“Romantic” once simply meant “of Rome.” Later it came to mean common language tales of knights, chivalry, and idealized heroes. The Romantic Movement (roughly 1770–1850) took up this meaning as a deliberate reaction against Enlightenment rationalism and the rapid changes then happening in commerce and society.
Thinkers as diverse as Rousseau, Herder, Schiller, and Burke applied “romance” to social institutions, economics, and political philosophy. They idealized pre-modern, organic communities (guilds, village life, feudal bonds), and nurtured nostalgia for agrarian life and artisanal economies as a way of critiquing industrial capitalism, which they found dehumanizing.
Get Liberalism.org in your inbox.
Conservative romantics, like Burke, Coleridge, and Carlyle, looked back in time to idealized medieval or agrarian social orders as a golden age. In Reflections on the Revolution in France, Burke praised custom and described loyalty to nations as if they were families with inherited traditions:
In this choice of inheritance we have given to our frame of polity the image of a relation in blood, binding up the constitution of our country with our dearest domestic ties, adopting our fundamental laws into the bosom of our family affections, keeping inseparable and cherishing with the warmth of all their combined and mutually reflected charities our state, our hearths, our sepulchres, and our altars.
Burke claimed that “prejudice renders a man’s virtue his habit,” defending people’s cherished but untaught emotional prejudices. The longer those habits had lasted, and the more generally they had prevailed, the more he valued them. Later romantic nationalists called these shared prejudices the Volksgeist, or national spirit.
Romantics of the left, especially early socialists, similarly criticized industrial capitalism, but they imagined utopian futures rather than idealizing the past. Karl Polanyi explicitly invoked the “romantic reaction” to commercial society—a defense of social embeddedness against the abstraction and alienation of market relations.
When “romantic” is applied to institutions or economics, it means an idealized, organic, pre-rational impulse that resists reason—nostalgia for an imagined communal past, or hope for an imagined communal future. It carries the original literary sense of “like a romance”—elevated, beautiful, but possibly impractical or illusory—and applies it to visions of how society should be organized.
I work in a branch of political economy called “public choice.” Public choice is politics without romance.
Hell is Other People
Near the end of Jean-Paul Sartre’s 1944 play Huis Clos, Garcin comes to a realization: “Hell is—other people!” Not because we have to depend on others, but because we must live with their apprehensions and judgments of us, “devoured” by their views.
Sartre could have been a public choice theorist. We romanticize the state and the process of government, but that’s a mistake. The state is just a particular technology—limited but powerful. As discussed in my previous essay, and as Jason Kuznicki laid out in Technology and the End of Authority, the mythologizing that moralizes and justifies state power has misled society for generations.
The power of choosing in groups is that it benefits group members; individual rights are not given by the state, but they are the justification for entrusting the state with limited powers. Much of public choice investigates the limits of group choice and collective solutions to social problems.
The first step is removing the romanticized moral status of “us.” Imagine the members of a group want to go to lunch. We enjoy each other’s company, but we have different tastes and dietary restrictions. This simple problem illustrates the core notion in public choice: politics as exchange. If we can make a choice we all like, we get to have a good lunch and a nice conversation. That’s a valuable goal, because we value doing things together more than doing things—even the same things—alone. This could be on the consumption side, like going to lunch, or on the production side, like national defense. In both, groups must choose.
The decision to “choose in groups” has three parts. In the lunch example, we have (1) Are we going to eat together or alone? (2) Which restaurant will we pick? (3) Once at the restaurant, what will we order? The order is crucial, because choosing to choose in a group determines many important aspects of what happens next. Our liking of any given restaurant could come down to the distance, the wait, the atmosphere, or the cost, as well as the menu. Membership in the group therefore matters: A given group will choose a particular restaurant. But if we add or subtract several people, quite a different choice is made, even if we follow the same procedures for choosing. And one of the core results of public choice is contingency: what we choose may depend heavily on how we choose. We could take turns—Alice chooses on Monday, Bob decides on Tuesday, and so on—or we could draw restaurant names out of a hat. We could vote, and different ways of voting also would “pick” different outcomes. This contingency matters more than is commonly recognized, and it’s important: method matters, and there is no obvious basis for saying one method is better than another.
