I think quite a bit about electoral law and electoral systems, and our MMP system in particular. So I'm somewhat surprised to find myself wavering over a number of the aspects of MMP that the Electoral Commission review will address. Hopefully the upcoming posts will help me form more considered views.
And, of course, thinking about these things, and seeing what others think, will hopefully encourage a few of you to make your views known as well. Which you can do here :-)
The review addresses a number of aspects of the MMP system: the two thresholds, dual candidacy, the rules for ordering party lists, the overhang, whether list MPs should be able to contest by-elections, the decreasing number of list seats (and whether this may present proportionality problems).
We've also got the option to raise others (as long as they relate to MMP, and not broader matters that could arise under any electoral system). The issue of whether we should retain the Maori seats is left to the constitutional review, along with (mostly) the number of MPs, but a few other issues are likely to get a look in: the ratio of list MPs to electorate MPs will be raised by a lot of people, and a few others I could probably never think of.
This first post looks at the threshold. Of all the issues clearly up for review, this is the one I've found easiest. I've known my views on this for a long time. My view on the electorate seat threshold did change not too long ago, but on the party vote threshold, I can't ever remember thinking differently. The arguments have changed, but my view has always been that the threshold should be as low as possible, and that we aren't even close to having the balance right.
Perhaps more than any of the other aspect of the MMP review (although open lists comes close), the debate over the thresholds is likely to bring out our political parties' partisan interests. So one thing I fear, is that the discussion on whether to change the threshold will become a debate between the current 5% threshold, and the 4% the Royal Commission recommended. On details like those we're discussing as part of this review, the status quo – or something close to it – almost always benefits the parties (and the MPs representing those parties in particular) more than anyone else.
The only difficult question I'm presented is whether to push for the complete absence of an artificial threshold, or to propose that it be kept, but lowered, as a more likely compromise.
New Zealand is among the longest-standing democracies in the world. We have had a functioning party system for over 100 years, and our two major parties are both well into their second half-centuries. I think that the New Zealand political system is mature enough that we could cope in the absences of an artificial threshold.
It may be that the time we have had with a 5% threshold has helped this, and that the absence of a threshold would have been destabilising at our first MMP election (although perhaps the likely effect would simply have been that one or more of the constituent parts of the the Alliance would have contested the election separately). I see no reason it would be partcularly destabilising now.
Almost inevitably, when the prospect of there being no artificial threshold is raised, someone will point out that Israel has no threshold, and that we don't want our politics to end up as fractured as theirs. My first reponse is, naturally, to point out that Israel has a threshold, and it always has – it started at 1%, was raised to 1.5% and is now 2%.
My second point is that the South African electoral system actually has no threshold, and they're not far short of a one-party state. Israel's fractured parliament arises from its fractured politics, not the other way around, and while a higher threshold in Israel could act to prevent the fissures in its politics spilling over into its Parliament, in countries without that highly fractured political climate, a highly fractured Parliament is unlikely.
When Israel formed, it's first parliamentary election was conducted under proportional representation, with a low threshold, so it never established the type of party system – with two major blocs – that people tend to view as providing stability. However, if we were to abolish, or substantially lower, the threshold, we would take our history of stability forward with us.
Supporters (and especially members) of National and Labour (and other parties too) who fear a fragmentation of our political system with a low threshold should be tasked with answering the question: why do you think that your party will disintegrate at the first opportunity?
And naturally, there are advantages to having multiple smaller parties in Parliament. One of the concerns many people have with minor parties is their exercising disproportionate power. But the greatest chance of that happening is when there is only a small number of small parties.
If there are multiple different options for a major party – if National can go with John Banks and Peter Dunne, or instead vote with the Maori Party ... if Labour can get support from the Greens, or New Zealand First, then the power of one minor party grouping is substantially diminished. The power New Zealand First exercised after our first MMP election in 1996 was because neither National nor Labour had a serious options that didn't include New Zealand First.
The Parliamentary Assembly of the Council of Europe has on several occasions considered the effects of different aspects of electoral systems on democracy:
In well-established democracies, there should be no thresholds higher than 3% during the parliamentary elections. It should thus be possible to express a maximum number of opinions. Excluding numerous groups of people from the right to be represented is detrimental to a democratic system. In well-established democracies, a balance has to be found between fair representation of views in the community and effectiveness in parliament and government.
It has also said:
Thresholds ... have a significant impact on the representativity of an elected body. This question of utmost importance has already been discussed in the Assembly on several occasions and the Assembly’s position is clear: in stable democracies legal thresholds over 3% are hardly [ever] justifiable. There is no reason to bar certain groups of citizens (minor parties) from access to parliaments.
We do not presently have the balance right. Whatever the rationale for a threshold (stability, effective government, effective Parliament, or something else) it should not be so high as to preclude a serious attempt to break into Parliament from the outside. And five percent has demonstrably failed in that regard.
