How often do snap elections end up in favor of the side that triggered them?

Question

Both France and the UK are about to head into a snap election. Presumably both Macron and Sunak decided to run one because they believe that their side will gain more seats compared to the counterfactual of running a scheduled election.

Historically speaking, how often have such bets paid off? I’m looking for research papers which have tried to analyze this question.

Somewhat related question: For countries with a runoff system, how often does the second place candidate end up winning in the second round?

Answer 1

united-kingdom

There were snap elections in 1923, 1931, 1951, 1955, 1966, 1974 2017 and 2019. These are elections that are more than a year early, and in situations which are not forced (for example when a government loses a vote of confidence).

Note that the 2024 UK election is not a snap election, in these terms, as it is more than four years since the 2019 election.

• 1923, Fail Stanley Baldwin (Conservative) calls the election and loses majority in Parliament
• 1931, Complicated Ramsey MacDonald (National Labour) forms "National Alliance" with Conservatives and Liberals and wins a massive majority, but the Conservatives win most seats, and Labour is almost wiped out.
• 1951, Fail Clement Attlee (Lab) loses slim majority to Conservatives.
• 1955 Success Antony Eden (Con) calls election just over a year early, and wins clear majority.
• 1966, Success Harold Wilson (Lab) calls election and wins large majority for Labour Party
• 1974 (Feb) Fail With just over a year, Heath (Con) loses majority and Labour minority government takes over.
• 1974 (Oct), Success Harold Wilson converts the Hung Parliament to a small majority for the Labour Party.
• 2017 Fail Teresa May (Con) loses seats, and converts a small majority to a hung parliament.
• 2019 Success Boris Johnson wins substantial majority for Conservatives.

So in the UK, over the last 100 years, there have been nine "snap elections". Four have been clear successes for the party and PM calling the election. Four have been clear failures and resulted in a change of PM. One is "complicated": A personal win for the PM but a massive loss for the Party that he had represented at the previous election.

Of course, what we can't know is the counterfactual. We don't know if Baldwin would have done better or worse for the Conservatives if he had not called the 1923 election.

In the UK, snap elections are sometimes successful.

Answer 2

TLDR seems to be that it works out to be advantageous on average (in Europe) but there surely are counterexamples in specific elections, and that might even apply to some countries as a whole.

(In 2019) I finally found something reasonably broad in Schleiter and Tavits, "The Electoral Benefits of Opportunistic Election Timing," 2016

In this paper, we present a comparative analysis of the effect of opportunistic elections on the incumbent‘s electoral performance. The existing literature on parliamentary dissolution and election timing generates contradictory expectations about the ability of incumbent governments to benefit from strategically timed elections. We evaluate these competing hypotheses, drawing on an original dataset of 321 parliamentary elections in 27 East and West European countries, observed from 1945 or democratization to the present. In order to causally identify the effect of opportunistic election calling on incumbent‘s electoral performance, we rely on instrumental variable regression. Our results reveal that opportunistic election calling generates a significant vote share bonus for the incumbent of as much as 5.5 percentage points. This finding suggests that by timing elections strategically, governments can significantly affect how voters vote. It therefore has powerful implications for our understanding the effectiveness of elections as instruments of democracy and accountability.

They even made an attempt at excluding some confounders. The raw correlation without correcting for those [confounders] is even greater.

Using crosssectional time-series data from Europe, we find that, compared to regular elections, opportunistic elections correlate with an 8 percentage point vote share bonus and a 10 percentage point seat share bonus for PMs. Such elections are also associated with 26%–30% greater odds of survival for the PMs.

[...] In our most complete instrumental variable regression that aims to address concerns of reciprocal causation, confounding, and alternative explanations (such as economic performance), we still find that opportunistic elections carry a vote share bonus of up to 5 percentage points.

(They only reported the latter, confounder-corrected figures in the abstract though.)

Somewhat oddly they don't cite any previous systematic work

Given the number of studies on parliamentary dissolution, election timing, and political business cycles, it is surprising that there is no systematic empirical study exploring this question.

Schleiter and Tavits alas did not publish any country-level fixed-effects. Those might not have been statistically significant given the small per-country sample, IDK.

But they do cite various papers on specific events in different countries. They actually do cite the 2004 book of Smith (and his 2003 paper) but only to say that

For example, Smith (2004) points out an intriguing puzzle: if governments can control the timing of elections, why are early elections not more frequent with more incumbents taking advantage of their popularity? He then goes on to provide a thorough theoretical account of why we should not necessarily expect opportunistic elections to pay off for the incumbent. [...]

In fact, according to one of Smith‘s central hypotheses, incumbents may even systematically lose as a result of opportunistic elections. Smith (2003, 399) cites examples from the UK and France of opportunistically called elections, which resulted in significant downturns for the incumbent at the polls.

Schleiter and Tavits also cite

Grofman and Roozendaal (1994) show with data from the Netherlands that parties which precipitate government termination and new elections may not yield any electoral benefit.

which does survey all the post-war Dutch snap elections till then.

I think Schleiter and Tavits consider this country-level data be rather anecdotal because of the small sample (of elections) involved in each, in contrast to their own work.

The UK case seems to be particularly contested in the literature. While Smith indeed appears to have found a disadvantage, in a subsequent paper that focuses on the UK, Schleiter and Belu (2018) found that there was a slight advantage to early election there too.

Our findings suggest that UK prime ministers made extensive use of their powers to time elections to favourable circumstances. Nearly 60% of the United Kingdom’s postwar elections up to 2015 were opportunistically timed, typically in circumstances that did not take voters by surprise. Incumbents who used this strategy realized vote and seat shares that outstripped those of their peers in other elections by 3.5% and 11%, respectively, on average.

I did not read these UK-focused studies in minute detail, so I'm not entirely sure what's the gist behind the different findings there. One issue appears to be which elections to categorize as opportunistic, e.g. we find this in a footnote:

Note that Smith also categorizes the February 1974 election, which the miners forced on the conservative prime minister Edward Heath, as unanticipated. However, [...] this was not an opportunistic election called by the government for partisan benefit.