Monthly Archives: May 2012

My MMP Review submission

Since the majority of people voted to retain MMP in last year’s referendum, they’re keeping it but they’ve launched a review into how it could be made better.

Public submissions are open until the 31st of May, which is ten more days as of today.

In a spate of procrastination I’ve written one, I didn’t take too much time but I think it represents what I think pretty well.

My submission is below.  You can do your own one (either a brief statement on one or two issues, or a longer document) at

Caleb Anderson – MMP Review Submission


  1. Basis for eligibility for list seats (thresholds)
  2. Overhang
  3. Proportion of electorate seats to list seats
  4. Other issues – By-elections and how they affect total party numbers


1. Basis for eligibility for list seats (thresholds)
Summary: The 5% threshold should be removed or significantly lowered – perhaps just enough for two MPs.  The electorate threshold would be less necessary if the 5% threshold was lower.

We still have a ‘two-party’ mindset to a large extent, which is perhaps inevitable when MMP is so new.  Currently the two biggest parties have an unfair advantage because they’re the only two who are actually guaranteed to be represented (recently the Greens have seemingly achieved this security also, but no other minor parties have managed to maintain a guaranteed presence).

This uncertainty about crossing the 5% threshold is also demonstrated in the way that minor parties (and allied major parties) are strategically using local seats to guarantee minor party representation when they cannot guarantee crossing the 5% threshold.

Because of the uncertainty of parties crossing the threshold (or winning a local seat), this leads to artificial contortions of party vote results, where recent polls affect the results far too much.  For example, in the last election ACT was not certain to win Epsom, so they may have lost voters worried about vote wastage.  NZ First polled just below 5% just before the election, so they may have gained voters who wanted to make sure they made it up over the line (and they ended up polling far higher than 5%).

The very idea of a threshold is arguably a symptom of a ‘major party’ mindset, a mindset which says that lots of small groups of people (or independent individuals) having their say is somehow detrimental to democracy, and that it’s better to have a few big groups controlling things.  Even if you do decide that it’s good to have some kind of threshold in order to keep things simpler, the experience of the last 16 years shows that 5% is far too high for New Zealand at this time.

The electorate seat threshold (the rule that when you win an electorate seat you can bring other members of your party with you) is also problematic.  The 5% threshold combined with the electorate seat threshold gives an unfair advantage to those parties who are able to secure a local seat (Maori Party, United Future, Mana, ACT, formerly Progressive Party) compared to those minor parties who cannot win a local seat but do have a significant amount of support across the country (Conservative Party, NZ First in the 2008 election, Values Party and Social Credit before MMP).

I tentatively support the electorate seat threshold currently and would not support removing this if the 5% threshold remains the same.  However, the electorate seat threshold would be unnecessary (or less necessary) if the 5% threshold was lowered or removed.

If every significant minor party could be guaranteed to meet one of the thresholds (ie: guaranteed that their vote would not go to waste), this would make party vote better reflect people’s actual preferences, and avoid giving unfair advantage to those parties whose presence is guaranteed or almost guaranteed.


2. Overhang
Summary: Difficult to know how to make it more fair.  Perhaps have ‘electorate-only’ MPs who cannot vote on issues outside the specific concerns of their electorate.

I tentatively support overhangs as I think they are more fair than taking MPs away from a party whose party vote earned them.

However, if there is a way of stopping overhang MPs having more influence than their party’s party vote earns them, I would support this too.  Perhaps ‘overhang MPs’ should not be allowed to vote on general issues, but ONLY issues directly affecting their electorate.

For example, if Peter Dunne regained his electorate seat, but United Future didn’t get enough of the nationwide party vote to justify one MP, Peter Dunne would not be allowed to debate and vote on national issues, only specifically Ohariu-based issues.  Or if the Maori Party wins three electorate seats but only enough party vote for two seats, only two can vote on ‘national’ issues and the third can only vote on what directly affects his/her electorate.

This kind of rule would avoid the entire country being affected by people who do not have support across the country but only people in one local area such as Ohariu.  It would also avoid major parties effectively earning an extra MP by conceding a local seat to member of an allied tiny party which has no hope of gaining a second MP (therefore most party votes will go to the allied major party rather than to the tiny party).

At the same time, hopefully this rule could be designed in such a way that it avoids crippling Maori representation, because electorates are populations of people.  An ‘electorate-only’ Maori MP could vote on all isues that affect Maori in his/her region, which would probably be a wider scope than a local MP in a non-Maori seat.

However, this kind of rule would be extremely hard to enforce – where do you draw the line between issues affecting your electorate, and national issues?

Perhaps an electorate-only MP would have to seek permission from the Speaker to speak and vote in his/her capacity as an electorate MP, explaining briefly why the issue specifically affects his/her electorate.  This could be a simple verbal process, but one which the Speaker could refuse if he/she felt that the request was unfounded.


3. Proportion of electorate seats to list seats
Summary: Perhaps a lower proportion of electorate seats to list seats; but only if local representation is kept the same or improved.

Some of the overhang issues could perhaps be avoided by having a lower proportion of electorate seats to party vote seats.

In principle I support local representatives having a significant say in national politics, but at the moment I suspect that the ‘local representation’ component of electorate MPs’ activity is dwarfed by national issues and party lines (even for MPs who are only in Parliament by virtue of winning an electorate seat).  Perhaps there could be a better way of achieving local representation in Parliament.


4. Other issues – By-elections and how they affect total party numbers
Summary: After by-elections I think seat allocations should be based on what would have happened if the general election had gone the same way as the by-election (especially if it’s in the first year since the general election).  Currently there is a loophole because by-elections bypass the party vote.

MMP is supposed to avoid parties earning a greater proportion of seats than their party vote deserves (with the exception of overhangs) by pegging total MP numbers to the party vote (regardless of how many of that party are electorate MPs and how many are list MPs).  However there is a loophole in by-elections, which still function based on a FPP system.  If I’m not wrong, whoever wins the by-election wins not only the local seat, but also adds one to their total party allocation, regardless of whether they deserve it based on their party vote.  This is an occurance that MMP prevents in general elections, but not by-elections.

Because of this I think that in by-elections the last general election’s party vote should come into play, especially if the general election was relatively recently.

In the case of a hypothetical Epsom by-election, for example, if a National candidate wins, I do not think National should win an extra seat no matter what.  I think the seat distributions should be determined based on what would have happened if National had won the seat at the general election a few months ago.  In this case that would mean three things:

  • ACT’s 1% of the party vote would be ‘wasted’, as would have happened if John Banks had not won Epsom in the general election
  • National would gain an electorate MP, but at the expense of a list MP
  • Obviously there needs to be an MP gained somewhere to make up for the MP lost, but this would go to the party whose party vote put them closest to earning one more.  In this case, according to my calculations it would be Labour.