Photo by Arnaud Jaegers on Unsplash

PCC turnout turned out to be a historic low and this is raising some questions about the legitimacy of elections with low turnouts.  What does it mean for an election if almost no-one takes part?

One thing that elections are claimed to do is ‘authorise’, but there tends to be fuzziness on exactly what that means. The general problem when defining what ‘authorising’ means in an election is that having a definition creates the possibility of elections that fail to meet that definition.  One possible escape from this is to use the argument that ‘authorising’ collapses quickly into tacit consent, where a typical series of moves might be:

Assuming majority rule is valid, if more than half of a population agree to authorize the same person, then the election has authorized that person. But wait, many elections are won by people without majorities - how can they be authorised?

Well, let’s say that by taking part in an election, they decide to accept the end of the process is authorised. Collectively over 50% agree to this, the 40% winner is authorised. But wait,  some elections have less than 50% turnout! How can that count?

Well, by opting to live in a democratic society people agree to abide by certain rules, which includes accepting the winner of elections as authorized.

Each stage makes the claim to authorisation more flimsy (I think it’s still valid as far as the second stage) and the trouble is this quick collapses into tacit consent makes all discussion of authorisation in elections pretty much meaningless. If absolutely every election can be said to authorize the winner, it really has no meaning.

The way out of this hole where we obviously have some elections that no reasonable standard could call authorising is that not all elections need to involve authorisation: elections can either be authorizing or consultative.  Authorising elections perform the function of the people empowering a person/chamber to act and take decisions on their behalf, whereas consultative elections are essentially very official polls.  The logic in the case of PCC elections would be roughly the following:

  1. Elections can either be exercises in authorization or consultation.
  2. Not all elections achieve 50% (notably local elections and referenda).
  3. An authorised chamber has the power to delegate its authority on certain matters to sub-chambers staffed by whatever process it chooses, including consultative elections.
  4. A chamber elected with less than 50% turnout can legitimately take decisions if its authority is established by a superior chamber with sufficient authorisation.
  5. The same logic allows referendums with less than 50% turnout to be acted on.

Under this logic, PCC elections are fine no matter how low the turnout, as they’re a consultative election - if they were appointed or picked out of a hat, they would be equally authorised.  The question is then if elections with low turnout are a good or cost effective method of picking the right people for the job, which an interesting and important question - but one detached from legitimacy and authorisation.

This passes a low bar for a definition of authorisation in that there are some possible elections that would not be authorised, elections to the superior chamber need to have at least 50% turnout (and as such the Manchester Central by-election result would definitely fall short). There is a final dodge available to MPs in the British system - that their elections can be seen as consultative as  their power comes from the crown -  but this is a bit of a unique case and is out of step with the expanded role of the British electorate in the last century.

This approach isn’t just saying that one set of elections needs 50% - in federal arrangements where different powers are sent to different places, multiple bodies would need to be elected by authorising elections. Likewise there’d be a strong case for requiring 50% turnout on constitutionally required referendums. If elections have an authorising role then turnout has to, at least sometimes, be practically significant. How this would actually be managed practically (delay November by-elections till May?) is a different question.