Woodrow Wilson and Warren Harding. Normal? Photographer: Topical Press Agency/Hulton Archive via Getty Images
“America’s present need,” the candidate declared, “is not heroics, but healing; not nostrums, but normalcy … My best judgment of America’s need is to steady down, to get squarely on our feet, to make sure of the right path. Let’s get out of the fevered delirium.”
The candidate was the Republican Warren G. Harding and the date was May 14, 1920. Six months later, Harding won a landslide victory over the Democratic nominee, James M. Cox, winning 60% of the popular vote and 404 Electoral College votes.
A return to normalcy: It’s an appealing prospect today, too, amid an ongoing pandemic, in the wake of an unprecedented economic shock, and after four years of political disruption. A century ago, to be sure, Americans had come through worse: the 1918-19 Spanish influenza, which killed around 675,000 people (the equivalent of 2.2 million today), and World War I.
A century ago, there was no incumbent to defeat, as Woodrow Wilson — having been struck down by the flu during the 1919 Paris peace negotiations and then by a severe stroke — was judged by his party to be unfit to run. (It remains to be seen if President Donald Trump’s admission to hospital for Covid-19 presages a premature exit for him.) But the parallel with today is still striking. In the so-called Red Scare of 1919-20, the country had been swept by strikes, protests and race riots. A severe recession had begun in January 1920. By November, what most Americans craved was indeed normalcy.
I have been thinking a lot about the election of 1920 in trying to predict that of 2020. Four years ago, chastened and educated by the experience of Brexit, I felt that Trump had at least an even chance of winning the presidency. Recall that in the week before the Nov. 8, 2016, the left-wing Daily Kos website put Hillary Clinton’s chances of winning the presidency at nearly 90%. According to The Upshot in the New York Times, the number was 85%. Betfair said 79%. Nate Silver said 65%.
So what do I think now, when even the ultra-cautious Silver puts Joe Biden’s chance of beating Trump at around 80%? Spoiler: I always said the half-life of populism was short.
Donald Trump is a classic populist, who offered disgruntled voters a heady cocktail of protectionism, nativism, easy money, isolationism and anti-elitism. Comparisons with European fascists between the World Wars always struck me as wide of the mark. Historically, it has generally been hard for mercurial figures such as Trump to win the highest political office, at least in the northern hemisphere. (I never bought former White House adviser Steve Bannon’s analogy between Trump and Andrew Jackson.) From Georges Boulanger to William Jennings Bryan to Huey Long, the history of populism is mostly of near misses — which was part of the reason most pundits assumed Trump would be a near miss four years ago.
When populists do get elected, they almost never deliver all they have promised to their supporters, and are often exposed as even more corrupt than the people they ran against. South America has a lot of experience in this regard, from Juan Peron in Argentina to Hugo Chavez in Venezuela. Latin American populists get re-elected not because they deliver higher living standards to their supporters (they may do so in the short run, but it always ends in some kind of financial crisis). They get re-elected by repressing their opponents and, when necessary, changing the constitution — a regional pastime.
To read the mainstream press, you would be forgiven for thinking something similar is about to happen in the U.S. According to Barton Gellman in the Atlantic, there is going to be voter suppression, voter intimidation, a declaration of emergency, the bypassing of election results in battleground states, and finally martial law. Trump’s going to steal the election somehow — and it may even be constitutional if he does, Fareed Zakaria has argued. Only a Biden landslide can save the Republic from violence and a constitutional crisis. Forget Hitler and Mussolini; now Trump is Richard III.
It must be said that Trump did everything possible to validate these narratives in last Tuesday night’s debate, short of opening with “Now is the winter of our discontent.” But, as Biden likes to say, “Come on, man.” Trump may have the instincts of a caudillo, but this isn’t Venezuela.
The debate would have mattered only if Biden had looked unmistakably senile. He didn’t. Instead, Trump came across as an insufferable bully. Even my ex-cop friends Mike and Gerry — who backed Trump in 2016 and were infallible guides to that year’s politics — felt their man had been too aggressive. And now it turns out that Trump was mocking Biden for wearing a mask, when he himself was probably already infected with Covid-19. (Not just mocking him, but yelling at him indoors from just six feet away. We won’t be sure for roughly a week that Trump didn’t infect Biden.)
