The Old vs. New Institutionalism: A Clear Split
Both schools put institutions at the center of economics, but they disagree about method, human nature, and the role of power. Understanding the split clarifies what each tradition can and cannot explain.
Two Traditions, One Word
The word “institutional” appears in the name of two quite different intellectual traditions in economics, and the shared label causes more confusion than it resolves. The “old” institutionalism, rooted in the work of Thorstein Veblen, John R. Commons, and Clarence Ayres in the first half of the twentieth century, saw economics as inseparable from culture, power, and evolutionary change. The “new” institutional economics, associated with Ronald Coase, Oliver Williamson, and Douglass North from the 1960s onward, used the tools of mainstream economics, particularly rational choice and optimization, to analyze institutions as solutions to transaction cost problems.
Both traditions agree that institutions matter. Both reject the textbook model of frictionless markets with perfectly rational agents as a complete description of economic life. But they disagree on nearly everything else: what institutions are, why they exist, how they change, what role power plays, and what counts as a satisfactory explanation. The disagreement is not merely academic. It shapes how economists think about regulation, corporate governance, development policy, and the relationship between markets and political power.
The Old Institutionalism: Veblen, Commons, Ayres
The old institutional economics never had a unified theoretical framework, which was both its strength and its weakness. What united its practitioners was a conviction that the neoclassical model of rational, utility-maximizing agents operating in competitive markets was a deeply misleading picture of how economies actually work.
Thorstein Veblen was the movement’s most provocative voice. Writing at the turn of the twentieth century, Veblen argued that economic behavior was driven not by rational calculation but by habits, instincts, and culturally shaped dispositions. His concept of “conspicuous consumption,” introduced in The Theory of the Leisure Class (1899), held that much of what people buy is motivated not by the utility of the goods but by the desire to display social status. This was not a minor qualification of consumer theory; it was a fundamental challenge to the idea that demand curves reflect underlying preferences that can be taken as given.
More broadly, Veblen distinguished between “industrial” and “pecuniary” employments, between the activities that produce useful goods and services and the activities that manipulate financial claims and market positions. He saw modern capitalism as increasingly dominated by pecuniary rather than industrial logic, with financiers and absentee owners extracting value from productive enterprise rather than contributing to it. This was not a Marxian class analysis in the strict sense, but it shared Marx’s concern with the relationship between productive activity and the appropriation of its fruits.
Veblen also introduced the distinction between “ceremonial” and “instrumental” behavior that would become central to the Ayres branch of institutional economics. Ceremonial behavior is governed by tradition, status, and vested interests; instrumental behavior is guided by the application of knowledge to practical problems. Economic progress, in this view, results from the growth of instrumental knowledge and its embodiment in technology, while economic stagnation results from the dominance of ceremonial institutions that protect existing power structures at the expense of productive efficiency.
John R. Commons took a different approach. Where Veblen was a cultural critic and evolutionary theorist, Commons was a legal economist and labor reformer. His central concern was with the legal foundations of economic activity: the rules of property, contract, and organization that define what people can and cannot do in markets. Commons argued that the “transaction,” not the commodity, was the basic unit of economic analysis, and that transactions were governed not by the impersonal forces of supply and demand but by the legal rules and organizational structures within which they took place.
Commons distinguished three types of transactions: bargaining transactions (between legal equals in a market), managerial transactions (between superiors and subordinates within an organization), and rationing transactions (between a collective body and its members, as in taxation or regulation). Each type involved different power relationships and different legal rules. Understanding economic behavior required understanding the legal and organizational context in which it occurred, not just the prices and quantities that resulted.
Clarence Ayres, working primarily from the 1940s through the 1960s, synthesized Veblen’s evolutionary approach with a philosophy of technology. Ayres argued that economic progress was fundamentally driven by the advance of technology (understood broadly as the application of tools and knowledge to human problems) and that the main obstacles to progress were the ceremonial institutions that protected existing patterns of privilege and power. His framework was optimistic about technology and skeptical of tradition, and it provided the old institutional economics with something approaching a theory of economic change.