But then, once the choice of a method for choosing is made, a restaurant is selected, and all those other factors are fixed: the only remaining choice is over menu items. And maybe how to pay.
This example illustrates the method of public choice: The choice of rules is a different choice from the options decided under those rules. Our “preferences” over different rules may be contingent on the actions we expect to take given those rules, just as choice of a restaurant is driven by what I will order if that restaurant is chosen.
Exit, or “going it alone,” is always an option, but many people prefer a group lunch at an acceptable restaurant over a solitary lunch at their favorite restaurant. That’s why choosing in groups is different from choosing as an individual: each person is simultaneously deciding whether to be a member of the group, and choosing to decide as part of the group. The romantic conception of the group requires only membership, as for example in Che Guevara’s “Socialism and Man in Cuba”; in this view, submerging individual wants into the identity of the group is just natural, because that is what any good person would want.
In earlier essays, I took up the theme of “seeing with two I’s”: information and incentives. The rules we choose dictate how we gather and use information, and they also determine the incentives that shape people’s actions. Collective, coercive choice is sometimes necessary. The problem is the temptation to extend it beyond the domain where such choices are required. In my restaurant example, we could decide, as a group, which restaurant to patronize, and then we could also “achieve consensus” on which menu item to order—we could make proposals, apply our rule for choosing, and enforce a single binding choice on everyone: liver and onions for all! Because “we” chose that.
That’s why Hell is other people: we want to be together, but making collective judgements imposes externalities. I am choosing for me, but I’m choosing for you, too. And vice versa. It can be a mess, and it’s very often unnecessary.
If we romanticize the collective, we risk having majorities decide that majorities can decide everything. That danger is not hypothetical: reproductive rights, marriage rights, and freedoms of speech, religion, and the press are all examples where, in just the last twenty years, American majorities have tried to pick one menu item and force everyone to eat from the same dish. But there is nothing about the bundle of technologies that make up government that forces us to use it to decide everything, or even most things, collectively. We need a way of judging when we should decide individually, and when we should choose in groups. That’s where public choice comes in.
The Public Choice Approach
Public choice rests on three interconnected premises: methodological individualism, behavioral symmetry, and politics as exchange. Together, they constitute not an ideology but a method—a way of thinking about how political institutions actually perform rather than how we imagine they might serve us.
Methodological individualism is the insistence that social phenomena can only be understood as the result of choices made by specific, identifiable people. Governments do not decide; people in governments decide. Bureaucracies do not act; bureaucrats act. These aggregated choices and acts may be impossible to untangle, in any given policy choice, but it is important to keep the base distinction in mind.
Methodological individualism is not ethical individualism; there is no claim that individuals are selfish, or that communities do not shape individual preferences—they obviously do. It is simply the requirement that any explanation of political outcomes must eventually cash out in terms of human beings with motives, constraints, and information. As Max Weber put it, social collectivities “must be treated as solely the resultants and context of the particular acts of individual persons.” Treating “the state” or “the market” or “society” as a unified actor with coherent goals is a kind of lazy metaphysics that obscures more than it reveals.
Behavioral symmetry rules out the assumption—dominant in mid-century welfare economics—that market actors are purely selfish, while state officials are purely altruistic. This is the “benevolent despot” model, in which we identify a market failure and hand the problem to a government presumed to be both perfectly motivated and perfectly informed. That model was never empirically defensible. Knut Wicksell, whose work James M. Buchanan credited as a key inspiration, put it plainly more than a century ago: legislators and executives “are not pure organs of the community with no thought other than to promote the common weal.” They are “in the overwhelming majority of cases, precisely as interested in the general welfare as are their constituents, neither more nor less.”