It may be that New Zealanders consider themselves reasonably well served by our political options, but a threshold of 5% has proved – and is likely to continue to prove – far too great a barrier to political alternatives. Unfortunately, a 4% threshold is likely to prove little better.
Not only does a threshold as high as 4% or 5% depress the vote, it may also artificially lower the vote of parties who are near the threshold. People who want their vote to count, and who are voting for a party whose support is non-negligible, shouldn't be placed in the position of wondering whether their vote (along with the votes of many tens of thousands of others) will count at all.
I reject David Farrar's comparison of my position to that of religious extremists who spit on children (seriously?). Supporters of the threshold need to do more than point to countries with low thresholds and argue we'll be like them. Not least because the experiences of democracies with low thresholds vary greatly.
Our nation is not historically based on the need to protect a long-maligned religion whose existence was threatened by the Holocaust, nor do any of our neighbours hope for our destruction, so maybe, just maybe, changing one aspect of our electoral system will not magically make us the Israel of the South Seas.
A threshold should be as high as is needed to meet whatever real threat we feel we need to guard against and absolutely no higher. At 5% – and I believe 4% as well – we will be denying democratic voice to people not because we fear for the stability of our political system, but simply because we don't want a few extra voices heard.
We have had a number of parties in Parliament with one or two MPs, and this does not seem to have posed any great stability problems. I can recognise that there is some legitimate benefit to avoiding a proliferation of small parties in Parliament: a large number of single MP parties could make governmental and legislative coalitions difficult to manage, but in effectively banning parties with one or two MPs, there is no need to ban parties that would have five MPs.
I'm putting forward a number for the threshold, and others are putting forward other numbers, and mostly we're just pulling numbers out of the air. I'd like the Electoral Commission to closely consider the pros and cons of the whole range.
What will it mean if we set it at a level where we can have single MP parties? How many are there likely to be? What will the likely effects be on government formation and the legislative process?
If the threshold is at around a two MP level, how well will such parties be able to perform the work we expect of parliamentary parties?
If a three-MP party is in the opposition, how frequently will it get to ask a primary question in the House? How many bills in the areas of legislation important to it will it be unable to subject to real scrutiny?
At what level is a parliamentary party likely to be too small that its MPs will be unable to do a substantial amount of the work we would expect even a small party to be able to do? Perhaps if a party is so small that it can do almost none of the things we would expect of people representing our voices in the House of Representatives, the arguments in favour of thresholds have some meaning.
If all that can be offered to the small groupings needed to elect one (or two) MPs is the illusion of effective Parliamentary representation, it may be legitimate to set a threshold at a level where we can have some confidence those who make it into Parliament will be able to be effective representatives.
Ultimately, the Commission should determine the size it considers a minor party is likely to be effective enough that the rationales for telling its voters that they can't be represented at all fall away.
Even though we've had 5% threshold, we have had parties in our MMP Parliament at every potential size for a minor party: ACT, Mana, the Maori Party, the Progressives, and United Future have been elected as single MP parties; ACT, the Progressives and United Future have had two MPs; the Maori Party and United Future have had three MPs, The Maori Party has had four MPs; ACT, the Maori Party and New Zealand First have had five MPs; and the Greens have operated with six MPs.
Over such a short period, we've amassed quite a useful selection of case studies: of minor parties that have worked closely with others in opposition, and that have worked alone in opposition, those that have sat on the cross-benches, and others been in government. There is the data to assess the contributions they made, the level of representation they provided those who voted for them, and the concessions they've exacted (for good or ill).
Looking at the experiences we've had, you might think, actually, a one-MP party can add to Parliament in a way that our democracy would have suffered were they not there. And if you don't agree with that, you might still look at our experiences with three MP and four MP parties and consider, yes, those parties were able to do a lot on the issues important to those who voted for them.
I want the Electoral Commission review to engage with these sorts of questions, in a way that the Royal Commission simply didn't (primarily because of scope – the Royal Commission made 71 numbered recommendations, of which recommendation 1 was MMP, including all the details such as the threshold). The Royal Commission recommended a 4% threshold (along with the one seat rule, and also having no threshold for parties primarily representing Maori interests), but its argument – at least as articulated publicly – was weak:
Before setting on a 4% threshold, the Commission considered alternative possibilities ranging from no threshold at all to a 5% threshold, as used in elections in the West German Bundestag. We are persuaded that if no threshold is set or if it is set too low, the operation of effective government is very likely to be frustrated. On current voting numbers and assuming 120 seats allocated by the modified Sainte Laguë method, the absence of a vote threshold would give a first seat in the House to every party recording around 25,000 votes. We think this is too low and could give rise to a proliferation of small parties with few seats in the House. The adoption of an appropriate threshold is a key element of our proposal, and we would view it as clearly undesirable to have no threshold . We only think it justifiable to waive the threshold in the very limited way which we indicated above.
