Despite Judith Collins’ claim, the caretaker convention – the understanding that in certain circumstances significant governing decisions should have a parliamentary majority, or be consulted on by the opposition – does not apply in the pre-election period. And for some perfectly good reasons, writes law professor Claudia Geiringer.

In July 1984, the Muldoon government was defeated in a snap election. The next day, senior Reserve Bank officials advised an urgent currency devaluation in order to avoid a run on the dollar. Prime minister-in-waiting David Lange supported the move. Muldoon initially refused.

Muldoon’s senior cabinet ministers were appalled, and they staged an intervention. Deputy prime minister Jim McLay told Muldoon there was a constitutional convention requiring him to act on the advice of the incoming prime minister. As McLay later admitted, no such convention really existed. But Muldoon swallowed the bait, and backed down.

In that moment, New Zealand’s caretaker convention was born. A convention, after all, is just a constitutional practice that political actors observe because they believe its required of them. (They believe it, so it’s so.) The caretaker convention’s been observed ever since, and it’s now authoritatively recorded in the cabinet manual – a document that’s endorsed by each new government at the start of its term.

Fast forward 36 years and the caretaker convention’s back in the news. Last week, Judith Collins lambasted the government for failing to consult her before moving Auckland to alert level three. She invoked a constitutional convention that, she said, required proper input from the opposition on major decisions of national significance during the pre-election period.

This was pure fiction. As others have explained, the caretaker convention does not extend that far. New Zealand governments have never been required to consult with the opposition during the lead-up to (as opposed to directly after) an election.

But should they? Should this be treated – as was 1984 – as an opportunity for the caretaker convention to be reimagined?

The idea does have some superficial attraction. Both the United Kingdom and Australia have pre-election “purdah” periods of the kind Collins envisages. One justification for that approach is that every general election carries with it the possibility of a change of government. Another is that, once parliament’s been dissolved, the executive can no longer be held to account for its decisions in the normal manner.

This is a point on which reasonable people may choose to disagree. A pre-election consultation requirement has, however, some significant drawbacks. Ironically, they’re graphically illustrated by recent events.

Let me explain.

The logic of the caretaker convention

We need to go back to basics. Under New Zealand’s democratic system, it’s critical that the head of state (in practice, her representative, the governor general) must always have an elected government to advise her. There must never be a gap between governments. For this reason, even after it’s been defeated at an election, the incumbent government retains full legal power until a new government’s sworn in. It continues to govern.

Important as that convention of continuity may be, it’s in tension with another key idea: in order to preserve their legitimacy, governments must retain the support (“the confidence”) of a majority of members of the house of representatives. When a government loses that confidence, it loses its democratic mandate. If it can’t re-negotiate its majority, it must either cede power to others, or advise the governor general to call an election.

The caretaker convention is, if you like, the bridge between these two foundational precepts – that there must be continuity between governments, and that governments must retain their numbers in the house. The convention has different limbs. But, in broad outline, the idea is threefold:

• If a government loses the confidence of the house or is defeated at an election, it continues to govern until a new government is ready to be sworn in.
• But it should avoid making significant or controversial decisions.
• If significant decisions can’t be avoided (or dealt with through temporary or holding arrangements), the incumbent government should consult to ensure its proposed intervention is supported by a majority in parliament. (If the composition of the new government is clear, it should simply act on the incoming government’s advice).

New Zealand’s caretaker convention does not apply in the pre-election period. As the cabinet manual explains, successive governments have nevertheless exercised a degree of abstinence during that time. Significant new appointments are generally deferred, as are advertising campaigns that could create a perception that government funds are being used to finance publicity for party political purposes. Importantly, though, neither the cabinet manual nor the Cabinet Office circular in which these habits of abstinence are spelled out in more detail mentions any circumstances in which pre-election consultation with the opposition is required.

Why we should resist the introduction of a pre-election consultation requirement

So there’s no convention in New Zealand of pre-election consultation. Not yet. But should there be? For me, the answer is no.

First, let’s be clear that such a consultation requirement could not take the same form as the current caretaker convention, nor could it be justified by the same logic. That logic, remember, is that significant decisions should have the support of a parliamentary majority. It’s that outcome – majority support – to which the expectation of consultation is directed.

It’s obvious that logic can’t work pre-election. Ardern’s government still holds the confidence of the house, and there’s no new set of numbers available to illuminate where the democratic mandate will lie during the upcoming term. Were we to analogise from the caretaker convention, we might decide that a government that’s suffering in the polls should exercise particular restraint in the pre-election period. But how we’d turn that moral precept into a workable convention isn’t at all clear. And if we did, it wouldn’t assist Collins. On the basis of this logic, her moral case seems particularly weak.

Collins, however, claims a far higher ground: consultation with the opposition (by which she means National) regardless of the likely electoral outcome. But would we really want our governments to be hampered in this way when important decisions must be made? Consider that our parliamentary system is an adversarial one which hinges (even under MMP) on a political contest between the two main parties. In the lead-up to an election, that contest is heightened as the parties compete for a share of the party vote. The pre-election period is exactly when we might least expect constructive collaboration between the government and the opposition.

That’s borne out, I’d suggest, by Collins’ and Gerry Brownlee’s recent behaviour. Over the last month, we’ve seen them searching methodically for an exploitable chink in Ardern’s armour. Their style has been colourful and combative; at times, controversial. It could hardly be called collaborative.

What the Covid crisis also illustrates is that “significant decisions” (whenever they arise) may be decisions on which life and death – quite literally – depend. Collins’ contention is that, at the very moment when strong and decisive government is at its highest premium, the Ardern cabinet should seek to collaborate with a political enemy that’s intent on its destruction. To me that seems like a recipe for disaster.

It’s fair to expect a government that’s approaching the end of its term to defer significant decisions where possible. A hardening of those expectations could well be a useful development. But sometimes significant decisions are unavoidable. When that happens, as here, we need the government to be allowed to govern. An expectation of cross-party collaboration at such a moment carries with it, in my view, a whiff of political naivety.

Does Her Majesty’s Loyal Opposition have a constitutional role to play in all of this? Of course they do! Their role, as always, is to hold the government to account (see here). In the exceptional circumstances we are now facing, it may well be appropriate for the government to facilitate that scrutiny by establishing a bespoke mechanism to operate during the period parliament is dormant (as, indeed, occurred during the initial lockdown).

There’s also a serious question to be answered as to whether the election should be delayed in order to allow opposition parties a fair opportunity to campaign. That call for electoral delay makes it all the more important, however, that we are well served in the meantime by strong and effective government. Until the electorate calls it otherwise, Ardern and her mob hold the Treasury benches. Let’s allow them to get on with the job.