I didn't actually vote for people

I tried, but MMP doesn't let you vote for people. Reflections on what I was doing instead, and my reluctant new voting philosophy

My dirty little secret is that I never actually did it. Six years of preaching voting for people, watching way too many interviews, forming obsessively detailed opinions of every politician who’s regularly on telly. I could tell you exactly who’d be in my dream Cabinet (cross-party, of course) and whom I’d boot from Parliament. And yet when I got to the voting booth, none of those performance evaluations made it into my vote.

The roadblock with “voting for people” in New Zealand is kind of obvious. In our MMP system, we don’t vote for people—we vote for parties.¹ Once I have my list of favourite and least favourite MPs, what do I do with it?

I imagined I’d run some sort of weighted average of each party’s list or something. But there are several problems with this:

1. How far down the list do I go? The minor parties have no hope of getting 30 MPs, so it makes no sense to take their 30th candidate into account. Do I guess their “relevant candidates” based on current polling?

2. Conversely, the first five on each major party’s list are virtually guaranteed to get in, and any (non-drastic) shift in polling won’t change this. Do I take them into account at all?

3. Taking this further, maybe I should just focus on the marginal MP, that is, the MP that on current polling would be the next to gain or lose their seat from the list. But because all parties have fantastic and lousy candidates on the same list, this would make my vote incredibly sensitive to current polling. And imagine trying to compute which list position that is! You’d have to take into account electorate predictions, most of which aren’t publicly polled.²

4. Actually, contrary to (2), the frontbenches are affected by votes—not for whether they’ll get in, but whether they get to be ministers or have to sit in opposition. And in practice, parties often speak through their frontbenches. Arguably, our votes are to decide how many backbenchers each frontbench should be backed up with. So maybe I should consider the top-of-list guaranteed entries after all.

5. Taking that further, how much weight should I give their leaders, in election campaigns that are increasingly “presidential” in nature?

There is no satisfying way to deal with all these questions.

One might even say that my voting philosophy goes against the very fabric of MMP. The whole point of a proportional system is to put parties at the centre. My obsession with individual candidates would make sense in FPP, or even STV³, but not here. Fifteen years of being an MMP fan, and I don’t even think like an MMP voter!

Photo by Jordan McDonald on Unsplash

⁂

It’s always questionable to extract a coherent voting philosophy from past behaviour. But if I were to pick one now, I’d say: For the last four elections, I voted for the party that I felt was most underweighted according to pre-election polling.

This is underspecified. What did I mean by “underweighted”?

I think I made some assessment of parties making “contributions to Parliament”, and on that basis had some idea of what my ideal “weightings” for each party in Parliament would be. (Before I eschewed policy-based voting altogether, this helped me get around the fact that I disagreed with all parties on at least one major policy.) Then I’d vote for the party whose current polling was the furthest below that ideal weighting.

That’s fuzzy as. Let me explain.

Over the last two decades, Act has see-sawed between being a respectable party of principle and an embarrassment to New Zealand politics. Its lowest days were between 2011 and 2014, when its only MP was John Banks. (After the 2011 election, an Act-aligned friend of mine described the party as “existing in name only”.) But after it appointed more honourable leaders—first Jamie Whyte, then David Seymour (hold your horses, I have more to say in a bit)—who espoused honest, judicious applications of libertarian principles, I felt they deserved more than one MP. I considered this to be an ideological mismatch, personally—I have some libertarian tendencies, but not nearly to the same degree as Act. But I thought Parliament as a whole would be better off for having their ideas in the room, so I voted to help make that happen. Then, in 2020, they started polling as high as 8%, so I decided they no longer needed me.

(After 2020, Mr Seymour’s tendencies changed. His popularity grew and his principles waned—I’m not sure which way round the causation was—and the Act of today flirts with populism on the regular and routinely conflicts with libertarian principles in confusing ways. It appears that for every one of me they lost they gained fourteen others, so they can have those fourteen others instead.⁴)

That year, 2020, it would’ve been hard to make the case that any incumbent party was underweighted. Parliament was already down to five parties, the fewest since we adopted MMP in 1996, and polling suggested it might well go down to four. (This was at NZ First’s expense, and I would not miss NZ First, but the trend of having fewer parties was nonetheless concerning to me.) The Greens and Act were both polling at 6–8%, which I considered to be pretty comfortable. Worryingly, one party was on track to get (and in fact got) the first single-party parliamentary majority in New Zealand’s MMP history.

The new kid on the block was The Opportunities Party, contesting their second election. (They’ve now rebranded as the Opportunity Party, or just “Opportunity”, but at the time they were often called TOP.) Here’s what I said about them at the time:

Part of TOP’s sales pitch is that their policy is “evidence-based”, which you would assume would appeal to someone like me. But their manifesto is surprisingly scant on evidence. […] The better case for TOP, and the case towards which they’ve pivoted for the campaign, is that they’re willing to take bold steps to tackle long-standing problems that the major parties have been too feeble to undertake.

They weren’t getting much media attention, which made it hard to assess them as thoroughly as the incumbents. But their ideas and ideals seemed laudable enough to be a worthwhile addition to Parliament—and you cannot get more underweighted than a party that isn’t even there.

So I ended up ticking their box in 2020, and then again in 2023 on nearly identical considerations. I wouldn’t have described myself as an enthusiastic voter—it was more, I couldn’t find a better use for my vote.

The snag is, I could only do this by unravelling my idea of “voting for people” at least a little bit:

Do I apply the same bar to TOP that I do to their incumbent rivals? Its candidates would doubtless chorus “yes”, believing they can meet it, but I think the answer is more subtle. Being a party outside Parliament is hard — no new party, other than splinters, has entered Parliament since the start of MMP in 1996. […] So as new entrants, it’s their potential I’m looking for, rather than their track record.

