Are midterm elections a referendum on the president?

There is evidence that suggests that the president’s performance can influence voting in the midterm elections even though they are not on the ballot:

A Presidential Referendum? - A Gallup poll conducted in 2014 found that 52% of registered voters planned on voting not based on the strengths of the congressional candidates, but to send a message to former President Obama. Around 20% considered their vote to be a message of support, but 32% wanted to send a message of opposition to the President’s policies. In the 2014 mid-term elections, the Democrats lost their majority in the Senate, while the Republicans increased their majority in the House of Representatives. It was very similar in the 2010 midterm elections, where 30% considered their vote as a message of opposition to Obama, and the Democrats lost their majority in the House. In 2010, the unemployment rate was over 9% and the President’s approval rating had fallen below 50%. The first two years of his Presidency saw the passage of several significant, but also very controversial bills. The Patient Protection and Affordable Care Act (2010) reformed the nation’s healthcare system, and the American Recovery and Reinvestment Act (2009), approved a $787 billion stimulus to help the economy. But, to the growing Tea Party movement, these laws were a sign of an overreaching government that was creating a worryingly large national deficit and debt.

On rare occasions, popular support for the president can appear to help their party to gain seats at the mid-terms. For example, in the 2002 mid-term elections, 28% considered their vote to be one of support for President Bush. The President’s approval ratings had increased significantly following the 9/11 terrorist attacks in New York, peaking at 90%. By the time of the November elections, a year later, his approval was still at 68%, far higher than the 51% he had prior to the attacks, and, correspondingly, the Republicans gained two seats in the Senate, and 8 seats in the House.

2018 Trump declared the midterms to be a great success even though the Republicans lost their majority in the House. He argued that thes results were better than Obama’s midterms and that there had been no ‘Blue wave’ i.e. sweeping Democrat victory in both Houses, The Republicans held on the Senate with a 2 seat net loss for the Democrats. However, this was a midterm the where Democats would find success difficult. The majority of Senate seats up for reelection were Democrat and as Nate Silver of FiveThirtyEight wrote that Democrats faced one of the most unfavorable Senate maps that any party had ever faced in any Senate election. Silver noted that ten of the seats Democrats defended were in states won by Donald Trump in the 2016 presidential election. Silver predicted that even a nine-point victory in the nationwide popular vote for Congress would not be enough to give Democrats a majority in the Senate. Some observers speculated that Republicans might be able to pick up a net of nine seats, which would give them the 60-seat super-majority necessary to break filibusters on legislation.

Republicans won a net gain of two seats in the Senate.

In the House Democrats defeated 29 Republican incumbents and picked up 14 open seats. Republicans did not defeat a single Democratic incumbent, though the party did pick up two open seats in Minnesota and one in Pennsylvania.

Is ‘All politics local’?

Nationalised campaigns - In recent decades, congressional campaigns have become increasingly nationalised, focussing on national issues, and the president’s performance on these issues, rather than local concerns. In the 2014 mid-term elections, Republican candidates focussed overwhelmingly on their opposition to President Obama, particularly his healthcare reforms and the economic stimulus. By October 2014, almost 100,000 TV ads had attacked the Patient Protection and Affordable Care Act (2010), and over 120,000 ads had targeted President Obama. Most of these ads came from Republican candidates, or conservative interest groups and Super PACs, but even some Democrats tried to distance themselves from the President’s performance and policies. Democratic Senator Mark Pryor ran an ad criticising the President’s proposed new gun controls, while Democratic Senate candidate Alison Lundergan Grimes ran ads opposing Obama’s energy policies, which she said threatened Kentucky’s coal industry.

However, there are many other, arguably more important, influences on the final result:

Some voters base their decision on local Issues and the strength of the individual candidates

US elections are generally seen as more candidate and issue centred than in the UK, where voters focus more on the parties than the individual candidates. The campaigns and policy positions of Democratic and Republican candidates can vary considerably around the country. Democratic candidates running in southern states tend to be more conservative than in the north, while Republican candidates running in the north tend to be more liberal than in the south. Therefore, even given former President Obama’s unpopularity in the south, it is possible that a conservative Democrat, who, for example opposes gun controls and strict environmental regulations, could still win in a more conservative state. However, this was certainly not the case in the 2014 Senate elections. Even though southern Democratic incumbents like Mary Landrieu (Sen. Louisiana) were much more conservative than President Obama, and spoke out on local issues like protecting coal and oil industries, they were still soundly defeated. The 2014 midterms were the first since 1980 in which more than two incumbent Democratic Senators were defeated by Republican rivals. In the Senate, 5 of the 17 incumbent Democrats lost their seats, but all 12 Republican incumbents were undefeated.

