Can parties sign up more people to postal votes with mailings? Some evidence

Welcome to the latest in my occasional series highlighting interesting findings from academic research. Today – how do party efforts to persuade supporters to sign up for postal votes work?

The campaign logic for parties appears to be clear and solid: people with postal votes are more likely vote, so it makes sense to get as many of your supporters signed up for postal votes as you can. Especially as when postal votes go up, turnout goes up – this isn’t just a displacement from one form of voting to another for people who would vote anyway.

But how well do the efforts parties make to sign up postal voters actually work? Not necessarily well is the conclusion from Joshua Townsley and Stuart J. Turnbull-Dugarte, “Can parties recruit postal voters? Experimental evidence from Britain”, Electoral Studies, https://doi.org/10.1016/j.electstud.2019.02.013.

They tested out a target letter postal vote recruitment exercise:

We carried out a pre-registered field experiment with the Liberal Democrats in Southwark, London during the United Kingdom local elections in May 2018. As our experiment represents the first of its kind in the UK, we replicate the letter-based approach that has been successful in US studies.

Ahead of the election, we compiled a list of 3340 registered voters that the party had identified, through prior canvassing, as being likely Liberal Democrat supporters…

Voters from this list were first clustered into their households. We then randomly assigned 1500 households to a treatment group, and the remaining 1583 households to a control group…

Seven weeks before the election, the treatment group was sent a personalised letter from the party’s local candidates in each ward, encouraging them to apply for a postal vote…

Recipients are asked to either apply for a postal vote online, or by completing and returning a postal voter application form that was enclosed with the letter.

The result? No increase in turnout:

There were no substantive differences between the control and treatment groups in terms of (1) postal voter registration, and (2) turnout. The control and treatment groups had a near-identical proportion of voters registered to vote by post (16.87% and 16.84%, respectively), indicating that the personalised letter and forms were unsuccessful in their aim to recruit party supporters to postal votes. The rate of turnout in each group is also similar, suggesting the letters also had no effect on mobilising subjects to vote.

That conclusion doesn’t necessarily mean such postal vote recruitment letters do not work. I say that not only because I’m pretty sure some I have done in the past have… but also because, as the study rightly highlights, the test was only of one particular postal vote recruitment message. The lesson may well be that care needs to be given to using the most effective message.

As they conclude:

The role that postal voting plays, and could play, for party campaigns, including how  parties can best mobilise supporters by recruiting them on to postal votes, remains of interest. The ‘prize’ associated with encouraging supporters to vote at a higher rate using postal votes (and to vote early) merits further research. If parties can use postal voting more widely to maximise their vote share, then this could radically alter how local campaigns are fought. Further, if future research finds that postal voting increases overall participation, then this could be an avenue worth pursuing by governments and electoral authorities as a means to increase voter turnout at elections.

And the more immediate lesson for campaigners: test your postal vote sign up messages carefully as you can’t just assume that a competent letter delivered in good time is enough.

You can read the other posts in the Evidence-based campaigning: what the academic research says series here.