The American people are pissed.
They are outraged at an economy that works for Wall Street but not for them. They are furious at a foreign trade regime that has devastated manufacturing. They are skeptical of immigration from Mexico, either as an unfortunate xenophobic consequence of economic despair (if you’re on the left) or an understandable response to large-scale social change that had gone unchallenged by the major parties before the rise of Donald Trump (if you’re on the right).
This is the election that has been embraced by those across the ideological spectrum. People are mad as hell, and they’re not going to take it anymore. That’s why Donald Trump is the Republican nominee. That’s why Bernie Sanders nearly beat Hillary Clinton.
There is just one problem: the facts.
The American people are not pissed at the state of the economy. At all.
They’re as happy as they were in 1984, when Ronald Reagan won a landslide reelection. They largely approve of their president. They overwhelmingly support free trade and oppose immigration restriction, and in both cases the public is becoming more pro-globalization, not less. Donald Trump won a Republican primary where turnout, as always, was low, and got as far as he did on the votes of a relatively small fraction of Americans.
And as we obsess over the faulty narrative, we miss some other things that really matter. This is just one aspect of the election that we’re getting very wrong.
Here are 21 facts about the election that just might alter your perception of what’s happening this year.
1) Americans are as happy with the economy as they were in 1995
This election has been widely covered as one where economic discontent has provoked deep anger in the electorate, enabling the rise of nontraditional candidates like Sanders or Trump. But as George Washington University's John Sides has noted, this narrative is hard to jibe with the Index of Consumer Sentiment, a long-running measure of America's feelings about the economy, which has fully recovered to its pre-recession levels. Americans feel about as good about how things are going as they did in the mid-2000s, or in the mid-1980s. They feel much better than they did in 2012.
Intriguingly, Sides also finds that the gap between high- and low-income Americans’ perceptions of the economy has shrunk in recent years. There was a lot of polarization in sentiment indices in the ’80s and ’90s. High-income people thought things were going great; low-income people less so. Now, while a gap persists, it’s considerably narrower.
2) Incomes grew across the board last year, especially at the bottom
Income gains across the board.— Justin Wolfers (@JustinWolfers) September 13, 2016
10th percentile +7.9% (ie poor)
90th percent +2.9%
95th percentile +3.7% pic.twitter.com/IgFUBDeyK8
Perhaps it shouldn’t be that surprising that Americans feel pretty good about the economy — the Census Bureau’s report on 2015 suggests that for actual families, income growth is faster than it’s been in years.
Not only did the typical household see its income rise by 5.2 percent, or $2,800 in real terms, but the growth was concentrated at the bottom. The poor saw incomes go up by 7.9 percent, compared with only 3.7 percent growth for people in the 95th percentile of incomes. If you adopt a more accurate inflation measure than the one the census uses, household median income has never been higher. As a final cherry on top, 2.7 million people fell out of poverty from 2014 to 2015, and the poverty rate, properly measured, fell by a full percentage point.
3) A majority of Americans think President Barack Obama is doing a good job
Another sign Americans aren’t actually outraged at the current state of affairs: The incumbent president is pretty popular — far more popular, in fact, than the candidates running to replace him. So much for an anti-establishment year.
Obama’s personal favorability — how much people like him, as opposed to approve of his job performance — is also positive:
This is striking, given that some presidents whose job performances gained good marks have also dealt with low favorability in their last year and early post-presidency. Notably, Bill Clinton’s combination of competent economic management and intern sexual harassment resulted in sky-high approval ratings and low favorability ratings.
4) Two-thirds of Americans like at least one presidential candidate
Yes, Hillary Clinton and Donald Trump boast among the highest unfavorables of any presidential nominees in history (the latter especially). But the fact remains that most Americans like at least one of the candidates.
According to polling Gallup conducted in June, only 25 percent of respondents report disliking both candidates; that's higher than the 11 percent of respondents in 2012 who disliked both Mitt Romney and Barack Obama, but it suggests that a large majority of Americans have at least one candidate they view favorably. Similarly, Morning Consult has found that only 23 percent of respondents dislike both Trump and Clinton.
5) Americans' support for foreign trade is higher than it's been for more than 20 years
This election featured the rise of two vociferously anti–free trade candidates in Bernie Sanders and Donald Trump, which helped force Hillary Clinton (long a fence straddler on the issue) to come out against the Trans-Pacific Partnership trade deal the Obama administration is currently pushing.
