As of this writing, most election analysts are predicting a “blue wave” election, in which Democrats win the presidency, the House, and possibly even control of the Senate. If the analysts are correct, the results could be a historic repudiation of a Republican Party that shifted hard to the right, away from the party’s more establishment, centrist norms.

Such a repudiation has happened before. In 1972, Democrats suffered a crushing defeat in the presidential race. Their candidate George McGovern had adopted positions far to the left of most of the electorate. In his race against Richard Nixon, McGovern won only 38 percent of the popular vote and 17 electoral votes.

Yet Democrats recovered, retained a partisan advantage within the electorate, and won the presidential election in 1976. They avoided long-term disaster for two main reasons: The party returned to more centrist positions, and the Watergate scandal seriously damaged the Republican Party.

Our analysis of American partisanship back to the 1950s finds the Democrats’ experience in the 1970s applies generally. Parties can recover from short-term swings to the ideological extremes. But if those shifts last for several elections, the party will lose partisans over the long term, destroying the loyalties of specific groups of voters. Partisan loyalty is what brings sustained power in local, state and national government.

Competing for partisans

Occasionally, U.S. politics shifts tectonically. That happens when masses of ordinary Americans’ loyalties shift, attaching to a different partisan tribe. More than 85 percent of voting-age Americans are willing to tell pollsters they are either Democrats or Republicans, or that they lean Democratic or Republican. In mature democracies, partisanship drives the vote more than other psychological factors. It also affects how people receive, respond to and organize information about politicians and current events.

What moves long-term partisanship?

Our research distinguishes between voters’ long-run partisan loyalties, which are usually stable, and short-term fluctuations in their votes and expressed partisanship because of economic conditions, scandals, presidential candidate characteristics and foreign policy mishaps. These short-term factors can influence partisanship and votes in specific elections without lasting effects.

We sought to understand partisan changes in major demographic groups by analyzing survey data from the American National Election Studies from 1952 to 2016. We analyzed shifts in party ideologies and in people’s own policy attitudes, as well as parties’ economic and foreign policy performance in office.

Partisanship, we learned, typically does not shift if Americans think a party failed to perform well. Nor does it typically shift because people move their attitudes toward or away from a party based on policy positions.

During this period, the most common reason people changed parties is they come to believe that either party moved away from or toward them on one of two broad issues: social spending on jobs and poverty reduction, and achieving racial justice. Ronald Reagan used to say he did not leave the Democratic Party, but that it left him. Similarly, Lyndon Johnson is reputed to have remarked that passing the Civil Rights Act of 1964 would cost the Democrats the South for a generation.

Southern Whites illustrate the importance of parties’ ideological shifts. Between the mid-1950s and 2016, Southern Whites showed little change in their attitudes about where and how the federal government should intervene on civil rights issues. However, their perceptions of the parties' positions on civil rights and racial justice changed dramatically between the mid-1950s and the late 1960s.

As a result, by the 2010s, Southern Whites have changed from being mostly Democratic to overwhelmingly Republican. That has endured far longer than the one generation Johnson predicted.

Future scenarios

After analyzing the historical data, we used our findings to predict 2020 partisanship and beyond, if the parties took various different positions on economic and racial issues in the future. This exercise is similar to how weather forecasters use past data to predict the path and severity of likely future storms.

If after this election Democrats stay where they are ideologically and Republicans continue to move further to the right, we find that key groups of both Whites and Blacks would shift toward the Democrats. Depending on what positions Republicans take, we predict an eight to 16 percentage point Democratic vote gain in national elections. This could happen, for instance, if the Republicans were to cut taxes for the wealthy even further, repeal the Affordable Care Act and its protections for preexisting conditions and continue to avoid condemning White racism.

However, if Democrats were to adopt extremely progressive positions — such as pushing for fully nationalized health care and for massive increases in social welfare spending — while Republicans stopped shifting further right, we find equally bad results for Democrats. Overall, the electorate would shift toward identifying with Republicans, and Republican votes would increase commensurately.

If both parties adopt centrist strategies beyond the 2020 election, we do not expect much change in overall partisanship.

How Trump has affected the Republican Party

We can see these patterns in action as recently as the 2018 and 2020 election cycles. Republicans have embraced Trump’s more extreme policies and personality, while Speaker of the House Nancy Pelosi (D-Calif.) has worked to keep Democrats focused on their traditional issues and positions. As a result, in 2018, Republicans lost 42 House seats and ceded control of the House. Current polling data suggest similar results in 2020, with Republicans potentially losing both the presidency and the Senate. The Democratic Party, despite vigorous efforts by its left wing, has mostly stayed centrist in the current campaign.

If the party of Trump goes down in a defeat like that of McGovern, will Republicans move again toward the center?

Ken Kollman is professor and John Jackson is professor emeritus of political science at the University of Michigan. They are co-authors of the forthcoming “Dynamic Partisanship” (University of Chicago Press, 2021).