I skimmed it before, and again just now, but that’s nationwide and not state-by-state, so it’s subject to the same pitfalls. I glanced at the tables which also do not have a state breakdown. I’m honestly not interested enough to request crosstabs.
ETA: Optimally I would want the table from p. 28, broken down by state, then weigh each state according to electoral college representation. That’s how one should, in my opinion, determine the importance of each demographic group to the election.
~Max
A state breakdown is still unlikely to tell you whether a white Protestant low-income person voted for Trump because they were low income, because they were white, or because of their religion (or, of course, due to some combination of two or all three). A high percentage of low-income people in the low population states may be all three of those things, because a high percentage of all people in the low population states tend to be all three of those things. So if a lot of low income people in, say, Montana voted for Trump: that doesn’t mean that they did so primarily because of concerns about their income.
That’s not the question I’d be looking to answer - the question is how important the white/Protestant/low-income/blue-collar vote was to Trump’s victory. If it turns out the blue-collar vote was critical, that’s an important premise in Cecil’s argument. There are many ways to argue that neoliberalism caused blue-collar workers to vote for Trump (all spurious in my opinion), but it’s all for naught if blue-collar workers weren’t actually key in his election.
~Max
Only if it was critical for economic reasons.
If what was actually critical was the distinction that shows up most clearly in the statistics – that white people voted far more heavily for Trump – there’s 400 years of American history behind that one, most of which has shit-all to do with neoliberalism.
Eh, no. I agree with msmith537 on that one. A blue-collar vote for Trump that is also a single issue vote on border security could be argued to have been the result of neoliberalism (not well, in my opinion, but still: NAFTA → Opioid epidemic → single issue of border security → Trump).
~Max
Aside from those being bad arguments: that difference between Black and white voters is highly unlikely to have to do primarily with either border security (maybe in 1800, not now) or with the opiod epidemic. Neither is a vote based on being a conservative Protestant likely to be based on either of those.
At least, unless you mean in the indirect sense that people insecure about finding themselves living in a country in which white conservative Protestants are the minority are easier to scare about border issues in general. That’s probably true. But again, their reasons for feeling insecure in that fashion don’t have much, if anything, to do with neoliberalism. And they go back much further than NAFTA. For that matter, politicians have been scaring people about border issues since the 1800’s.
The column argues that neoliberalism (e.g. globalization) “sucked” for blue-collar folks, and therefore (???) gave us " Trumpism and much that led up to it – catastrophic job losses, the disintegration of communities and the problems that entails, widening income inequality, an inadequate social safety net, you name it."
It does not, in my read, argue that most blue-collar Trump voters cast their votes with the economy in mind. Nor does it need that premise for blue-collar workers, white voters, or any other demographic group.
In my read, the conclusion that neoliberalism gave us Trump is based on two premises:
(I find the first premise spurious, you seemed to want to dispute the second)
ETA: Your counterargument, as I understand it,
That’s why I would want to look at exit poll data to see which groups voted for Trump, and weigh by state to determine how important those groups actually were to his election. Not why each demographic group voted - there probably isn’t going to be a uniform reason in polling data, or Trump would have used that to win 2020 too.
~Max
But for Trump to be neoliberalism’s fault, it must be arguing that they wouldn’t have so voted if neoliberalism hadn’t happened.
My argument is that the differences in vote which the statistics actually do show refer to issues that long pre-date neoliberalism, and so obviously weren’t caused by it.
Given the recent budget in Canada from the Liberal government, some austerity and restraint on spending might have been welcome. Hopefully the considerable environmental investments mimicking American policy will be helpful.
Canada is a nation dependent on international free trade, particularly with the US, which is fortunate since we amazingly still do not have unrestricted trade between provinces. I am not implying other parties would have done better, but the idea austerity is out of fashion and increasing debt is unconcerning is a risky idea. It only works until it doesn’t. Would that the Liberals were more neoliberal.
I don’t agree with your assumption. There is a large conservative segment of society that enjoys country music, Jeff Foxworthy’s humor, education with parental input, science without political agenda, and a view of what ‘Real America’ should look like that might be different from urban view points. Isn’t that how it’s supposed to work?
I think you misread that post. msmith537 was saying there’s a bunch of conservatives who reject things that don’t fit their view of “Real America”. Country music, Jeff Foxworthy’s humor, education with parental input, and science without political agenda are things a stereotypical American conservative would say fits their view of ‘Real America’.
~Max
People still follow “Jeff Foxworthy’s humor”?
<gerry fleck>I did not know that.</gerry fleck>
Stranger
The data that needs to be considered here is not the 2016 Trump vs. Clinton election but the change from that to the 2020 Trump vs. Biden election. That’s obviously more indicative of the Trump base.
I finally found a good, low-level accounting of the variation between the elections.
