Liberalism, Property, and the Means of Production


David Singh Grewal is Professor of Law at UC Berkeley School of Law and co-founder of LPE Blog. 

Jedediah Purdy (@JedediahSPurdy) is the William S. Beinecke Professor of Law at Columbia Law School.


David Singh Grewal is Professor of Law at UC Berkeley School of Law and co-founder of LPE Blog. 

Jedediah Purdy (@JedediahSPurdy) is the William S. Beinecke Professor of Law at Columbia Law School.

This is part of our symposium on Hanoch Dagan’s book, A Liberal Theory of Property. For a concise version of Dagan’s argument, see this restatement. Image credit: Sam Abell, National Geographic.

Hanoch Dagan’s A Liberal Theory of Property presents a thorough, thoughtful, and to our minds, convincing take on what a specifically liberal justification of property entails. It deserves to be in the modern canon of property theory. Dagan argues that a liberal justification of property requires structuring entitlements so as to enhance individual autonomy and he lays out what autonomy means by elaborating what he calls “structural pluralism” and “relational justice.” (One of us advanced a similar normative theory of property law in a 2010 book and earlier articles published in 2005 and 2006, although without the synoptic richness that Dagan has developed over several decades of work.)

In this post, we specifically consider liberal defenses of private property in the “means of production”. This focus allows us to put the liberal defense of private property into dialogue with Marxism, with which it shares a broad humanistic heritage and many particular normative framings. Our focus also connects liberal property theory with a variety of later critiques of private property in certain productive resources, including those of the progressive and realist lawyers who generated American doctrines concerning “public utility” and other modes of resource governance that are neither strictly “private property” nor strictly matters of state control.


In his chapter on “Just Markets,” Dagan studies the role of property in modern market settings and explores the conditions of its appropriate use. Dagan rightly recognizes that “liberal property and markets are so deeply connected that a liberal theory of property cannot ignore the market. A liberal theory of property must explain how property can remain loyal to its liberal commitments in the context of large-scale economies heavily reliant on the operation of markets” (p. 179, emphasis his). For us, the important word in this passage is not so much “liberal” as “scale,” which we will want to use in a more precise way than Dagan perhaps intended.

To first sketch Dagan’s central argument: he spends much of the chapter focused on just liberal markets of a particular kind – labor markets – because of the particular threat to individual autonomy that cooperation in a labor market represents. Individual autonomy is Dagan’s lodestar, and the question is how market exchanges enhance or degrade it (with possible regulatory objectives thereby clarified). This is not a libertarian argument but something more developmental and “Millian”: “Understanding contract and property as empowering devices places self-determination, not negative liberty, at the market’s moral core” (p. 188). The main benefits of markets for self-determination consist in their tendency to promote “mobility” and to “expand options.” A liberal defense of property thus requires seeing how markets can do that well, and without sacrificing autonomy (relational justice) in other respects.

Crucially, this requires seeing markets as products of law, and as constantly subject to background conditions that the market itself cannot specify. Two important passages in which Dagan makes these points are worth citing at length, both for their intrinsic intellectual contribution (of particular interest, again, to readers of this blog) and because both point to a possible problem with private ownership of the means of production.

First, on markets as constructs of law (pp. 223-224):

Property, contracts, and money play key roles in the constitution of contemporary markets. Without implying that markets cannot exist without law – law-like social conventions can, and have, facilitated markets – this simple observation does mean that, where law exists, its prescriptions provide at least some of the foundations of the market. Indeed, appreciating the heavy reliance of markets on law (especially at present) denotes that, as property theory is a species of legal theory (as I claim in Chapter 2), so is a theory of the market. It also explains why the claim that, absent market failures, law should refrain from intervening in the market is conceptually confused.

Second, on what kind of background regulation markets require, and why (pp. 188-189):

Understanding the market as a means for self-determination also implies that markets cannot function well as a stand-alone autonomy-enhancing device. A liberal polity enlisting markets in the service of self-determination should thus be committed to the creation of a background regime that secures the social and economic conditions (or capabilities) enabling individuals to be self-determining. This prescription … is particularly urgent due to some (unfortunately inherent) harsh consequences of the market’s operations, beyond the externalities generated by specific market interactions…. Because these effects often cannot be properly addressed within the law of the market, they need to be remedied by the background regime, which thus becomes essential to the market’s legitimacy or, at least, to its normative desirability

