Talking Past Each Other
A Crisis Dispatch
Derek Thompson posted on X this week that he is glad the author of “Rent Control Is Fine, Actually” calls themself Unlearning Economics, because it’s good to just state things clearly, such as the open animosity that many left economic populists have for the field of economics and economists themselves. The post links to a Nathan Robinson piece in Current Affairs. The framing is the structural problem.
Where I stand on the substance: I generally like Abundance. Thompson and Klein have correctly identified a real cluster of pathologies in how progressive governance has actually operated for fifty years — the proceduralism that produces nothing, the veto-cracy, the NIMBY architecture, the way state capacity has been systematically degraded by the very people who claim they want to wield it. The reactions from the anti-liberal left that paint Abundance as a corporate Trojan horse — the Sam Seder version, the Nathan Robinson version — are overwrought. Most of them have not engaged with the actual book. Abundance is closer to the opposite of what they describe. It is an argument for the state to be able to do things again, in service of the kind of polity working people can live in.
Sam Seder is wrong on the substance, and the wrongness matters because Mamdani — who is currently mayor of New York and who is, on every account I have read, operationalizing significant parts of what Abundance recommends — is the figure who is actually showing how the supposed dichotomy between Abundance-style state-capacity reform and democratic-socialist values politics resolves. It does not resolve as a contradiction. It resolves as a synthesis. The mayor of the largest American city is doing the work Abundance describes, while calling himself a democratic socialist, while operating from the values position that Abundance is supposedly the corporate enemy of. The dichotomy collapses on inspection.
So far Thompson and I are on the same page.
The post about rent control is the place where I want to stop and say: Derek, you are wrong about what kind of argument you are in.
Thompson’s framing is that economics is a field, the field has reached substantial conclusions about the second-order effects of rent control, and the people who reject the conclusions are operating in open animosity to the field. The framing positions Thompson as the defender of the science against the populists who reject the science. The economics shows what works. The populists reject what works. Thompson’s job, as he says in the post, is to popularize the ideas that I think are good. The work is to bring the public to where the science already is.
The economics is roughly right. Rent control, in its strict price-freeze forms, does have the second-order effects Thompson describes — landlord conversion to condos, deferred maintenance, market exit, marginal supply contraction. The contraction tends to hurt the next generation of renters trying to enter the city more than it helps the current generation already housed there.
But the argument is not about the consequences. The argument is about which consequences are worth which trade-offs. That argument is not an economic argument. It is a normative argument about what a city is for, who it is for, and which goods it wants to maximize at the expense of which other goods.
Hume named this two hundred and fifty years ago. You cannot derive an ought from an is. No matter how complete your description of the empirical world, you cannot extract from that description a normative conclusion about what should be done. Reason, in Hume’s account, is the slave of the passions. The passions specify the goal. Reason figures out how to get there.
Modern technocratic thinking, in its standard operating mode, has forgotten the distinction. The technocrat looks at the empirical evidence — rent control reduces supply over time — and concludes that rent control is bad policy. The conclusion has slid across Hume’s guillotine without acknowledging the crossing. The slide assumes that more housing supply is the good being maximized. The assumption is hidden inside the framing. If you ask the technocrat to make the assumption explicit, the technocrat will say something like of course we want more housing, this is the entire point of housing policy, as if the of course did the work the assumption needs to do.
It does not. More housing is one possible normative goal among several. Other goals: protecting existing residents from displacement. Maintaining the cultural and social character of established communities. Keeping children in their schools. Keeping elderly tenants in their apartments. Preserving the immigrant social networks that took two generations to form and that disperse in months when the rents move. Maintaining racial diversity in neighborhoods that gentrification reliably whitens. Holding the line against the asset-class transformation that converts housing from a place where people live into an instrument of capital extraction.
Each is a real good. A polity that ranks any of them above supply maximization is not making an economic error. It is making a normative judgment about what its housing system is for. The judgment may turn out to have costs the polity did not fully anticipate. Every normative choice has costs. The question of which costs to accept in service of which goods is the question politics exists to answer.
⁂
The technocratic move I am pushing back on is what happens when this normative deliberation gets reframed as a question of economic literacy. The reframing removes the question from the deliberative arena and relocates it to the expert arena. It tells voters that their preference for rent control is not a value choice the polity gets to make, but a mistake the polity needs to be educated past. It positions the technocrat as the authority who knows what the polity should want, with the polity’s actual stated preferences treated as ignorance to be corrected through better popularization.