The timing for this turn against trade, however, is puzzling, given that the share of Americans saying trade is an “opportunity” rather than a “threat” hit in 2016 its highest point in Gallup’s polling since the organization began asking the question in 1992. And that’s not just a counter-reaction to Trump and Sanders: The popularity of foreign trade has been high since 2013.
6) More Democrats than Republicans have favorable opinions of foreign trade
More intriguing still, Americans' split on the issue by party has changed over the past couple of decades. While Democrats used to be more skeptical of trade, and Republicans more sympathetic, now Democrats are likelier to say it's an opportunity.
A March 2016 poll from Pew found the same thing — and it’s not just an artifact of Trump supporters echoing their candidate. A plurality of John Kasich supporters also said that free trade agreements have historically been a bad thing for America, and supporters of Trump, Ted Cruz, and Kasich were all less likely to call trade deals a good thing than backers of Clinton or Sanders:
Some of this is likely a matter of partisan alignment; the current president pushing trade deals is a Democrat, so it’s natural that Democrats and Republicans would switch places. Still, the scale of the effect is remarkable, as is the fact that even Sanders supporters expressed positive feelings toward free trade pacts.
7) Bush’s free trade czar is running way ahead of Trump in Ohio
Compare this chart:
With this chart:
Hillary Clinton is currently slightly ahead of Donald Trump in Ohio. That's to be expected: Ohio is the archetypal purple state, and was quite close in 2012 as well. But Republican Sen. Rob Portman is absolutely crushing former Democratic Gov. Ted Strickland in the Senate race. He’s running far ahead of his party’s presidential nominee, despite the fact that a lot of his fellow Republicans up for reelection in swing states (Richard Burr in North Carolina, Kelly Ayotte in New Hampshire, Pat Toomey in Pennsylvania) are faltering, either running closely ahead or closely behind their challengers and doing about as well as Trump.
This is a kind of confusing situation. Portman — a former lobbyist who served as both George W. Bush’s budget director and his trade representative — theoretically represents everything voters in general, and Republicans in particular, are rejecting in this strange election year. The political establishment. Lobbyists. The Bush administration’s betrayals of fiscal conservatism. Free trade deals. But he’s running far ahead of Trump in Ohio.
8) Americans are more supportive of racial diversity than Europeans
America is considerably more racially heterogeneous than most other rich, developed countries. While the 2016 election, and the white nationalist rhetoric of Donald Trump and many of his backers, has shined a light on the racial conflict and tension that white anxiety about that diversity can bring, in general Americans embrace it.
Polling from the Pew Research Center earlier this year found that while 58 percent of Americans say that having many different ethnic groups present makes their country a better place to live, the numbers are far worse in countries like Sweden, Germany, Italy, and Greece. In the latter two countries, a majority of respondents said that diversity made their countries a worse place to live.
9) Nearly 60 percent of Americans oppose reducing immigration
Trump has made immigration a more important issue in this election than any other presidential race in recent memory, which can obscure a remarkable fact in public opinion: Support for immigration is actually rising. While the share of Americans calling for reduced immigrant inflows peaked in 1993 and 1995 at 65 percent, according to Gallup, the share has plummeted to a mere 38 percent. Meanwhile, the share of Americans calling for immigration to stay constant has grown somewhat, but the real growth is in the minority that wants more immigration.
The General Social Survey and National Election Surveys also find that immigration opposition has fallen sharply in the past two decades:
The New York Times’ polls found the same:
This isn't an artifact of question wording, either. There's a similar trend when you ask about legalizing unauthorized immigrants:
Or about whether immigrants are a boon or burden:
Or about whether Americans should welcome some, all, or no immigrants:
As David Bier wrote in a report for the Niskanen Center last year, “The belief that Americans reject immigrants or are fed up with current policies could not be further from the truth.”
Further, there are huge generational gaps in responses to these polls, according to work done by the Pew Research Center:
While 76 percent of millennials say that immigrants strengthen the country, only 48 percent of baby boomers do. If those attitudes persist as millennials grow older, and younger generations similarly exhibit pro-immigration attitudes, attitudes toward immigrants in the country as a whole could grow yet more positive.
Donald Trump’s campaign also appears to have hurt the anti-immigrant cause. While in December and January the RAND Corporation found that 48 percent of respondents backed a border wall, that had fallen to 38 percent by July and August.
