Trash the Polls
Last month I turned the television on to CNN and resolved that I would watch the entire Democratic debate, from start to finish. That was before I realized it was going to go on for three hours. Before long, I was driven to distraction. That is, I opened my laptop. Who among us has not resorted to Twitter while watching these debates, so as to replicate the jeering camaraderie of the high school cafeteria?
We were well into the posturing of the Midwestern Moderates—Mayor Pete Buttigieg and Senator Amy Klobuchar, who now believe their only hope to break out of the pack is to hector Senators Bernie Sanders and Elizabeth Warren with “how ya gonna pay for it?”—when something on Twitter caused my fingers to click on a link to the CNBC website. I landed on a post with a jarring headline: “Trump is on his way to an easy win in 2020, according to Moody’s accurate election model.”
Huh. An “easy win.” According to the “accurate election model” developed by Moody’s Analytics. Huh. In fact, the story said, Moody’s Analytics had developed three models: one based on “pocketbook” issues, one pegged to the stock market, and one tied to the unemployment rate. Trump wins in all three. Here’s the gist as presented in the CNBC article:
“If the economy a year from now is the same as it is today, or roughly so, then the power of incumbency is strong and Trump’s election odds are very good, particularly if Democrats aren’t enthusiastic and don’t get out to vote,” said Mark Zandi, chief economist at Moody’s Analytics and co-author of the paper along with Dan White, the firm’s director of government consulting and fiscal policy research, and Bernard Yaros, an assistant director and economist. “It’s about turnout.”
That’s what you might call a fully hedged prediction: it’s based on an “if” and a “particularly if.” Moody’s Analytics is a company, by the way, that “helps capital markets and risk practitioners worldwide respond to an evolving marketplace with confidence.” So, if you are a “risk practitioner” you can now proceed with confidence, knowing that if certain favorable conditions prevail Trump has a good chance of being reelected.
Huh. An “easy win.” According to the “accurate election model” developed by Moody’s Analytics.
It might seem odd that a company that specializes in understanding risk does not have a model that factors in several other potential risks to Trump’s chances. But the modeling game, of course, is played by taking one or more important knowable and constant factors at play in every election from the recent past—say, the unemployment rate—and then “backtesting” those elections in which the result is also known. Thus, you learn that if the unemployment rate is low, or the stock market is booming, or people feel good about “pocketbook issues,” the incumbent president (or party) always wins. (The exception being 2016, when the Moody’s model predicted a Hillary Clinton victory, but anomalies blah blah blah unexpected events blah blah blah.)
The obvious difficulty in predicting the 2020 election is that there will be some, let’s say, irregular conditions—for which there is no dataset! There have been exactly zero presidential elections conducted with an incumbent president under the cloud of impeachment. (Bill Clinton was impeached in 1998, in his second term.) Nor has there ever been an election featuring a president whose dishonesty, racism, corruption, and graft are all so openly part of the show.
But, as Moody’s Analytics noted, “the power of incumbency is strong” and—high-priced analysis here—much will also depend on whether Democrats are enthusiastic about their candidate and whether that candidate can inspire a strong turnout, especially in states like Wisconsin, Michigan, and Pennsylvania. Which brings us back to those twelve hypothetical future presidents on the CNN debate stage in October and whether any of them can inspire the kind of turnout it will take to dislodge Trump.
The question can be simplified. Leaving aside the varying qualifications and quirks of the many individuals who seek the Democratic nomination, and leaving aside that all come forth with unequal charisma and rhetorical gifts, the essential Democratic debate is about whether the centrists in the party should follow the lead of the party’s left wing, or whether the left should fall in line and follow a centrist. In the minds of those on the left, a centrist will never excite the party’s base and will not generate the necessary turnout. Centrists are sure the left will lead Democrats off a cliff, and a vast sea of voters in the moderate middle will have no choice but to vote for Trump. As always, most of the Democratic Party establishment believes in the latter scenario. Which is to say, they believe that if their party offers some variant of European-style social democracy, too many voters will opt instead for the corruption, incompetence, and white nationalist agenda of the current mountebank in the White House.
This is pretty much the only argument for the Joe Biden candidacy, unless you also want to contend he will benefit from “identity politics” because the nation’s large bloc of elderly voters will identify with his shambolic elderliness. And now “don’t scare away moderates” has become the argument for those who hope to become younger versions of Biden. During the October debate you could easily imagine the conversations that must have gone on in the Buttigieg and Klobuchar prep sessions, in which savvy campaign operatives explained to them the dynamics of the race: Sanders and Warren have the left wing of the party locked down, so your only hope is to position yourselves as the “realistic and reasonable moderates” and be ready to fill that slot if Biden falls down and can’t get up.
