New York City’s mayor is a democratic socialist. So is Seattle’s. And Washington, D.C. looks poised to join them: Janeese Lewis George won the Democratic nomination for mayor on what the New York Times called “the once-forbidden ‘S word,’” and a promise to expand child care, build tens of thousands of homes, and widen rent stabilization. Zohran Mamdani, the figure the rest of the movement looks to, took office in New York pledging to “replace the frigidity of rugged individualism with the warmth of collectivism.”
So socialism is back, and the commentary has arranged itself in the usual formation. On the right, alarm: the reds are at the gates. On the left, a banner: the future has arrived. What is striking is how little of the excitement and the horror attaches to anything the new mayors actually propose. The same rent freeze, the same municipal grocery, would pass almost unnoticed under another name. It is the word that thrills the one side and frightens the other, and so the question everyone hurries to argue is a question about the word. Is this really socialism, or just New Deal liberalism with sharper slogans? Does a city-owned grocery store cross the line, or is it ordinary municipal housekeeping?
Get Liberalism.org in your inbox.
The fixation is a mistake, and an instructive one. Whether Mamdani’s program “is socialism” tells you almost nothing about whether any piece of it is worth having, and focusing on the question is one of the surest ways to think badly about the policies it attaches to. It is worth seeing why, because the confusion is not special to socialism.
* * *
Begin with the most basic confusion. When Pew found in 2022 that more than a third of American adults (and 44 percent of those under thirty) hold a favorable view of socialism, many observers were horrified. Had people forgotten—or never learned—the horrors of Stalinist Russia or Maoist China? Did people not realize that central planning simply couldn’t work? This reaction is understandable. But it was based on a misunderstanding. As an earlier Pew study revealed, people who profess to support socialism mean something very different by that word than the people who say they oppose it. Ask a critic of socialism what he has in mind and he points to the Soviet Union or Venezuela, where the state owns the economy and plans it from the center. Ask an admirer and she points to Denmark or Finland, market economies with private firms, open trade, and unusually large welfare states.
The distinction matters because the strongest objection to socialism applies to only one of these two sets of examples. Mises and Hayek showed that a centrally planned economy cannot allocate resources rationally, because without prices and private ownership, its planners are flying blind. The argument is decisive, but for the purposes of the current debate it is simply beside the point, because Mamdani and the others are not proposing anything like central planning. An argument against an authoritarian, centrally planned economy is not an argument against a rent freeze or a free bus. When someone invokes the graveyard of twentieth-century communism to condemn Mamdani’s transit policy, he has let the word do his thinking for him. The catastrophe and the bus share a name and nothing else.
* * *
That confusion is a special case of a more general problem with ideological labels. In The Myth of Left and Right, Hyrum and Verlan Lewis distinguish two theories of what an ideology is. On the essentialist theory, the positions associated with a label cohere because they flow from a single underlying principle. You start from the essence, deduce your views, and only then do you join the tribe that shares those same beliefs. The Lewises argue that this is backwards. An ideology’s views cohere because a tribe holds them, not the other way around. The tribe comes first. People anchor into a tribe for reasons of family, peers, or a single galvanizing issue, and then take on the rest of the package. “Ideologies do not define tribes,” the Lewises write; “tribes define ideologies.” A label tells you reliably who someone stands with. It tells you very little about what they stand for.
I leaned on this argument in an earlier piece about my own side, where libertarians, who all claim to be defending liberty, turn out to disagree about immigration, foreign policy, and nearly everything else. Socialism is no different. Consider Nithya Raman, a democratic socialist on the Los Angeles City Council who has broken with the Democratic Socialists of America on housing, backing the development-friendly policies many of her comrades resist. Her allies in the movement want to protect tenants chiefly by restricting landlords. Raman wants to build. It’s the same label, but opposite answers to the most consequential question in urban policy. Merely knowing that Raman is a socialist tells you nothing about where she lands on it.