Far from being in peril, I would guess, the Constitution is about to do what it was designed to do: Having successfully constrained a demagogic president throughout his term, in the usual ways — courts striking down executive orders, Supreme Court appointees acting independently, midterms handing the House to the Democrats — it is going to allow voters to eject him from the White House and install in his place dear, old Joe Normal.
For any of the “end of the Republic” scenarios to happen, this election needs to be close — close enough for the results in multiple states to be challengeable. But I struggle to see how this could come about.
If Jimmy Carter couldn’t get a second term after the small recession of January to July 1980, and if George H.W. Bush couldn’t get one after the comparably minor recession of July 1990 to March 1991, how on earth can Donald Trump get a second term after the disaster that has befallen the U.S. this year? Who gets re-elected after a pandemic that has killed more than 200,000 Americans and a recession that sent unemployment up to 14.7% in April, compared with peaks under Carter and the elder Bush of 7.8%? Trump’s latest jobs report has unemployment at 7.9%.
Even with the recovery that’s occurred since the lockdown low-point back in the spring, the U.S. economy is still on course to shrink by 3.8% this year, according to the Organization of Economic Cooperation and Development. The only example I know of a democratic leader getting re-elected under such economic conditions is Angela Merkel in 2009.
Now let’s look at the polls, where Biden leads Trump by an average of around 7 percentage points. The remarkable thing here is the consistency of Biden’s lead: Over more than a year he has never been less than four points ahead. We’ve seen nothing like this in our lifetimes — in most presidential elections since 1968, the polls have bounced around, sometimes wildly.
Moreover, Biden’s lead right now is not just bigger than Clinton’s four years ago at this stage in the race; it is also bigger than Barack Obama’s 32 days out in both 2008 and 2012. If the news on the pandemic and the economy is bad between now and election day, Trump could end up where John McCain did, seven points behind. Or, if the news improves, he could somehow claw his way back, as Mitt Romney did in 2012, and still lose. What I struggle to imagine is Trump getting close enough to rerun the George W. Bush-Al Gore standoff of Florida 2000 in multiple states.
Remember what happened in the final phase of the 2016 campaign. First, a relentless stream of negative news about Clinton throughout October ate away at her lead. Second, state polls seriously underestimated Trump’s support in Michigan, Pennsylvania and Wisconsin. Finally, third-party candidates took 6% of the vote in 2016. Maybe there’s a secret stash of toxic opposition research waiting to be unleashed against Biden this month, but I doubt it.
Instead, what we’re getting is a relentless stream of negative news flow about Trump, not the least of which is the New York Times expose of how insanely little income tax the guy has been paying. (As my friend Mike told me last week, “I heard a couple of blue-collar workers today, cops and firemen, talking about the Times story about him not paying any taxes ... it was the first time I ever heard anything negative about Trump from this base.”)
And that’s not the only bad news for Trump. He’s only one of nearly 300,000 Americans who tested positive for Covid-19 last week. The hospitals in Wisconsin are filling up with new Covid-19 cases. And Covid-19 is the main reason Trump is struggling with older voters, a key demographic for him four years ago.
Trump clearly wanted to announce a successful vaccine before the election. That seems less and less likely. Jared Kushner wanted the economy to be “rocking” by now. But the refusal of the pandemic to “go away, like a miracle” is clearly having some adverse effects on the economy, preventing mobility from returning to normal in the most affected states, and slowing the recovery of the labor market. Finally, the finances of the Trump campaign appear to be in disarray (though being outspent did not stop him winning four years ago).
Even if you allow for polling errors as bad as 2016’s in the key states, Trump is going to struggle to get above 240 Electoral College votes, 30 shy of victory. For all these reasons, I am inclined to think he is going to be a one-term president, and that the election result won’t be close enough for full-scale GOP lawfare to save him.
