One of the great paradoxes in the research on corruption in democracies—and one of the great sources of frustration for anticorruption activists—is that while large majorities of voters consistently claim that they detest corruption and would be less likely to support corrupt politicians, nonetheless politicians credibly accused of corruption regularly win elections. There are many possible explanations for this, including the possibilities that voters lack sufficient information about corruption allegations against candidates, or that voters ultimately prioritize other factors. Yet another possibility—similar to yet distinct from these familiar explanations—is that even if voters are generally aware of corruption allegations against certain politicians, when the time comes to vote, other issues are more salient in many citizens’ minds, and integrity concerns fade into the background.

That last explanation implies that if concerns about politicians’ integrity were made more salient shortly before the election—even if the focus was on political corruption generally, or on corruption in some other jurisdiction—then voters would be less inclined to support politicians suspected of corruption. In a recent article, titled Can Institutions Make Voters Care about Corruption?, Omer Yair, Raanan Sulitzeanu-Kenan, and Yoav Dotan find that this may indeed be the case, and further suggest that if high-profile institutions—such as courts—take actions that raise the salience of corruption and integrity issues shortly before an election, this can lead voters to place more weight on such considerations when casting their ballots.

The authors focus on the 2013 Israeli municipal elections, which took place shortly after the Supreme Court of Israel issued a pair of unprecedented rulings that ordered the instant dismissal of three incumbent mayors accused of corruption (see here and here). The study focuses not on the elections in these three municipalities, but rather on elections in other municipalities. In some of these municipalities, the incumbent mayor had been formally investigated by the police for corruption. (The authors label these mayors “‘low-integrity’ incumbents.”) In other municipalities, the incumbent mayor had not been formally investigated for corruption. (These are the “‘high integrity’ incumbents.”) The authors then compare the gap between the vote shares of low-integrity incumbents and high-integrity incumbents (controlling for a variety of other factors) in the 2013 election cycle to the low-integrity/high-integrity vote share gap in the two preceding election cycles, 2003 and 2008. (The three municipalities directly affected by the Supreme Court’s 2013 decisions are excluded from the analysis.) The main finding is that the vote share gap between low- and high-integrity incumbent mayors was considerably higher in 2013 than it was in 2003 and 2008. The authors interpret this as evidence that the 2013 Supreme Court decisions raised the salience of mayoral corruption in the minds of many voters, making those voters less inclined than they otherwise would have been to vote for incumbents who had been investigated for corruption, even though the Supreme Court decisions didn’t directly pertain to their city or mayor.

Of course, there could be other reasons why voters seemed to care more about corruption allegations in 2013 than they had in 2008 or 2003, but the authors find additional support for the hypothesis that the Supreme Court decisions increased the salience of corruption issues, and hence reduced support for low-integrity mayors, in a controlled experiment in which voters in seven municipalities (none of which were directly affected by the 2013 Supreme Court rulings) were asked to rate their satisfaction with the incumbent mayor. Some of these voters were first exposed to a short excerpt describing the 2013 Supreme Court decisions. The mayoral performance satisfaction gap between respondents from cities with a low-integrity incumbent mayor and those from cities with a high-integrity incumbent mayor was wider among those who were reminded of the 2013 Supreme Court ruling—even though, again, that ruling did not directly pertain to the mayors of any of the respondents’ cities. The researchers interpret this as further evidence that a high-profile judicial ruling about corruption can “make voters care about corruption,” by emphasizing the importance of integrity.

I highly recommend that readers check out the full paper, which is admirably concise. There is much to ponder. I was left with a number of open questions. Here are some that I found myself most curious about:

• First, as the authors mention in the article, the three mayors who were removed from office due to the Supreme Court’s rulings were still authorized by the Court to re-run, and in fact all three of them were reelected. This is itself a puzzle, and certainly complicates the otherwise encouraging result that the Supreme Court’s decisions seemed to increase the salience of corruption to Israeli voters generally. Is it possible that these three mayors were simply popular enough in their respective cities to overcome the adverse effects of the Court’s rulings? Or could it be that judicial anticorruption rulings influence directly-affected residents differently from residents of “third party” cities?
• Second, I wondered to what extent the effect the authors found depends on the public support for, and legitimacy of, the Israeli Supreme Court. While the Court still enjoys relatively high public trust compared to other Israeli government institutions, that support is now lower than it was in 2013. (Some of this probably has to do with the negative influence that Prime Minister Benjamin Netanyahu’s corruption investigations, prosecution, and trial have on many of his supporters’ trust in the legal system.) What effect would such changes have on the ability of judicial anticorruption rulings to influence voters’ behavior? If public trust in the Court were to drop far enough, would there be a risk that its anticorruption decisions might actually encourage voters to vote for low-integrity candidates, as a kind of “no confidence vote” against the Court?
• Third, I wondered whether what matters most is the outcome of the judicial decision, or whether the language in the opinion matters more. Suppose, for example, a judicial decision contains language that emphasizes the importance of integrity, but the decision itself does not impose any sanction for corrupt (or allegedly corrupt) behavior. This is not a purely hypothetical possibility. Consider the highly-publicized May 2020 Israeli Supreme Court decision that held that Benjamin Netanyahu is legally authorized to form a government despite being charged with bribery, fraud, and breach of trust. Some of the justices were vocal about their discontent with this situation, with one justice calling it “a social crisis and a moral failure by both society and the political system.” Yet most of the newspaper headlines focused on the fact that the Court did not prevent Netanyahu from continuing to serve as a Prime Minister (see here, here, and here). How would a ruling like this affect voter behavior? Would it still increase the vote share gap between high-integrity and low-integrity incumbents, on the logic that the decision still increases the salience of corruption issues and contains strong anticorruption language? Or would the fact that the decision ultimately reached an outcome that did not sanction the allegedly corrupt politician actually decrease the salience of corruption accusations for voters?

These are only a few of the questions that future research may explore. Meanwhile, anticorruption advocates and scholars may be encouraged by the fact that it does seem possible that, with appropriate prompting, voters do seem to care about the integrity of candidates.