Two days after the election, on November 5, House Democrats held a private conference call to air grievences. Before the election, the party was expected to win a majority of 43 seats according to the FiveThirtyEight model, which would be a pick up of 7 seats, but it was clear by that Thursday that the party’s majority would be substantially narrowed. Abigail Spanberger, a Blue Dog Democrat from Virginia and member of the Problem Solvers caucus inspired by centrist political organization No Labels, blamed progressive Democrats in heated terms. She perceived that the problem was that left-wing messaging, specifically regarding “socialism” and “defund the police”, was unpopular. “If we are classifying Tuesday as a success from a congressional standpoint, we will get fucking torn apart in 2022.”
Progressive members of Congress, such as Seattle congresswoman Pramila Jayapal, co-chair of the Congressional Progressive Caucus, argued that left-wing ideas energize the party base. Rashida Tlaib, one of Detroit’s congresswomen and member of the progressive quartet “The Squad”, argued back, saying “To be real, it sounds like you are saying stop pushing for what Black folks want.”
This meeting sparked days of arguments between progressive groups and establishment groups. One skirmish was between former Republican governor of Ohio John Kasich, a Biden supporter, and Alexandria Ocasio-Cortez, outspoken progressive member from the Bronx and member of “The Squad”. Kasich urged establishment Democrats to put progressive Democrats in their place, saying “the Democrats have to make it clear to the far left that they almost cost [Biden] this election”. Ocasio-Cortez responded in an interview with the New York Times, “If the party believes after 94 percent of Detroit went to Biden, after Black organizers just doubled and tripled turnout down in Georgia, after so many people organized Philadelphia, the signal from the Democratic Party is the John Kasichs won us this election?” She continued to make her case that progressives actually did alright on a CNN interview with Jake Tapper, “Progressives have assets to offer the Party that the Party has not yet fully leaned into…Every single swing seat member that co-sponsored Medicare for All won their re-election, and so the conversation is a little bit deeper than saying anything progressive is toxic.”
So who carried who? Did progressives drag the Democrats down, or did they energize the base and lead Democrats to victory despite Biden’s moderate impulses? Did blue dogs like Spanberger win because their centrism was forgivable to a redder electorate, or did they almost lose because they weren’t progressive enough?
I wanted to look at this with actual voting data. I wanted to compare the margin of victory (or defeat) for Biden with the margin of victory (or defeat) for the Democratic congressional candidate in each district. If what Spanberger was saying is true, one would expect moderate, centrist, blue dog Democratic candidates to outperform Biden’s margin and progressive Democrats to underperform. If, however, Ocasio-Cortez is correct, one would expect progressive Democratic candidates to outperform Biden’s margin and blue dogs to underperform.
The first step was determining who counts as a progressive and who counts as a blue dog. There is not a simple cut-and-dried way to categorize members of congress as left-leaning or right-leaning, especially when perceptions of candidates are generally crafted by the candidates in their separate districts (or their opponents) and these may vary widely across geographies. But there are certain national indicators of progressiveness or centrism that I chose to follow. These are the indicators I used to determine who counts as a centrist Democrat:
— Membership in the congressional Blue Dog Coalition
— Endorsements from Blue Dog PAC
— Being in the upper right area of Democrats in this DW Nominate score plot
— Membership in the congressional Problem Solvers Caucus
— Endorsements from the US Chamber of Commerce, which almost never endorses non-Republicans
— Democrats with a grade C or better from the National Rifle Association
These are the indicators I used to determine who counts as a progressive Democrat:
— Endorsements from Ocasio-Cortez’s Courage to Change PAC
— Endorsements from Justice Democrats
— Endorsements from the Progressive Change Campaign Committee
— Being in the bottom left area of Democrats in this DW Nominate score plot
— Endorsements from Bernie Sanders
— Membership in the Congressional Progressive Caucus
— Co-sponsors of H. Res. 109, Ocasio-Cortez’s Green New Deal resolution
— Co-sponsors of HR 1384, Jayapal’s Medicare for All Act of 2019
Folks who were not categorized as a progressive or a blue dog were by default categorized as establishment, though I recognize this may be an overly broad and inaccurate categorization for some.
