Research-based perspectives from MIT 

Democracy in the balance? | Devin Caughey 
Associate Professor of Political Science

Graphic via iStock

Perspectives for the 2022 Midterm Election
Main Page

Devin Caughey is an associate professor with tenure in the MIT Department of Political Science. His research focuses on American politics, often from a historical angle, but he also works on political methodology and dabbles in comparative and international politics. His book, The Unsolid South: Mass Politics and National Representation in a One-Party Enclave, was published by Princeton University Press in 2018. 

Politics, unlike physics, does not have many scientific laws (Maurice Duverger claimed one of the few). One empirical regularity that comes close, however, is the tendency for the U.S. president’s party to lose ground in midterm elections. This tendency, known as “midterm loss,” is most visible in Congress. Since 1914, the first time all U.S. senators were elected by popular vote, the median midterm loss has been four seats in the Senate and 28 seats in the House of Representatives. Only twice (in 1934 and 2002) has the president’s party gained seats in both chambers, a paltry 7% success rate.

Midterm loss is not confined to Congress, however. The president’s party suffers down ballot as well, in races for governor, state legislature, and other state and local offices. For example, in Bill Clinton’s first midterm election (1994), his fellow Democrats lost control not only of Congress, but also of 11 governorships and 14 state legislative chambers. Indeed, as Steven Rogers has shown, partisan swings in congressional and state legislative seat share are tightly correlated in midterm as well as non-midterm elections.

Once Joe Biden was elected president in 2020, then, the 2022 midterms were destined to be a promising election cycle for Republican candidates at all levels. Given the Democrats’ narrow congressional majorities (222–213 in the House, 50–50 plus the tiebreaker in the Senate), even a below-average midterm loss would hand Republicans control of Congress in addition to augmenting their existing dominance in the states. The main question is not whether Democrats will lose seats but rather how severe those losses will be.

See-sawing indicators

The signs are mixed. On one hand, President Biden is relatively unpopular. His 42% approval rate is about where Donald Trump’s was in 2018, when Republicans lost 40 seats in the House (but gained two in the Senate). The last president to gain House seats at midterm, George W. Bush in 2002, had an approval rating 20 points higher. Moreover, the state of the economy is middling at best. At 3.5%, the unemployment rate is low by historical standards, but inflation is higher than it has been in decades, resulting in declining real wages for most workers.

On the other hand, the Democratic Party seems to be more popular than President Biden is, at least compared to the Republican Party. Since August, Democrats have enjoyed a slight lead in the generic congressional ballot. This represents a reversal of the state of affairs earlier in 2022, when congressional Republicans consistently polled ahead of Democrats. The Supreme Court’s June decision in Dobbs v. Jackson Women’s Health Organization to overturn Roe v. Wade may partly explain this change of partisan fortunes. There is good evidence that one of the drivers of midterm loss is partisan “balancing”: some voters react against policy shifts to the left (under Democrats) or right (under Republicans) by throwing their support to the opposing party. As a sharp rightward shift in abortion policy, Dobbs undermines the usual public perception that policy moves left under Democratic presidents, weakening the logic of balancing. Dobbs also seems to have galvanized pro-choice Americans, particularly young women, potentially mitigating the turnout drop-off that Democrats in particular often face in midterm elections.

Republicans are also facing a repeat of problems they faced in 2010, President Barack Obama’s first midterm, when inexperienced and extreme candidates may have cost them several winnable seats (and control of the Senate). As Republican Senate Leader Mitch McConnell acknowledged this August, Republicans’ chances are better in the House than in the Senate due to poor “candidate quality” in states such as Pennsylvania and Georgia. Many of these Senate candidates were endorsed by former President Donald Trump and prevailed over more moderate or conventional candidates in the Republican primary. Republicans’ odds of capturing the Senate are still very good, but if Democrats have a chance of defying the midterm curse, they will likely do so by holding onto the Senate.

Democratic institutions on the line

What are the stakes of the 2022 midterms? At the national level, a Republican takeover of either chamber would mean a barrage of investigations on topics ranging from President Biden’s son Hunter to the president’s decision to withdraw from Afghanistan last August. It would also likely mean a return to the sort of political hostage-taking and brinksmanship characteristic of the later Obama years (see here for a preview), with last-minute deals negotiated among party leaders. Legislating would continue in this “unorthodox” form, but large policy changes would be largely off the table. If the Democrats hold the Senate, President Biden will at least remain able to fill administrative and judicial vacancies, including on the Supreme Court should any appear.

At lower levels, the stakes are more complicated and possibly much more consequential. In recent years Republicans have made so many gains in state legislatures that they have only a few pickup opportunities, including chambers in Maine, Minnesota, and Alaska. The party does has a decent chance of flipping Democratic governorships in Oregon, Nevada, Kansas, Pennsylvania, and Wisconsin, though it is on the defensive in Arizona. The outcome of these elections may affect whether, for example, states take advantage of Dobbs to pass new restrictions on abortion.

The 2022 elections also have more fundamental implications. According to the Washington Post, a majority of Republican nominees have denied or cast doubt on the legitimacy of the 2020 presidential election. These include, for example, the Republican nominee for Secretary of State of Michigan, the office responsible for overseeing elections, who touts baseless claims that fraud won the state for Biden in 2020. The election of such candidates to statewide offices, or to a critical mass of state legislative seats, poses a deep and possibly existential threat to American democracy. In 2020, a constitutional crisis triggered by Trump’s refusal to concede was averted thanks in large part to lack of cooperation from Republican officials like Georgia Secretary of State Brad Raffensperger. Whether we can expect such commitment to the rule of law in the 2024 election may depend on what happens this fall.


Suggested links

MIT Political Science

Devin Caughey's MIT webpage

The Unsolid South: Mass Politics and National Representation in a One-Party Enclave (2018)


Series published by SHASS Communications
Office of the Dean 
Alison Lanier: Web Publishing; Senior Communication Associate
Leda Zimmerman: Principal Editor