What Good is MMP Anyway?

November 11, 2009

This is the next in several posts on MMP and the debate around whether it should be retained or done away with in the referenda to take place over the next few years. Here I ask: What good is MMP anyway?

The traditional wisdom on proportional systems versus majoritarian systems (or MMP vs. FPP) is that proportional sytems are ‘nicer’ and more inclusive, whereas majoritarian systems are more effective. The logic behind this is pretty straightforward: The more parties in government, the more people get to have a say; with only one party in government decisions get made in a quick, decisive way.

But does this logic hold? Actually, in 1999 it was blown out of the water. Political scientist Arend Lijphard categorized 36 countries as consensual or majoritarian, but he also asked the ‘so what?’ question, and he found that on the whole the traditional wisdom was unfounded: there is no trade-off between quality and effectiveness of democratic government.

1. Economic development is at least as good under proportional (MMP) systems as under FPP. Successful macroeconomic management needs a steady hand as much as a strong one, and MMP systems are in fact better at managing inflation than FPP systems (though this is probably because they are associated with independent central banks).

2. Consensual systems like MMP are associated with less political violence like riots.

3. Consensual systems provide for better democracy: Higher voter turnouts, higher representation of minorities and women, more equal distribution of political power and less corruption.

4. Consensual systems are ‘nicer’: they have less people in prison and are less likely to use the death penalty, they have better social welfare systems, they are better at protecting the environment and provide more foreign aid.

Obviously you will have to read the book and look at the statistical analyses, but taken at face value these conclusions rob opponents of MMP of one of their main arguments: We can have our cake and eat it too – there is no trade-off between inclusiveness and quality of government and effectiveness or economic growth.


Who Will Stop MMP Reform?

November 5, 2009

To continue the analysis I started in my previous post on the apparent unpopularity of MMP in NZ and its possible demise I ask: Who will try and stop a move back to First-Past-the-Post electoral procedures? Let’s take a purely utilitarian approach to this question.

In Parliament, the opposition must surely be opposed to a government which seems to seek to change the electoral ‘rules of the game’ to its advantage. And I believe the Labour party will oppose a move away from MMP, ostensibly for the reason I mentioned previously: Labour has traditionally had more allies with which to form coalitions under MMP, and can thus consider the current system to be to its advantage.

Because the matter will be decided by a public referendum, one could assume that Labour supporters and supporters of all small parties would vote against a move away from MMP and as a bloc form a majority. National did not win a clear majority in the last election after all.

However as the last post showed, not only National supporters are advocating a change back to First-Past-The-Post. Why? And without a Labour-Small Party bloc who will support MMP?

I suggest that National’s success in wooing the Maori Party and United Future has changed some Labour supporters’ view of the ‘rules of the game’ of MMP in that they no longer see themselves as a priviliged player under MMP compared to National. If National can also form coalitions with three or more smaller parties, what are the benefits to Labour from MMP (over and above those for National, it goes without saying)?

There are none. Labour supporters are therefore put in a position the same as National supporters were up to now: they have no reason to support proportional representation when they could govern alone under First-Past-the-Post (except ideational reasons which I may look at later). By this reasoning, if voters vote according to their parties’ interests at the referenda only small party supporters will support PR and MMP should be shown the door.

How to explain politicians’ continued support of MMP? They are hedging their bets, maintaining their popularity in the eyes of smaller parties, who they will have to work with until 2014 at least, and perhaps beyond if the referenda fail.

There is a very interesting poll on the NZ Herald website today which shows most New Zealanders would like to change their electoral system away from MMP. 49% said they would vote against MMP at a referendum, 36% said they would vote for it and 15% said they didn’t know which way they would go. There is to be a referendum on whether to keep MMP at the next general election.

To compare these figures to the results of the election a year or so ago: Then only 14% of voters voted for a small party – that is not for National or Labour (but excluding those who did not make it into Parliament). So MMP obviously has support outside its core demographic of voters who do not subscribe to the two main parties.

However, a larger proportion of respondents are against MMP than voted for National at the last election: 49% versus 45%. I find this a little surprising, as my first instinct on MMP reform was that it was just a case of the victors being able to change the rules of the game to ensure their continued electoral success: National has always had a lesser number of willing coalition partners than Labour and thus would prefer a return to FPP. (Yes, I maintained this belief despite John Key’s sincere utterings that most NZers want to stick with MMP – which have been proved to be wrong).

But this result seems to indicate something a little deeper than that: It is not just National, but also Labour, supporters wanting to get rid of proportional representation. The flip-side of the 14% of people voting for smaller parties are the 79% of people voting for either Labour or National, as shown below.


As this graph makes visually clear, if supporters of the two main parties decide that they no longer want proportional representation, and would rather govern alone without coalition partners, they will always be able to push this through.