I segmented the incumbent progressives into two mostly-but-not-completely overlapping categories, Candidates Marketed as Progressive, and Candidates who Vote as Progressive. Candidates Marketed as Progressive are those who have chosen to join the Congressional Progressive Caucus and welcome the endorsements from left-wing groups and Bernie Sanders. Candidate who Vote as Progressive are the co-sponsors of progressive legislation and those whose votes are left-wing according to two axes of DW-Nominate scores. Most Democratic incumbents who wish to be perceived as progressive also vote in a progressive manner, but there are a handful of members of the Congressional Progressive Caucus who did not co-sponsor either of the two biggest progressive tentpole pieces of legislation of the 116th congress and who do not regularly vote in a progressive way (“Fake Progressives”). Similarly, there are actually quite a few Democrats who do not seek to be identified as progressive but who do vote in a progressive manner and have co-sponsored Medicare for All or the Green New Deal (“Sneaky Progressives”).
This is a diagram of how I categorized Democratic incumbents in this analysis. This is only a partial list and includes only those districts where I have presidential voting data and where there was a single Republican opponent.
This is a diagram of Democratic challengers. Note that because none of these people voted in the 116th Congress, it was not possible to distinguish candidates who would actually vote in a progressive manner. It is also noteworthy to point out there are many more establishment candidates than either blue dogs or progressives in this group. These are people who, for the most part, are competing in districts held by Republicans, and so most chose to avoid association with left-wing ideas and endorsements. But the handful of candidates in swing districts who went after the mantle of progressive identity are important data points.
The second step was acquiring district-level data on presidential vote tallies. This is not as easy as it should be, since it requires someone to go into certified election results county-by-county, precinct-by-precinct, and add up votes for each candidate. I used data from Daily Kos Elections, Bloomberg reporter Greg Giroux, and my own painstaking analysis of certified precinct-level vote totals in Florida, Texas, Kentucky, and Michigan. Of the 469 senate and congressional races on the ballot in November 2020, I have data on 219 of them (47%).
The results show that Spanberger may be more correct than Ocasio-Cortez. While Joe Biden beat Democratic congressional candidates by an average of 1.6 points, Blue dogs/centrist Democrats outperformed Biden on average by 2.3 points.
One of the key things about this analysis however is that incumbent Democrats of all types did better than Biden. Incumbent Democrats are mostly found in districts full of Democratic voters who maybe are more in step with Democratic messaging. So saying, as Ocasio-Cortez did, that “every single swing seat member that co-sponsored Medicare for All won their re-election” is more of an indication that incumbents won than that progressives won. In fact, blue dog democratic incumbents outperformed Biden to a greater extent than did establishment candidates or progressive candidates. And the type of incumbent that did the poorest in relation to Biden’s margin were incumbents who sought to be known as the progressive choice.
Breaking it down even further, the progressives that did the best were the progressives that strove not to be identified as progressive. The progressive incumbents that did the worst were those who both voted for progressive ideas and identified themselves as progressive.
Democratic challengers mostly underperformed Biden’s margins fairly significantly. Though the number of challengers who courted the blue dog label were rather small, they seemed to do better than other challengers as a group (though none of them actually won).
There were 117 races that I designated as competitive. I defined competitive races as those that the FiveThirtyEight deluxe model showed as the winner having a 95% chance or less of winning, or that Larry Sabato’s Crystal Ball showed as likely or lean Democrat or Republican. Of those 117 competitive races, I have presidential vote tallies in 71 races. In competitive races, progressive candidates did far worse than Biden. The only category where the Democratic candidate did better than Biden was blue dog Democrats.
In non-competitive races, progressives did slightly better than Biden. But blue dogs did much better than Biden.
Now, I happen to identify as a progressive. I think that generally progressive ideas like Medicare for All, the Green New Deal, reforming policing, all of these are good ideas for the country. If people responded en masse to these policy ideas, progressive candidates would have been more popular than Biden. But this data shows me that the country seemed to prefer Joe Biden’s messaging over progressive messaging.