Unlike most articles using national polls, this article looks at county-level data. Doing so it allows understanding of the rural/urban areas and the concentration of like voters into like areas that has been the trend in America for years.
Some results:
Education
You can see the steady increase in the Democratic vote (blue bars) as the percentage of adults with college degrees increases. There is a 28 percentage point difference in the Democratic vote in 2020 between the most educated group of counties and those counties with the lowest percentage of people with college degrees.That gap widened by 4 percentage points from 2016 to 2020.
Income
Again, there is a nice stairstep up in the Democratic vote as the median income of the county increases. In 2020, Joe Biden won nearly 60% of the vote in the counties that had the highest median income. Donald Trump won only 38.6% of the vote in those counties.Trump, however, won 54.6% of the vote in the counties with the lowest income; Biden won 44.2% of the votes in those counties.
Democrats increased their share of the vote in the richest group of counties from 55% in 2016 to 59.4% in 2020.
According the 2020 census, “Median household income was $67,521.” The charts in that article show a shift from Trump to Biden in counties where the median household income is $60,000-$70,000.
The Trump base is in rural, low-education, low-income counties. Any other conclusion is a myth.
If you mow your front yard and find an automobile, you might be a redneck.
Yeah, I remember 1996, as much as I’ve tried to blot it from my memory like a character in a Charlie Kaufman movie.
Stranger
Not if you’re asking how important votes were in 2016. Cecil asserted that neoliberalism gave us Trump. We got Trump in 2016. In 2020, we got Biden, so statistics for that election are not directly relevant.
Though 2020 is highly relevant to the more forward-thinking question of who Trump’s base is.
~Max
I think you’re right, I did misread the post. I guess you could say that most people reject things that don’t fit their own view of Real America. My bad.
The column referred both to Trump and to Trumpism. I don’t want to conflate them, though. Trumpism was not a full force in 2016. The movement developed over Trump’s term in office, from the things he said and did while in power.
Examining the demographic change from 2016 to 2020 reveals Trump’s true base, which is smaller than the Republican vote in 2016. Were the grievances of that base caused by globalism, which the column pretty much equates with neoliberalism? Partially. But Trumpism depends on social issues, not economic ones. Trump understood and exploited that. For every minute he spent promising to bring back American jobs, he spent an hour fomenting hate toward his and their enemies. The lunatic right that rides the crest of that wave do not put forward solutions to economic problems. They wave guns, hate gays, support Russia, ban books, praise the police, and want to bring back Jim Crow. Their only stab at an economic policy is to keep out immigrants who would gladly take the jobs the movement members spurn.
Globalism is at fault only in the sense that the forces that came together in Trumpism are so numerous and varied that they can be called global. Job loss was caused by capitalism, which dips its tendrils into our every action. Capitalism embraced globalism because it made money for capitalists and also because it was wildly popular among the public. The very people who complain loudest today are the same people who would have complained loudest if the goods that filled every niche in their lives weren’t constantly available and affordable. Or even just when they are available: guns ain’t cheap but they sure are symbolic.
Many have taken issue with your columnist’s contention that Trump’s election could be blamed on neoliberalism and offered their own explanations. Clearly a followup column is called for. Consider this response a rough draft.
A sampling of competing views:
thorny_locust then cites a Guardian story entitled “White and wealthy voters gave victory to Donald Trump, exit polls show.”
Let’s take this step by step.
First, to acknowledge the obvious: Trump had a much better grasp of the mood of the electorate than Hillary Clinton did. Not saying that was their only difference, but it was pivotal. More on that later.
Second, Trump reflected a worldwide populist trend. A research group that maintains a database of these things says global populism is near a 30-year high. Populist leaders and movements can be found at both ends of the political spectrum, but the ones that have gotten the most attention are on the right - Bolsanaro in Brazil, Orban in Hungary, Duterte in the Philippines, Modi in India, the Brexit movement in the UK, and many more. The fact that Trump was part of a global phenomenon strongly argues against a strictly U.S.-centric explanation of how he wound up as president.
Academics commonly attribute the worldwide rise of populism to globalization, but they disagree on exactly how the latter gave rise to the former. Researchers tend to fall into two camps - one emphasizing economic insecurity (offshoring of jobs threatens workers’ livelihoods), the other cultural insecurity (outsiders threaten the country’s traditional - in the U.S., white Christian - way of life).
A good case can be made either way, and you need to look at the question from both angles to get a complete picture. Political scientist Yotam Margalit draws a useful distinction between outcome significance and explanatory significance. Outcome significance is (often) a relatively small change that produces outsized results. Margalit concedes that, in 2016, economic insecurity had major outcome significance, in that it enabled Trump to win three key states - Michigan, Pennsylvania, and Wisconsin - and thus the election. He cites a study demonstrating as much.