We want to draw from these important passages several implications for the right of private property in the means of production. As Dagan notes in his reference to antitrust law, the issue of who owns such  assets, and which powers owners may exercise over those resources—and, hence, over other persons working with them—is an important feature of how a property regime realizes the value of reciprocal autonomy. He also recognizes (following Joe Singer), that it might be necessary for a liberty property regime to (re)define owners’ rights in ways that recognize the claims of workers and communities. He argues (p. 199):

Nor does ownership of the means of production, as Joseph Singer correctly argued, need to imply that owners’ decision to close a plant leaves workers remediless. I cannot consider here the specific rights – from notice and information to supervised negotiation and first refusal – that may ameliorate the possible devastating effects of plant closures and similar dramatic decisions on the autonomy and welfare of workers, without unduly burdening owners. But in a liberal system of property, where the private authority of owners of means of production is carefully delineated and workers’ claims to relational justice are properly acknowledged, these rights cannot be easily dismissed.

In other words, we might draw an analogy between private ownership of the means of production—a situation in which one individual’s law-enabled actions can have devastating effects on others—and the more familiar ways that property law manages the interdependent interests of neighbors (with separate property rights) or partners in a common tenancy (with shared ones). The factory example provides an instance of Dagan’s general view that, “Structuring markets so as to serve autonomy also requires universal participation (or opportunity to participate) and, more broadly, compliance with the prescription of reciprocal respect for self-determination, meaning that interactions in and around the market must be governed by the maxim of relational justice.” (pp. 192-93).

We are very sympathetic to this ideal. We are attracted to Dagan’s liberal-realist view, in which a “market” is not a generic institution, with a timeless and context-free essence, but rather a congeries of rights, powers, etc.—a thoroughly legal creation, ultimately traceable to a political judgment that itself should be subject to considerations of legitimacy. The ideal, then, is to “embed” a market—to borrow a term from twentieth-century political economy—in a web of legally specified relationships, expectations, and values that define and discipline the power exercised within property relations and link that power to a vision of a tolerably legitimate social order, such as Dagan’s egalitarian and liberal pluralism.


Our question follows straightforwardly: whether, and in what ways, markets based on private property—particularly “capitalist” markets predicated on private ownership of the means of production—exhibit intrinsic tendencies not susceptible to such stable regulatory discipline or “embedding.” If there are such tendencies, then an otherwise liberal regime of property is constantly at risk of becoming illiberal in ways that Dagan (and we) would wish to avoid: marked, for instance, by concentrated market power, oligarchic concentrations of wealth, and the domination of owners over workers and everyone else.

We want to consider one key dynamic: the phenomenon of “increasing returns to scale” or “economies of scale.” It will be recalled from Econ 101 that a “proportionate” transformation of “inputs” into “outputs” exhibits is defined as “constant returns to scale;” that a greater-than-proportionate transformation of “inputs” into “outputs” is defined as “increasing returns;” and a “less-than-proportionate transformation of “inputs” into “outputs” as “decreasing returns.” In Econ 101, it may have also been noted rather sternly that “normal” markets exhibit constant or decreasing returns to scale, perhaps with a nod to the fact that products based on “technology” sometimes exhibit increasing returns. The question of scale can be addressed at either the level of the whole “economy”—which many classical political economists thought, in its “progressive” mode, showed increasing returns (such that an increasing “social surplus” was generated above and beyond the requirements for reproduction or replacement of each factor of production) or, more commonly, at the level of particular sectors or firms.

By contrast with its neoclassical successor, classical political economy began with an intense interest in the question of increasing returns to scale at both the “economy-wide” and the firm level. Adam Smith’s description of the “division of labor” in a pin factory is one of the two most famous passages in his Wealth of Nations. Smith’s example became canonical in discussions of production across the nineteenth century. The possibility of more efficient production at an ever-larger scale—what we now call “increasing returns to scale” (or “economies of scale”)—was seen as the result of the division of labor, usually assisted by (sometimes predicated upon) the use of new machinery in production.

The conceptual difficulty with increasing returns to scale is, as Nassau Senior already noted in 1836, that it precludes economic calculation: “The price of those commodities which are comprehended in the first class [i.e., competitive markets] appears to be subject to laws capable of accurate investigation” (p. 174), whereas “[t]he prices of the commodities comprised in the second, third, and fourth classes [i.e., those subject to market power] are but little governed by any general rules” (p. 176). If the efficient level of production in such a firm depends on external factors (e.g., the size of the whole market) rather than internal ones (costs of production), there are many possible equilibria, none of them determinate.