This is, of course, a simple case of majority rules and exactly what MMP was brought in to prevent. Should the change to MMP therefore be regarded as an anomaly; a short period where the majority were so disillusioned with politics that they were prepared to allow minorities a more significant say, which has now passed?

Of course we will have to see what the results of the actual referenda are, but this initial poll result would suggest that supporters of both main parties are a little more cold-blooded in their attitudes towards minorities than one might have expected.

Three weeks after the Bundestag election in Germany and the discussions between CDU, CSU and FDP continue. It is still not clear what the exact policy directions of the new coalition government will be, or who will occupy which ministerial/government posts.

I see an interesting difference in the coalition bargaining between the parties and what has gone on in NZ after MMP elections since 1996.

The German parties are driving hard bargains. Both the CSU and FDP are sticking to their guns and placing demands on Merkel’s CDU that they make concessions so the smaller parties honour their pre-election commitments. There have been several examples of this as the discussions have gone on, for the FDP most notably in measures protecting personal privacy.

But as predicted, the big one sticking point is tax cuts: both smaller parties want them, the CDU knows they can’t really afford them. Merkel doesn’t want to be seen as fiscally irresponsible and run large deficits just to finance tax cuts (and the CSU will not allow this anyway). The result is a sort of Mexican stand-off. This has been going on for weeks. It has finally been decided that there will be tax cuts, the question is how big they will be.

The important point here compared to NZ is that the coalition bargaining is very detailed, and very contested. I am not an expert on NZ’s coalition agreements, but NZ moved away from both these aspects after the first experience with coalition government in 1996.

Then, Winston Peters and NZ First undertook to draw up a coalition agreement which set out most important policy directions in writing. Peters drove such a hard bargain that he was seen to be holding the country to ransom, and the discussions took so long that the country was without an effective government for two months. NZ First ended up going with National, which was the opposite of what they had alluded to pre-election. The coalition didn’t last very long, and neither did NZ First’s (or MMP’s) popularity.

Since this negative experience with detailed, drawn-out coalition bargains NZ has moved to a more flexible approach based on agreements on confidence and supply (settling other issues on a more ad-hoc basis), which would seem to be more in keeping with the country’s Westminster/common law tradition. Helen Clark’s Labour did this with the Greens in 1999/2002 and with NZ First/United Future in 2005. John Key has a similar confidence and supply agreement with ACT, United Future & Maori Party.

The Germans, however, seem happy to persist with longer, more contested coalition bargaining phases in which major policy issues are hashed out in detail after the election. This is more in line with the German tradition of constitutional consensual government, and the German civil law.

MMP: What about Politics?

September 25, 2009

Due to the upcoming referendum on the MMP electoral system in New Zealand, there have been a raft of blog postings on the performance of the present system, what people would prefer in a new electoral system, what the effect of the 5%  threshold is, and different types of electoral system.

These all make for interesting reading, if you are interested in arcane electoral mechanics, mundane spreadsheet calculations and the functioning of democracy in obscure countries (like NZ).

However they all miss one aspect of the debate around MMP, and that is Politics, or ‘Who gets what, when and how’.

They seem to assume that NZ politics will be solely defined by its electoral system and miss other things such as public opinion, political parties, interest groups, political culture, the list goes on.

It is therefore worthwhile asking: Would having a different electoral system change NZ politics? The answer is, not very much. The changes would be on the fringes: with perhaps slightly higher representation for smaller parties in Parliament, and perhaps the creation and representation of a new party or two.

Due to the inate tendency of party policy to tend to the middle, it can be safely assumed that the majority of the votes cast will be gathered by two large parties who battle it out for the chance to form a government.

Sound familiar? Yep, it sounds like democracy pretty much all over the world. That’s the beauty of not examining an electoral system in isolation but in its context, Politics.

Remember back to before NZ’s November General Election and the scrutiny of National’s leader and future Prime Minister John Key? He was labeled as ‘Labour-lite’ and centrist; his party was careful not to wage war on some of Helen Clark’s more popular policies; the differences between the two main parties were small, negligible even.

And for those who follow US politics – you must have been struck by the partisan divide in both society and politics between the Democrats and Republicans. It is, in the words of a columnist, ‘The 50/50 Nation‘. Heck, the two camps even have their own separate news channels!

Over in Germany, people ask themselves: What is the SPD doing in opposition? (oops, sorry, the election isn’t for 2 weeks. I meant, Grand Coalition.) They could form a government with the Greens and the Left Party tomorrow!

Discussion of all these issues, and some others which I have speculated upon here previously, can be informed by a powerful, but simple, political science model (originating in economics) called the Median Voter Theorem.

Briefly, the theorem is a model of majority decision making which states that the median voter’s preferred policy will beat any alternative in a vote; in a stronger form it also says that the median voter will almost always get his/her most preferred policy. The rationale behind it is too much for this blog and is explained briefly and clearly here, but the consequences are far-reaching.

