Skip to main content Accessibility help


  • Access

  • Economic Theory and the Roman Monetary Economy
  • Online publication date: February 2020
  • pp 1-19


      • Send chapter to Kindle

        To send this chapter to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle.

        Note you can select to send to either the or variations. ‘’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

        Find out more about the Kindle Personal Document Service.

        Available formats

        Send chapter to Dropbox

        To send content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about sending content to Dropbox.

        Available formats

        Send chapter to Google Drive

        To send content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about sending content to Google Drive.

        Available formats
Export citation


All historical applications of formal economic models require justification – not merely within their own closed system of logic, but in a wider historiographical context which includes serious and thoughtful substantivist critiques of the formalist enterprise more generally and especially of applied economic theory. Even if new institutional economics is not the solution, are there other ways Roman economic historians might use economic theories to better understand the economic choices made by the inhabitants of the ancient world as well as the embedding contexts which channeled such choices? History and economics, despite fundamental differences embedded in each discipline, can meaningfully and symbiotically intersect. Economics offers Roman historians valuable and helpful organizing concepts, so long as these concepts are used within an agenda of historical understanding.

Chapter 1 On Writing Roman Economic History

… it is our critical intelligence which is prior to the sources.

Keith HopkinsFootnote 1

History-writing is a subjective art. The historian is an agent – an individual who selects, arranges, interprets and emplots a series of ‘facts’ within a narrative determined, whether consciously or subconsciously, by ideology.Footnote 2 There is a fundamental difference between the past – events which empirically occurred in physical, temporal reality – and what historical theorist Hayden White calls “the historical past” – “a construction and only a highly selective version of the past.”Footnote 3 Even the historical ‘facts’ themselves – the building blocks of historiography – are inescapably subjective, even intangible. Neville Morley pithily observes “a fact is a verbal statement, an idea, with no empirical existence outside people’s minds.”Footnote 4 Although not all historians are willing to embrace such a deconstructionist or otherwise broadly postmodern dissonance between ‘history’ and ‘the past’, there is broad agreement in the mainstream of the discipline that history-writing is a process of argumentation, interpretation and weighing plausibilities. History-writing, therefore, cannot be called ‘objective’.

Not so with many of the theories of modern economics – a result of many developments, but especially the early twentieth century’s distinction between ‘value-free’ (wertfrei) and ‘normative’ economics as well as the widespread adoption of Chicago School economist Milton Friedman’s concept of “positive economics” in the 1950s.Footnote 5 Now central within the overarching methodology of modern economics is the idea that economic propositions are ‘testable’ in the same way as scientific statements about the natural world. Economic propositions – accompanied scientific-sounding terminology – hold the status of hypotheses which can and must be either ‘confirmed’ or ‘falsified’ against observations.Footnote 6 Friedman won a Nobel Prize, in part because his use of mathematical models, combined with simplifying assumptions about the efficiency of markets and the economizing and utility-maximizing rationality of all human beings, reified economics’ abstract logic into measurable and objective ‘science’. Prior to these developments, economics in fact had much more in common with sociology; hence, economic models were often employed with the purpose of understanding and explaining human behavior, whether in the present or the past. Now, however, economics is a collection of scientific ‘laws’ which have been verified by repeated testing – laws which offer the tantalizing promise of predicting human behavior, whether in the future, present or past.Footnote 7

How should this ‘objective’ economics accommodate the ‘subjective’ art of history-writing? Perhaps anticipating this conundrum, Friedman himself offered a solution which is still widely practiced: economic historians are to be predictors of the past:

the ‘predictions’ by which the validity of a hypothesis is tested need not be about phenomena that have not yet occurred, that is, need not be forecasts of future events; they may be about phenomena that have occurred but observations on which have not yet been made or are not known to the person making the prediction. For example, a hypothesis may imply that such and such must have happened in 1906, given some other known circumstances. If a search of the records reveals that such and such did happen, the prediction is confirmed; if it reveals that such and such did not happen, the prediction is contradicted.Footnote 8

Something like the approach Friedman advocates is at the core of what many call ‘new economic history’; sometimes it is called ‘cliometrics’, most often by those who oppose it.Footnote 9 Roman historians became late adopters of the new economic history, but it is now safe to say that many who study the Roman economy do so after the fashion of Friedman’s ‘predicting the past’ approach – testing models against evidence in one form or another, whether or not formal economic theory is explicitly involved. There are a few nuances to such an approach. The historian, for example, must assume that the model deployed is both internally consistent and historically applicable. Also, sufficient evidence must be available for testing the model. If the model is ‘confirmed’ by the evidence, however, then the historian can more plausibly state what ‘actually happened’ without necessarily having empirical evidence for what ‘actually happened’ – a prospect as intriguing as it is unsettling.

