What Should Democrats Do?
David Shor and others say look to the polls for the answer, but they ignore personality and contingency at their own peril.
I’m Michael A. Cohen, and this is Truth and Consequences: A no-holds-barred look at the absurdities, hypocrisies, and surreality of American politics. If you received this email - or you are a free subscriber - and you’d like to subscribe: you can sign up here.
It’s Never Stopped Being 1968
In 2016, I wrote a book titled “American Maelstrom,” which chronicled the 1968 presidential election. While researching and writing, I kept coming across a palpable tension in the story I was trying to tell. On the one hand, the 1968 election was a reflection and culmination of the nation’s worsening social, cultural, economic, and racial trends. White privilege was under attack in both the North and the South as Black activists demanded social change - and whites feared losing the significant economic gains they’d made in post-war America. An emerging counterculture and shifting cultural mores (sex, drugs, and rock ‘n roll) created a notable and increasingly toxic generational clash. These powerful forces are a compelling explainer for America’s political transformation in 1968, the victory of Republican Richard Nixon, and the ascendancy of the conservative movement in American politics.
However, the margin in the 1968 presidential election was extraordinarily narrow Richard Nixon beat Hubert Humphrey by half a million votes. Had a mere 42,000 votes in three states from Nixon to Humphrey, no candidate would have gotten 270 electoral votes, the election would have gone to the House of Representatives, and Humphrey would have likely won. In an election that close, personality and contingency loom large. For example, had Bobby Kennedy not been assassinated in June of that year; had Humphrey come out against the Vietnam war sooner than five weeks before Election Day; had Lyndon Johnson begun peace talks with the North Vietnamese months earlier than he did; had Senator Gene McCarthy not challenged Johnson for the Democratic nomination for president … 1968 could have turned out very differently.
Don’t get me wrong: I’m as big a believer in empiricism and data when explaining politics, but we ignore contingency and personality at our peril. Today that feels more true than ever — because there is no “personality” who more clearly defines the cross-currents of modern politics than Donald Trump. Increasingly, I’m of the view that the best way to understand what’s happening our present political moment is understanding voter reactions to Trump and Trumpism, in general.
“Popularism”
All of this is a very long way of getting into my take on Ezra Klein’s deep dive into pollster David Shor’s analysis of what Democrats need to do to win elections today. In a political strategy that Shor calls “popularism,” he argues that Democrats, because of the structural impediments of the American constitutional system, must do a better job of attracting white, non-college-educated voters and stemming their decline among non-college-educated Black and Hispanic voters. In Shor’s view, it’s not enough to mobilize Democratic-friendly voting blocs. The party needs to reassemble the political coalition of minority voters, young people, non-college-educated whites (some not all) and affluent liberals, which put Bill Clinton into office in 1992 and 1996 and Barack Obama in 2008 and 2012. To do that, Shor says, the party should focus less on unpopular policies, like immigration, and focus instead on those that poll well, like lowering Medicare drug prices. That means coming up with better and more compelling messaging that will help them get through to these non-college-educated voters — and give them a fighting chance in purple or red states and districts.
Short’s strategy makes logical sense. But I’m not sure it works in the real political world.
First of all, campaign themes are messaging are not decided in a vacuum. The other side gets a vote. If Republicans focus entirely on immigration or crime, then Democratic candidates can’t simply ignore it. They need to have an adequate response. Shor would argue that when Democrats are defined by activists’ calls to “defund the police,” that’s a major political problem for them because it plays into GOP talking points. This is especially problematic in a period of extreme polarization because voters may be more inclined to lump all Democrats and Republicans into one big bucket. But in 2020, Joe Biden denounced “defund the police,” and few Democratic candidates embraced it. But since Republicans went all-in on attacking what was, in effect, a slogan from liberal activists, Democrats were forced to respond. I don’t care how good a candidate is at messaging; if your opponent is hammering you repeatedly and accusing you of holding a position you don’t, you need to address it. And as much as Democrats might like to control what progressives say and do … they don’t and I can assure you, they never will.
Second, Shor puts significant, and I would argue, misplaced emphasis on the power of messaging. According to Michael Podherzer (quoted in Ezra’s piece), the political director for the AFL-CIO and a leading Democratic voice on election strategy, “There’s no argument that saying unpopular things is better than saying popular things. My argument is it’s not close to being an important enough factor to warrant attention. If the object is for Democrats to win, that’s a tertiary, at best, factor.”