But he goes on to say economic insecurity has only modest explanatory significance - it doesn’t really tell you why 63 million people voted for Trump despite his manifest shortcomings. Margalit believes cultural insecurity offers a better explanation for this astonishing fact. I acknowledge the truth of this, and here attempt to provide a more nuanced account.
With Margalit’s approach as a lens, let’s consider the competing explanations for Trump’s victory:
Voters were rejecting the neoliberal policies advocated by the leaders of both political parties. This is the critique from the left, as described in the OP. Even if you don’t buy this argument in its entirety, elements of it deserve to be taken seriously - in particular, the claim that the neoliberal emphasis on free trade laid the groundwork for a massive offshoring of U.S. manufacturing jobs, which in turn led to a political backlash in Rust Belt states.
Scoffers may say: that’s giving neoliberalism too much credit. Globalization was inevitable - globalization here understood to mean outsourcing of manufacturing to foreign firms, with a corresponding loss of U.S. factory employment. It was made possible by the growing productive capacity and lower costs of developing countries coupled with transportation improvements such as containerized freight. It would have happened regardless.
A reasonable response to this would be: Sure, it would have happened eventually. It didn’t need to happen so abruptly. The fact that it did can be at least partly blamed on what now seems a naive faith in the virtues of free trade.
Free trade had long been advocated by Republicans. During the Clinton administration it was embraced by Democrats as well. An early step was ratification of the North American Free Trade Act in 1993. But the most dramatic changes occurred following the admission of China to the World Trade Organization in 2001, a U.S.-led move that resulted in a substantial reduction in trade barriers. The chart below, from the Bureau of Labor Statistics, shows what happened next:
In the 1980s and 1990s, U.S. manufacturing employment had been relatively stable, but during the 2000s it fell off a cliff. Roughly 5 million American manufacturing jobs - one-third of the total - were lost in just 10 years.
This trend was mirrored in Michigan, Ohio, Pennsylvania, and Wisconsin. In February 2020, prior to the pandemic, these states had a million fewer factory jobs than in 2000.
In the study cited by Margalit, the economic dislocations caused by the China shock led to a shift in votes that resulted in Hilary Clinton’s loss of Michigan, Pennsylvania, and Wisconsin, which Obama had won in 2012. From the standpoint of outcome significance, then, economic insecurity stemming from globalization is why Trump won.
If China had been kept out of the WTO, could the China shock have been prevented, or at least slowed down enough to reduce it to a China bump? There’s no way to know. What we do know is that the Obama administration was reasonably successful in getting things turned around during the depths of the Great Recession, when U.S. manufacturing jobs reached their low point. Over the next few years it launched various initiatives to rebuild manufacturing, invest in new technology, bring jobs back, and so on. Arguably these efforts had some impact, both on the economy and how people felt about it. Manufacturing jobs ticked up modestly. Obama won re-election by 5 million votes in 2012.
So if the job situation improved after 2010, why did Clinton lose in 2016? Here we turn to what Margalit believes is the factor of explanatory significance - cultural insecurity, which I acknowledge is needed to fully understand what happened. To see how this played out, it’s instructive to look at some of the other reasons suggested for Trump’s victory.
The majority of white people voted for Trump. The argument here is that whites had more money than the rest of the electorate and wanted to keep it that way. A related claim is that Trump was elected by racists, presumably white.
Let’s posit that Trump received 58% of the white vote in 2016. (Pew Research came up with a different, lower number in a later analysis, but we’ll skip that for now.) What this overlooks is that Republicans have gotten the majority of white votes (or plurality, in elections with three candidates) in every presidential election since 1976, when exit polls began categorizing respondents by race:
From this we can conclude that most white people will vote for the Republican candidate no matter who it is. That doesn’t tell us anything about why they voted for Donald Trump specifically. In short, this hypothesis lacks explanatory significance.
What about the argument that Trump was elected by racists? Many of those who voted for him surely were. We lack detailed data on how whites voted before 1976, but the “solid South” reliably voted Democratic prior to the civil rights era. After Democrat Lyndon Johnson muscled civil rights legislation through Congress, southern states began voting Republican. From this one may reasonably conclude that a significant percentage of Republicans are racist, and it follows that many Trump voters were racist. But by the same logic, many of the voters for any Republican candidate are racist - that’s not something specific to Trump. So the racist argument also lacks explanatory significance.
Does racism have outcome significance? Let’s look at the vote totals in the decisive states, Michigan, Pennsylvania, and Wisconsin. Here are some graphics from a Washington Post analysis of the results from 2012 through 2020:
These states voted for Obama by wide margins in 2012, went narrowly for Trump in 2016, then swung back into the Democratic column in 2020. It’s implausible to suggest the electorates in these states were temporarily seized by racism in 2016. We need to look elsewhere for an explanation.