It is worth noting these same considerations in Marx’s critique of capitalism. Monopoly had long been regarded as a legal prerogative, or, if not a matter of formal privilege, at least a problem of collusion. But the “monopoly” to which Marx thought capitalism tended was neither a legal privilege nor a product of conspiracy against competitive conditions, but the economic outcome of free contract—not a violation of the free market, in other words, but its fulfillment in conditions of returns to scale. Marx on monopoly capitalism thus differs crucially from Smith on “monopolies” (for which his exemplars were the vast overseas trading companies given royal charter or collusions of merchants). Following through Marx’s historical arguments—occupying the middle parts of the first volume of Capital (detailed empirical observations on nineteenth-century factory conditions and output that many contemporary readers simply skip over)—we can see that his notion of the “general law of capitalist accumulation” (Vol. 1, Ch. 25) is predicated on a geometric increase in the ratio of “constant” to “variable” capital (meaning fixed instruments of production to labor), with productivity increases that accrue (as profit) to the capitalist. The “immiseration” of the worker is the flip side of this private ownership of such vastly productive instruments, as political economists before Marx had noted: the worker, “set free” from production, nevertheless lives in a world that is ever wealthier and, accordingly, has ever less need of all they have left to sell.

Political economists were quite attuned to the problem that scale economies presented to a conception of freedom and equality based on a system of property interests in the means of production. Already, in his Principles of Political Economy, John Stuart Mill offered the basic analysis that American legal reformers of the Progressive Era would adopt several generations later, suggesting that for reasons of “public utility” the government would need to either nationalize or regulate any increasing returns production process, which would otherwise tend toward natural monopoly (Principles I. IX. 3). And Marx’s argument concerning ownership of the means of production—private monopolies under capitalism to be managed collectively under communism—tracks the same essential problem, already well understood by middle of the nineteenth century. That property theory would need to contend with scale dynamics was thus obvious to reformers and revolutionaries alike, for the most important kind of property—the productive assets on which commercial “progress” was predicated—would necessarily, and to an increasing extent, become the institution through which people were subjected to arbitrary will (a recurring liberal concern) rather than shielded from it.

Yet returns to scale have disappeared from liberal accounts of property, including Dagan’s (or, before him, Rawls’s). Why? Two shifts—one theoretical and one empirical—mark the century that separates Rawls from Mill, a shift that has not since been repudiated.

Theoretically, the neoclassical economics of the late 19th and early 20th centuries ignored and even formally denied the possibility of increasing returns to scale in production. The reason goes back to Senior’s diagnosis of the essential incalculability of economic equilibria influenced by monopoly power in the market.

The neoclassical shift from production-anchored theories to exchange-anchored theories emphasized formal modeling of competitive markets producing optimal equilibria. Following the lead of Francis Edgeworth (and Vilfredo Pareto), articulating determinate equilibrium conditions was the project for most of the first half of the twentieth century, culminating in the general equilibrium theory of Kenneth Arrow and Gerard Debreu, which precluded increasing returns to scale to make the models work. The “return to increasing returns”—as the title of one book put it (Buchanan and Yoon, 1994)—began only in the late 1970s, driven by interest in technological change and international trade (for, by then, the US was losing its industrial edge to erstwhile rivals it had built up in the post-war). The return to increasing returns did not bring back with it, however, a sense that production as such in commercial society was an increasing returns process (though why not remains unclear) and instead focused on particular markets, as if they were the exception to the rule.

The empirical contribution to the sidelining of increasing returns was the post-war assumption that capitalist growth could be both high and widely shared—pace Marx (and even Mill, who assumed that the division of labor in industry would necessarily bring about ever-concentrated corporate control). This more optimistic conclusion was the result of careful studies of income accounts by Simon Kuznets of the experience of the post-war decades.

 As we have both noted elsewhere, the famous “law of capitalism” into which Piketty distilled this finding (r>g, or the rate of return to capital exceeds that of economic growth generally) is simply an untheorized generalization from data, with no account of how such returns accrue to forms of capital as disparate as enslaved persons in the antebellum U.S. South and the bonds (in a different sense) of the rentier class in Britain’s nineteenth-century commercial empire. An account of increasing returns to scale as a feature of a capitalist market  would be a start to a more elaborated “law of capitalism” than Piketty provides. Such an account might take essential cues from industry-specific studies, such as Lina Khan’s analysis of Amazon’s market-dominating platform-based commercial strategy, and the broader move to theorize the place of power in “informational capitalism” that Amy Kapczynski has recently marked out. This much is clear: concentrations of capital tend to grow, out of proportion to all other economic growth, and today, as was true a century ago, both new and established industries display dramatic concentration of wealth, market share, and, increasingly, market-shaping power.


