Each week, In Theory takes on a big idea in the news and explores it from a range of perspectives. This week, we’re talking about polarization in politics. Need a primer? Catch up here.

David Broockman is an assistant professor of political economy at the Stanford Graduate School of Business.

Academic hand-wringing rarely changes political institutions. But political science researchers’ recent obsession with political polarization seems to be having some impact. These researchers argue that polarization has “put the nation at risk” and are championing reforms meant to stem it.

The remedies proposed are many — from reforming campaign finance to changing the primary system — and all share a common theme: By shifting more political power to everyday people, academics believe that extreme politicians would be encouraged to heed the public’s centrist demands.

The widespread view that citizens are more centrist than politicians seems obvious, but it’s actually misleading at best. And worse, it’s endangering our democracy.

Voters Are Not All Centrists

Most voters support some liberal policies and some conservative policies. Academics have long taken this as evidence of voters’ underlying centrism.

But just because voters are ideologically mixed does not mean they are centrists at heart. Many voters support a mix of extreme liberal policies (like taxing the rich at 90 percent) and extreme conservative policies (like deporting all undocumented immigrants). These voters only appear “centrist” on the whole by averaging their extreme views together into a single point on a liberal-conservative spectrum.

This makes those who celebrate voter centrism rather like the fabled statistician who drowned in a river that was 2 feet deep on average. Even if voters are centrist on average, they can be quite extreme on many particular issues.

The result? Reforms that empower voters may not push politicians further to the center — instead, they may encourage politicians to pander to extreme views popular among voters. Indeed, where they have been enacted, many changes that reformers favor — like public funding of elections and top-two primaries — have resulted in politicians doing just this. By seeking to further empower voters in the name of reducing polarization, well-meaning reformers may actually be encouraging dangerous extremism.

Donald Trump: The Favorite Son of  ‘Moderate’ Voters

Donald Trump’s rise exemplifies these dangers.

Political scientists and pundits alike argue that it would improve governance to devolve political power from the political elites who know the most about politics and policy to the voters who know the least. Polarization scholars hold these uninformed voters in the highest esteem because they look the most centrist on a left-right spectrum. They are also Donald Trump’s base.

Yes, you read that right. Political scientists have long exalted the centrist wisdom of those who now constitute some of Trump’s strongest supporters — the poorly educated authoritarian xenophobes who are attracted to a platform suffused with white supremacy, indulge in unapologetic nationalism and use violence to silence opponents. As commentator Jacob Weisberg has written, these extreme voters’ views are a mix of “wacko left and wacko right” — the key credential one needs to qualify as centrist by scholars’ most popular definition.

Trump’s popularity among these voters should have been no surprise. Their penchant for extremism on many issues has been visible in public polling for years. Indeed, it has been visible for decades: Modern public opinion research itself was founded scholars who had just witnessed World War II and had good reason to be fearful of mass publics’ extremist tendencies.

Yet despite our own country’s populist past, many scholars today resist acknowledging public extremism is even possible. (Reviewers of my research, for example, dismissed as simply too “doubtful” the idea that voters might have some views more extreme than those of elected officials.) A simple conceptual error has blinded them to the true character of the Trump coalition.

Forgetting the Founding Fathers’ Wisdom

Dismissing the public’s susceptibility to extremism is not only naive — it is dangerous. With anti-majoritarian institutions like six-year terms for senators, the Founding Fathers empowered elites to serve as a check on unwise popular passions.

Reformers may find the Founding Fathers’ ideas outmoded. After all, if the public is a homogenous centrist mass, why be afraid of unchecked popular control?

Trump’s rise brings the founders’ wisdom into sharp focus. However, reformers have by now already eliminated nearly every institutional safeguard that would once have stood in Trump’s way to the Oval Office. In previous eras, party elites would be able to more easily foil his nomination at the Republican convention; financial contributions to his competitors would be less limited; and state legislators could have refused to pledge their state’s electoral votes to him. As with insurance plans canceled right before a catastrophe, the upside of these safeguards is only fully apparent when a demagogue like Trump is already doing damage.

This is not to say that elites always have innocent or correct intentions. Outright oligarchy is no better than outright direct democracy, and conservative elites may even have inadvertently abetted Trump’s rise. The point is that the Founding Fathers were right to allow elites and voters to serve as checks and balances on each other. Today’s reformers have altogether forgotten the merits of one side of this equation.

How Extreme Are Today’s Politicians?

One reason scholars play down the benefits of allowing elites to serve the function the Founding Fathers envisioned is that they characterize today’s political elites as highly extreme. The empirical evidence for this view is also much flimsier than many realize. Today’s politicians strategically exaggerate their disagreements, but on many issues these disagreements are substantively small by historical standards. Indeed, where yawning disagreements existed 50 years ago around issues of trade, regulation, race, taxation, redistribution and foreign policy, there is now a Washington consensus on such matters to which both parties largely subscribe.

Today’s elites certainly take some extreme positions. But the prospect of President Trump does put the allegedly wholesale extremism of today’s elites in perspective. Trump seeks to shatter countless tenets of the Washington consensus in the name of shameless nationalism and overt xenophobia. If he gets his way, the United States would be plunged into a deep recession and millions of lives would be thrown into disarray. That is real extremism.

An Artificial ‘Disconnect’

Students of polarization and pundits alike bemoan a “disconnect” between extreme elites and centrist voters, arguing that voters must urgently be empowered to resolve it. But what if voters would actually like to see today’s politicians become more extreme on many issues? What if many voters actually do, at the end of the day, agree with Trump? In that case, reforms that empower voters may have less appealing effects than supporters argue.

Advocates of reforms would be wise to consider this possibility as they upend centuries-old institutions in the name of democracy. Faithful representation of public opinion is scarcely the sole standard to which representative democracy aspires.

Explore these other perspectives: