The battle between grassroots Democratic activists and Washington-based party leaders continued to unfold Tuesday night, with the national party notching some rear-guard victories and local forces delivering the party its second high-profile setback in as many weeks.
Through all of these contests, national party leaders have argued that their decision-making is not personal or ideological. They believe in the same progressive values as the grassroots activists, goes the argument, but more moderate candidates are needed to be able to win the general election and take the House back from Republicans.
That argument was made most explicitly earlier this month in the New York Times, by Brookings senior fellow Elaine Kamarck, who endorsed the practice of political parties intervening in primary elections. Kamarck was responding to The Intercept’s coverage of House Democratic Whip Steny Hoyer attempting to push a candidate in Colorado out of a House race by appealing to party elites’ superior savvy (emphasis added):
Are party leaders always right? Of course not. But they are different from the activists who often dominate the party primaries because they are more concerned with electability than with ideological purity. Party leaders have the job of winning nationally; Democrats are painfully aware that not all congressional districts are Berkeley, Calif.
Her contention, which mirrors conventional wisdom, is that party leaders — the loose network of campaign committees, consultants, elected officials, and key donors — are simply more strategic than activists, refusing to let ideology get in the way of their laser focus on winning elections.
That’s an assertion of fact, not opinion. And according to new political science research, it is incorrect.
A paper in this month’s edition of the peer-reviewed Legislative Studies Quarterly analyzes a decade’s worth of federal elections, finding that party organizations boost moderate candidates across the board, whether the general election is expected to be competitive or a long shot. In other words, party support for moderates does not appear to be strategic, but sincere. “They’re not doing this to have a better shot at winning elections,” said the paper’s author Hans Hassell, assistant professor of politics at Cornell College in Iowa.
The evidence points more to the conclusion that party elites “have strong incentives to prefer loyalists who can be trusted to implement its preferred policies after the nomination,” Hassell writes.
The study not only breaks with other political science findings, but decades of rhetoric from party leaders. It’s obvious from the most casual survey of primary elections that parties support moderates, but the races that observers tend to watch closely are competitive contests in swing states, so it stands to reason that a moderate in such a district may indeed be the smarter strategic play. Indeed, in a series of high-profile battles with progressive activists, the Democratic Congressional Campaign Committee has consistently positioned itself as being pragmatic, willing to bend on its progressive principles if doing so can lead to victory.
Hassell’s work expanded the field of vision, looking at races in which the Democratic nominee is likely to cruise to victory. The full scope of the research indicates that party leaders are actually committed to elevating candidates with a narrow range of beliefs.
If party elites were merely strategic actors, the data would show higher support for moderate candidates in swing races, while not showing as much support in seats that were either safe or out of reach. That’s not the case. In Hassell’s findings, parties consistently supported the more moderate primary candidate, regardless of the expected outcome of the general election. Even after excluding incumbents — which party committees almost always support — support for moderates holds. It’s also consistent regardless of party. And while this data set used Senate races, for his book Hassell also measured House races, finding the same result.
In Hassell’s findings, parties consistently supported the more moderate primary candidate, regardless of the expected outcome of the general election.
“Party elites are not systematically showing any preference for more moderate candidates in competitive districts,” Hassell writes. In fact, the pull for moderate candidates is stronger in noncompetitive districts. “This shows that parties are not strategically moderating their preferences in attempts to win competitive districts.”
Kamarck’s use of Berkeley to make her point is instructive to this end. If Hassell’s research is right, we’d expect to find elites even in Berkeley lining up behind the more moderate candidate, even though a communist is more likely to be elected there than a Republican. And indeed we do. Former Obama campaign aide Buffy Wicks is running for an open state Assembly seat, receiving large donations from the likes of Obama’s billionaire former Commerce Secretary Penny Pritzker. The majority of her donations for a down-ballot Assembly seat came from out of state in the initial reporting period. This is precisely the type of party elite donations that Hassell tracks to prove establishment support for moderates, regardless of the makeup of the district. (Her opponent, Richmond City Councilperson Jovanka Beckles, meanwhile, has the support of the Working Families Party, the local Our Revolution, and a host of other progressive groups.)
