Elections are not about neoliberalism
If people don’t vote on policy, they definitely don’t vote on abstract elite catfights.
Since time immemorial (but especially since 2016), progressive and moderate Democrats have bickered about election strategy. Here in the home stretch of election season, the two sides have a temporary truce, linking arms behind Kamala Harris. But no matter the election’s outcome, I assure you they will soon be at each other’s throats again to explain why that outcome proves their side right about how to win elections.
This post is not about which side is right on the policy merits.1 It’s not even about which is right about politics. It is rather about how politics are stupid and chaotic, such that both sides are full of hot air on the subject.
A recent example of the fight I’m describing took place on Substack between Matthew Yglesias (representing the moderates) and Freddie de Boer (representing the progressives). In my view, both men are guilty of the pundit’s fallacy, which is ironic because Yglesias coined the term and de Boer teases him for that without applying the same scrutiny to his own thesis.
As part of a broader, mostly-reasonable series defending neoliberalism from its progressive critics, Yglesias argued President Biden’s more interventionist approach to economics “has proven, in practice, to be less politically popular than the old way,” citing Biden’s low approval ratings.2 This fits his broader hunch that moderate centrism gets results and wins elections, whereas radical progressivism scares voters and loses them.
But it was a dopey thing to say, and de Boer called him out on it. After drilling home the important point that the poles move the center of the Overton Window (more on that at the end) de Boer points out that Biden’s approval ratings likely tanked due to factors unrelated to policy, like his age. He writes:
Looking at Biden’s recent polling struggles and concluding that the problem lies in his onshoring efforts and stances on green energy does not make a lot of sense to me, especially given that I doubt more than 5% of Americans know the first thing about his industrial policy at all. His problem is that he’s an old man who looks and sounds like an old man, and Americans care about vigor and looking presidential and all that jazz. Joe Biden’s approval rating in 2024 is not a public referendum on neoliberalism, sorry.
This is exactly right, and to his credit, de Boer extends the insight beyond Biden:
There are of course all kinds of dynamics that contribute to a given presidential candidate’s success and popularity that have nothing to do with policy. (I would argue, in fact, that policy is not the biggest part of success in presidential elections, which is buttressed by the inconvenient fact that swing voters simply are not political moderates as defined in the Beltway.)
Later, he puts it even more succinctly: “Psst, I’ve got a secret: presidential elections are not about policy.” Again, I mostly agree.
The trouble is that de Boer then dives into a very similar claim in reverse, in service of his rival instinct that bold economic progressivism wins elections while moderate liberalism brings electoral catastrophe. In fact, that argument is the entire premise of his post’s title: “Clinton & Obama Gave Us Trump.”
In his hyperbolic way, de Boer claims Trump won in 2016 because the neoliberal order embodied by Barack Obama and Bill Clinton “abandoned the working class” and “shanked labor by passing NAFTA, casting millions of people into economic devastation in crumbling Rust Belt towns.” As a result:
“the people in those crumbling Rust Belt towns weren’t too keen on rewarding the Democrats with their votes. They watched their communities slip into poverty and addiction thanks to a ‘pro-growth’ economic plan and total indifference from big shot Democrats…to any American citizens without fancy college degrees.”
What Freddie de Boer gets wrong
This narrative – that neoliberalism gutted the heartland until it got fed up and voted for Trump, because warts and all, at least he wasn’t neoliberal – is simultaneously too broad and too narrow. Too broad because it wishfully stretches Trumpers’ limited gripes about immigrants and globalization into an imaginary hostility to neoliberalism writ large, when the truth is closer to the opposite. And too narrow because it ignores the many non-economic explanations for Trump’s victory, which likely played at least as large a role.
It is fair to frame the 2016 election as an angry Rust Belt repudiation of elites and their shibboleths. It’s also fair to argue that NAFTA contributed to this anger, so long as we recognize it as just one contribution among many. But neither elites nor NAFTA can be swapped out for this uselessly broad “neoliberal” term the left insists on using. There are at least seven reasons it’s silly to insist that neoliberalism, Obama, or (Bill) Clinton “gave us Trump,” much less to infer from this a recipe for political success moving forward.