The New Institutional Economics: Coase, Williamson, North
The new institutional economics began, in a sense, with two of the most influential papers in the history of economics. Ronald Coase’s “The Nature of the Firm” (1937) asked why firms exist if markets are efficient, and answered that firms economize on the transaction costs of using the market. His “The Problem of Social Cost” (1960) argued that, in a world without transaction costs, the initial assignment of property rights would not matter for efficiency, because parties would bargain their way to the efficient outcome regardless. The implication, which Coase himself emphasized, was that in the real world, where transaction costs are positive and often large, the assignment of property rights matters enormously, and institutions that reduce transaction costs are central to economic performance.
Oliver Williamson built on Coase’s insights to develop a theory of economic organization based on transaction cost economics. Williamson argued that the key factors determining organizational form were the frequency, uncertainty, and asset specificity of transactions. When transactions involve relationship-specific investments (assets that are much more valuable within a particular relationship than outside it), there is a risk of “hold-up”: one party can exploit the other’s sunk investment to extract a disproportionate share of the surplus. Firms, contracts, and governance structures exist to manage this risk.
Williamson’s framework was powerful and precise. It could explain why some transactions are organized within firms and others through markets, why vertical integration occurs in some industries and not others, why franchise arrangements take the forms they do, and why complex contracts include the specific provisions they do. It was testable, and a large empirical literature confirmed many of its predictions.
Douglass North extended the transaction cost framework from the organization of firms to the organization of entire economies. He argued that the institutional framework of a society, its rules, norms, and enforcement mechanisms, determined the transaction costs of economic activity and therefore the range of economic exchanges that were feasible. Societies with institutions that protected property rights, enforced contracts, and limited predatory behavior by political elites had low transaction costs and could sustain complex, specialized, long-distance exchange. Societies without such institutions were trapped in simple, local, low-productivity economic activity.
What the Labels Hide
The labels “old” and “new” suggest a simple chronological succession, with the new replacing the old. This is misleading. The old institutionalism never disappeared; it continued to develop, particularly in the Association for Evolutionary Economics and the Journal of Economic Issues, and it remains a living tradition with active practitioners. The new institutional economics did not grow out of the old; it grew out of mainstream neoclassical economics, as an extension of price theory and rational choice to questions about institutions that the mainstream had previously ignored.
The two traditions share a concern with institutions, but their intellectual roots, methods, and assumptions are profoundly different. Understanding the differences clarifies what each can contribute and where each falls short.
Methodology. The new institutional economics operates within the mainstream methodological framework: individual rationality (bounded, in North’s later work, but still recognizably rational), equilibrium analysis (institutions as equilibrium solutions to transaction cost problems), and formalization (mathematical models and econometric testing). The old institutional economics is methodologically eclectic, drawing on evolutionary biology, cultural anthropology, pragmatist philosophy, and historical analysis. It tends to be skeptical of mathematical formalization, not because math is inherently wrong but because it often forces assumptions (like rationality and equilibrium) that distort the phenomena being studied.
Human nature. The new institutional economics assumes that individuals are rational, or at least boundedly rational, and that institutions are the result of purposive design or evolutionary selection among purposive agents. The old institutional economics, particularly in its Veblenian form, sees human behavior as driven primarily by habits, instincts, and culturally transmitted dispositions. Rationality is not denied, but it is understood as a product of institutional context rather than as a pre-institutional given. People do not bring fixed preferences to institutions; institutions shape their preferences, beliefs, and cognitive frameworks.
The role of power. This is perhaps the sharpest divide. The new institutional economics tends to explain institutions as efficient solutions to cooperation problems: property rights exist because they reduce the costs of transacting; firms exist because they economize on market transaction costs; governance structures exist because they manage opportunism. Power and distributional conflict enter the analysis, particularly in North’s later work, but they are not the primary explanatory variables.
The old institutional economics places power at the center. Institutions are not primarily efficient solutions; they are the crystallized outcomes of power struggles between groups with different interests. Property rights do not just reduce transaction costs; they determine who gets what. Corporate governance does not just manage opportunism; it determines the distribution of income between workers, managers, and shareholders. Regulation does not just correct market failures; it reflects the balance of political power between regulated industries and the public.