This long-running argument between the two wings of the party sometimes resembles a debate between members of religious traditions that worship different deities: nothing can be proven and a lot is taken on faith. Both sides put their faith in a mythical swing voter who will be the key to victory. For conservative and centrist Democrats, this voter is imagined as the white working-class guy in the Midwestern diner. For the left wing of the party, the voters who can swing the election are younger, also working-class but not necessarily white and male, and will need to be mobilized, energized, and given a reason to vote. What you do is, you get so many of those voters that the old white guys in the diner don’t matter.
Both sides put their faith in a mythical swing voter who will be the key to victory.
In truth, the Democratic coalition can’t be easily imagined in terms of any one kind of archetypal voter. A Democratic victory depends on winning a huge majority of non-whites in cities and suburbs, and then not losing too badly among the white population, which is still around 80 percent or more of key states like Michigan and Wisconsin. Does that coalition only respond to some kind of cautious centrism à la Hillary Clinton who offered a return to the Bill Clinton years, or now à la Joe Biden who offers a return to the Obama years? Or would the coalition be supercharged by a vision of a bold new swerve into activist government that addresses health care and climate change and inequality? It’s a test of faith! We are in an evidence-free zone.
Nevertheless, we are in for a year’s worth of political punditry that will constantly conjure the all-important swing voter, and this swing voter will almost always be a stand-in for the pundit’s own predilections. The prototype of the genre appeared this summer on the New York Times op-ed page, written of course by columnist David Brooks, under the title “Dems, Please Don’t Drive Me Away.” But a more interesting version was published last month in the Boston Globe. Written by John Ellis, the column made an argument for “How Elizabeth Warren can win.” Ellis is not your typical Warren booster: he is a cousin of George W. Bush and was once described by Politico as having “developed something of a cult following in the political class.”
And what is Warren’s roadmap to victory, according to Ellis? She should stay true to the tough-on-Wall-Street stances but downplay some of the riskier proposals such as Medicare for All, and then—here it comes—reassure swing voters she will be a steady and reliable president by choosing a “retired general or admiral as her running mate.” Ellis concedes that “the base of the Democratic primary electorate wouldn’t be happy with this arrangement, but who cares what they think, really?” She will have won the nomination by the time she announces her running mate. “Swing voters will look at her differently the moment she announces that her running mate is a highly decorated military officer,” he projects.
A couple of questions were not addressed in the John Ellis column: Would Bernie Sanders also be advised to choose a highly decorated military officer as his running mate if he were to win the nomination? Also: If Warren did choose a military man for vice-president, would this satisfy David Brooks so that he could in good conscience vote for the Democratic ticket?
There’s an elephantine fact obscured by the calcified pundit perspective: Trump himself did not win the presidency by moving toward the center or gearing his campaign to please moderates. He ran from the far right nether-regions of the Republican Party. That doesn’t automatically prove that a Democrat could win by consolidating left-wing support and winning the day for socialism. But if Warren or Sanders emerged to lead the Democratic campaign, perhaps the general election would not really turn on voters’ considered judgments about whether Warren or Sanders has a fully formed plan to finance a Medicare for All system or on whether either could fine-tune a Green New Deal with our unrepresentative Congress. It’s unlikely that putting a military officer on the ticket would close the deal, since foreign policy seldom registers in the minds of American voters.
Trump must have won with some other method than by finding accord with the policy preferences of the American electorate. What might that method be? Could it be the particular way that Trump stoked the racial fears and resentments of white voters? And if so, isn’t the crucial question about how the Democratic candidate might offer something powerful enough to counteract the coming retooling of the 2016 Make America White Again campaign into the 2020 Keep America White campaign?
You would not know it from listening to anyone in the Democratic Party establishment or the mainstream punditocracy but for several years a few shrewd analysts working with labor unions have been focused on the way Republicans have used racism as a divide-and-conquer tactic. In this analysis, the use of racial fear and resentment is a class weapon that helps the GOP hold onto power while defending and preserving the rule of the wealthy.
One of the leading proponents of this view is Ian Haney López, a law professor at the University of California, Berkeley. López has worked closely with the AFL-CIO and several of its affiliated unions and with Heather McGhee at Demos and DemosAction on what he ultimately came to call the race-class narrative project. Now he is out with a book called Merge Left: Fusing Race and Class, Winning Elections, and Saving America, published last month by New Press.
In Merge Left, López offers a nuanced and sophisticated look at how racism works in American politics. Building on his previous work in the 2014 book Dog Whistle Politics, López discusses the right’s use of “coded racism.” The trick—and Trump has been the master of it—is to play on fears but to also deny there is anything racist going on. Republicans have benefited from dog-whistle racism for a long time (think back to Lee Atwater and the first Bush campaign’s use of the Willie Horton ad) but Trump has perfected the techniques. “Trump made his racist narratives obvious to his critics and a constant focus of media storms,” López writes. But he also realized he could benefit from the accusations of racism. When Hillary Clinton attempted in August of 2016 to link his ideology to the racism of the alt-right, Trump returned the fire, telling his supporters: “She lies and she smears and she paints decent Americans—you—as racists.”