1. Neoliberalism barely happened.
This is the subject of the first post in Yglesias’ series. In many important respects - including environmental, housing, disability, and healthcare policies – the U.S. economy is now much more regulated than it was 50 years ago. Over the same time, tax revenues have remained steady and government spending has gone up as a portion of GDP, including for social welfare programs. Yglesias elaborates:
“Since Bill Clinton’s time, Social Security and Medicare has grown due to an aging population. Medicaid has expanded several times. The ACA exchanges were created. SNAP benefits have become more generous. For that matter, Ronald Regan only very partially rolled back the Great Society.”
Like “isolationism” or “trickle-down economics,” the word neoliberalism is a lazy pejorative for a wide spectrum of beliefs, only some of which Obama and Clinton shared, and even fewer of which were enacted into law. The only neoliberal policies you can plausibly blame for the 2016 election are free trade and reduced labor unions/protections, which already narrows the critique significantly.
2. Those aspects of neoliberalism which did happen produced broad benefits as well as costs. Forgoing those benefits would have hurt different people in partially offsetting ways.
I said I’d steer clear of the policy fight, this post is getting long, and this point is inessential to my argument, so I’ll move it to a footnote.3
3. If voters were actually penalizing neoliberalism, they should have penalized Republicans just as much.
Even if neoliberalism was entirely to blame for the Rust Belt’s deterioration, it would have been dumb to hold Democrats in particular accountable for it. For the entire period in question, Republicans were the more neoliberal party of the two. They championed free trade and small government for decades, including during Obama’s term. They were union-busting the whole time. They had a whole Tea Party movement about small government, which directly countered the Occupy movement and segued into MAGA.
The idea that Trump won because Obama-era Democrats were too economically conservative would have come as a surprise to the great majority of people who voted for Trump, who famously believed Obama was a socialist. (This actually reaffirms Freddie’s earlier, much stronger point that people don’t vote on policy.)
4. Even in 2016, Donald Trump was more neoliberal than Hillary Clinton on most issues.
Donald Trump wanted lower taxes than Hillary Clinton. He wanted less regulation than her. He wanted a less generous social safety net than her. He wanted lower minimum wages than her. He was more “pro-growth” than her. In what way was he a protest candidate against neoliberalism?
The only plausible policy answers are trade, immigration, and foreign policy. Let’s be real about which of these three – immigration - exerts the most pressure on election outcomes. Let’s also be real about what animates the average Trump voter’s views on immigration. It’s not their concern for how much power labor has relative to capital. It’s not their deep familiarity with the economic literature on how immigration impacts wages or employment.4 Psst, I’ve got a secret: it’s racism. The most relevant way Trumpers opposed neoliberalism was as a euphemism for guarding “our” jobs, women, and soil from being “taken” by immigrants who weren’t Mexico’s best.5
Since de Boer agrees with me that elections are not about policy, maybe the strongest version of his argument would concede that Trump voters did not coherently oppose neoliberalism in their own words, nullifying my points 3 and 4. Rather, he could argue their communities and lifestyles were uprooted by neoliberalism in practice, which left them angry, desperate, adrift, and abandoned. Perhaps that anger and desperation is what made them lash out for Trump – not because of his policies, but because he was a bull in a China shop who promised to shake things up. This has a kernel of truth to it.
But for this argument to have political implications that are not redundant to the policy debate – for it to bear at all on how Democrats can win elections moving forward - the direction in which voters lashed out needs to at least be the less neoliberal one. Otherwise, democratic accountability is fundamentally broken, and the lesson is just to be a bull in a China shop for the foreseeable future. Even if we trust voters to accurately attribute blame or credit for their quality of life (which I do not), policy debates only become politically relevant if one side can hold power long enough to noticeably improve the decisive voters’ lives. Four years of pro-union policies apparently isn’t long enough.
5. Hillary Clinton won the popular vote decisively.
This seems relevant to the idea that neoliberalism is a political loser. If de Boer’s theory of what 2016 was a referendum on is correct, he concedes neoliberalism made most people happy. If it made other people angry and resentful, it was only in particular areas, which may not be the swing states forever. In a more democratic country with no electoral college, Obama and Clinton would have given us Hillary.
