For the last few months, the Trump presidency has presented analysts with a weird puzzle. The news has been chaotic -- Congress has made multiple attempts to pass new health care legislation, stories tied to Russian interference in the election seem to come out daily, various key White House roles have been vacated and filled and the president has made multiple foreign trips. Yet despite this roller coaster of events, his poll numbers have been shockingly stable. Specifically, Trump’s job approval has hovered around the 40 percent mark for over two months, almost never deviating by more than one percentage point.
So why are these numbers so stable when events are anything but? The world of political data has been busy debating this topic, so rather than simply lay out one view and argue for it, I'm going to describe five different views that have popped up. (Note: Some of these ideas were generated simultaneously by multiple people, adopted and edited by others and generally have changed in ways that make them difficult to trace back to one author.) I’ll also explore what each view might predict about the future. In the coming months we'll be able to check what happened against each theory’s predictions and get a better sense of which one, if any, was right.
Theory 1: The GOP Base and Strong Democrats Are Immovable Objects
One way to explain Trump’s stable approval is to simply look at who does and doesn’t approve of him and how strongly they hold those views. The situation is simple: Trump has a high approval rating among the GOP base and low approval ratings among the Democratic base, and unless there are truly exceptional circumstances, neither group will change their mind.
When political scientists and wonks talk about this theory, they usually invoke “polarization.” The basic idea is that over the last few decades, the two parties have become more ideologically distant from each other (there’s still disagreement about whether the GOP alone pulled right or if Democrats also moved left). As that happened, the majority of voters sorted themselves into one of these partisan camps, began to increasingly distrust and dislike the other party and thus gave each party a reliable base.
In other words, Trump’s 40 percent approval rating might represent something close to his floor. On Election Day 2016, 37.5 percent of voters viewed candidate Trump favorably, yet he won 46 percent of the vote. Since then, Trump may have lost some of those general election voters by pushing an unpopular health-care bill (or through some other actions they disapproved of), but this 40 percent approval rating represents party stalwarts sticking with him.
The polarization theory comes with a few predictions about the future. As long as there aren’t any truly out-of-the-ordinary events (e.g. a terrorist attack, a recession, a historic economic boom, impeachment, etc.) Trump’s approval rating will stick within a relatively narrow band. He won’t lose his loyalists but he also won’t convert many outside that group. This theory doesn’t rule out a Democratic midterm wave or a second Trump term, but it does suggest that there are some guardrails that keep either party from scoring the sort of landslide wins we saw in the last century.
Theory 2: Nothing “New” Is Happening
It’s also possible that despite the roller coaster of events in Washington, voters have seen little new information that would change their mind about Trump.
This might seem like a strange argument. News junkies remember James Comey’s testimony before Congress, Donald Trump Jr.’s emails, the ups and downs of the Senate’s efforts to repeal Obamacare and the whole host of related breaking news stories over the past few months. All of this information is new in the sense that nobody could have predicted the exact contours of the health-care debate or the content of Comey’s testimony.
But it appears that this information isn’t moving voter preference much. Suppose a “normal” voter -- someone with a job, a family, friends and a typically busy life -- knew a few basic facts about the House version of the GOP health care, disapproved of it and had a diminished opinion of the president as a result. News about Senate debates on a similar bill might not change their mind on the subject. Or consider a loyal Trump voter, someone who believes the president is more credible than mainstream media outlets; for that person, it’s possible that a new story from CNN on ties between Russia and the Trump campaign wouldn’t alter their opinion. In other words, not all events that merit news coverage will end up moving public the polls.
Pollsters who have long records of repeated interviews with the same voters would be best equipped to figure out how news events figured into public opinion. It’s possible that these events did change voter preferences in important ways. But none of these major storylines caused Trump’s topline approval rating to change much. And the other storylines of the summer (special elections in Montana and Georgia’s 6th Congressional District, terrorist attacks at an Ariana Grande concert, Trump pulling out of the Paris climate agreement, Trump fighting with the media, etc.) might have been either too short-lived to change public opinion or served as new information that did nothing to change old impressions of the president.
If this theory is correct, we should expect Trump’s approval rating to change only if truly new events or information are introduced to the electorate. If the economy suddenly boomed or tanked, if Trump decided to push a popular infrastructure bill instead of a relatively unpopular health care bill, if the situation in North Korea escalated and Trump had to respond to it -- basically if the national political conversation changes in a meaningful way -- Trump’s support might change.