In a normal political world, the efficacy of political messaging is, at best, limited. At our current moment of intense polarization, it seems borderline fanciful to believe that it’s the key to electoral success. I don’t care how well a particular policy message scores in polls and focus group, for most Americans, what will matter far more is the party affiliation of the person stating it.
Even policy outcomes don’t appear to matter to voters much anymore, as presidential approval increasingly no longer tracks with the economy's strength. As I wrote last year, “There is a surprisingly wide gap between the president’s approval ratings and the electorate’s views of the economy. That was true before the pandemic when the economy was strong. Usually, if people feel positively about the economy, they give credit to the president. But not anymore. Americans are increasingly judging the strength of the economy by their political affiliation — yet another byproduct of intense polarization.” This gap was particularly wide during the Trump years, but it also existed at the tail end of Obama’s presidency, as the economy improved, and is happening now with Biden.
The places where policy outcomes may still matter is on parochial issues. Take a look, for example at the 2018 midterms. The election was, in key regards, a referendum on Trump, but local issues played a role as well. In late 2017, Republicans passed a major tax cut that included an cap on the state and local tax deduction (SALT). The SALT cap disproportionately affects Americans who live in more affluent communities with higher property taxes. A host of House Republicans in California, New Jersey, New York, and Illinois voted for the final bill. They were largely wiped out. According to an analysis from the Tax Policy Center, “the largest shift towards Democratic house candidates occurred in the 20 percent of counties with the greatest percentage of tax filers taking the SALT deduction.” Taking the opposite position didn’t help much either. Twelve GOP members went against their party to protect their constituents from tax increases. Five lost reelection, and two retired.
Shor is focused on a set of macro political issues and their efficacy in moving voters to the Democratic side — like allowing the government to negotiate lower drug prices through Medicare. But it’s quite possible that a more parochial issue like raising the SALT cap, as part of the president’s Build Back Better plan, will do far more to effect Democratic fortunes next year.
Third, as John Sides, Michael Tesler, and Lynn Vavreck have written in their excellent book “Identity Crisis,” identity has become a far more significant driver of voter decision-making than in the past. It’s not just racial identification. It’s also political, and class identification, and these self-images are increasingly difficult to overcome. Indeed, one of the more striking elements of the Trump years was how quickly and passionately Republican voters switched their views on policy issues. On trade, confronting Russia, cutting government spending, and tackling the growth of social insurance programs and so-called family values, Republicans did a virtual 180, largely because of Trump. There’s an argument in political science research that people’s view on any particular policy is driven less by their understanding of the issue and more by the cues they receive from political elites. The Trump presidency offered strong support for that perspective. Indeed, Elliot Morris’s newsletter directed me to a fascinating study that found when voters were told a candidate’s views on a host of policy issues it didn’t change their mind about voting for the candidate — it changed their position on the policy issues.
In addition, identity runs both ways. Shor argues for focusing on issues that appeal to non-college-educated white voters, but downplaying the cultural issues that matter to Black and Hispanic voters could undermine that effort and diminish their enthusiasm. Smart politicians know how to play both sides of the fence and use messaging and political appeals that work with different groups of voters. Indeed, perhaps the best counter to Shor’s argument is that Democrats don’t necessarily need better messaging, they need smarter and more disciplined candidates.
Cult of Personality
Finally, there is the aforementioned issue of personality and contingency. On the contingency front, it’s worth pointing out, for example, that if not for James Comey’s last-minute intervention in the 2016 campaign, Hillary Clinton probably wins the White House, and we’re having a very different conversation about the state of American politics today.
But putting that aside, I was struck by a recent series of tweets from Seth Masket, a political scientist at the University of Denver. He tweeted the percentage of the Black and Hispanic vote received by Democratic candidates since 2000:
Black percentage of the vote received by Democratic presidential ticket:
2000: 90
2004: 88
2008: 95
2012: 93
2016: 89
2020: 87
Latino percentage of the vote received by Democratic presidential ticket:
2000: 62
2004: 53
2008: 67
2012: 71
2016: 66
2020: 65
Two things stand out from these numbers. First, the share of the Black vote received by Democratic candidates was virtually identical in 2000, 2004, 2016, and 2020. It jumped by 5 to 7 points in 2008 and 2012. Second, among Hispanic voters, support for Democratic candidates jumped in 2008 and 2012 before slipping in 2016.