This is what Thompson means when he says if policies that work are less popular than policies that don’t work, that doesn’t mean it’s smart to ignore the difference between good and bad affordability ideas. It just means I have more work to do. The sentence sounds humble. It is not. It assumes that good and bad affordability ideas are settleable on the empirical evidence, that the popularity of bad ideas is a defect in popular understanding rather than a clue about what people might be valuing that the technocratic frame is missing, and that the technocrat’s job is to keep producing content until the public catches up.
The deliberative posture is different. It says: the public’s preference for rent control is data. The data is not data about what works. It is data about what people value. They are telling you that they value the protection of existing residents enough to accept costs at the margin. The economics tells you what those costs will be. The economics does not tell you the people are wrong to accept them. Whether the costs are worth it is the question the polity is having, and the question is not closed by the economics.
This is the same distinction Aristotle drew between techne and phronesis. Technical knowledge is the knowledge of how to produce a specific outcome reliably. Practical wisdom is the knowledge of which outcomes are worth producing. The two are different kinds of knowledge. The first can be possessed by experts in narrow domains. The second is the knowledge a polity full of co-citizens develops through the exercise of self-government, and it cannot be transferred from technical domains because it requires the lived experience of co-citizenship to develop.
Thompson is operating in the technical register and treating it as if it covered the practical register. The treating is the move. It is what Abundance‘s critics are reacting to, even when their reactions are otherwise wrong. Sam Seder is wrong about almost everything in his anti-Abundance content, but he is not wrong about this. There is a structural posture in Abundance-adjacent commentary that treats the policy preferences of the working class as obstacles to be overcome rather than as positions to be engaged with.
⁂
The technocratic posture, applied to politics, generates a specific failure mode. It produces commentators who are right on the empirical questions, increasingly frustrated that the public is not converging on the empirically correct answers, increasingly inclined to attribute the non-convergence to the public’s economic illiteracy or media-induced confusion or animosity to the field. The commentators are not lying about their frustration. They are genuinely puzzled. Why won’t the public adopt the policies that the evidence supports?
Because the policies the evidence supports do not exist as freestanding objects. There are no policies that are simply good or bad in the abstract. There are policies that maximize specific values at the expense of other values, and the values they maximize are visible to the people who care about those values, and the values they sacrifice are visible to the people who care about those values. The technocrat who tells the public that the good policies are unpopular is, without realizing it, telling the public that their values do not count. The telling is heard. The hearing is part of why the technocrat keeps losing the argument they are sure they should be winning.
This is what the populist class on the left and the right have correctly identified, even when their alternative proposals are worse than what they are critiquing. They are not wrong that the technocratic class has, for several decades, treated its own value commitments as the neutral baseline from which deviations are pathological. The populists are wrong about a lot of other things. On this point, they are tracking something real.
⁂
The technocratic posture, taken to its logical conclusion, is structurally similar to the Yarvinite posture. Not identical. Not even close in moral character. But structurally similar in this specific way: both treat the deliberative apparatus of liberal democracy as a kind of failure mode, a noisy and inefficient process that gets in the way of producing the correct outcomes. The Yarvinite says: dissolve the apparatus, install the CEO-king, let the technically competent rule. The technocratic liberal says: keep the apparatus formally, but treat its outputs as legitimate only when they align with what the experts already know is right. The methods are different. The underlying frustration with deliberation is the same.
The deliberative posture rejects both. The apparatus is not a failure mode. It is the political form humans have developed for working out what they value when they disagree about what they value. The technical experts inform the deliberation. They do not replace it. Their authority is real but bounded — bounded by the layer above them, which is the layer of value-commitment, which is the layer the polity itself owns. When the experts try to claim the upper layer as well, what they produce is not better policy. They produce the slow erosion of the deliberative legitimacy that gives the lower layer its actual political force.
This is what is happening to Abundance-adjacent commentary right now, in real time, even when the underlying argument is correct. The argument is right. The framing is corrosive. A thoughtful supply-side liberal like Thompson finds himself in open animosity with progressive populists who, on closer inspection, share most of his actual policy preferences but cannot tolerate the register in which he advances them. The talking-past is what the framing produces.
⁂
When you write about rent control, you are not just writing about supply curves. You are writing about whether the city you are advocating for is a city that prioritizes the entry of the next generation of renters over the protection of the current generation of tenants, or vice versa, or some specific blend that the polity gets to choose. The choice is the political content. The supply-curve analysis is an input to the choice. Treating the choice as if it were already made by the analysis is the move that makes the populists right that something is being concealed in your framing.
A polity gets to choose to protect its existing residents at the cost of slower supply growth. The choice is not stupid. It is a substantive value commitment about what kind of place the polity wants to be. A polity gets to choose to prioritize supply maximization at the cost of significant displacement. That choice is also not stupid. It is a different value commitment about what kind of place the polity wants to be. Different polities, in different cities, with different histories, will make different choices. The choices are theirs to make.