* * *
These are problems about what the word picks out. There is a further problem, and a deeper one, about what even a settled principle could tell us. I came to this problem while reviewing Matt McManus’s book on liberal socialism. In that book, McManus defines socialism in terms of a set of moral principles: the equal worth of persons, individual freedom, community, and so on. The problem is that nothing in the logic of these principles renders them incompatible with markets, private property, and limited government. In other words, nothing in them dictates a socialist economy. As I put it in my review, just as economic institutions cannot be read off moral principles, so one cannot read moral principles off economic institutions. The label sits in the gap between the two, doing no work.
It is worth adding, because it bears on the whole argument, that McManus and other self-described liberal socialists have over the past few years been more consistent than a good many libertarians in naming the illiberalism of the Trump administration for what it is. If you had wanted to know in advance who would defend liberal constitutional norms when it counted, the label “libertarian” would have steered you wrong, and the label “socialist” would have steered you wrong, in opposite directions. The labels, in other words, are a poor guide to the thing you actually care about.
* * *
So set them down and look at the policies. They sort out very differently than the banner suggests. Rent control, which several of these candidates favor, is a genuinely bad idea, but not because it is “socialist.” It is bad because ceilings on rents reduce the supply and quality of housing, a conclusion economists on the left and right have shared for half a century. Building more housing, which Raman favors, is a good idea for precisely the same reason, even when it flies under the same flag. In between are proposals with serious arguments behind them that have nothing to do with socialism at all. There is, for instance, a respectable economic argument for subsidizing child care. James Heckman’s work on high-quality early-childhood programs finds returns on the order of thirteen percent a year, part of it through the higher earnings of mothers freed to work. Even fare-free buses, which sound like a pure giveaway, have a surprising efficiency argument that can be made on their behalf. On many bus systems, the cost of collecting fares rivals the revenue it brings in, and dropping fares speeds boarding enough to improve service and lift ridership. The label “socialism” bundles a clearly bad policy, a probably good one, and several genuinely open questions. The bundling tells you only which team to root for.
A skeptic will object that I have stacked the deck by focusing on easy cases. There’s something to this charge. The platform of the Democratic Socialists of America runs well beyond rent freezes and free buses. Among other things, it calls for abolishing the police, for bringing whole industries under public ownership, and for taking housing off the market altogether. These are ideas considerably worse, and considerably more radical, than mere rent control. Still, the practical significance of all this is dubious. Platforms are largely symbolic. They are written to signal whom a movement stands with and what it longs for. They are written to thrill the faithful rather than to survive contact with the realities of budgets and political compromise. What matters at the end of the day is actual governance. And when socialists have been forced to make the hard choices involved in governing, they have mostly ignored the platform. Mamdani ran under the banner and then kept the police commissioner he inherited, the one who had just presided over a record drop in crime. Raman wears the labels and builds.
Even granting all of this, one might argue that the label itself is a political fact with consequences. A coalition gathering under a banner changes what is sayable and who is electable, and this banner carries a long and bloody history behind it. To wave away the word is to wave away a real and reasonable worry about where the country is being pulled. I think this is a legitimate worry. But the lesson here is that we should be more careful in how we use the word, not less. Part of why the word “socialism” has lost its bite is that it has been used so loosely, first by a right that has called every welfare program socialism for generations, and now by a left that’s content to claim the name for Denmark. A word stretched to cover everything ends up warning against nothing. If some of the ideas under the banner are really dangerous—as I think they are—the way to fight them is to name them and show, on the merits, why they are bad. The answer to bad, fuzzy ideas is to be more careful, clear, and rigorous in our analysis. Not less.
Tyler Cowen once offered a rule of thumb worth keeping: “every time you’re telling a good vs. evil story, you’re basically lowering your IQ by ten points or more.” Is it socialism? is a good-versus-evil story in the costume of an analytical question. Set it aside, and the real questions come into view, the ones with answers. Does rent control work? Can a municipal grocery break even? What policies actually result in new housing getting built? Those are the questions worth our time.
