

**members of the Dáil, the lower and more important House of the Oireachtas

The crazy world of US election rules

The BBC has an interesting article by Larry Sabato (politics professor in the University of Virginia) on one of the more arcane features of the US constitution: The US election nightmare scenario. He is talking basically about what would happen if the vote in November produces a tie in the electoral college, something that is not outside the bounds of possibility. Apparently the Senate then gets to elect the vice president (on the basis that the VP is the Senate's chair), and the House of Representatives picks the president. However, the House picks the president not by a straight vote, but by one in which each State's representatives have one vote between them (with a majority of the state's representatives deciding which way the state's vote goes). In such an election, California would have the same clout as Delaware, and the result could easily end up being completely random and bearing no relation whatsoever to the way the popular vote fell. Were this to happen, we may perhaps be spared the prospect of US leaders lecturing other countries on the benefits of democracy.

Ireland's Berlusconi?

As you know, the Lisbon Treaty has been rejected by Irish voters in a referendum. This has happened despite the treaty being backed by something like 90% of the members of the Dáil, Ireland's directly elected parliamentary chamber. European treaty referendums have always seen a higher proportion of No voters than votes in the Dáil. This disconnect is even more apparent now that the treaty has been soundly rejected in a referendum with a relatively high turn-out.

One thing that was suggested about the No vote on Thursday was that people were expressing their distrust of the Irish political establishment. This may well be the case, and even if people voted on a careful weighing up of the proposals contained in the treaty, you would have to think that No voters must be somewhat dissatisfied with a political establishment that has solidly endorsed Lisbon. Ireland is, however, parliamentary democracy, and it was only last year that the Irish electorate voted in the people they now so distrust. It could be that events since the election have led to a massive erosion of trust in our political elite. My suspicion, though, is that a great many people do not really see elections as having anything to do with producing a government. Ireland has a constituency-based electoral system. My feeling is that many Irish people vote for local characters they either have a fondness for or whom they think will bring in cargo for the area or for them personally.

Lisbon's failure nevertheless suggests a considerable degree of dissatisfaction with Ireland's political elite. It may be that the country is ready for someone to tap that dissatisfaction. If that someone could make people register that it is in their power to remove the political elite, then it would be possible to mount an insurrectionary electoral campaign that would shatter the established pattern of Irish politics.

Declan Ganley of Libertas is surely the person best placed to ride the tiger. There were other players in the No campaign, but they were from fringe political movements that do not look like credible challengers for the political big time. Libertas, though, look like a political party in waiting, and it is striking how some of their posters seemed to campaign against the establishment ("Don't Trust Them!") as much as against the treaty. Ganley seems not to have ruled out the idea of running for public office, so maybe we will next year be seeing Libertas try to establish an electoral base.

People tend to think of populism as something you get in funny Latin American countries. However, many European countries have in recent years seen the emergence of populist parties led by charismatic leaders railing against the cosy consensus that dominates their countries' political life. These populist movements have enjoyed different levels of success in different countries, but in several (including Poland, the Netherlands, & Austria) they have spent some time in government. In Italy, meanwhile, such a party is now the dominant party in that country's governing coalition. It is perhaps not for nothing that, writing in yesterday's Irish Times, Stephen Collins wrote of Declan Ganley becoming the "Silvio Berlusconi of Irish politics".

I am not entirely sure that Ganley's political prospects are quite so good. My impression is that populist challenges work best where a leader can easily affect a direct relationship with the electorate. This is easy in the kind of presidential systems they love in Latin America, where people can directly vote for the populist leader. It is also something you might see in countries with parliamentary government where the electorate votes for a nation-wide list that the leader can head. It is a bit more difficult in constituency based parliamentary systems. In such countries, a populist leader has to find credible candidates to run in the constituencies, and faces always the possibility that their party cohorts may put down local roots and not function loyally as their creatures. Ireland's electoral system therefore provides some institutional blocks to Ganley's sweeping to political power.

Some Links

In lieu of a substantive post, here are some links:

International Development

My old friend and quaffing partner "wwhyte" discusses international capital flows, based on a chunky PDFed article by Brad DeLong to which he links. With the US economy in trouble, there could be net capital flows to Third World countries. This may or may not be a good thing. Actual capital investment (you know, money coming in to build productive resources etc.) can often be the kind of thing less developed economies need. Hot money (fast moving liquid cash) and portfolio investment (purchasing of shares and bonds) can however be problematic in such environments, as they create instability in a system unable to manage it.

d'Hondt or D'Hondt

Matthew Søberg Shugart ponders The Great Debate – how do you spell the name of the guy who invented the d'hondt system for seat allocation in proportional electoral systems. Like most people in Ireland, I first became aware of d'hondt in the context of government formation in Northern Ireland. During the years of peace process & political paralysis, there kept being a lot of talk of "triggering d'hondt", meaning that the parties would get an allocation of the places in government based on their representation in the Assembly. They then picked the government positions one by one; I'm not sure if d'hondt covers this.

A Sensible Path on Iran

Zbigniew Brzezinski and William Odom talk about Iran, arguing that the more pressure the country is put on, the more likely it is to develop nuclear weapons. This is hardly an orginal argument, but it is well-stated here. Brzezinski & Odon also discuss countries that gave up their nuclear weapons programmes, typically without external pressure. I suppose part of what makes this all interesting is its broadly Realist bent, and its sense of the limitations of US power.

Fringe Thoughts

The brainy sociology blog of people I know. Sociology is an interesting discipline… looking at it from the outside, it all seems very leftist. That's not a bad thing in and of itself, but it might marginalise sociology somewhat. I wonder are there any right-wing sociologists (by which | mean not Nazis but people who aren't leftists - you know, people do not believe in the desirability or possibility of radical social change).

I am basically revealing my ignorance of sociology here, as I have not really engaged with it since I left it behind for the questionable benefits of political science and economics. I don't even know what sociologists argue about with each other. Maybe they have transcended disagreement.

The Biter Bit

I have been reading about electoral systems in the Middle East. These are typically used by regimes to mimimise the amount of seats won by the opposition in elections (usually in conjunction with other measures, like throwing opposition activists or candidates in jail, banning some of the opposition parties, or straightforwardly rigging the election results). My current favourite regime manoeuvre is one used in Egypt at one point. They had a list system of proportional representation, but with a high threshold of 8% - any party getting less than that proportion of the vote got no seats in the parliament (this at a time when the opposition was rather fragmented). The stroke of genius was a ruling that any votes for parties getting less than the threshold were to be counted as votes for the goverrning party.

Turkey either is or is not in the Middle East, depending on who you talk to. They are having some interesting political stuff at the moment. The Turkish Prime Minister Recep Tayyip Erdogan wants to make Abdullah Gul, one of his party colleagues the country's president. As the president is chosen by the parliament, and as Erdogan's Justice & Development party has a handsome parliamentary majority, there should be no problem here... except that Justice & Development is an Islamist party, and Turkey's army does not like the idea of an Islamist president and prime minister.

Some have apparently taken to complaining that although the governing party has a massive majority in parliament, they achieved this on on something puny like only 30% of the popular vote. Amusingly, they managed this because the country has a comedic electoral system specifically designed to weaken the Islamists.