Change and evolution. The new institutional economics tends to model institutional change as a response to changes in relative prices, technology, or the political environment. When the costs and benefits of existing institutions shift, agents have incentives to renegotiate, reform, or replace them. The old institutional economics sees change as an evolutionary process driven by the interaction between technology (which generates new possibilities) and ceremony (which resists change to protect existing interests). Change is not smooth or efficient; it is contested, path-dependent, and often incomplete.
Why the Split Matters for Policy
The practical implications of the split are significant. Consider the question of corporate governance reform. A new institutional economist might analyze the problem in terms of principal-agent relationships and incentive alignment: how can shareholders (principals) design compensation contracts and monitoring mechanisms that induce managers (agents) to maximize firm value? The solutions would emphasize better incentive structures, stronger boards of directors, and market mechanisms like hostile takeovers that discipline underperforming managers.
An old institutional economist would ask different questions. Who defines “firm value”? Why are shareholders’ interests privileged over those of workers, communities, or the environment? How did the current distribution of corporate power arise, and whose interests does it serve? The solutions might emphasize stakeholder governance, labor representation on boards, or regulatory constraints on executive compensation, not because these are more “efficient” in the narrow sense but because they reflect a different understanding of what corporations are for and who should have a say in their governance.
Similarly, consider development policy. A new institutional economist might recommend strengthening property rights, improving contract enforcement, and reducing transaction costs through legal and regulatory reform. These recommendations are not wrong, but they may be incomplete if the old institutionalist concern with power is warranted. If existing institutions serve the interests of a powerful elite, then reform requires not just better rules but a shift in the distribution of political power. And if habits, beliefs, and cultural norms shape economic behavior as much as formal rules do, then institutional reform requires engagement with the cultural and cognitive dimensions of human behavior, not just the incentive structures.
Can They Be Reconciled?
The question of whether the old and new institutionalisms can be reconciled has been debated for decades, with no consensus. Some scholars have argued for a synthesis that combines the new institutional economics’ analytical rigor with the old institutional economics’ attention to power, culture, and historical process. Geoffrey Hodgson, one of the most prominent advocates of this position, has argued that the concept of “habit” can be given a rigorous, evolutionary foundation that is compatible with (but richer than) the bounded rationality of the new institutional economics.
Others are more skeptical. The methodological differences may be too deep to bridge. The new institutional economics derives its analytical power from the assumption of rational choice within constraints; the old institutional economics derives its critical power from questioning the rationality assumption itself. If you give up rationality, you lose the ability to generate precise, testable predictions. If you keep rationality, you may be assuming away the very phenomena (habits, power, cultural inertia) that the old institutionalism considers most important.
A pragmatic middle ground might be to use different approaches for different questions. When the issue is the design of specific governance structures, contract provisions, or regulatory mechanisms, the transaction cost framework of the new institutional economics is often the right tool. When the issue is the political economy of institutional change, the distributional consequences of institutional arrangements, or the cultural foundations of economic behavior, the old institutional economics has insights that the new institutional economics tends to miss.
What both traditions agree on, and what distinguishes them collectively from mainstream neoclassical economics, is that institutions are not a sideshow. They are not the stage on which the “real” economic drama of supply, demand, and equilibrium plays out. They are the script, the casting, and the direction, as well as the stage. Understanding the economy without understanding its institutions is like understanding a game without understanding its rules, which is to say, not understanding it at all.
The Stakes
The debate between old and new institutionalism is not just a question of intellectual history or academic classification. It shapes how economists and policymakers think about some of the most important questions of our time. Is corporate consolidation a problem of market power or of political power? Is persistent poverty in developing countries a result of bad incentives or of bad institutions shaped by colonial legacies and power imbalances? Is the appropriate response to technological disruption better incentive design or a fundamental rethinking of the institutional frameworks within which technology is developed, deployed, and governed?
These questions do not have simple answers, and the tension between the two institutionalisms is, in many ways, the tension within economics itself between the aspiration to scientific precision and the recognition that human societies are messy, contested, and historically contingent. Acknowledging this tension, rather than resolving it prematurely, may be the most honest and productive response.