6. Economics is clearly not all voters care about.
Think back to 2016. What was floating around Facebook? What was hyping people up?
I remember racial resentment about Colin Kaepernick taking a knee, and about BLM after the Freddie Gray riots. There was the Pulse nightclub shooting, which every conservative insisted was solely an issue of radical Islamic terrorism, and every progressive insisted was solely an issue of gun control and homophobia. There were Hillary Clinton’s emails. There was a lot of fake news and Russian misinformation. There was strident criticism of Clinton’s foreign policy, by people of all economic persuasions, because she was an unrepentant warmonger her entire career. There was an audiotape about pussy grabbing, and an endless stream of media freakouts about Trump’s offensive political incorrectness, fanning the flames of the culture-war fights he thrives on.
For 2016 of all years, “it’s the economy, stupid” is dopey reductivism. “Clinton & Obama Gave Us Trump” ignores dozens of factors that sure seemed politically salient at the time to pretend that the 2016 election was about the North American Free Trade Agreement signed in 1992. Stop it. People were not angry and resentful because of vaguely pro-growth policies. They were angry and resentful for a complex convergence of reasons exacerbated by social media and phone addiction, and they lashed out at whichever side happened to be the incumbent at the time. Maybe because they were elites, but not because they were neoliberals.
7. The anti-neoliberal narrative does not fit any other election in my lifetime, nor even 2016 counterfactuals.
Reducing politics to economics and economics to trade and unions lets de Boer present Joe Biden in an unfamiliar light: as an unabashed leftist, who never runs to the center like those timid weak-kneed Obama neoliberals. It should give him pause that this is the exact opposite of Biden’s general reputation, including on immigration, race, abortion, Israel, drugs, and crime. For most of his final year in office, Biden was tripping over himself to run to the center on these issues.6 For all of 2020, Biden ran as a safe, moderate, return-to-normal Democrat to sharpen the contrast with Trump.
While pretending that Joe Biden somehow doesn’t qualify as a big shot Democrat of the existing neoliberal order in which he spent 36 years as a centrist Senator, de Boer makes a bold claim:
“Joe Biden absolutely would have beaten Trump in 2016; he was four years younger than in 2020, when he won, and enjoyed good favorability ratings, unlike the candidate Obama anointed.”
But hold on now. It took an earth-shattering pandemic for Joe Biden to beat Donald Trump in 2020 – and even then, he barely won. So barely that it took a week to count all the votes proving he won. And in 2016, Biden could hardly have campaigned as a crusader against the existing neoliberal order after eight years as Obama’s VP, any more than Harris can run against Biden now. If the 2016 election was the repudiation of Obamanomics that de Boer wants it to be, how can he be so confident Biden wouldn’t have been saddled with them too?
Because, in his words, “Any generic Democrat would have beaten Trump. It took nominating the historically unpopular Hillary Clinton to screw that race up.” This, I actually agree with. But you know who else “anybody” probably includes? Barack Obama, and probably even Bill Clinton, had they been allowed to run again. Again, this problematizes the narrative that Hillary’s loss repudiated “the consequences of the political dynasty she belonged to.” It repudiated her individually, and the tone-deaf campaign she ran.
The vote totals affirm this. In the past four presidential elections, the Democrats’ shares of the popular vote have been 52.9, 51.1, 48.1, and 51.3. Republicans’ shares have been 45.7, 47.2, 46, and 46.9. Trump was not an outlier on the right; Hillary was an outlier on the left. Biden’s done about as well as Obama. Can we really read into this some searing indictment of Obama’s economic agenda?
Nor do the elections before and after – mysteriously lacking a massive blowback against neoliberalism – fit this pattern. Here’s what de Boer has to say about Obama winning twice by comfortable margins:
Obama was elected in 2008 because he followed a president who had dragged us into Iraq, fiddled while thousands of people drowned in New Orleans, and utterly failed to predict or prevent the biggest financial meltdown since the Great Depression; he also ran as a figure of transformational ambitions, hope and change and all that, and then very conspicuously governed as one of the most timid and anxious presidents in American history. He was able to survive in 2012 because we largely pulled out of Iraq (which we would have done anyway) and because following 2009 the economy could hardly do anything but improve. Also, a lot of Americans secretly harbor prejudice against Mormons, which didn’t hurt.