10) Over the past 10 years, more people have immigrated from the US to Mexico than vice versa
One reason Americans aren’t as worried about immigration is that on the southern border, it’s slowed to a trickle, and then actually gone negative. It's certainly true that in the 1990s, millions of Mexicans moved to the US. But more recent numbers show the trend is going the other way.
Analysis by Pew Research Center found that from 2005 to 2014, net migration between the two countries was about 160,000 in Mexico's favor. In other words, over 100,000 more people went from the US to Mexico as the reverse.
The Mexican immigrant population in the US peaked at 12.8 million in 2007 and declined to 11.7 million by 2014, a decline mirrored by the subset of Mexican immigrants who are unauthorized. Remarkably, in 2014, for the first time in recorded US history, there were more Border Patrols apprehensions of non-Mexicans than Mexicans.
This trend started with the recession, but we haven’t seen a return to pre-recession levels of Mexican immigrant inflows since growth started back up again. At this point, it’s worth wondering whether building a border wall would actually increase the undocumented population by preventing people who might want to return to Mexico from doing so.
11) Your odds of being killed by a refugee terrorist are 1 in 3.6 billion
Donald Trump and many of his GOP primary rivals raised objections to allowing Syrian refugees into the US, arguing that this ran the risk of increased Islamist terrorism on US soil. Ted Cruz and Rand Paul voted for a Senate amendment banning immigration from dozens of Muslim countries. Jeb Bush called for Christian refugees from Syria to get preferential treatment because of their religion. Marco Rubio called for a shutdown of Syrian refugee admittance after the Paris attacks, allowing some exceptions, providing as an example, tellingly, "a well-known Chaldean priest."
So the Cato Institute's Alex Nowrasteh took it upon himself to calculate, based on terrorist death tolls from 1975 from 2015, the odds of an American being killed in a terrorist attack in which an immigrant or refugee was involved. The odds of death in an immigrant-perpetrated attack were 1 in 3.6 billion: lower than the odds of someone dying by being hit by a train or their clothes catching on fire.
It’s almost enough to make you wonder whether the refugee freakout is about safety at all, or if it’s really about racist backlash politics.
12) Fear of terrorism is at its highest point in more than a decade
While the objective danger Americans face of falling victim to a terrorist attack is basically zero, that hasn’t prevented the rise of ISIS from prompting a hysteric panic over terrorism among the public of a kind unseen since the aftermath of 9/11.
Gallup polling in December 2015 found that 51 percent of Americans say they're very or somewhat worried that they or someone in their family will become a victim of terrorism. The last time Gallup recorded a combined total that high was October 11 through 14, 2001. More recent polling from CNN reached similar conclusions, with 71 percent of respondents in June saying they thought it was very or somewhat likely the US would face an act of terrorism in the next several weeks (a more reasonable response than those to the Gallup question, given that it's not about personal safety). That was the highest point CNN recorded since March 2003, when the Iraq War began.
13) The murder rate is half of what it was in 1991
Trump has also campaigned on the claim that President Obama is overseeing a massive increase in the murder rate, asking attendees at a rally in Florida, “Do you know it was just announced that murder is the highest it’s been in our country in 45 years?”
As my colleague German Lopez notes, this is not even remotely close to true. While the murder rate ticked up mildly in 2015, from 4.4 per 100,000 to 4.9 per 100,000, it's still half of what it was in 1991, and less than half of its 1970s and 1980s peak. We're still enjoying the fruits of a rapid, massive decline in homicide rates that began in the mid-’90s.
14) Only 6.2 percent of Americans backed Donald Trump in the primaries
It’s easy to lose sight of the fact that because US presidential primaries have really, really, really ridiculously low turnout, a pretty small fraction of Americans can wind up selecting a nominee. Case in point: While in the 2012 general election, Barack Obama earned support from 30.6 percent of the voting-age population, and Romney got 28.3 percent, Hillary Clinton and Donald Trump won their parties' respective nominations with votes totaling 7.5 and 6.2 percent of the US voting-age population, respectively.
This is something to keep in mind if you’re despairing that a candidate like Trump could get nominated: 93.8 percent of Americans didn’t vote for him in the primaries.