By itself, this theory doesn’t posit a floor or ceiling for Trump. It’s possible to imagine a situation where he moves on to more popular legislation, passes it later this year or next year and thus improves his approval rating substantially. Or it’s possible to imagine voters punishing the GOP for passing an unpopular health care bill by immediately knocking down Trump’s approval and electing Democrats in the 2018 midterms. The real test of this theory is how Trump’s approval changes when the national political conversation inevitably shifts.
Theory 3: It’s the Domestic Policy, Stupid
It’s also possible that Americans don’t really care about Russia and that Trump’s approval rating is simply a result of the public’s mixed opinion on his domestic record.
It’s not hard to find polling that would flesh out this theory. The RealClearPolitics average shows that Trump has a 44.7 percent approval rating on economic issues, which isn’t so far off the 46 percent he won in the 2016 popular vote. Yet Americans are currently focused on health care. A July Bloomberg poll showed that 35 percent of respondents (a plurality) said that health care was the most important issue facing the country. And although it’s tough to compare polling results that differ the wording of the question, most polls show that the current GOP bill is unpopular.
This adds up to a dynamic where a decent economy is trying to pull up Trump’s job approval numbers, but his policy focus is pulling those numbers down.
This theory would lead to similar predictions as Theory 2. If Trump moved off health care and onto something more popular, he might improve his polls, and if he signs a health-care bill or moves onto equally unpopular legislation, his numbers might suffer. The biggest predictive difference is that if the Russia issue somehow disappeared overnight, public opinion might not shift drastically under this model.
That being said, Theory 3 does paint a very different portrait of voters and their concerns than Theory 2 does. Under this version, news about Russia doesn’t matter too much. Very few voters think it’s the most important issue facing the country, and some of them would disapprove of Trump on other grounds if the Russia issue somehow dissipated. It also holds that voters probably care more about their personal day-to-day concerns (e.g. keeping up with the cost of living, having good health care, sending their kids to college, etc.) than about issues that don’t affect them directly.
Theory 4: This Is the New Normal
It’s possible that there isn’t much to explain about Trump’s 40 percent approval rating. Someone could argue that this low, stable rating is almost unavoidable in our current political moment.
To see how this argument would work, imagine that Hillary Clinton had swung enough voters away from Trump in Wisconsin, Michigan and Pennsylvania to become president. She would have come into the office with a low favorability rating and likely wouldn’t have had much of a honeymoon period. She might accomplish some of her goals via executive action, but it would to be hard to get her policy priorities through a Republican-held Congress. Others have examined this possibility more thoroughly, but the point is that in some alternate universe there are other presidents who would also have a low approval rating.
It’s unclear exactly how far you could extend this argument. A President Rubio or a President Biden might take some similar actions to Trump or Clinton, respectively, yet have better approval ratings due to sheer likability. But some facts of our current political moment -- a highly polarized public, a GOP base pushing (as party bases or presidents that have just returned to power sometimes do) for legislation that isn’t broadly popular and a Republican Congress strongly opposed a Democratic president -- could damage the approval of many hypothetical commanders-in-chief not named Trump.
From here, someone could argue that the stability of Trump’s approval ratings is simply a product of partisanship. Strong Republicans will provide a floor for presidential approval (as would strong Democrats for Clinton) and independents and members of the opposing party would disapprove of the new president.
The key difference between this theory and the more orthodox version (Theory 1) is that someone could start here and use results like the Georgia 6 run-off to offer a more bullish assessment of the GOP’s chances in 2018. The argument would run something like this: Trump’s approval is more a product of timing and personal unlikability than his policy actions. When presented with a clear choice in an election with more typical turnout, some of these Trump-skeptical (or, in our alternate universe, Clinton-skeptical) voters will come home, allowing the incumbent party to defend its territory more successfully than presidential approval alone suggests they should.
Theory 5: We Don’t Know What’s Going On
This might sound like a cop-out, but it’s important to acknowledge the possibility that none of our other theories is right. Many in the media/analysis/political science space (myself included) had theories that had to be adjusted significantly or tossed out entirely after watching the Trump phenomenon play out.
So it’s possible that we simply don’t yet know why Trump’s approval is low and consistent. And if his approval evolves in a way that invalidates all these theories, we may have to go back to the drawing board (something that’s become familiar to many of us in the wonkier side of the political world).
So Which Theory Is Right?
I don’t know which theory is correct. I have some reservations about Theory 4 and, if pushed, I might slightly favor Theory 2 or 3 over the alternatives. But my goal isn’t to place a bet on one theory or claim that any single explanation tells us everything we need to know. It’s to set up a few different theories and detail what they predict so that we can assess which one was best supported by the data in the coming months.
For the time being, we just have to wait and see.