Median_voter_modelIf we apply the model to the Left-Right spectrum above, we can come to the following conclusions:

1) if voters cast their vote for the candidate closest to their ideal policy, then the candidate who stands nearest to the median voter will win. In this case, this would be candidate A, who is closest to point M.

2) if the candidates are free to choose their policies, they will choose positions which will place them closest to the median voter’s at M; so candidates A and B will move in the directions of the arrows towards the middle of the graph and the median voter’s preferences will be fulfilled.

The theory puts the median voter in the lucky position of always getting what they want, but it also has other consequences – many more than will be expanded on here, but to go back to the examples above:

Key vs. Clark in NZ: parties will choose policies which are moderate and middle of the road; no wonder that they look so similar.

Democrats vs. Republicans in the US: if parties continue to act as above for many years they could create blocs, each accounting for approximately half of the population (see the column from Mickey Kraus).

The SPD and the Left Party in Germany: parties will be very careful not to take extreme positions which could alienate voters closer to the median.

So there you have it. A brief but hopefully not-too-incomplete rundown of the Median Voter Theorem in five minutes. A simple, powerful analytical tool for analyzing party politics.

This is the third post on this topic where I first asked: Is NZ’s MMP system Mature? I decided this wasn’t a very fruitful line of enquiry and asked the more interesting question: What type of Democracy is NZ?

This was a little more informative – it seems that through the introduction of the MMP proportional voting system NZ’s democracy has been split between a Majoritarian system and a Consensual system, with some parts belonging to one category and some to another. In my mind this could be a large part of a disconnect within NZ’s political system  and perhaps part of the reason behind continuing discontent with MMP.

But I think rather than simply looking at the political system, it could also be interesting to look at broader NZ society and how it interfaces with it. In other words, does NZ’s political system fit its society?

A lot of things which I originally intended to mention here were actually picked up by Lijphart and are therefore in the previous post: Interest group mediation (Majoritarian); Lack of a written constitution/constitutional court (Majoritarian); Independent Central Bank (Consensual). Although I am not a sociologist (and would value a contribution from one here very much) I will take a rough stab at some other aspects of NZ society which I can think of and might tell us something about how it fits with the political system.

1. Political Socialization: With this I quite simply mean the way large groups of people in NZ grew up thinking about politics. For a lot of people still, this was under a Majoritarian, two-party, first-past-the-post situation which undoubtedly still shapes their thinking about politics.

2.Minority & Ethnic Groups: I would make the cautious assertion that here NZ society tends towards the Consensual. Basically, the more homogenous the society the better suited it is to a Majoritarian system. Interests do not diverge substantially from group to group and today’s minority can easily become tomorrow’s majority, leading to balanced public policy. I would cautiously say that NZ has, firstly, always had minority groups (eg Maori) represented in Parliament and well-organised out of Parliament. Secondly, I would add that NZ society has, especially through immigration, become more ethnically diverse over time. Both these points speak for NZ society fitting a Consensual democratic system better than a Majoritarian one.

For lack of other aspects which occur to me I will end the analysis here and confess that I do not see a clear pattern emerging. A lot of institutions in NZ are fitting with a Majoritarian democratic system. However, some are closer to a Consensual system.

As for society, my analysis above is too short and not detailed enough. However, with time the generations who grew up under the purely Majoritarian system will pass away. And with further immigration NZ society will become less homogenous and therefore less suited to a Majoritarian political system.

It seems that at the moment NZ society may be split, like its political system, between a more Majoritarian and a more Consensual system. As time goes on it may move to being better suited to the Consensual model.

I finished my previous post by stating that there is a disconnect between NZ society and the system through which it elects its parliament and I said the disconnect was that NZ society is pluralist and majoritarian, and the electoral system is consensual. This idea needs to be more fully developed.

As we all know, there are different types of democratic electoral systems all over the world. Prominent political scientist Arend Lijphart has looked at them and he sees two types of democracies: Majoritarian and Consensual.

Majoritarian Democracies are those like in Britain which are also called Westminster democracies. They are characterised by first-past-the-post voting, two-party dominance of parliament, powerful Prime Ministers/Cabinets, and unitary (not Federal) governance.

Consensual Democracies are like those in Germany and are characterised by proportional voting systems, multi-party coalition governments and federal governance.

So if there are these two types of democracy, which does New Zealand fit into? Lijphard actually addresses NZ in his book Patterns of Democracy and for him it was the best example of Majoritarian Democracy, until the introduction of MMP in 1996.

Since then the picture is a little clouded. Perhaps you might have thought NZ had swung all the way over to a Consensual Democracy, but that is not the case. Lijphard works on 2 Dimensions: Executive-Parties and Unitary-Federal.