So what happened in 2008 and 2012? The Democratic Party nominated a Black man to be its party’s nominee for president. That certainly would explain the party’s boost in support among Black and Hispanic voters, as well as the drop off in 2016 and 2020. Without Obama on the ticket, enthusiasm for the Democratic Party among minority voters declined. Conversely, polling data in 2016 and 2020 shows a gender gap among Black and Hispanic voters — men were more likely to support Trump than women. Perhaps, this is a function of men finding Trump personally more appealing than a female candidate — or just being attracted to him because of his background and wealth.
Podhorzer gets to this point:
“A lot of things Trump did were grossly unpopular but got him enormous turnout and support from the evangelical community. Polling is blind to that. Politics isn’t just saying a thing at people who’re evaluating it rationally. It’s about creating energy. Policy positions don’t create energy.”
In 2008 and 2012, Obama created energy among Black and Hispanic voters. He also convinced a broad base of white working class voters to support his candidacy, but he had the added benefit of never having to face off against Donald Trump, who was far more effective at motivating Republican voters than Mitt Romney or John McCain.
In 2016 and 2020, Trump had the enthusiasm and energy of those non-college-educated white (and some Hispanic and Black) voters, rank-and-file Republicans, and many Americans who didn’t regularly vote. In places like Florida and Pennsylvania, Trump brought out hundreds of thousands of new or occasional voters, who hadn’t cast a ballot in 2012 for Mitt Romney, but did for him.
In 2018 and 2020, Trump created similar levels of enthusiasm … among his opponents. Democrats who stayed home in 2016, didn’t in the next two elections — and in 2020 the party was even able to bring out new and occasional voters. Suburban voters moved decisively in the direction of Democrats. In all three elections, policy messaging arguably took a back seat to ones feelings about Trump — both for better and for worse. Arguably, Republicans did better down ballot than they would have otherwise if not for Trump being the party nominee.
Is Trump Unique?
All of this raises perhaps the most essential question about our current political moment: how much of the results of the past five years are unique to Trump? In other words, can a different Republican politician mobilize Republican voters who turned out in 2016 and 2020 because of Trump? And conversely, can Democrats mobilize the voters who spurred them to victory in 2018 and 2020 if Trump isn't on the ticket?
As noted above, if you look back at 2016, Trump vastly outperformed Romney in two states (Florida and Pennsylvania), where Clinton either did better than Obama or nearly matched his 2012 numbers (this is also true in North Carolina). Will those Trump voters turn out again in 2022 and 2024 if he’s not on the ticket? Would a different Republican presidential nominee replicate the gender gap among Hispanic and Black voters?
In Michigan and Wisconsin, Democrats lost in 2016 because they didn’t get their voters out (Trump basically matched Romney’s numbers in both states). They did in 2018 and won major statewide victories. Can they do that again in 2022 when Trump is not on the ballot?
None of us know the answer to these questions, though I think that much of Trump’s support is unique. If he leaves the scene, I am skeptical that there is a Republican who can unify the party and mobilize his core supporters as effectively as he did. However, as long as he is around, he is the primary driver of voter attitudes, and that works in the Democrat’s favor. Trump remains a deeply polarizing and unpopular figure. Considering the limited benefits of policy messaging and the experience of the last few cycles, I think Democrats should dance with the proverbial girl who brung them. An anti-Trump message worked in 2018 and 2020, and with Trump increasingly looking like the likely 2024 nominee, it can work again.
Fear is a huge motivator in politics. It’s the root of Trump’s political appeal. But Democrats need to turn the tables by linking every Republican politician to Trump and warning that a GOP victory in 2022 will lead to a return of Trumpism. After all, this is what Republicans are doing — linking themselves to Trump because they believe that the enthusiasm he generates among Republican rank-and-file voters is the key to them winning back the House and Senate.
I want to be clear in saying that while I think personality is an underrated factor in our current political debates, it’s not the end-all, be-all. As noted above, messaging on parochial issues can help in congressional races, and along the margins, in close races, it can be decisive. There aren’t many swing voters left, but there are enough to make a difference in a few states and districts. National trends matter too, and when prominent liberals use slogans like “defund the police,” that can have a damaging trickle-down effect on other Democratic candidates, particularly those in close races. Data and polling are still probably the best way to understand the American electorate and strategize for an election. But it’s not everything. Personality and contingency weigh heavily and in the Trump era, perhaps more so than ever before in modern American politics. If Democrats want to win elections in 2022 and 2024, it needs to be front and center.
Took a while to get to this...well done. I have to hope that Trump sticks around for the midterms and the Dems can make him the major issue. If they can do that, and retain control of the legislative branches, he'll be gone for the '24 election. That will lead to a bunch of candidates trying to be the next Trump. Could be the first contested convention in a long, long time!