President Trump likes to trumpet his "tremendous" support and strong base, but polls show that his approval rating is declining, even among key demographics that voted for him in 2016. (Jenny Starrs/The Washington Post)

News about the unpopularity of President Trump has been metronomic since he took office. Week after week, new polls showing new lows in Trump’s approval rating; month after month, new stories reporting how America views Trump more negatively than any president in the era of modern polling.


The constant drip can be misleading — Trump’s approval rating slump hasn’t been that dramatic compared with other presidents — but the kernel of those stories is accurate. Trump is uniquely unpopular, and it’s not getting better.

Trump and his supporters insist that the numbers are meaningless, that, just as Trump was underestimated in the 2016 election, he is being underestimated now. Of course, that argument skips over the important details that (a) Trump led in Republican primary polling for more than a year and that (b) national polling suggested a narrow win in the popular vote by Hillary Clinton, which happened.

But it is the case that polling suggested that Clinton would also win the electoral college, which she didn’t. Trump can justifiably claim to have beaten the odds, and no observer of the past two years could claim to know exactly what might happen next.

With that in mind, we decided to ask a number of polling experts what they thought Trump’s low approval ratings might actually mean for his presidency. Is it important? Does it affect his policy goals? Will it undercut the Republicans next November? Their responses, sent independently over email, are excerpted below.

The experts

Ariel Edwards-Levy, director of polling at HuffPost.
Harry Enten, senior political writer and analyst for FiveThirtyEight.com.
Charles Franklin, director of the Marquette Law School poll.
Margie Omero, co-host of the podcast “The Pollsters” and executive vice president at PSB Research.
Sean Trende, senior elections analyst for RealClearPolitics.

1. What would we normally expect low approval ratings for a president to mean?

The Post: It’s certainly not without precedent that a president should be unpopular. Trump’s low approval ratings are unusual this early in an administration, but other presidents have seen worse.

So, Trump aside, what would such low approval ratings normally mean in terms of a president’s policy agenda?

Franklin: No matter how late in a term, approval ratings in the mid-30s or below are a signal that the president lacks widespread support and normally Congress becomes less willing to advance those legislative priorities.

Enten: Presidents who have higher approval ratings generally have an easier time getting Congress to go with them on big ticket items (e.g., tax reform). That said, with the growing polarization within Congress, even a less popular president can get big stuff passed without being all that popular if his party holds the majority.

Trende: I think if his job approval were in the 50s, or even the high 40s, things would look very different for Trump. For starters, you’d have a large number of Democratic senators up for reelection in states where he was popular — in some cases, extremely popular. This would make them less willing to vote against him. Likewise, his power over the Republican Party would be increased, as they would see him as a competent leader aiding their own reelection efforts. As it stands, Democrats don’t really fear him.

The Post: What would this normally mean for the midterm elections?

Enten: The president’s party often does worse in midterms when the president’s approval rating is low. The margin of error is wide on that, though.

Enten has written about this, including the graph below.


Franklin: Presidential approval at the midterm is one of the strongest predictors of how many seats the president’s party loses (as they almost always lose seats in midterms). The lower the approval the more seats the party typically loses. Only two postwar presidents have faced midterms with approval below 40 percent, Truman in 1946 at 33 percent in Gallup’s final pre-election poll, and George W. Bush in 2006 at 38 percent. Democrats lost 55 seats in 1946, and Republicans lost 34 seats in 2006.

Trende: Congressional elections are referendums on the party in power, and with a job approval in the 30s, substantial losses for Republicans are close to inevitable. The one thing Rs have going for them is that they don’t have much exposure; only eight members are in districts that are Democratic-leaning, and Trump carried 231 districts.

The Post: What would such low approval numbers generally mean for a president’s leadership more broadly?

Edwards-Levy: In a normal political environment, you’d probably expect those sorts of numbers to be a source of concern for the White House and a potential albatross for the president’s party heading into the midterms.

Enten: I think we’re seeing it right now. Republicans are more and more willing to challenge Trump. It’s not just [Arizona Sen. Jeff] Flake or [Arizona Sen. John] McCain or [South Carolina Sen. Lindsey] Graham. We’re seeing [Tennessee Sen. Bob] Corker and [South Carolina Sen. Tim] Scott being willing to do it, too. Now part of that has to do with what Trump said around Charlottesville, but it also has to do with the fact that they see him as vulnerable.

2. Do those normal expectations hold for Trump?

Enten: Well, isn’t this the question?

Edwards-Levy: An approval rating under 40 percent, objectively, looks pretty bad for a president this early in his tenure.