More bluntly: I was willing to hold their people to lower standards because Parliament as a whole would have been the better for having them in it.

⁂

I never understood what parties meant by “values”. Friends of mine would sometimes cite values as their reason for joining a political party, and I didn’t understand that either. Values often seemed attractive but vague: “equality”, “dignity”, “freedom”. They rarely offered meaningful guidance on which policies to choose. Where they did, parties never stuck by them consistently, and with good reason: all values that are meaningful enough to discriminate between policies should only be applied some of the time.

(Example: National’s values include “competitive enterprise”; Labour’s include “co-operation, rather than competition” in economic relations. Which is better? It depends! In building standards you want co-operation, to prevent a race to the bottom. In hospitality you want competition, because that’s what drives vibrancy and variety.)

TOP also had values. They wanted to move past the ideological left vs right divide, propose “evidence-based” and “common sense” policy, and (from 2019) put forward bold ideas that established parties lacked the courage to consider.

Did they live up to those values? Not always. I critiqued the party’s universal basic income policy in 2020, not for being wrong, but for not being evidence-based. There couldn’t be much evidence; no one else had tried it! Rather, it was bold—a different value of theirs.⁵ TOP’s adherence to values was about as elastic as every other party’s.

This year, the renamed Opportunity Party has pivoted its messaging to hammer on about values more and policy a little less. Their new leader, Qiulae Wong, is trying to tap into frustration about stagnation and flip-flopping between National and Labour. She talks about “leaving ideology at the door” and points fingers at “tribal politics”. It seems to be working. They’re polling on the verge of 5% for the first time in their history.

I’m embarrassed to say it, but I’ve been paying more attention too. At a public meeting, someone asked why a party with evidence-based roots had shifted its pitch to be more feeling-driven rather than analytical. Ms Wong explained that they had tried analytical in previous campaigns, and it hadn’t got them very far. This time they’re trying something different, and all their engagement metrics are higher than they’ve ever been. This is maybe disappointing, but it’s hard to argue with it. At that moment, I realised I’m part of this statistic, and probably for the same reason as everyone else: even for a wonkish technocrat like me, values strike a chord in a way that details don’t.

It was like finding the missing piece. It wasn’t that values are hollow. I never liked values because there was no party that held mine. I don’t subscribe to left-wing or right-wing ideology as values. I think those ideas sometimes work and sometimes don’t. I worry that “left-wing” or “right-wing” groupthink blinds you to more practical solutions. I think it takes work, dialogue and some experimentation to figure out what policies are best. In some sense, all that is a value, just not one that sits neatly on the conventional political spectrum. I sometimes think of it as a meta-value—a value that sits “above” more substantive values—but it’s a value nonetheless.

And once a party started preaching those values, everything else fell into place.

⁂

Writing this now, I’m reminded of all the times I heard, “I don’t agree with all the things my party does, but…” and I’d wonder, just how much would you have to disagree with before you stop aligning with that party?

I imagine everyone has a breaking point somewhere, but I think I now have a better idea of why this isn’t a live question for the party faithful. There’s a lot I have to criticise about Opportunity (a topic for another day). I think several of their policies don’t live up to their ethos. But it seems like I’m willing to trade that away for a party that wants to do politics the way I think politics should be done.

I feel a bit dirty writing that. And all my worries about partisanship remain. I continue to be irritated by groupthinky partisans who paint their opponents as uniformly evil. It continues to baffle me how some partisans seem to swing their opinions on policy whenever their party does. But I also have friends in parties who aren’t like this—who do have minds of their own—and after a lifetime of not getting it, I feel like I’ve finally caught a glimpse of how they reconcile their own thoughts with adherence to a party.

I should take a moment to say that “voting for people” isn’t entirely lost. Ms Wong speaks more eloquently in interviews than any of her predecessors—and as thoughtfully as my favourite MPs—and that’s almost certainly helped drive my confidence in Opportunity. At the same time, the most noteworthy thing about her performance is also an illustration of values: holding the line on finding common ground even as other parties try to drag hers into partisan mud-slinging.

And so I resign myself to the next phase of my voting philosophy. On the one hand, I fear becoming exactly the thing that I can’t stand—a partisan who gave up the ability to think independently. On the other hand, perhaps a new appreciation for a side of politics that I had always shunned awaits.

1. In MMP, we have two votes: a party vote and an electorate vote. The latter is for a person—candidates to be your local MP—but this is widely considered to be the less important of the two votes (except in some edge cases, which don’t include my electorate). I do assess candidates as people for my electorate vote, but this post will only be about the party vote.

2. Assuming you have the polling data for it, you’d have to (1) compute how many seats each party would get proportionally, (2) predict who would win each electorate, (3) remove predicted electorate winners from the party list and (4) figure out how many list seats each party would get to bring list + electorate MPs up to their proportional totals. The actual mechanics aren’t hard (it’s just how MMP works), but doing this on predicted winners—with all the uncertainty, and with some swing seats being close—would be a mess. Normally, political commentators in New Zealand don’t think about this too much, because unlike in FPP systems, it doesn’t affect the overall party composition in Parliament. But it matters a lot for identifying the marginal MP in each party.

3. I continue to hold that STV is not a proportional system.

4. Act’s share of the party vote went from 0.50% in 2017 to 7.59% in 2020.

5. I’d even suggest that there’s a fundamental tension between evidence and boldness: bold ideas are ones that haven’t yet been tried much, and therefore have a limited evidence base.