Today’s critics of populism often place their trust in political elites, expecting mainstream politicians to respect American democratic institutions. But in my latest work, Elitism versus Populism, I directly test whether today’s politicians are any more likely to support fair political competition than members of the public. Instead of focusing on high-profile examples, like Republican Senators’ hypocritical manipulation of Supreme Court vacancies, or the Biden administration’s work to censor social media, I conducted experiments with politicians and voters to abstract away from specific political controversies.

The reason these experiments are valuable, and the problem with focusing on individual examples, is that people can always justify their side’s actions and find plenty of reasons to condemn those of their opponents. But in my experiments, everyone evaluated the same institutional changes—partisan gerrymanders, restrictions on minority‑party input, even bans on opposition candidates. The twist was that half the respondents saw a version where the policy change clearly helped their own side; the other half had no idea which party would benefit.

Support for the rule changes increased by about 15 percentage points when the partisan payoff was explicit, and that increase was about the same for the public and for officeholders. Democrats and Republicans broke the rules at the same rate, and strong liberals and strong conservatives were both worse than moderate partisans.

One surprising finding is that legislators badly misjudge each other: they think about 70 percent of the opposite party would rig the rules if given the chance, even though fewer than half of either party actually would. That perception gap probably helps explain why legislatures engage in pre‑emptive power plays—“do it first before they do it to us.” The takeaway isn’t that democracy is doomed; it’s that waving a “populist threat” flag misses that mainstream elites share similar temptations and face similar incentives, so reforms probably need to check the excesses of whoever happens to be in power.

I was excited to learn that the project will be published as the 2000^th Element, and working with Cambridge has been great. From the start, Jamie Druckman has been an enormously helpful editor. Jamie provided great feedback on my initial proposal, solicited helpful reviews, and was there every step of the way with thoughtful comments and insights to improve the research. I am also grateful to the anonymous peer reviewers for their help improving the manuscript, and to the Institute for Humane Studies, who provided funding to publish the Element open access so that more people can read it. Finally, Sasirekha Sanjivi kept everything organized and efficiently oversaw copyediting and production.