Franklin: To be this low this early is simply unprecedented. Ford fell to 37 after pardoning Nixon early in his term and Clinton fell to 37 as well by June 1993 but both rapidly rebounded after that low, Ford into the mid-40s and Clinton into the 50s. So their time below 40 was quite limited and was followed by more middling-to-good approval levels. … If his support continues to decline at the current pace, he will almost certainly face a midterm with the lowest approval of any modern president.

Trende: Forty percent [approval] seven months in, absent a recession, is really not good. It suggests that something is broken in the way the administration is conducting itself. It doesn’t mean that you can’t recover — Bill Clinton hit 36 percent in Gallup in June of his first term — but it is a tough row to hoe.

Enten: He’s still leader of the Republican Party, but not as strong as he would be if his approval rating was say 10 points higher. … [P]olarization as a whole can be a good thing for Trump if it keeps Republicans in line who are afraid to lose a primary. It could save pieces of legislation that might not otherwise pass.

Franklin: Most presidents have a clear first year agenda based on priorities from the campaign and a well prepared set of legislative initiatives. That has been uniquely missing in the Trump administration so far. This makes it hard to separate the current effects of low approval from those of lack of legislative success due to lack of clear proposals.

Enten: We know Trump won the presidential election, despite being the least liked major party candidate of all time. That’s why I think we need extra caution. … That said, caution is one thing and outright ignoring the past is another.

Trende: After all of the craziness from last year, the election actually ended up pretty typical. That is to say, our models suggested it would be a close popular vote race, and it was. Our models suggested that Republicans would probably lose two or three seats, and they did. I think there’s a chance that Republicans don’t get punished fully by the American voters, but I think it shouldn’t be our expectation.

Franklin: On midterms the biggest question is whether the historical record of seat change will be matched in the current highly polarized distribution of seats. It is quite possible that modern seats will not exchange hands as readily as they have over the past 70 years or so. … Usually an unpopular president makes it harder for his party to recruit strong congressional candidates, as they see the electoral forces working against them. We don’t yet know if that will be true this year.

Omero: [A] problem Trump faces that you can’t see in just approval ratings is the enthusiasm on the left. The surge of Democratic candidates, Democratic voters in the special elections, in Democrats (particularly women) who say they’re following news and politics, the new grass-roots organizations, and so on. There’s very little Trump can do at this point to put that genie back in the bottle, and he seems unwilling to try.

Enten: The big difference is Hillary Clinton isn’t on the ballot in 2018. Trump and House Republicans won the vast majority of voters who didn’t like either presidential nominee in 2016. Now, Republicans are the only game in town.

Franklin: [W]e are in nearly uncharted waters going into 2018.

3. To what extent are Trump’s poor approval ratings just a function of the polarized electorate?

Trende: [H]e has had a very rough first seven months, in large degree of his own making. I do think that if he had made some better choices (as I say to my 4 year old), he would be closer to where he started out — in the mid-40s, or perhaps a touch higher. He didn’t get a honeymoon, but that’s as much a function of his low favorability ratings — the other side just never gave him the benefit of the doubt.

Edwards-Levy: On the vast majority of political issues I’ve seen polling on this year, especially those related to President Trump, the divides along partisan lines are sharper than basically any other demographic. … That being said, while anyone in office at the moment would probably struggle with the same forces to some extent, Trump’s ratings have only fallen since he was sworn in, and the 40 percent ceiling he’s been hovering beneath since spring would seem to be to a large extent of his own making.

Franklin: Trump’s first seven months have looked a lot like Obama’s middle years, 2-7. GOP support has held in the 80s, Dem support has been at 10 or below and independents in the 30s. If Reps never desert Trump, as Dems never deserted Obama, then this could provide something of a floor.

Enten: The thing to remember about Obama is that the economy was doing okay (not great), and he wasn’t going on Twitter or television and making a bombastic statement every day. It was the perfect formula for his approval rating to stay steady.

Omero: His ratings with Democrats have been terrible and are unlikely to improve. We can’t just blame it on “partisanship” because aside from encouraging the Rhode Island Teacher of the Year to [include] his fan into an official photo, Trump hasn’t done a single thing to reach out to Democrats. Everything he does is a base play, and voters outside his base have gotten that message loud and clear. This is beyond simply “the climate.”

Edwards-Levy: It seems unlikely Trump is going to win over his opponents any time soon, but his base support has softened somewhat and there’s reason to think that whatever his exact floor is, he hasn’t hit it yet.

Enten: There was plenty of polarization during the [George W.] Bush years, and his approval rating rose like a rocket and dropped like rock falling from a high building.

I guess what I’m saying is that I’m skeptical that Trump cannot see the same thing occur.