1. Executive-Parties Dimension: Here NZ seems to be on the Consensual side. Since 1996 we have only had multiparty coalitions in government; we vote according to proportional voting rules; in the end, it is not easy for one party to take control of government.

The only place where we diverge is in interest group mediation, where NZ is more pluralist than corporatist, i.e. interest groups agitate on their own behalf for influence and do not reach compromises within their own institutions.

2. Federal-Unitary Dimension: On this one NZ would seem to be more on the Majoritarian side. We do not have independent federal states with wide-ranging competence and independently elected parliaments; we have no independent upper house of parliament; we have no written constitution and no constitutional court. In effect, once in power the government has great control of actual policy.

Here we diverge only in that we have an independent central bank.

So it looks like since the introduction of MMP New Zealand has gone from a majoritarian democracy to a bit of a mixture between majoritarian and consensual democracy.

Or, if we look at which aspects have changed from majoritarian to consensual we see that only those factors relating to the electoral system have changed: Which parties are represented in Parliament; How many parties are in Government; How people vote in elections. Everything else has remained the same, staunchly Majoritarian.

This seems to me to be part of the disconnect: Half the political system has been wrenched from Majoritarian to Consensual Democracy while the other half has continued on as before.

A ‘Mature’ MMP System

August 16, 2009

The MMP electoral system adopted by New Zealand in the 1990s is even now not without controversy. There is some speculation in New Zealand that voters are yet to come to terms with their new electoral system and that NZ therefore does not yet have a ‘mature’ MMP system. Often, rates of vote-splitting (between parties) and ‘strategic voting’ in NZ MMP elections are taken as positive indicators of voters becoming more familiar with the system, which itself would then be ‘maturing’ through this process (from memory I attribute such statements to Nigel Roberts, VUW).

Hence one could conclude that there are deep changes to come for New Zealand politics as the public adapts to the proportional representation system. I hope that some of the previous posts on the German electoral system (60 years MMP and counting) and the different factors influencing electoral success might have gone some way towards informing such a debate. Their main points were:

1. There are glaring differences between which parties are represented in the Bundestag and the NZ House of Representatives. Some (esp. the lack of a kingmaker party in NZ) are meaningful from a NZ perspective.

2. However, there are different factors influencing parties’ success: Electoral systems, Socio-economic structures and political culture.

The debate over whether NZ has a ‘mature’ MMP system, or is used to its new way of voting, is therefore of little importance. The electoral system in a country is only one factor influencing the country’s politics, and because of this the German example does not offer a set path which NZ is bound to follow (as we have seen in previous posts). Debating whether the MMP system is ‘mature’ or not does nothing to predict future political trends, and says very little about how well the electoral system articulates the preferences of the NZ population.

Far more important are the underlying social, economic and cultural trends which politics inevitably follows. Electoral systems will change as societies change; in Germany for example, they are currently debating whether they should do away with the five-percent-threshold to allow smaller parties into the Bundestag.

In particular, looking at whether voters split their votes between parties at MMP elections does nothing more that ask whether voters understand the electoral system within which they exercise their democratic right. It says little about voters’ actual interests and how they should be articulated in Parliament. This is where any real change would, and should, come from – rather than from spontaneous further electoral reform to skew party representation one way or the other.

A continuing discontent with MMP amongst voters and politicians, or what some commentators see as a lack of ‘maturity’ on the part of New Zealand voters, is not a lack of understanding the NZ electoral system, but a symptom of a more fundamental and long-lasting disconnect, concerned with the other, deeper factors influencing electoral success: Socio-economic structures and political culture.

It is the problem of trying to fit the ‘square peg’ of a pluralist, majoritarian society into the ’round hole’ of a consensus-democracy electoral system.

I return to the comparison of which parties are represented in the German and New Zealand houses of parliament and seek to draw some conclusions about what influences party representation in Parliament. Drawing on my previous posts, and bearing in mind that this is just a blog and not an academic paper, I put forward three possible factors influencing party success in elections:

1. Electoral System. A very important factor in a party’s success is the rules under which it is playing the political game. Just ask the Greens, who were never represented in the House before the introduction of MMP but have been ever since.

2. Socio-Economic Factors. According to a typical left-right spectrum, this should be the key influence on party formation and influence, with socio-economic groups organising themselves politically in order to represent their interests in parliament. Perhaps the deciding factor behind the lack of a powerful extreme left party in NZ.

3. Political Culture. This is a very difficult concept in political science: Trying to define political culture has been described as being like ‘nailing a pudding to a wall’. However it is a concept which has lasting influence and appeal. It used to be referred to as ‘socialisation’ – the way that people were initiated into the way politics is done in their society. Phenomena like distrust of  ‘Kingmaker’ parties in NZ, caused by negative previous experience with them, can only be put down to something like political culture.

And as for which factor is most important: As you can see from the brief examples above, they all seem to be important – in different ways and at different times.