These Political Scientists Surveyed 500,000 Voters. Here Are Their Unnerving Conclusions.

Episode Summary

Executive Summary: Ezra Klein interviews John Sides and Lynn Vavreck about their book on the 2020 election, arguing U.S. politics is both highly calcified and tightly balanced: voters rarely switch sides, yet tiny shifts can swing power. They show COVID, protests, economy, and Trump’s behavior were absorbed into polarized identities, making 2020 a continuation of 2016 and a warning about democratic fragility.

Main Topics: Calcification and partisan parity (Priority: 5/5): The core framework: voters are increasingly sorted and less movable, while parties are nearly evenly matched, so small changes can produce large governing shifts. Issue polarization and identity conflict (Priority: 5/5): Issues like immigration and abortion moved from low-polarization to highly polarized, with identity-inflected politics replacing older New Deal-era conflicts. Negative polarization and misinformation about the other side (Priority: 4/5): People’s feelings toward their own party changed less than their hostility toward the opposing party, reinforced by exaggerated stereotypes and media coverage. 2020 shocks: COVID, George Floyd protests, and Trump’s response (Priority: 5/5): The book finds these events did not directly realign votes as much as expected; instead they were quickly reinterpreted through partisan lenses. Voting behavior, policy prioritization, and popularism (Priority: 4/5): Survey experiments show voters prioritize the most identity-charged, conflict-heavy issues, limiting the power of moderate policy appeals. Economy, approval, and election forecasting (Priority: 4/5): 2020 economic indicators pulled in opposite directions, and forecasts based on fundamentals and approval did better than many polls; the Biden economy discussion raises 2022/2024 concerns. Democratic backsliding and refusal to accept losses (Priority: 5/5): The biggest alarm is not disagreement over policy but losing candidates’ willingness to change rules or subvert outcomes after defeat.

Key Arguments: American politics is calcified: people are much less likely to switch parties because party coalitions are more internally coherent and more ideologically distant than in the past. Partisan parity amplifies the effects of calcification: with control of institutions often decided by tiny margins, small shifts in turnout or persuasion can change national power. Immigration and abortion illustrate rapid polarization, with positions shifting dramatically in just a few years once Trump re-centered the conflict. Negative polarization is intense: voters increasingly dislike the other party and see it as morally and patriotically suspect, often through distorted stereotypes. Local COVID deaths and George Floyd protest activity did not produce large direct vote swings; instead, these events were absorbed into preexisting partisan conflict. Voters increasingly prioritize the issues most tied to party identity and conflict, not the issues politicians or commentators might assume are most important. Popularism/policy literalism likely has limited leverage because the electorate is already highly sorted and many issues are symbolically and identity loaded. 2020 reinforced ideological sorting across racial groups: Trump did better among conservatives of all races, Biden among liberals of all races. The major democratic danger is not just polarization but losers refusing to accept legitimate defeat and trying to alter the rules of future elections.

Data Points: NationScape interviews: More than 500,000 interviews - Large data collection used by the authors to analyze 2020 opinion and vote choice. Survey cadence: 6,000 people per week for 87 weeks - Nationscape fieldwork described by John Sides. Perception of major party differences in 1952: 50% - Share of voters saying Democrats and Republicans had major differences. Perception of major party differences in 2020: 90% - Share of voters saying Democrats and Republicans had major differences. Immigration attitudes among Democrats vs Republicans before Trump: About 30% of Democrats vs 11% of Republicans wanted to increase immigration - Illustrates widening gap on immigration on the eve of Trump’s candidacy. COVID deaths vs vote choice: Higher COVID-death counties were slightly more likely to vote Trump - County-level analysis found no direct punitive effect against Trump from local pandemic deaths. Protest effects on vote choice: No clear direct effect; if anything, slight Biden benefit - County-level analysis of George Floyd protest activity and voting. Trump approval during early pandemic: Approval briefly rose in March 2020, then declined - Authors note a short-lived bump when Trump treated COVID more seriously, followed by decline. Economic indicators in 2020: GDP fell, unemployment spiked, incomes rose sharply - Why the 2020 economy was electorally ambiguous. Consumer sentiment in 2020: Roughly as pessimistic as 2012, not as bad as 2008 - Used to contextualize how voters evaluated the economy during the pandemic recession. Forecasts for Trump in 2020: Around 47% - Fundamentals-based forecasts approximated Trump’s eventual vote share. House forecast: About 40 Democratic seat losses - Model-based expectations for a difficult midterm environment discussed in relation to 2022. Demographic misperceptions: Republicans thought almost half of Democrats were Black; Democrats thought almost half of Republicans earned $250,000+ - Example of exaggerated perceptions of out-party composition.

Pivotal Quotes: "calcification. Makes our politics rigid and inflexible." — John Sides: Defines the book’s central concept for why dramatic events failed to change vote behavior much. "people know what kind of world they want to live in. And they don't want to live in the other side's world." — John Sides: Explains why partisan identity has become deeply resistant to persuasion. "the biggest threat to democracy [is] ... if people who lose free and fair elections will not accept that outcome and will actively work to undermine it or subvert it" — Lynn Vavreck: Closing warning about election denial and institutional erosion.

Implications: The episode suggests U.S. elections are driven less by persuasion than by sorting, identity, and tiny margin shifts. That makes outcomes volatile, policy swings large, and democratic institutions more vulnerable when losers reject defeat.