Trumpism is characterized by hatred and fear of The Other, be it other races, gays, Muslims, immigrants, or what have you. This gets close to the truth. It’s essentially the cultural insecurity argument advocated by Margalit, and is surely of explanatory significance not only for Trumpism but for populism around the world.
Briefly put, Trump voters tended to be less educated and lived in smaller cities and towns. They felt looked down on by the educated elites who lived in big cities and voted for Democrats. In common with the followers of populist leaders elsewhere in the world, they believed their country was slipping away, feared immigrants, and were susceptible to demagogic appeals.
The evidence here is abundant and well understood:
Nate Silver, in “Education, Not Income, Predicted Who Would Vote For Trump,” pointed out that, in the 50 counties with the highest percentage of college grads, Clinton improved on Obama’s performance in 48 and won the great majority outright. In the 50 counties with the lowest percentage of college grads, she did abysmally - worse than Obama in 47 cases. In contrast, income was a poor predictor of Clinton’s performance. In well-educated, middle-income counties, she did just fine.
The bigger the city, the more likely it was to vote for Clinton. Smaller towns and rural areas went overwhelmingly for Trump, as this NPR analysis shows:
mixed evidence that economic distress has motivated Trump support. His supporters are less educated and more likely to work in blue collar occupations, but they earn relatively high household incomes and are no less likely to be unemployed or exposed to competition through trade or immigration. On the other hand, living in racially isolated communities with worse health outcomes, lower social mobility, less social capital, greater reliance on social security income and less reliance on capital income, predicts higher levels of Trump support.
I could go on, but you see my point. To use Margalit’s terminology, economic insecurity is a factor of outcome significance - it’s why Trump won three key states and thus the presidency. Cultural insecurity offers more explanatory significance. Clinton, with her talk of deplorables, made it clear she didn’t understand cultural insecurity. Trump did, and that’s why people voted for him. Biden’s admirers think he also gets it, and that’s plausibly why in 2020 he got the presidency back.
Economic insecurity and cultural insecurity aren’t mutually exclusive arguments. On the contrary, you need to understand both to fully grasp what happened - that seems like the surest way of avoiding Trump 2.0. I gave inadequate attention to this distinction in the OP, and hope I’ve now put the matter right.
Trump and other populists generally don’t look at issues in detail from all sides, determine their beliefs and stick to these principles. Instead, they tend to be good at determining what people fear, stoking these concerns and gaining power through intimidation. Switching opinions when convenient.
Some manufacturing jobs were lost because wages and benefits are lower in other countries… But some were lost due to environmental concerns, changing trends, outdated business practices and improved technology. To a populist, economic and cultural insecurity are two sides of the same coin and both serve to win votes by creating fearful scenarios. The difference is irrelevant. In the cases of the populists mentioned, which the Trump movement itself stoked, these also include crime, corruption, accepting refugees and other jingoistic national identity issues like immigration, language and religious issues.
Trump is indeed good at reading the electorate and did so better than Hilary, but also better than the media and all his other challengers. He continues to use the media which is still eager to cover his every passing thought for short-term gain despite Trump’s contempt. He offered an alternative from politics as usual, and it is true major political parties are often more similar than different. He is currently twice as popular as DeSantis and will likely win the Republican nomination, or proceed as an Independent if he does not win and is able to run. He owns the party, even if many in it (and just where are “the adults in the room” “in on the joke”?) dislike his methods or lack of principles or policy knowledge. His sons and his daughters are beyond your command.
But he has not talked about any of the issues this time around. Not even made the pretence. People interested in policy and politics often assume others share this interest. Most don’t. And this is why Trump’s success is a personality thing rather than a rejection of neoliberalism. He talks and acts like a wrestler.
So most politicians probably wouldn’t stalk people during debates, give insulting nicknames to opponents, mock heroic veterans and the disabled, deride their national security services and military, insult Mexicans and many other countries, mock world leaders and international forums, minimize colleagues, ignore tradition and democratic principles, incite violence, kowtow to dictators who hold human rights in contempt and who are considered American nemeses, break the law, fail to address a major health crisis putting millions at risk, show genuine signs of sociopathy, engage in blatant nepotism, use current high office for crass money-making, assume without study or reason they know complex issues better than skilled academics and advisors or allegedly cheat on their pregnant wives.
It is not my place, as a Canadian, to judge Trump for doing so or many Americans for their tolerance of these things. But they are not debates about issues, even neoliberal ones. They do invite free and plentiful media coverage, and that is enough for Trump. What issues has Trump seriously addressed in the last month for his campaign?
(As a Canadian, I tend to strongly support free trade and think it ridiculous so many barriers exist between our own provinces. Trump talked about trade. But what changes did he actually make? Fortunately for everybody concerned, precious few.)