What am I missing? What could make me wish I’d stuck to my contrarian position of four years ago?
After all, only last December another populist, Boris Johnson, won a much bigger victory in the U.K. general election than almost anyone (including me) expected, sweeping a bunch of traditionally Labour-voting working class constituencies in the north of England. Could there be an equivalent surprise in this year’s U.S. election?
Leaving aside the potential Covid-19 impacts on the candidates’ health, I can think of nine reasons why the polls might be even more wrong than last time.
First, a striking 11.7% of Republicans say they would not report their true opinions about their preferred presidential candidate on telephone polls, while 10.5% of independents also fall into the “shy voter” category.
Second, the law-and-order issue really matters to those shy voters. (It also gave Trump his best debate moments.) Polls give us a sample of voters’ stated preferences. Revealed preferences are in many ways more reliable. According to Small Arms Analytics, gun sales in August 2020 were 58% higher than in August 2019, continuing a surge of purchases (especially of handguns) since the spring. In 2016, gun ownership was very closely correlated with voting for Trump.
Third, the resumption of the so-called culture wars this summer was a godsend for Trump. Judging by Tucker Carlson’s ratings — not only on cable but also on YouTube — there are at least five million Americans who share his skepticism about the Black Lives Matter movement, to say nothing of “critical race theory.”
Fourth, check out whose Facebook posts have been getting shared the most this year. On the day of the first presidential debate, five of the top 10 posts were by conservative firebrand Ben Shapiro, not an unusual occurrence this year.
Fifth, the third vacancy on the Supreme Court in as many years was another stroke of luck for Trump. Conservative voters care more about the makeup of the court than liberals, so Amy Coney Barrett was a near-perfect pick to boost Republican turnout.
Sixth, Hispanic voters seem unenthused about Biden and indeed about voting generally. That matters in Florida, obviously, but there are 11 other states where Hispanics are more than 10% of eligible voters, including Arizona and Texas.
Seventh, Republicans are winning the voter-registration game in key states, notably Pennsylvania and Wisconsin.
Eighth, as a significant percentage of mail-in ballots tend to be rejected because of errors, Trump should benefit from the higher proportion of Democratic voters intending to vote that way.
Finally, don’t underrate the economy. A third-quarter bounce as big as the one projected by the Atlanta Fed’s GDPNow would give Trump a second term if you simply plug the number into the wonderfully parsimonious model devised many years ago by Yale’s Ray Fair to predict U.S. elections with economic variables.
Usually, if you can think of nine reasons why a hypothesis might be wrong, it’s probably wrong. And yet, even when I add all these variables together, I still don’t think Trump can salvage the situation. There is a lot of overlap, after all: Most gun purchasers probably owned at least one firearm already, and they may be the same people watching Tucker Carlson, liking Ben Shapiro and rooting for the confirmation of Justice ACB. In terms of new votes in swing states, and therefore Electoral College votes, my nine reasons to be doubtful may sum to zilch.
The probability of a repeat of 2016, when the votes of fewer than 40,000 people got Trump over the line in Michigan, Pennsylvania and Wisconsin, is simply too low. The probability of another 2000 or another 1876 (when the results in four states were contested) is also low. The probability of a contingent election — when no presidential or vice-presidential candidate receives an absolute majority of Electoral College delegates — is even lower: We haven’t seen one involving the presidency since 1824. None of these scenarios is remotely as probable as a victory for the “normalcy” candidate who has been out in front every single month of this annus horribilis.
The irony is that if a Biden victory is accompanied by a Democratic majority in the Senate, then it could suddenly be the turn of Republicans to cry “Republic is in peril,” as projects such as packing the Supreme Court, getting rid of the filibuster in the Senate, and giving statehood to the District of Columbia and Puerto Rico (i.e., packing the Senate) will suddenly seem feasible to the progressives.
But that’s the trouble with voting for normalcy. Remember, Americans did just that — overwhelmingly — a hundred years ago. What they got was the Roaring Twenties, followed by the Great Depression, followed by World War II.