The message López drives home in Merge Left is that Democrats will not defeat Trump by calling out “deplorable” white voters for supporting a race-baiting Republican. Nor will they succeed by falling silent about race in hopes that a focus solely on pocketbook issues—“colorblind economic populism,” in his term—will re-direct voters’ attention to “what really matters.” The approach that works, according to his extensive testing of political messages in focus groups with a range of voters, is to combat the right’s “core narrative” with a new story that fuses race and class.
“Core narratives differ from policy platforms,” he writes. The core narrative is the “skeletal story”—“the hidden scaffolding that undergirds campaign speeches, political ads, and slanted media reporting.” Over the last fifty years, and especially in the Reagan era, the GOP has perfected a story with three elements, López says: “1. Fear and resent people of color; 2. Distrust government; 3. Trust the marketplace.” It’s those elements that the opposition must speak to. López writes:
Narratives do not collapse simply because they don’t fit reality. The question is less the actual fit than the perception of whether they fit. Core narratives crumble when people come to believe that they are false. That, in turn requires a counternarrative that offers a new way to see what’s happening in society. . . .
What is the counternarrative? It starts with the understanding that “racial fear and resentment pervade politics and are so powerful they have been the major weapon used to divide working people for the last five decades, if not from the beginning of the country.” It proceeds with an understanding that even though there is an unwinnable base of racists who are aligned with Trump (and a bloc of unshakeable white evangelical Christians) there is also a group of “persuadable voters” who are not Trump cultists. López says his research shows that
the majority of whites who respond positively to stories of threatening nonwhites are not consciously committed to white superiority. On the contrary, though moved by narratives of racial threat, they also seem remarkably open to messages of economic populism tied to cross-racial solidarity.
Most of us who have spent time outside of the coastal bubbles—or who have family members in the heartland—know that what pundits always imagine as a thoughtful and moderate middle can also be described as a mushy middle, full of people with a mix of liberal and conservative views, often engaging in “contradictory thinking that can be tugged in either direction,” as López writes. “Three in five voters are in the convergence zone between the clashing Right and Left weather fronts.”
There is no inevitable force that has put Trump on course for an “easy win.”
What López describes in Merge Left is a kind of Solidarity Populism or Unity Populism. Such terms don’t even exist in the American political lexicon, because populism is always understood as divisive—“us against them.” But López is describing an “us and us against them” approach, in which the “them” is specifically defined as those who use and benefit from “strategic racism.” In other words, a campaign to unite whites and non-whites against “the nation’s most powerful economic and political forces” that have “a vested interest in public division.” He finds that a blanket condemnation of the upper class, in fact, is less convincing to voters in the middle than a specific rejection of the “greedy lobbyists” or the manipulative plutocrats represented by the “Kochtopus” network of political groups funded by the Koch Brothers.
It’s easy to despair at this point in the cycle, when we know what a political shitstorm we’ll face through 2020, and when we know how slow the Democratic Party has been to understand what it is up against, and when we imagine someone like Biden bumbling his way through a presidential campaign. This factoid produced by López doesn’t uplift the spirits: “No Democratic candidate for president has won a majority of the white vote since 1964.”
We also know how hard it is to knock off an incumbent president, no matter how bad they are: Nixon won in 1972 even though he was thoroughly dishonest and corrupt; Reagan won in 1984 even though it was clear by then he was taking his party—and the nation—on a hard right turn that would lead to lower taxes on corporations, vast increases in military spending, and greater inequality; Clinton won in 1996 despite being, as they say, “personally undisciplined”; and Bush won in 2004 despite his disastrous and incompetent foreign policy.
But there is no inevitable force that has put Trump on course for an “easy win.” We see the Democratic candidates grasping for some kind of unifying appeal. Biden seems to believe he can charm the Republicans into a return to bipartisan comity. Buttigieg declared in the CNN debate last month that “the purpose of the presidency . . . is the unification of the American people.” What López is describing is not some abstract “unity” or bipartisanship that no one believes in. He’s talking about a cross-racial alliance that connects racial and economic justice by appealing to people’s better angels but also to their self-interest. Are working-class whites damaged by racism? If it is used to bolster the GOP coalition that then serves the interests of the 1 percent, yes.
No election model should be trusted at this point. Polls now are as useless as the ones that gave Hillary Clinton a 99 percent chance of victory in the months and days before the 2016 election. Trump will continue to be the dominating force in American politics right up until the moment it all comes crashing down. How might that happen? “For democracies under demagogic assault,” Lopez writes, “the most effective defense is to vigorously promote social solidarity.”