Kamarck’s reference to Berkeley may simply have been meant as a rhetorical flourish, but it ended up undermining her central claim. Hassell’s paper, which builds off his 2017 book, “The Party’s Primary,” includes interviews Hassell conducted with Republican and Democratic state party chairs, staffers, donors, and candidates, to see if what they say matches what they do. The interviews are inconclusive. While some parroted the line that the party network focuses more on winning, others highlighted splits with lower-level activists. “There absolutely is a disconnect between the elites — party leaders and donors — and party activists,” said one former state party chair who was unnamed in the paper. “They’re focused on different things. They’re different types of people.”
This ideological leaning can be best seen in how parties target viable candidates within their narrow networks. As a former party staffer puts it, “[The party’s elite] are all connected to each other. … And if they don’t know each other, they all know somebody who knows somebody who knows them. It’s a small group where information is shared.” So the candidate search cannot help but reflect the preferences of that small, insular group; it’s like looking under a streetlamp for your keys because that’s the only place where you can see.
Much of how party insiders coordinate on candidates happens under the surface and can be difficult to measure: endorsing, fundraising, supplying staff and polling support, sending hopefuls to candidate schools, presenting them before donors and PACs, and discouraging rivals. To quantify this, Hassell lists 199 primary races for U.S. Senate between 2004 and 2014, examining which candidates received the most donations from individuals who also gave to the main party committees, the Democratic Senatorial Campaign Committee and the National Republican Senatorial Committee.
Candidates with higher concentrations of shared donors line up with candidates who receive the most party support. “I’m positive that this is picking up the general trend,” Hassell said. For example, in Jane Norton’s 2010 Republican Senate primary in Colorado against Ken Buck, both candidates raised similar amounts, but Norton’s fundraising was five times as “connected” as Buck’s. This matches the media narrative that Norton was the insider candidate. (Norton lost that primary, and Buck lost the general election to Democrat Michael Bennet.)
The Database on Ideology, Money in Politics, and Elections also uses donor contributions to map candidates along a liberal-conservative spectrum. This has proven accurate in determining candidate ideology, even among candidates who never reach political office. Hassell supplemented this with more traditional scoring from congressional and state legislative voting behavior, where applicable.
Party elites can point to evidence showing that moderates have a better chance in tight races. Research from Stanford University using the same DIME data finds that parties nominating “extreme” candidates could lose between 4 and 7 points in a general election, because they mobilize opposition party turnout. The question of what counts as “extreme” — like Kara Eastman’s support in Nebraska for the right of a woman to choose her own medical care and broadly popular universal background checks for gun buyers — is obviously a critical variable.
The performance of the national party this cycle, when it comes to winning elections, has not been confidence-inspiring.
The performance of the national party this cycle, when it comes to winning elections, has not been confidence-inspiring. In March, Texas held its first round of primaries, a bloodbath for elite-backed candidates. In the closely watched 7th Congressional District in Houston, the party’s preferred candidate, moderate Alex Triantaphyllis, failed to advance to the runoff after the party tried but failed to stop Laura Moser. (A candidate with 50 percent of the vote would win the nomination outright; any less required a runoff against the second-place finisher.)
Moderate Ed Meier, a former Hillary Clinton staffer who had heavy Washington backing, also failed to advance past the first round in Dallas, where two progressives moved forward. In San Antonio, Jay Hulings had big money, the backing of the Blue Dog PAC, Rep. Steny Hoyer, Caucus Chair Rep. Joseph Crowley, and the Castro brothers, but came in fourth, with two progressives, Gina Ortiz Jones and Rick Treviño, moving to the runoff.
The Democratic Congressional Campaign Committee, the party’s campaign arm, regrouped, and picked its favorite of the runoff contestants across Texas, and in each case, the party-backed candidate won the nomination — including Lizzie Pannill Fletcher, who beat Moser on Tuesday night by 2-1. She and the other party-backed candidates were helped by huge margins in the absentee and early vote, the place where party support and organizational muscle can be most decisive.
In Kentucky, the DCCC’s hand-picked candidate, Jim Gray, was beaten handily by Marine fighter pilot Amy McGrath.
The finding that party networks support moderates in races expected to be won by far bigger margins than 4 to 7 points blows away the theory of strategic support where necessary to win races.
