Incumbents in France are less often re-elected while all candidates can be reimbursed for their personal expenses
Campaign spending in the 2020 races for President and Congress totaled more than $14.4 billion, more than double the 2016 level. Embedded in all of this spending is the assumption that a campaign that raises more money than its competitors has an advantage.
Among players who care about competitiveness, this is seen as effectively giving holders a 10-yard head start in a 100-yard sprint.
While the power of money has been decried for years, there is still little evidence of the causal impact of financial regulations. In a working paper published by the National Bureau of Economic Research, Nikolaj Broberg of the European University Institute, Clémence Tricaud of UCLA Anderson and Vincent Pons of Harvard offer causal evidence that an electoral process that reimburses candidates for their personal campaign helps level the playing field.
The causal case of the reimbursement of candidates’ expenses
The goal of representative democracies is to elect leaders who are most aligned with people’s preferences. The mechanics of elections – who can run for office, how they are funded, who can vote, where and how voting takes place – greatly influences the representativeness of election results. Research by Tricaud and others seeks to measure the causal impact of these regulations.
The authors compare elections in French constituencies of just over 9,000 people, where campaign rules apply, to constituencies just below that population threshold, where campaign spending and contributions are not regulated. They find that incumbents are less likely to win when their challengers have the financial backing of knowing that the personal funds they have committed will be repaid.
The French government structure includes 101 regional departments responsible for the management of culture, local development, social assistance, education, housing, transport and tourism. Each department is broken down into electoral constituencies in which the inhabitants elect the members of the departmental council. A candidate who receives at least 50% of the votes in the first round is declared the winner. If no one reaches the 50% threshold, the top two candidates in the first round advance to the second round, along with any candidate who achieves a certain tie-vote threshold.
Since 1990, all electoral districts with a population of at least 9,000 have had limits on election expenses. In 1995, the French electoral commission also introduced a campaign reform that allows all candidates to recover their personal expenses. The French state reimburses candidates up to 50% of the value of the spending limit and as long as they obtain 5% of the votes in the first round.
The researchers analyzed the elections between 1998 and 2011, corresponding to nearly 10,000 departmental elections and more than 50,000 candidates.
While they found no impact on the number of candidates in any given election, they find that elections to which campaign rules apply are 11 percentage points less likely to be won directly in the first round. .
Looking at the impact on the identity of the winner, they show that the odds of a previous finalist winning increase by 5.2 percentage points and the odds of a total newbie winning increasing by 9.2 percentage points.
Given the zero-sum nature of elections, the big losers are the incumbents. The researchers find that campaign settlements decrease an incumbent’s likelihood of being re-elected by 14.5 percentage points. This is partly because incumbents are 7.4 percentage points less likely to run for office and those who run again are less likely to retain their seats. Conversely, challengers become more likely to run again and win provided they run.
The staggered implementation of the two types of electoral reform gave researchers two election cycles (1992 and 1994) to study the effect of spending limits in isolation. They find no regulatory impact in those two years, suggesting that spending caps are not what moves the needle of competitiveness in France. This can be explained by the fact that few candidates — before and after the 1995 reform — were close to the expenditure ceiling. Instead, these results suggest that the main effects are driven by candidate reimbursement, an interpretation that is further supported by analysis of the composition of candidate contributions before and after the introduction of reimbursements in 1995.
The researchers also performed the same analysis at the more local municipal level, but found no impact. They suggest that this is likely due to the fact that French municipal elections use a proportional list system, in which campaign expenses can be distributed among the various list members, instead of being borne by a single candidate.
To the left
The authors also explore the political orientations that benefit the most from campaign regulations in departmental elections. While extreme candidates do not become more likely to win, leftist candidates win the most. Consistent with this finding, the authors note that historically in France, left-wing candidates have collected less than half of the private donations pocketed by right-wing candidates, and that their personal spending has not made up this gap. Once personal expenses became reimbursable, left-wing personal expenses increased relative to expenses of right-wing candidates.
“Our results indicate that, on the net, campaign finance regulations level the playing field and diminish the incumbent advantage,” they conclude.