Contrasting Strategic Voting under FPTP and TRR: A Lab Experiment

What is the story?

In a previous blog entry, we reported the results of an online experiment conducted within the context of the 2011 Ontario elections and pointed out that the overall amount of strategic voting varies depending on the electoral system in use. In particular, we showed that strategic desertion is higher under first past the post (FPTP) than under alternative vote (AV).

Along these lines, another research has been conducted to contrast the importance of strategic voting under FPTP and two-round runoff system (TRR). This time, the laboratory experiment design has been chosen. The strategic paradigm assumes that voters attempt to maximize their expected utility, which is based on a combination of preferences and expectations about likely outcomes, and it is very difficult to isolate these parameters. As a consequence, such models are usually not testable directly using observational data.

The experimental protocol

8 experiments have been conducted between December 2006 and February 2007 in Paris, Lille and Montreal. For each experiment, 21 participants were submitted to 8 successive elections, 4 under FPTP and 4 under TRR (order varies). In each election, there were 5 candidates, located at 5 distinct points on a left–right axis that goes from 0 to 20: an extreme left candidate, a moderate left, a centrist, a moderate right, and an extreme right candidate (see Figure 1 below).




According to theoretical models, two factors are crucial to characterize strategic voting. First, there is the benefit associated with electoral outcome. To measure it, the participants were informed from the beginning that one of the 8 elections will be randomly drawn as the “decisive” election. They were also told that they would be paid €20 (or $20) minus the distance between the elected candidate’s position and their own randomly assigned position. For instance, a voter whose assigned position is 11 would receive €10 if candidate A wins in the decisive election, €12 if E wins, and so forth. In the experiment (as in real life), it was in the voter’s interest that the elected candidate be as close as possible to his or her own position. Second, the perceived viability of each candidate is also crucial. When casting his or her vote, the participant was asked to rate, on a 0-to-10 scale, each of the candidates’ chances of winning the election.

Finding: Similar strategic considerations under the two electoral systems

As expected, the experimental data shows that the propensity to vote for a candidate increases with the benefit associated with this candidate and with his or her perceived viability. As reported in Table 4, the related regression coefficients are both positive and highly significant. What is more surprising is that this pattern holds for both FPTP and TRR (although the effect of viability is stronger in the former). In other words, and contrary to what Duverger (1954) argued in his seminal book “Political Parties”, strategic considerations are no different under one-round and two-round elections.

As a consequence, an alternative explanation needs to be found to account for the existence of a multitude of parties in some democracies using TRR such as in France. Further research will be conducted within the framework of the Making Electoral Democracy Work project to elucidate this paradox.







Further details may be found in Blais, André, Labbé St-Vincent, Simon, Laslier, Jean-François, Sauger, Nicolas, and Karine van der Straeten. 2011. “Strategic Vote Choice in One Round and Two Round Elections:  An Experimental Study.” Political Research Quarterly, 21: 637-646.

Follow us on Facebook and Twitter.

This entry was posted in Blog. Bookmark the permalink. Both comments and trackbacks are currently closed.