On the other hand, we view a 5% threshold as too severe. Under such a proposal a party would need almost 100,000 votes to gain one list seat. In our view this would, in New Zealand, be too great an obstacle to the development of new and emerging political forces.
An only slightly shorter version would be: “no threshold is too low, 5% is too high, let's go with 4%.” Why was 2.5% too low? Why was 3% too low? Or 2%? You can read the several hundred pages of the Royal Commission's report and won't be any wiser. It simply never offered a reason. The Electoral Commission – with it's much narrower focus – is well-placed to do so.
The way that people tend to look at this is to consider the effect on parties: for example, in 2008, the 5% threshold meant that New Zealand First wasn't represented in Parliament. This is a fundamentally flawed way to approach thresholds. I don't care about parties. I care about voters. The threshold wasn't unfair to the New Zealand First Party, but it was unfair to the 95,356 people who gave it their party vote. By having a threshold, and in particular, by having a high one, we are telling a lot of people that they have no place in our democracy, and that their views matter less because they voted the wrong way. In creating a threshold, we are deciding that the voices of some voters just aren't worth hearing. 95,000 voters are enough to given any party 5 MPs, or any party 5 MPs more. That's a lot.
My simple point is that the threshold should be as low as is needed to achieve whatever it is we want to achieve by having a threshold, and absolutely no higher. So I'm stumping for 2.5%. Whatever anyone wants to achieve by having any threshold at all, I consider it will be achieved with a threshold at this level. Any number will have a whiff of arbitrariness about it, but I think this has a bit going for it if we are going to have a threshold.
In a 120-seat House of Representatives, a party which has the support of 2.5% of voters, has fully earned 3 MPs (with rounding, a party could get three MPs with somewhat fewer votes – as little as 2% will sometimes be enough). Although the case can be made (and I'm quite amenable to it) that two MPs is large enough to have a positive effect on Parliament, when we're talking about 3 or more likely 4 MPs (which cuts in at around 2.8% ~ 3%), we're really talking about a significant and useful bloc (of voters, and of MPs). At 4% or 5%, you're saying that a some groupings of five MPs are too small to bother with, and their voters justifiably ignored, which is at least a couple of steps too far.
Now, you can legitimately argue that even a 2.5% threshold is too great an imposition on the the principle that all voters should be equal, and there's something in that, however I'm a pragmatist, and because I think it highly likely that this debate will end up being an argument over whether the threshold should be 5% or 4% and because I think both of these numbers are far too high, I'm happy to compromise.
Guaranteeing that the smallest party (other than one with a strong electorate presence) will have at least three MPs will ensure our politics doesn't fracture unless one of the major parties does something particularly egregious. Each party will be able to sit on a range of select committees, and will be able to cover most of the important bills; the party will have a primary question each week, and a caucus large enough to ensure a party presence in the House, with sufficient support staff to enable them to effectively operate. A party of three MPs is large enough not be viewed as a joke, and large enough to provide a meaningful level of representation to those who voted them into office.
A lot of people do have concerns about our politics fracturing, and while I think that those concerns are exceedingly likely to prove entirely unfounded, it is only proper to recognise that for many, those concerns are legitimately held. The problem with pushing for the total absence of an artificial threshold is that if that argument is lost, and it is decided we should have a threshold, the debate may default to: should it be 4% or 5%?. For me, I'm willing to give up the slight chance of no threshold, for an increased chance of a threshold that is not so high as to stifle political innovation.
So that's why I'm pushing for 2.5%. I don't appear to be alone, which is good, and I'm hopeful for a good hearing. But I recognise that I have no particular claim to possession of the right answer on a matter like this, which is why I consider that it is imperative that the Electoral Commission assess the arguments for an against setting the threshold at a range of levels: none, none with modified Sainte-Laguë, 0.833% (or 1/120th of the vote), 1%, 1.5%, 2%, 2.5%, 3%, 4%, 5%, 6%, 8%, and maybe even a few others.
It should look at what this would mean the smallest parliamentary strength a party could have under each scenario, and how many votes a party would need to get, as well as how many wasted votes there would likely end up being, and a whole host of the other likely effects. If it's particularly brave, it might even like to model what effect these changes might have on enrolment and turnout (something that the triennial select committee review of the general election – which is seeking submissions as well – and it is particularly concerned with the drop in turnout might like to consider). At 2.5% (or lower) we may give many more potential voters certainty that their vote will count.
Of all the issues being considered in the review, getting this one right has the greatest prospect of improving our democracy. At 5% the status quo political parties – government and opposition – are protected. Even at 4% they will probably be reasonably confident too that a new political movement claiming to better represent their core voters will be unlikely to ever sustain a level of support needed to grow over time to be a major political force.
The threat, not of losing an election or two, but of being supplanted by someone else, is what should ensure whatever party we support is true to its supporters. The high threshold we currently have makes this an empty threat.
If you wish to make an oral presentation to the MMP Review, you should forward your written comments by Thursday 5 April. Otherwise, you have until 31 May to let them know what you think.