It is almost always a good idea to approach dramatic new poll numbers with caution. Americans are fickle, and polling them is often like trying to take the temperature of a 4-year-old: You can get a useful result, but it will take some wrangling — and it may not be useful for long.
So when Gallup released updated data on partisan identification on Monday showing a big swing to Republicans in 2021, my initial response was to recall that this metric in particular moves around a lot. When Democrats gained ground against Republicans in 2015, I cautioned against overinterpreting things. When they opened up another gap a year ago, I warned that this was not evidence of “a significant trend reshaping party identification.” Coming into this new data, then, I had the same caution: A big change in a volatile metric may not be as significant as a casual observer would assume.
In this case, though, there are two reasons that the shift measured by Gallup is important. The first affects the short term: this year’s midterm elections. The second is about the argument that Democrats are making about their opponents — an argument that appears not to be landing.
When considering Gallup’s measurements of party identification, it’s useful to recognize the patterns we expect to see. The first is that, over the past 15 years or so, most of the movement has been among independents, a group that’s mostly made up of voters who align with one party or the other. Below are the average identifications by quarter since 2004. The dark red and dark blue segments — those who identify explicitly with one party or the other — hold fairly constant in size. The shifts are usually in the middle.
That’s particularly obvious in the four most recent quarters — that is, the four quarters of 2021 that are the focus of Gallup’s new report. The first quarter of 2021 was particularly good for Democrats and Democratic-leaning independents. The last quarter was particularly not good.
From the first to fourth quarters, there were shifts in actual party identification as Republicans gained a bit and Democrats lost a bit. The shifts among leaning independents were bigger.
The shift is obvious when we look at the whole quarterly trend. (The dots on the graph below indicate individual poll results.) Since 2004, the right’s advantage on identification has never been larger. In the first quarter of 2021, the left’s advantage was the largest it had been since 2012. That’s a big change.
The timing is obviously quite bad for Democrats. The trend in party identification is pretty useful in predicting how House elections will turn out. When there’s been a big shift toward one party before an election (2006, 2010, 2014), that party tends to do well. The shift over the course of 2021 is the sharpest change recorded in the past 15 years.
That trend may not continue, of course. In 2020, there was a shift to the GOP in the first quarter of the year, a move that paralleled an uptick in approval for President Donald Trump linked to his initial response to the pandemic. That was fleeting. Should President Biden’s approval rating improve, it’s likely that partisan identification might shift back to the left. Whether that happens fast enough to limit damage to the party in November remains to be seen.
But this is a bigger problem for Democrats in the moment than simply one centered on the election. In the wake of Trump’s efforts to retain power after losing the 2020 election, Democrats argued fervently that the GOP posed an explicit risk to American democracy. There was a flurry of stories about how Republicans had turned on the party in the wake of the Jan. 6, 2021, attack at the Capitol, stories that were themselves overblown, but ones aimed at the idea that the Republican Party had become toxic to some segment of the population.
Gallup’s new data undercuts that idea severely. Americans don’t appear to be particularly concerned about the Republican Party’s response to 2020, particularly given the significant role that Trump still plays in setting its direction. Democrats have repeatedly hoped that Trump would prove so poisonous that the electorate would turn against the GOP. It worked in 2018, when the midterms served as a repudiation of Trump’s politics. It didn’t work in 2016, though, when Trump first won, and it offered only limited utility in 2020, when Trump earned significantly more support than he had four years prior, even while losing the popular vote by a wider margin. Democrats had unified control of government — but only barely.
And that was before Trump and congressional Republicans tried to subvert Biden’s victory. There are a lot of reasons for the swing back to the right over the past year, most of which center on Biden, not Trump. But Democratic efforts to cast the GOP as hostile to democracy itself either aren’t landing — as polling has suggested — or aren’t compelling.
In other words, Gallup’s data suggests both that Democrats are poised to lose ground this year and that a central argument against their opponents isn’t having a political effect. That bodes poorly for the left over the short term and the long term.
The good news for the party is that this metric is volatile.