See how this works? The 2008 election was largely about foreign policy and disaster relief. The 2012 election was largely about foreign policy and prejudice about Mormons. Obama’s economic achievements were inevitable from circumstance, not the result of his policies. But the 2016 election? That was clearly a pushback to Obama-era economic policies – not to his race, or to still being in Afghanistan, or to everything in point 7.
The same goes for 2020. Biden did not win primarily because of his economic proposals; that’s retrofitting order onto chaos. He won because COVID was a black swan event that affected everyone and made Trump look even more out of his depth than usual. And then we had BLM summer, and everyone was anxious and heated and scared, and the sense we were on the wrong track became overwhelming. And even then, Biden won barely, because tribalism’s a hell of a drug.
What Freddie de Boer gets right
For all the ink I’ve spent refuting his narrative on 2016, I actually agree with the most actionable takeaways of de Boer’s post: that presidential elections are not mainly about policy, and that chasing the center is the enemy of progress. Taken together, this means politicians should update in the direction of ignoring polling and in favor of saying what they believe with conviction and integrity. Just “pull the rope,” as Freddie puts it.
Here is my crude theory of the case, for which I have only intuitive evidence, and which is cleverly and hypocritically broad enough to be unfalsifiable.
Most voters are rationally ignorant about policy issues and outsource their political thinking to trusted pundits in their ingroup. As tribal creatures in a polarized, fragmented, and disorienting media landscape, they’re increasingly inclined to believe anyone who flatters their biases and blames people unlike them. But being uninformed and non-ideological, their policy opinions (distinct from their broad values) are weakly held and highly malleable by those they trust. They take cues about policy from whichever public figures signal allegiance to their tribe.
This creates an annoying cycle. Something goes wrong in America, and pundits on all sides blame their preferred boogeyman: immigration, billionaire greed, wokeness, neoliberalism, the deep state, patriarchy, cancel culture, white supremacy, the decline of religion, etc. With neither the training nor the incentive to carefully reason or research, everyday Americans pick whichever narrative flatters their biases. They decide they don’t like X boogeyman either, not because they have some unshakeable opinion on the matter, but because that’s what they were told by the people to whom they outsource their political thinking.
Soon, polling data shows that X is now unpopular – and the biased pundits pounce on that data as proof that the boogeyman not only produced bad outcomes, but is also a political loser. The more complicated the problem is, the more everyday people need to outsource their views, and the more malleable their causal theories become.
This is roughly what happened to Republicans after the 2008 financial crisis. It’s what happened to Trump after COVID and BLM, and to Biden after inflation, Ukraine, and (most clearly) the southern border.7 It’s what happened after October 7th, and what happens on most foreign policy issues. Incumbents will always be blamed for bad outcomes, whether they deserve it or not; but that’s less because their policy positions were innately unpopular than because the right people scapegoated them at the right time, which is easier than ever to do in the post-truth era.
Annoying as this cycle is, it also has a more optimistic implication: leaders can make forceful and persuasive arguments for what they believe to be true, and if enough pundits on their side echo them, roughly half the voters will nod along. If the new argument differs from what this tribe used to believe, many voters will silently change their minds. The polling will not reflect this until you commit to the argument, though - so you can’t be too scared away by the polling. Trump certainly wasn’t, and Republicans dropped their old positions like a hot potato.
“A leader is not a seeker of consensus, but a molder of consensus.” – Martin Luther King Jr.
I do not mean to imply that presidential elections have nothing at all to do with policy positions. Changing public opinion takes time, and picking the right issues at the right moment probably picks up crucial voters at the margins. I mean only that a) the impact of policy is typically dwarfed by the impact of unrelated factors, from age to charisma to the timing of a pandemic or foreign war, and b) even within policy, there’s no universal formula, so the party as a whole might as well tell the truth. Thoughtful Americans have a wide range of views on economic policy, while decisive portions of the electorate have no clear, firm, or consistent views at all. There’s little reason to pander to soft opinions on things these voters don’t understand if they’re going to vote on vibes anyway.
