A reminder courtesy of this new WSJ/NBC poll that the protests in Michigan and elsewhere this past weekend don’t yet reflect majority opinion. But that’s changing. Public opinion on this question is inevitably a moving target. With each new gruesome unemployment report and each new data set showing slowing growth in the number of COVID-19 deaths, the balance will shift towards reopening sooner. In fact, there’s evidence that it already is.

It can be tricky comparing different polls on this subject since pollsters don’t always ask about the same thing. Some ask about stay-at-home orders; some ask about a national stay-at-home order; some don’t mention stay-at-home orders at all, choosing a gassier term like “restrictions”; some refer more generally to “social distancing.” All but the most hardcore COVID truthers see the need for some social distancing. The debatable policy question is how assertive and aggressive state governments should be in locking down the local economy to achieve it.

The phrasing in today’s poll closely resembles the phrasing in a recent Pew poll, though, which means we should be able to compare one to the other to see if opinion has moved lately as Trump has begun clamoring to reopen. In a poll taken between April 7-12, Pew asked people if they’re more concerned that restrictions on public activity by state governments will be lifted too quickly or not quickly enough. Result:

Majorities of both parties worried that we’d move too soon, before the outbreak had been mitigated, than too late. But what a difference a few days can make. Between April 13-15, WSJ/NBC asked whether people worried more that “the U.S.” would move too quickly in lifting restrictions or move too late. Result:

From 66 percent in the Pew poll worried we’d relax restrictions too soon to 58 percent now, with Dems and GOPers each less likely to say they fear relaxation. The change in both groups is small and among Democrats it’s negligible, from 81 percent in Pew to 77 percent here. But it’s more significant among Republicans. Whereas a small majority of the party in the Pew poll worried about restrictions being relaxed too early, a plurality in the WSJ/NBC survey now worries about them being lifted too late. Just 39 percent of GOPers say they’re more concerned about moving too quickly to reopen versus a near-majority of 48 percent who say they’re more worried about moving too slowly. Earlier this month (April 2-6), when New York cases were still growing, 68 percent of Republicans told Quinnipiac they’d support a *national* stay-at-home order to contain the outbreak. Right-wing opinion has come a long way since then.

I think May 1 will be a crucial date to gauge opinion just because Trump’s done so much to focus attention on it as a target for widespread reopening. A YouGov poll taken last week showed 56 percent of Republicans already in favor of reopening businesses by or before May 1 versus 60 percent of Democrats who thought reopening should happen afterward. That’s a sign of how much partisanship (and the urban/rural divide among D’s and R’s) is driving opinion, and with Trump egging on Democratic governors to reopen soon no matter what, it’s a cinch that it’ll get starker near-term. The mystery is how long it’ll take for Democratic voters to follow Republicans’ lead and start shifting meaningfully towards reopening as well. Would a big spike in testing capacity convince them? New data showing a major decline in deaths nationwide? It has to happen. Economic necessity leaves them no choice but to shift to the “reopen soon” position eventually.

Relatedly, here’s another new poll showing a meaningful partisan divide:

Not much change there in whether people think we need continued social distancing but there’s a big change in how many think the worst is still to come. Majorities of Democrats (68 percent) and independents (57 percent) believe that it is, but just 43 percent of Republicans do. Presumably as deaths continue to decline in the short term even Dems and indies will grow more optimistic, and even if we reopen too soon and trigger a second wave of the outbreak, it may take awhile before there’s enough data to make that claim irrefutable and trigger a new wave of pessimism. Even after we have the data, it’s unimaginable that Americans will tolerate another de facto national lockdown in the name of containing the disease. The next public debate as cases begin rising again will be whether and how to isolate cities and maybe even whole regions where the disease is raging from ones where it isn’t.

Political support and overall tolerance for social distancing is also fairly high now. It may not always be. So that may be another argument for "keep doing it now, while you can, recognizing that it may not be as feasible to maintain this in the long run."

— Nate Silver (@NateSilver538) April 18, 2020

Trump’s advisors understand the political danger involved in him making himself the face of the “reopen ASAP” push, though, even if more of the public is destined to side with him in the next few weeks:

But allies and advisers inside and outside of the White House are urging caution, telling the president there is more potential risk at going too fast than too slow, citing polling that shows Americans are looking to take a more measured approach…

[L]ifting social distancing guidelines too soon, or without the proper preparation, could lead to a rebound in infections and deaths, public health officials have warned. While Trump’s allies believe they can make the case that the initial outbreak wasn’t his fault, urging Trump to place the blame on China or the World Health Organization, it would be a harder case to make that he doesn’t bear the responsibility for a resurgence in infections should he encourage restrictions to be lifted.

“It would be a disaster if it opened up too early, and again, we have a double curve and a double spike in cases and in deaths and we are back to where we were before,” said the White House official.

Trump’s own lack of credibility is doubtless making it easier for Democrats like Gretchen Whitmer to wave off his demands to reopen soon. The WSJ/NBC poll finds just 36 percent say they generally trust what he has to say about the virus versus 52 percent who don’t. By comparison, 66 percent say they trust their governor and 60 percent say they trust Fauci. Usually any poll question you ask about Trump will find a baseline of 43 percent or more supporting him or his position, reflecting his overall job approval throughout the country among hyper-loyal Republicans. For him to be in the mid-30s on a question about COVID-19 credibility is proof that even some righties who normally give him the benefit of the doubt aren’t doing so this time, making it that much harder for him to galvanize a majority in favor of reopening soon.

In any case, it’s silly to conduct national polls on when to reopen and will seem even sillier going forward. Fauci and Birx acknowledge that less hard-hit regions will be able to move more quickly in resuming business than harder-hit ones will. So what’s someone supposed to say when they’re asked by a pollster if “the U.S.” should lift restrictions soon? Might be fine in Kansas! Not so fine in New York. We need more state polling, especially in swing states, to see how this issue is playing in a meaningful way. Besides, asking “when should we reopen” is asking the wrong question. It’d more insightful to ask people what they want to see happen in order to make them more comfortable with reopening. How many are watching trends in the daily death toll for guidance? How many are focused on progress in testing and contact tracing? How many are listening to Trump? How many are listening to Fauci? There’s a cacophony of informational noise bearing on people’s assessments. Let’s see if we can find a signal.

In lieu of an exit question, study this interesting — and ominous — graph from Dr. Jeremy Faust, who posted it in a long thread about why people are overestimating how safe it is to lift restrictions. It feels safe, argues Faust, because we’re at a stage where most Americans personally know only a very few people who’ve had the disease. Getting sick feels like a longshot, which makes socializing seem less risky than it’s been cracked up to be. But that’s an illusion if many more of us are actually infected than the number of confirmed cases reflects (which everyone believes, including scientists). And the more actual asymptomatic cases there are, the more even small gatherings pose a considerable risk of infection. If you’re going to protest Whitmer at the Michigan capitol, at least be smart and wear a mask.

This graph, by @joshuasweitz shows just why these numbers are scary. The graph shows the likelihood of being in the same place at the same time with SARS-CoV-2 positive person, given two facts: 1) The number of cases in the US and 2) The number of people at the event you’re at. pic.twitter.com/aHgQHuSNKV

— Jeremy Faust MD MS (@jeremyfaust) April 18, 2020