If you tell me markets fail because people are selfish, but that governments succeed because officials are altruistic, you have not done political economy; you have written a morality play, and you are trying to import romance into politics. Stop it.
The third premise, politics as exchange, is the most important, because it defines the reason why we want to tolerate the hell of “other people” in the first place. Politics is not a romantic goal to be pursued for its own sake, but neither is it simply predation. At its best, choosing in groups is an enterprise to obtain mutual benefit. People enter political arrangements for the same reason that they enter commercial ones: because the arrangement makes each of them better off than they could be alone. Without this premise, public choice degenerates into what Buchanan called “a counsel of despair.” We would have nothing to offer except a catalog of the ways political processes disappoint us.
The idea of politics as exchange does not mean that political bargains are always legitimate, or that majorities cannot oppress minorities, or that the outcomes of political competition are always good. It means that the reason for having politics at all—the reason even self-interested actors would choose to submit themselves to a common set of rules enforced by coercive authority—is that the alternative is worse. Covenants without the sword are but words, as Hobbes observed. Buchanan’s insight was that accepting this Hobbesian baseline does not require accepting the Hobbesian conclusion. The sword that enforces agreements can itself be constrained by constitutional rules, and those rules can reconcile coercion with liberty—if, but only if, they reflect genuine consent.
The Two I’s, Applied
This is where the “two I’s” framework becomes indispensable: the first choice is whether to make a decision collectively or individually. Any institutional arrangement—state or market, club or commune—must solve two problems at once: it must gather and process information about what people need and what resources are available, and it must align incentives so that those doing the gathering have reason to do it well.
Markets solve these problems imperfectly, but the degree of imperfection can be hard to see. We’d like to measure the real world against an ideal, the perfectly efficient alternative that would be produced by an omniscient, benevolent dictator. But that hypothetical alternative (1) can’t be identified, because the government is not omniscient, and (2) probably wouldn’t be selected anyway, because the state is not benevolent, but is composed of self-interested individuals. The standard welfare model compares markets as they are to a romantic vision of a state, one that is profoundly misleading, even fraudulent. Romantic politics gets around the “pretty pig problem” by imagining a different situation altogether.
Prices aggregate dispersed information in a form no central authority could replicate. In a democracy, the alternative source of information is voting. In the romantic vision of “epistemic democracy,” voter-supplied information will be more accurate than prices, and electoral incentives—plus bureaucratic insulation from being fired—will outperform the system of profit and loss.
Austrian School economists have focused on the information failure of centralized direction, and it is real and serious. Public choice scholars acknowledge this, but they focus more on incentives. Even if a central planner could somehow obtain all relevant information, she would face a principal-agent problem of staggering proportions. Those charged with implementing the plan are not those who designed it, and their private interests diverge from the public goal in ways that are difficult to monitor and nearly impossible to correct. Political incentives also compress time horizons severely: politicians must produce visible results before the next election. For members of the U.S. House of Representatives, that means caring only about the next two years, beginning on the day after the election.
None of this means state action always fails. It means that the drawbacks of choosing in groups are real, systematic, and predictable. Romantics put their faith—literally, faith—in the claim that “we just need to elect better people!” They don’t recognize that the problem is a product of information and incentives, not one of setting the right intentions.
When I hear a friend make a romantic claim that “the state should be able to make people do _________,” I suggest what I call the “Munger Test.” It’s simple: I take out the word “state” and replace it with “Trump” (or “Harris”; I’m not trying to make a partisan point). Then I ask if they still believe their claim.
Many romantics, of course, simply deny the premise: “We shouldn’t elect bad people! That’s the problem!” Of course, your “bad person” may be my hero. “Politics without romance” requires that we design institutions so that even the wrong people, pursuing their own interests, are channeled in directions that serve the public.
If we use the state as the technology to solve human social problems, the consequence—without romance—is that we may all have to have the same thing for lunch. Even if it is something you don’t like.