So on one hand, no, neoliberalism did not give us Trump in the way de Boer implies. On the other hand, no, Democrats do not need to run to the center; and yes, continually doing so will give Republicans free policy wins without earning enough votes to justify it.
Conclusion
Zooming out from de Boer and Yglesias, this whole recurring debate between moderates and progressives reminds me of my favorite XKCD:
Both sides accuse the other of being out-of-touch elites who can’t see what everyday working people can. Both accusations are at most half right. The accused may really be an out-of-touch elite who can’t see the truth – but that’s not the problem. The problem is that the everyday people can’t see shit either. They’re equally adrift on the same sea of incomprehensible information as us uppity college kids, except that they’re even less trained on how to swim in it.
The humble ones admit they are perpetually confused, while the arrogant ones are wildly overconfident in primitive or fantastical theories – but either way, the direction of their theories point in opposite directions as often as not. Nobody knows what the hell is going on at any given time, and every four years we have an election that increasingly resembles a coin flip.
It’s the least bad system I know of, but hoo boy…that doesn’t make it great.
For full disclosure, I typically think progressives are better on foreign and social policy while moderates are better on economic policy and sounding like a normal person.
If you don’t know what neoliberalism is, that substantiates the point I’m making in this post. Most voters don’t know either.
Because of free trade, a wide range of goods – like TVs, software, toys, and cellphone service – became much cheaper for all Americans in ways that directly benefited their quality of life. Families could afford more Christmas gifts for their kids. Seniors could retire earlier or more comfortably off the backs of a strong stock market, which benefits a majority of American families. Even if the drawbacks exceeded the benefits (or at least, were more politically salient than the benefits) it’s begging the policy question to only count one side of the ledger.
Had Clinton not passed NAFTA and Obama not…done TPP? or whatever neoliberal thing he’s accused of...the resulting rearrangements to the pie could have just pissed off a different group of people who also vote. And the Rust Belt might have rusted anyway, only slower. Pro-union presidents may not have been able to save it – they might have just caught equal blame for its irreversible decline, while also being saddled with a more sluggish economy with higher prices. Even those who prioritize labor may not be rewarded for it at the polls.
Studies finding that immigration improves the wages, income or living standards of native-born citizens can be read here, here, here, here, and here; that immigration improves native employment, here, here, here, and here; that it improves native occupational mobility, here. A consensus of 88% of economists polled at a 2013 IMG forum agreed that the average American would be better off with more high-skilled immigrants (0% disagreed), while 52% said the same for low-skilled immigrants (with 28% uncertain – only 9% disagreed). Even reviews finding a negative impact on natives concede “most studies find only minor displacement effects even after very large immigrant flows.” As one study put it, “evidence that immigrants have harmed the opportunities of less-educated natives is scant.”
To push on this: according to de Boer’s own theory of the case, the rebellion against neoliberalism in our country overlaps pretty closely with the movements for nativism, anti-intellectualism, fascism, and populist demagoguery. That’s not who Freddie is, to clarify; but if his theory of 2016 is right, it’s who most of the people in the country upset about neoliberalism are. Insofar as that’s true, that seems like a reason to be cautious about Democrats hewing too far in that direction. Just as intelligent, cosmopolitan, bleeding-heart libertarians have learned to steer clear of the Libertarian Party because it turns out to be full of racist idiots, progressives should be wary of aligning with such malignant bedfellows.
Except for abortion, but for most of his career he was pro-life too.
In May 2020, 28% of Americans wanted to reduce immigration levels. In July 2022, that figure was 38%. In July 2024, it hit 55%, despite border crossings being lower than in 2022 and less than half what they had been a year prior. Breathless 24/7 conservative TV news coverage of the “crisis” and “chaos” at the Southern Border did not respond to over a quarter of the population spontaneously changing their mind; it caused the population to change its mind. It pulled the rope. Republicans messaged loudly enough for long enough to convince the American people that this was a terrible crisis responsible for crime and inflation, while Biden refused to fight back with any positive case for immigration precisely because the polls told him he shouldn’t touch the issue. He sought consensus, they molded consensus, and now the consensus is just as dumb as they are.