For decades, low levels of voter participation in U.S. elections have been a source of worry and disappointment. The 2020 presidential election and its aftermath in January show that high political engagement can be a mixed blessing. On the one hand, the 2020 election generated the highest turnout in over 100 years, with two-thirds of eligible voters casting a ballot. On the other, the post-election disputes and the Capitol invasion meant that nobody is celebrating the state of American democracy. And the two go together. Historically, high voter engagement in the United States has come during periods of high democratic dysfunction.

Turnout was high because a lot was at stake

Why was turnout so high (at least by U.S. standards) in 2020? The easy answer is this was an incredibly high-stakes, competitive election that consumed remarkable mind-space, and both campaigns worked hard to convince Americans that the fate of the nation was up for grabs.

For once, the easy answer is also true: High-stakes, competitive elections generate higher turnout, especially among lower-propensity voters. That’s because highly competitive, highly consequential elections generate lots of media coverage. Campaigns spend colossal sums of money (like 2020’s $14 billion record) on advertising and get-out-the-vote efforts. And citizens feel like their choice is consequential, and so they vote.

In other words, marginal voters turn out at much higher rates when they feel like their vote really matters and something important is at stake. Furthermore, high voter turnout is much more closely linked to motivation than to opportunity. Reformers have spent decades making it easier to vote (recent setbacks notwithstanding), through “convenience voting” reforms (e.g., early voting, the motor-voter act). In general, they had at best minimal effects, and at worst, have only made life easier for already engaged voters. For sure, it is possible that convenience increased voter turnout by a few percentage points — among the convenience voting reforms, voting from home appears to have had the greatest effect on voter participation. But that is probably all.

High turnout and toxic politics can go together

High turnout might be a hallmark of a responsive democracy with an engaged and knowledgeable citizenry. However, while the all-or-nothing high-stakes hyper-partisanship of 2020 might be good for high turnout, it is not good for American democracy. Closely contested winner-take-all elections make losers extremely dissatisfied and thus exacerbate polarization. This is especially dangerous when party allegiances divide along urban-vs-rural lines, as is currently true. This binary divide generates a kind of “pernicious polarization” in which winning becomes everything, and, partisans will look past, or even actively support, democratic transgressions and power grabs in order for their side to win. Sometimes, tight elections are how democracies die. Such extreme polarization is probably one of the factors that explain the Capitol invasion.

Politics was highly partisan in the late 19th century, the last time the United States had a similar period of high participation. Back then, like now, there was high inequality, democratic backsliding and deep distrust of institutions. However, late 19th-century partisan fights centered more on patronage jobs than substantive national policy disagreements. As reforms curbed patronage and the secret ballot made it harder for parties to put pressure on voters, participation (and partisanship) declined.

It may be that the same underlying forces that generate high turnout — high polarization and closely contested high-stakes national elections — also weaken democracy. Another interpretation is that many Americans prefer to leave politics to the politicians, but reluctantly take part when politics degrades. That sounds more optimistic than it is — the “Stealth Democracy” thesis suggests that the more marginal participants who enter into politics are most likely to hold unrealistic expectations for the ability of politicians to solve problems without disagreement. That makes them more open to authoritarian demagoguery. Both explanations are probably relevant.

High turnout can go together with satisfied voters — but only under different rules

Does this mean that high participation is always a danger sign for democracy? Not necessarily. What looks like a record-high turnout election in the United States still puts America in the lower-middle range of advanced democracies. This country has long been a laggard on voter turnout.

One reason for America’s exceptionally poor performance is that democracies that use single-member districts, like the United States, Britain and Canada, simply wind up with lots of uncontested legislative elections, which lower voter turnout. In more proportional democracies, votes count equally regardless of district, which increases voter turnout. Proportional democracies also typically have more political parties. This increases the likelihood voters will find a party they feel enthusiastic about. It also means more parties and candidates to reach out to different voters. In such democracies, high turnout is the norm. And it doesn’t require high-stakes partisan hatred. As recent comparative studies of affective polarization show, proportional democracies have lower levels of out-party hatred.

In summary, the 2020 election should change how we think about voter turnout. First, it tells us that motivating voters with competitive elections is far more effective than just making it easier to vote, confirming what scholars have long known — the most productive way to increase turnout is to make elections more competitive.

Second, it reminds us that increasing voter turnout does not guarantee a higher-quality democracy. Widespread participation may be an important goal, but the reasons for it matter, too.

Third, it reveals that increased voter turnout does not benefit Democrats as clearly as both Republicans and Democrats had long assumed. It’s at least possible that this will cause partisans on both sides to rethink their assumptions about how to win elections and open up new options for improving the quality of our democracy.

Lee Drutman is a senior fellow in the Political Reform program at the think tank New America, co-host of the “Politics in Question” podcast and author of the book “Breaking the Two-Party Doom Loop: The Case for Multiparty Democracy in America.”

This article is part of a seminar on the 2020 election and its aftereffects organized by the Stavros Niarchos Foundation Agora Institute at Johns Hopkins University. See below for the complete list of contributions to the seminar.