According to most forecasts, the Democrats probably will lose control of Congress in the upcoming midterm elections. That would be consistent with the pattern for nearly 30 years: When one party controlled the presidency and both houses of Congress, voters divided it up again in the 1994, 2006, 2010 and 2018 midterm elections.
Given that looming possible midterm loss, progressive and centrist Democratic politicians are once again arguing over electoral strategy. Rep. Alexandria Ocasio-Cortez (D-N.Y.), for instance, asserted recently that the real reason for the Democrats’ dismal midterm prospects is that they’ve failed to pass bold legislation like the Build Back Better spending package.
But our research suggests the opposite. A strongly progressive strategy would more likely alienate moderate voters and jeopardize Democrats’ ability to sustain a united government.
Lessons from the 2010 midterm elections
Consider evidence from the midterms in 2010, for example. In March that year, a Democratic Congress had passed President Barack Obama’s Affordable Care Act, which Republicans attacked as a risky departure from the existing system of private health care. In the figure below, you can see the Obama 2008 vote in the districts of Democratic incumbents by different levels (45 percent and below, 46-50 percent, 51-55 percent, 56-60 percent, above 60 percent) along with the percentage of House Democrats who won despite voting for the ACA at each level.
The higher the prior level of support, the more likely the Democratic incumbent won, despite voting for the ACA.
Centrist voters determine elections
That’s because moderate and conservative Democrats and true independents (meaning independents who say they don’t lean toward one party or the other) defect after a bold progressive move. To determine this, we looked at the YouGov/Economist survey data from the week before the elections of 2010, 2014 and 2018. Here we looked at who liberal, moderate and conservative Democratic voters said they intended to vote for during a period of unified Democratic government (2010); divided government with a Democratic president (2014); and unified Republican control (2018).
The Economist YouGov poll is done weekly and is a national sample of 1,500 adult citizens surveyed via the internet. The polls are weighted by age, gender, race, education and previous presidential vote. The data show that moderate and conservative Democratic voters turned against the party in 2010 after it passed the ACA, as you can see in the figure below.
However, by 2014, some of the moderate and conservative Democrats of 2010 began to return as they witnessed Republican control of the House and Senate. And the unified Republican government under Donald Trump brought back even more moderate and conservative Democrats by 2018.
Moderate House members pay the electoral price
Why, then, are some progressives more willing to risk losing power by dragging their party to left of the median voter?
One answer is they are far less likely to suffer at the ballot box. The figure below compares the Americans for Democratic Action (ADA) scores of winning and losing candidates from 2000 to 2020. ADA scores members of Congress on the basis of how they voted on 20 or so roll call votes. A perfect score of 100 indicates a liberal voting record. (You’ll see that the figure doesn’t include 2006 or 2018 because no Democratic incumbents lost to Republicans in those years.)
In the nine elections in which incumbent Democrats lost to Republicans, the average ADA score for defeated Democratic incumbents was in all instances lower than the average House Democratic ADA score. In other words, Democrats who lost general elections were not as progressive as the average Democratic member of the House.
To put it another way, progressive incumbents don’t have to worry about losing general elections — because they typically represent safe seats in which the only plausible threat to a very liberal Democrat comes from someone even further to the left.
When Democrats control the presidency and both houses of Congress, moderates’ seats are at risk
When Democrats take control of Congress, the party promotes policies that are further left — even though the party has taken control by adding new Democratic members who flipped Republican seats, meaning that they tend to be more moderate. At the same time, losing the flipped seat shifts the Republican Party to the right, making it even harder to win their votes for Democratic legislation — and leaving Democrats solely responsible for those policies.
Expecting midterm defeat can create a “grab and go” mentality on the left: Grab the policies that they can get and accept loss of power as the price. Centrist Democratic members are in a difficult position: Defy the progressives and face a primary challenge, or go along to get along and face a difficult reelection fight in November.
The more that progressives believe that losing government control is inevitable or out of their hands because of circumstances such as inflation, the more rational it is for them to try to take what they can get in the short run. They can count on the fact that divided government will eventually place Republican dysfunction back in the spotlight, enabling Democrats to unite again around what they oppose instead of what they agree on, get back into power, and replay the grab-and-go cycle.
Centrist voters and the 2022 election
Where does this leave centrist voters and candidates?
Since incumbents will be tied to the most left-leaning proposals regardless of whether they supported them, united Democratic control looks like a bad deal to them. Unsurprisingly, many prefer divided government. In the figure below, a more recent YouGov survey reveals liberal, moderate and conservative Democratic voters’ preference for control of the House in 2022. Only 69 percent of moderate or conservative Democrats, who make up 40 percent of party ranks, want the House of Representatives to be controlled by the Democrats; almost 90 percent of liberal Democrats do.
Unified government control is a bad deal for moderates, as they are seeing more and more over time. Following progressive electoral logic risks leading to Republican unified control — which would enable them to enact policies that move in the opposite direction from where all factions of the Democratic Party want to end up.
David Brady is an emeritus professor of political science at Stanford University.
Bruce Cain is a professor of political science at Stanford University.