We know that midterm elections are almost inevitably referenda on the incumbent president. One way to think about it is that in November of a president’s second (or sixth) year in office, voters go to the polls looking to vote for or against that president and, not finding the name, simply vote for or against every candidate wearing the same color jersey as that incumbent president. The deeper into a president’s term we get, the more meaningful his job-approval rating becomes, and the greater its predictive value.
As Gallup’s Frank Newport and Lydia Saad recently wrote in Public Opinion Quarterly, “The strong relationship between presidential approval and both presidential and midterm elections is fascinating and impressive given the simplicity of this question devised more than 70 years ago.”
The bottom line: Democratic hopes of retaining their slim majorities in Congress are almost entirely dependent upon President Biden not sinking them.
Biden’s approval ratings in the Gallup poll have ranged from a low of 50 percent (in their most recent survey earlier this month) to a high of 57 percent (in their first poll of his presidency, taken over his first two weeks in office). Gallup’s average over seven polls is 55 percent, with 41 percent disapproving.
Even a cursory look at presidential approval ratings in this period of ultra-partisanship underscores how monolithic each party is. Among Democrats, his approval has ranged from 90 to 98 percent; among Republicans, he’s been between 8 and 12 percent. There is little question how partisans on each side would vote; the only question is how many of them will show up.
The mood of independents demands special attention. They are the voters untethered by partisanship, typically very fickle, not liking or trusting parties or politicians, and not even following politics as closely. Biden’s approval rating among independents has ranged from as low as 48 percent in the most recent poll, to 61 percent in that first poll since taking office.
This column has noted many times the lack of volatility in Biden’s approval numbers (and for that matter Donald Trump’s), caused by the extreme partisanship on both sides—near universal support for a president among those of the same party, almost uniform opposition among those in the other party. Over the weekend, CNN’s Harry Enten pointed to the recent erosion in Biden’s numbers—nothing precipitous but certainly worth watching.
In a subsequent Q&A on Gallup’s website, Newport noted that “our data show that George H.W. Bush had a first-term average of 61%, which seemingly should have made it easy for him to sail to a second term—but, of course, he was defeated by Bill Clinton in the 1992 election. So the averages across four years, we assume, are less important than where the approval rating is in the months before Election Day. For Bush, his high ratings sank like a stone in the summer and fall of 1992.”
While their discussion focused more on presidential reelections, the lesson is just as good for measuring the buoyancy, or lack thereof, of a president in midterm elections.
Indeed, if there’s a single, commonly asked polling question that could predict the outcome of the elections next year, Biden’s approval rating—particularly among independents—is probably it.
Emory University political scientist Alan Abramowitz makes the case for watching the generic congressional ballot test question instead. His point is well taken, but in my opinion the question is not asked often enough to allow really close scrutiny of the ups and downs.
One smart Democratic analyst privately argues that midterms are less a referendum on the incumbent president’s party than a reflection of the fact that whichever party loses the presidential race is bound to go into the next election with much higher levels of enthusiasm. He urges caution in a party becoming overconfident about a state it just narrowly won.
Midterm elections have a lot of moving parts. There is never just one thing to watch. But there’s no better baseline than Biden’s approvals.
This article was originally published by the National Journal on July 27, 2021.
Image credit: AP Photo/Andrew Harnik
Our subscribers have first access to individual race pages for each House, Senate and Governors race, which will include race ratings (each race is rated on a seven-point scale) and a narrative analysis pertaining to that race.