One political scientist consulted by The Intercept gave a reasonable alternative explanation for this finding. “It could be sincerity or it could be looking to the future,” said Gary Jacobson, a professor of political science at the University of California, San Diego, who supervised Hassell’s Ph.D. work when he was a student there. “If the party wants to advance forward, their image may be helped by someone closer to the center, closer to the district.” Similarly, in an easy victory, party leaders may believe that their national image can be hurt by a so-called extremist candidate.
But if parties pick moderates strategically in close races, and also pick moderates strategically in a blowout, what does “strategically” mean? “Sincerity and strategic behavior can be virtually indistinguishable,” Jacobson acknowledges.
Other experts who read Hassell’s work found it compelling. “I would say it’s very credible,” said James Shoch, a professor at California State University, Sacramento. “His method allows him to analyze a larger number of cases than a purely qualitative study could.”
Seth Masket of the University of Denver added that the data Hassell uses to generate his results “are put to a lot of different testing and they’re pretty good. … It’s a clever analysis.”
Hassell said that he believes party insiders come to their beliefs honestly: “They really view the way to a general election victory as their ideological stance.” To be fair, so do progressives, who often insist that running on big ideas like free college and single-payer health care will net electoral success.
Elaine Kamarck said that she would have to dive deeper into the methodology of the study before she could assess it properly. But she said the finding that party elites often back moderates even in deeply blue districts was plausible. “A lot of times, it’s not just partisanship, not just ideology — here’s so many other things, sometimes the moderate candidate is simply the better candidate,” she told The Intercept.
“They’re intervening in a very small number of races,” said Kamarck. “Party leaders have a right to try and win control of the House, which is how you get progressive policy, unless you just wanna feel good and have a bunch of goddamn Republicans running the House.”
The silver lining for activists in Hassell’s findings is that party support did not remain static over time.
The silver lining for activists in Hassell’s findings is that party support did not remain static over time. For example, the makeup and the politics of the party apparatus can change. In an Orange County, California, House race where Democrats are at risk of missing the top-two primary, the DCCC coordinated with local Indivisible leaders in choosing Harley Rouda, despite the state party supporting a different candidate.
“If you get a different set of individuals in that network, you get a different set of preferences and different people who they have connections with,” Hassell said. After 2010, when the tea party swept in more conservative Republicans in the House, party support, while still favoring relative moderates, shifted to the right. So if Democrats in office shift to the left, the ideology of the party network has the likelihood of shifting with it. For a mirror image of how that works, just watch Republican primaries, where candidates who were once clones of John Boehner or Mitt Romney now compete to out-Trump each other. The establishment is not a fixed thing.
But for now, the Democratic insider network is dominated by the Elaine Kamarcks of Washington. Her own ideology is a mix of sincerity and strategic business logic. She’s a founder of the New Democrat movement and manager of the “re-inventing government” initiative in the Clinton administration. She has parlayed that into consulting gigs with the centrist group Third Way and the RATE Coalition, a group of major corporations — including Walmart; Altria, the tobacco company; as well as insurance companies, telecoms and defense contractors — organized to lobby for corporate tax cuts.
Any battle between the populist-progressive wing of the party and its center necessarily revolves around the role of corporate power in shaping policy, but Kamarck said that she was under no obligation to disclose her consulting work. “I have no dog in the fight. That’s ridiculous,” she said, noting that the RATE Coalition was formed in 2010 during the Obama administration. “The position of the Democrats in the Obama administration was that you should have a corporate rate cut. … If somehow being for jobs for American workers makes me a corporate tool, I plead guilty. That has nothing to do with the analysis of party intervention in primaries.”
An expert on political campaigns, Kamarck also writes academic-looking papers for hedge funds locked in lobbying battles in Washington. In 2016, for instance, she waded into a housing policy dispute on behalf of a campaign to force taxpayers to bail out the federal housing agencies on behalf of Wall Street speculators. The paper, which argued for a large infusion of funding for affordable housing to go along with the hedge fund bailout, misspelled Ginnie Mae twice. Her co-author Robert Shapiro, also nowhere near an expert on housing policy, misspelled Fannie Mae in a discussion of the paper. The paper was written for the dark-arts lobbying firm the DCI Group as part of her relationship with the economic consulting firm Sonecon, though Kamarck told The Intercept that she was a subcontractor and was under the impression it was paid for by a low-income housing group.
It’s unclear if the views presented in the paper were sincere or pragmatic.