15) Most of the time, when the polls change, it’s not because people actually changed their minds
The above chart takes a second to parse, but once you do the ramifications are pretty profound. It comes from a paper published earlier this year in the Quarterly Journal of Political Science by Columbia's Andrew Gelman, Stanford's Sharad Goel and Douglas Rivers, and Microsoft Research's David Rothschild, who in the last month and a half leading up to the 2012 presidential election surveyed more than 345,000 respondents on Xbox Live (oh, brave new world that has such survey data in it).
Like many polls, the Xbox Live data (after it had been properly demographically weighted) showed Obama taking a hit following the first presidential debate, when many observers perceived Obama as having lost to Mitt Romney. The natural way to interpret that is that the poor debate performance drove voters away from Obama and toward Romney.
But a closer examination reveals that interpretation to be fallacious. The left-hand chart shows Obama's share of the two-party vote in the Xbox survey; the right-hand chart shows the fraction of respondents who identified with a political party who identified as Democrats, in a given day's poll. You will notice that they look nearly identical.
That implies that the shifts in polling had less to do with people switching sides from Obama to Romney and more to do with demoralized Obama supporters being less likely to respond to polling questions. So while the polls show the race changing dramatically, in reality people aren’t changing their minds all that much. The authors estimate that only 0.5 percent of respondents actually switched from Obama to Romney, while 0.2 percent switched from Romney to Obama. That just isn’t a very large net effect at all.
The simplest explanation is that these mid-campaign shifts are a polling artifact, not a sign that millions of minds are being changed based on what happens on the campaign trail. This jibes with other research as well, like a 2004 paper by Columbia's Robert Erikson, Fordham's Costas Panagopoulos, and UT Austin's Christopher Wlezien, which found that shifts in 2000 election polling owed more to changes in sample composition than people changing their minds.
16) Only about 5 percent of Americans are actual swing voters
Back in the olden days, about one in eight Americans was a swing voter, or "floating voter" in political science parlance. Michigan State University political scientist Corwin Smidt, using the American National Election Survey, estimated that from 1952 to 1980, about 12 percent of respondents, on average, voted for a presidential candidate of the opposite party from the candidate they voted for the time before. By 2012, that percentage had fallen dramatically, to only 5.2 percent.
By contrast, "standpatters" — a term Smidt borrowed from political scientist V.O. Key for people who vote for the same party year after year — swelled in numbers over that period.
Strikingly, Smidt notes, "Pure independents were more stable in their party support across 2000–04 than strong partisans were across 1972–76 and about as stable as strong partisans across 1956–60." In other words, independents today are no more likely to switch their votes from election to election than hyperpartisan Republicans and Democrats from the 1950s and ’70s. Today's independents are as partisan as yesterday's partisans.
The reason for this, Smidt argues, is increasing awareness of the parties’ differences. The share of voters — both partisans and independents — saying that they feel like they understand the differences between the parties' stances on the issues has shot up dramatically over time. Greater awareness of the differences led to greater partisan loyalty. After all, if you don't change your position on the issues between elections, and you know where the parties stand on them, why are you ever going to change which of the parties you vote for?
17) If only men voted, Trump would win in a massive landslide; if only women voted, Clinton would win in an even bigger one
Gender divisions among voters sometimes get less emphasis than economic or racial cleavages, but they can be crucially important.
Here's what the map would look line if only women voted: https://t.co/sjVY67qouE pic.twitter.com/rrc3GuXmGl— Nate Silver (@NateSilver538) October 11, 2016
And here's if just dudes voted. pic.twitter.com/HjqJzIVwc4— Nate Silver (@NateSilver538) October 11, 2016
Case in point: If only men were allowed to vote, then according to FiveThirtyEight's polling average, Trump would win, 350 electoral votes to 188. He'd pick up every swing state, some reliably blues states like Minnesota, and even one of Maine's congressional districts.
By contrast, if only women voted, the result would be a truly massive and unprecedented Clinton landslide, with Democrats picking up Texas, Kansas, South Dakota, Alaska, and South Carolina.
18) Gary Johnson is the most successful third-party candidate in at least 20 years
Former New Mexico Gov. Gary Johnson has seen his Libertarian campaign for the presidency flounder due his apparent ignorance of the war in Syria and inability to name basically any world leader he admires, but the fact remains that so long as he gets at least 2.7 percent of the vote on November 8, he’ll be the most successful third-party candidate for the presidency in 20 years.
