Research
Research
Are preferences or actions more important in determining how prosocial we are? Evidence from UK politics
People tend to behave less prosocially towards members of an out-group as compared with members of an in-group and there is evidence that this extends to the case where groups are defined according to political identity (Fowler & Kam, 2007; Iyengar & Westwood, 2015). There are, however, different ways to define groups based on political identity; we consider two relevant dimensions. A group could consist of members who all (i) prefer or (ii) vote for the same political party. For tactical voters, these groups do not coincide. This paper uses a dictator game to assess prosocial behaviour towards others shortly after the 2024 UK general election, in which a significant number of voters were expected to vote tactically (YouGov, 2024). By alternating the recipient and providing the dictator with information about the party the recipient most preferred and voted for, we establish that (i) less money is given to a recipient that is a member of an outgroup across both dimensions and (ii) less money is given to a recipient that voted differently than to one who preferred a different political party. We therefore argue that the other’s actions (how they voted) are a stronger determinant of what constitutes an out-group than the other’s preferences.
The Role of Opinion Polls in Coordination Amongst Protest Voters: An Experimental Study with Arthur Schram and Randolph Sloof (revise and resubmit at The Economic Journal)
In an election, protest voters signal their discontent with the party they traditionally support in different ways. This paper examines a specific form of protest voting in which voters choose an anti-mainstream party over their true first preference, the mainstream party, as a way to signal discontent with mainstream policies or influence future policy decisions. Protest voters face a trade-off stemming from a coordination problem. Too few protest votes mean that the strength of the protest is insufficient to affect the mainstream’s policies; too many protest votes may result in an anti-mainstream victory, which is a sub-optimal outcome for the protest voter. One way to address this coordination problem is through opinion polls. In this context, polls serve a dual purpose: they provide information about the challenges protest voters face (information channel) and function as a coordination mechanism, allowing voters to adjust their behaviour based on poll results to resolve the coordination problem (coordination channel). We test, experimentally, the extent to which each of these channels increases the likelihood that the protest is successful and find that both channels are significant.
How do Politicians Respond to Protest Voting? The Case of UKIP
When deciding how to cast their vote in an election, the tactical protest voter has to think strategically: they want a party from the political mainstream to win the election, but they want the protest party to do sufficiently well so that the protest signal registers. Protest parties often campaign on a single issue, making the intended effect of the protest signal clearer. An example is votes cast for UKIP in the 2010-15 UK Parliament – many of their voters wanted the Conservative Party to win elections during this period, but to implement more Eurosceptic policies in office. In this paper, we use votes for UKIP in local elections, in which much protest voting takes place, to study whether higher proportions of voters for UKIP in a Member of Parliament’s local area encourages them to publicly shift their policy position on EU-related issues. We use both roll call votes and parliamentary speeches to estimate MPs’ positions and estimate the effect of UKIP votes on Conservative MPs’ positions using a continuous treatment difference-in-differences approach. Preliminary results find no significant effect of votes for UKIP on either measure of MP-level Euroscepticism.