The technocrat’s job is to tell the polity what the consequences of its choices will be, with the most rigorous evidence available, in the most accessible language possible. The technocrat’s job is not to tell the polity which consequences it should accept. When the technocrat slides from the first job to the second, the polity is right to push back. The push-back may take the form of populism, which is sometimes coherent and sometimes not. The incoherence does not make the push-back wrong about the slide. It makes it wrong about other things, while still being right about the slide.
⁂
Mamdani is interesting because he is, so far, refusing to make the false choice. He is operationalizing supply-side reform — densification, permitting reform, public construction — while remaining in the values-language of democratic socialism. He is doing the Abundance work without buying the framing that Abundance-adjacent commentary often arrives in. He is treating the technical reforms as instruments of a value commitment, and the value commitment as the thing that has to be worked out through democratic deliberation rather than through expert popularization. The model is right. Whether his execution lives up to the model is a separate question we will be able to assess in two or three years. The model itself is the right one for serious liberal thought to be developing right now.
Values from the polity. Technique from the experts. Neither layer collapsing into the other. The technical experts informing the deliberation but not replacing it. The deliberation owning the values but not pretending it can do without the technique.
This is harder than what Thompson is doing, and it is harder than what Sam Seder is doing. Most of contemporary American political commentary has chosen one of the two collapsed versions — either the experts know what the policy should be and the public needs to catch up, or the experts are corporate stooges and only the people’s authentic voice matters. Both are wrong. Both miss the relationship between the layers that is the actual political work.
The actual political work is the deliberation about what the polity is for, conducted by the polity itself, informed by the best available evidence about consequences, with the experts as participants in the deliberation rather than as authorities standing above it. The deliberation is messy. It is slow. It produces outcomes the experts often disagree with, and the disagreement is part of how the deliberation works rather than evidence that the deliberation has failed.
⁂
A working-class tenant in New York who supports rent control is not economically illiterate. They are valuing the protection of their community over the maximization of aggregate supply. The economics tells them what the trade-off costs. The valuation is theirs. The polity is supposed to be the arena where the valuation gets adjudicated against other valuations, with the consequences understood, in the deliberative form that gives the adjudication its legitimacy.
When that arena gets reframed as an expert-popularization process, the legitimacy drains out of the adjudication, and what is left is technocrats wondering why the public will not catch up. The public is not behind. The public is having a different argument than the one the technocrats think they are in. The argument the public is having is the right argument. It is the argument about what the city is for, who it is for, what it values, what it is willing to trade.
The technocrats are supposed to be in that argument too, as participants. When they posture as the arbiter of the argument’s correctness, they exit the deliberation. The exiting is the problem.
Sam Seder is wrong about most of what he says. The Sam Seders of the world are wrong about most of what they say. On the question of whether Thompson is in an argument about science or in an argument about values, they have noticed something the technocratic frame is built to obscure.
Economic science is a tool. It measures. It predicts. But it doesn’t decide.
Those are my thoughts on the matter.
Good to get past the posturing grown up around this subject. As an urban planner, I remember this got started by environmentalists trying to prevent more elite abuse that was crammed down urban throats in the 60s -- mostly inner city neighborhoods and rural pasturage. Of course, the more supercilious earned a canned disparagement from actual elites, and NIMBY effectiveness was seized by elites to control growth for themselves, so enlightened were they. It's time to return to the original purpose and to trust ordinary people with decisions over their own lives, deliberatively.
Well put. Thanks for the clear presentation of the absurdities so many argue by.
Yet there's an underlying assumption to the claim "reason is the slave of the passions," that reason and the passions are functionally separate, separable realms or capacities -- no intertwining as reasonable passion nor passionate reason.
Our attention to the world steers as largely by our passions and by our reasons. Even when our eyes are directed by passion, we reason about what we then perceive. And when with reasoning we construct alternatives, our passions attract or divert us among them; where we are directed by reason, we assess in part with passion. Hume would be right in that famous phrase IF reason and passion were ever separate, separable realities of our psychology. Extreme examples may suggest they sometimes are, but most often, individually and socially, they are an admixture; just as walking standardly involves two feet, despite that we may hop.
If we take Hume to be advising we never idealize a reason devoid of passions, that's good advice. But intertwining the two as passionate reason, and reasonable passion is hardly to make a "slave" -- despite how the hyperbole makes for a memorable phrase. Your passionate reasoning is a fine example of intertwining, so undercuts the stronger version of Hume's claim.