Admittedly, the bar isn’t that high. In the past three presidential elections, no third-party contender has topped 1 percent. (Johnson himself came very close last go-around, with 0.99 percent.) Before that, you have Ralph Nader with 2.7 percent in 2000, and then Ross Perot with 8.4 percent in 1996 and 18.9 percent in 1992, the latter of which remains the high-water mark for third-party bids in the modern era and isn’t likely to be dislodged anytime soon.
But Johnson has at points in this election topped Perot’s 1996 numbers, as FiveThirtyEight’s Harry Enten has pointed out. He’s similarly often come out ahead of John Anderson’s eventual vote share from 1980 (though Anderson’s polling overestimated his actual support).
If Johnson's current estimated support level of 6.6 percent holds (and it might not), he’ll fall below Anderson and Perot, but he’ll have more than doubled Nader’s total, increased his share from 2012 nearly sevenfold, and have easily posted the best result in the history of the Libertarian Party. That’s a big deal.
19) Republicans are absolutely crushing Democrats in state legislatures
Republicans control 30 legislatures across the US right now, compared with merely 12 that are in Democratic hands (and that's counting New York as Democratic-controlled, when a coalition of conservative Democrats and Republicans in fact run the state Senate). Seven states are split, with one chamber in each party's hands. Nebraska's legislature is technically nonpartisan, but self-identified Republicans form a supermajority.
The situation is even worse for Democrats when you consider states like Maryland, Massachusetts, New Jersey, and Illinois, where they control the legislature but not the governorship. If you count New York’s control as split, and Nebraska as Republican, then Republicans had total control of 23 states, 20 are divided, and only seven (plus Washington, DC) are in Democratic hands. Republican-controlled states have a total population of about 149.5 million; split states of about 120.4 million; and Democratic-controlled states and DC of about 51.5 million, the vast majority of which is accounted for by California alone.
As it stands, Democrats’ continued success nationally has tended to obscure their very deep losses at the state level.
20) State legislature voting is based mostly on presidential approval
It may sound weird to say that Democrats’ odds of retaking state legislatures depend almost entirely on the party’s performance in the presidential elections. Don’t people vote based on how their actual representatives in their state House and Senate perform? Don’t they base opinions on individual candidates’ track records and positions?
Short answer: No, not really. In a recent paper for the Annals of the American Academy of Political and Social Science, St. Louis University's Steven Rogers found that state legislative election outcomes are much more a consequence of national factors than local ones. Take, for instance, the above chart, showing Democratic seat changes in the US House and in state Houses. They're almost perfectly correlated. That makes zero sense if voters are basing their decisions on the actual job performance of their specific representatives.
Rogers looked at the 2008, 2010, and 2012 elections, as well as New Jersey elections from 1973 onward. He correlated respondents' reported votes in state legislature races against their approval ratings for their legislature, governor, and the US president. He found that presidential approval has a significantly larger effect on state legislature voting choices than either of the other, more relevant approval ratings.
Here, for instance, is the relationship his analysis finds between respondents’ stated approval of their state legislature in 2012 and their odds of voting for the dominant party in that legislature (the solid line), and the relationship between respondents’ approval of Barack Obama and their odds of voting Democratic in state legislature elections (the dashed line):
Respondents’ opinions of Obama play a much bigger role in their state legislative voting than their opinions of their legislature. All politics is national.
21) Sometimes county officials represent more people than US senators
Michael Enzi and John Barrasso, the US senators from Wyoming, represent a state of 586,000 people. That's a little bit smaller than Milwaukee and a little bit bigger than Fresno. By contrast, Sheila Kuehl represents about 2 million people through her elected role in government. Kuehl is one of five members of the Los Angeles County Board of Supervisors, and because that massive county (America’s largest) has 10 million people in it, each supervisor has 2 million in his or her district.
And because California gives a considerable amount of power to counties, that makes supervisors very, very powerful. Indeed, Hilda Solis, another member of the LA County Board of Supervisors, stepped down as US secretary of labor to run for it. You read that right. Solis — a former Congress member to boot — resigned a Cabinet position because she wanted to take a job in county government. And it makes sense — she represents more people than a senator from a small state.
The same principle applies to state legislatures. The 435 members of the US House represent about 737,348 people each, on average. That’s smaller than the average district of a state senator from California or Texas, two massive states with small Senates.