Gerrymandering and wasted votes mean that few seats are actually genuinely competitive

In most states, redistricting is done by the state legislature, so when one party has the majority they can gerrymander the boundaries to help their party. Gerrymandering works by purposely wasting votes for your opponent, either by packing all of their support into just a couple of districts, or by thinly spreading opposition support over a large number of districts, so that they lack the plurality needed to win. As congressional elections use First Past The Post, the seats won by each party is rarely proportional to their national vote share. In 2012, the Republicans won 234 seats in the House with only 46.9% of the national vote, while Democrats won 201 seats, with 48.3% of the national vote. In 2014, the Republicans had 51.2% of the national vote, while the Democrats received 45.5%. But this 6% difference gave the Republicans 247 seats, and the Democrats just 188. FPTP also results in large numbers of safe seats, with very few competitive districts that could actually change hands. In 2014, over 380 seats were considered to be safe Republican or Democrat seats, making it harder to argue that the result reflected support or opposition to the President.

The advantage enjoyed by incumbents means they are less affected by the president’s performance

Despite Congress having an approval rating of around 13%, over 96% of House incumbents were re-elected in 2014. In the House, only 11 Democrats and 2 Republicans were defeated. Incumbents enjoy a number of advantages. They can send free letters to their constituents without incurring any costs. They have large teams of staff who can professionally manage their election campaign. They have far greater name recognition and media attention. They have a voting record with which they can prove their commitment to certain issues and show where they have ‘brought home the bacon’ for their district or state. And, importantly, they find it much easier to fundraise, as donating to incumbents can be seen as a wiser investment than donating to an outsider who is unlikely to win. In the 2014 midterm elections, seats were far more likely to change hands if the incumbent had resigned, suggesting that the number of open seats being contested can have a significant influence on the final result.

Differential turnout/ abstention can affect the final result even more than actual votes cast

The turnout or abstention of different social groups in presidential and mid-term elections can also greatly impact the final results. Turnout in the 2012 elections was around 58%, but, in the 2014 mid-terms it was just 36%, the lowest for 70 years. The reason why this can have such a significant impact on the result is that particular social groups tend to have much higher rates of abstention in mid-term elections. For example, younger voters, who tend to vote for Democratic candidates, are much less likely to vote in the mid-terms. Youth turnout (18-29) was estimated to have been around 45% in the 2012 presidential election, but it dropped to just 21.5% in 2014. The electorate in mid-term elections tends to be overwhelmingly white, male and older than in presidential election years, and, as these groups are now more likely to vote Republican, this can have a significant impact on the results. In the 2012 election, around 16% of voters were over 65 years old, and 72% were white. In 2014, 22% were over 65, and 75% were white.

The absence of presidential coattails might be as significant as the president’s performance

During presidential election years, congressional candidates can benefit from a coattails effect, where the excitement surrounding a presidential candidate increases turnout and support for other candidates from the same party. When these coattails are then absent in the mid-term elections, the president’s party loses seats. This could explain why the presidents’ parties have lost congressional seats in all but three mid-term elections in the last 100 years. In 2008, there was tremendous excitement surrounding the Democratic candidate Barack Obama, while the previous Republican President, George Bush, had record low approval ratings following the controversial wars in Iraq and Afghanistan, and the 2007 financial crisis. These conditions helped Democrats to win in a number of conservative districts and states. However, the majority of these ‘Blue Dog Democrats’ were then swiftly defeated in the 2010 midterm elections. Without an exciting new presidential candidate on the ticket, and without broad dissatisfaction at former President Bush, many voters in these districts returned to voting for Republican candidates. For example, in 2008 the Democratic Senate candidate Mark Begich defeated six-term Republican Senator Ted Stevens in Alaksa. But in 2 014, without Obama’s coattails to help him, Begich was easily defeated after just one term.

In 2020 Trump lost the election but his party did well. There was a kind of coattails effect as Trump manged to energise nearly 6 million more Republicans to vote and, unfortunately for him he also motivated more Democrats- while this cost him the election the greater Republican turnout, coupled with safe seats, gerrymandering and the structural advantage which Republicans have- i.e. it takes more votes to elect a Democrat than a Republican- meant that the party of the president did well.