What sort of legal and political strategies can check the tendency to concentration in the private ownership of the means of production, let alone bend it toward a liberal and egalitarian ideal of autonomy? One attractive suite of answers is, of course, the kind of thing Dagan indicates—regulation of the concentration, and the power, of capital ownership, through antitrust, corporate law, labor and employment law, and other means. Let us pause, however, to consider the countervailing forces implicit in the dynamic we have described. To the extent that the returns-to-scale dynamics we have posited do in fact hold (as imperfect empirics suggest they do), the marketplace will tend to become (A) itself a site of domination, populated by large-scale enterprises and disempowered or dispossessed workers—a reality that is not far to find in the current pandemic economy, in which Amazon and the stock markets have flourished while unemployment and precarity have soared among line workers who increasingly work for a few huge enterprises; (B) a site of market power that may reinforce scale dynamics—e.g., as Khan has shown, enterprises such as Amazon and Facebook can lock in their scale advantages by shaping the playing field against competitors and new entrants; and (C) a source of political feedback and capture that supports oligarchy and harnesses the state to the regulatory reinforcement of scale.

This last tendency may be reinforced, ironically enough, by political institutions that have come to be associated with liberal pluralism. We have in mind, for instance, (1) the anti-democratic tendencies of the US constitution, which impedes legislation and the formation of national governing majorities, thus tending to leave the play of market dynamics as the “default setting” of shared life; (2) judicial review, which in the U.S. has produced a notorious series of cases protecting political spending as speech (thus promoting conversion of economic to political power), and in the European Union has upheld the free movement of capital across borders, favoring the logic of markets over that of political regulation; (3) cost-benefit analysis, often treated as an essential form of expert rationality, but most frequently deployed in support of market-making or market-mimicking policies that tend to reinforce returns-to-scale dynamics; and (4) perhaps more diffusely, the long-running (but recently faded) ideological association between liberalized international trade and the growth of liberal pluralism, which has bolstered institutions that facilitate global capital mobility, from fast-track provisions for trade agreements to procedural and substantive protections for investors in those same agreements. The same institutions and ideas that are often associated with liberal pluralism also promote and intensify specifically capitalist economies, and serve to ensconce those politically, in ways that may help both to produce and to intensify the illiberal effects of increasing returns to scale.

This prospect raises a series of practical questions. When we talk about “regulation” of capitalist markets, how strong a set of policies and institutions must we imagine, to take the problem seriously on its own terms? Do certain key platforms (e.g., Amazon) need to be treated as public utilities rather than private enterprises, or as a regulatory hybrid of a kind once central to progressive visions of law, as Sabeel Rahman and others have reminded us in recent work? What kind of political control over, and agenda for, the money supply itself is necessary to a stably egalitarian political economy? Will the autonomy of the owners of production’s means stay reconciled with those of employees and consumers without strong labor unions and other institutions of collective action, embedded in the legal system itself, as glimpsed in the First New Deal, early visions of the National Labor Relations Act, and radical unionism more generally, going back to the idea of the “cooperative commonwealth”? Must owners’ autonomy be curtailed to the degree of reestablishing controls on capital movement and confiscatory progressive taxes?

The more we are inclined to say “yes” to any of these, the closer we move from what Rawls called a “property-holding democracy” to what he called “liberal socialism”, which would involve the abolishment of private property. And the more we move toward liberal socialism, the more tension there is between Dagan’s rich account of the normative requirements of a liberal property regime and the lived reality of liberal capitalism.

These questions also raise the artificiality of trying to think about the economy without thinking at the same time about the state, and in equally basic terms. It may be that what is necessary to pursue an economy of egalitarian, reciprocal autonomy is also a certain kind of democratic state, one characterized by stronger means of majority-formation and legislative action, more restricted judicial review, and fewer concessions to regional and global institutions of neoliberal economic integration. To be a liberal egalitarian in a capitalist economy, it may also be necessary to be a political democrat of a strong kind. Where liberals put their faith in certain kinds of anti-political technocracy and juristocracy—arguably, even, in “property”—do they risk undermining the political preconditions of the economy—the political economy—they say they want?

The institutional depth and moral ambition of Hanoch Dagan’s argument has led to these questions, which seem to us to be at the heart of the pursuit of broadly liberal aims under today’s ever-intensifying market conditions. We thank him for the chance to consider them.

Related Content