Divisions and Debates

The use of the new economic history by Roman historians has not met with universal approval. Current disagreements over methodology, however, are merely the latest skirmishes in one of the most enduring and inimical disputes in all of scholastic history. As what would become neoclassical economics was emerging in the late nineteenth century, a clear division formed in the writings of several prominent German intellectuals. The economist Karl Bücher, building upon the foundations of nineteenth-century economist Johann Karl Rodbertus, argued that ancient economies were culturally and conceptually bound by the oikos – the autarkic household – and, therefore, were qualitatively different than the modern, exchange-dominated economy of the nation-state (Volkswirtschaft).Footnote 10 It was the ancient historian Edward Meyer – an ardent believer in the comparability of classical antiquity and modernity – who first labeled Bücher and others “primitivists.”Footnote 11 Meyer’s views were bolstered with the publication of Michael Rostovtzeff’s Social and Economic History of the Roman Empire (1926) – a work many now see as the height of what would become primitivism’s antonym: “modernism.”Footnote 12 For Rostovtzeff, like Meyer before him, the differences between ancient and modern economies were merely quantitative.Footnote 13 “Ancient capitalism,” as Rostovtzeff named it, was different from modern capitalism only in terms of scale, techniques and technologies such as “factory mass production with the use of complicated machinery.”Footnote 14 Rostovtzeff’s work roundly discredited primitivism. The modernist perspective, in turn, dominated the next several decades of scholarship on the ancient economy.

It was not until the 1950s and 60s that the primitivist remnant found an unlikely ally in anthropologist Karl Polanyi, whose work emerged in ideological, theoretical and analytical opposition to the modernizing tendencies in his own field.Footnote 15 In The Great Transformation, Polanyi argued that economic activity prior to the Industrial Revolution was “embedded” in society – that is, that the economizing action of individuals was itself submerged within the political, religious and cultural aspects of society. The “great transformation” which took over the agendas of many industrialized societies in the nineteenth century, however, was a complete subversion of the historical position of the economy. The autonomous, self-governing market that efficiently regulates all human activity was not merely a theoretical abstraction, but a utopia to be adopted. Attempts to bring about such a system made society “an adjunct to the market” – a result which Polanyi saw as not only undesirable but unsustainable.Footnote 16

Polanyi’s skepticism of capitalism’s excesses was not mere polemic. A central and lasting tenet of his interpretive framework was the observation that human societies have provided for their material well-being in three primary modes of economic interaction, two of which were non-capitalistic: redistribution (typically by political structures), reciprocity (in tribal, familial or gift-giving relationships) and exchange (by market institutions).Footnote 17 According to his analysis, the first two modes were dominant in the ancient Mediterranean world.Footnote 18 Despite the grand scale of economic activity in the Roman world, “it formed no exception to the rule that up to the end of the Middle Ages, markets played no important part in the economic system.”Footnote 19 The only responsible means of studying pre-industrial societies, therefore, was through “substantive” approaches which accounted for the material realities of embeddedness.

By the middle of the 1960s, Polanyi’s main theses were applied to the Roman economy by Arnold Hugh Martin Jones and then in the 1970s with Michael Crawford’s Money and Exchange in the Roman World. Jones’ and Crawford’s work in turn prepared the scholarly landscape for the emergence of Moses Finley’s foundational monograph The Ancient Economy, a book based upon his Sather Lectures given a few years prior at the University of California, Berkeley. Finley produced an account which crafted, mostly through cleverly selected anecdotes, an image of antiquity in which market exchange was important but ultimately constrained by overriding cultural and social frameworks.Footnote 20 Homo economicus – the idealized actor of neoclassical economics, ‘perfectly rational’ and self-interested in his pursuit of maximum profit, whether psychic or material – was not to be found anywhere in the classical world.Footnote 21 The actors in the ancient economy, according to Finley, played by their own unique and embedded set of rules, constrained by the pursuit of status and other non-economic goals and incentives.Footnote 22 Elements superficially shared by both antiquity and modernity – trade networks, monetization, prices and markets – did not self-evidently imply the historical ubiquity of a market system.Footnote 23 “Trade,” for example, Finley revised as “redistribution” or “administered trade.”Footnote 24 Finley’s rich conceptual framework – illustrated through carefully selected references from primary sources – showed the folly of the previous generations of modernizers who had uncritically classified and described the economic structures of antiquity within modern capitalistic concepts and constructs.

The embedded economy framework inspired a new generation of ancient economic historians. Almost concurrent with economists’ embrace of Friedman’s positive economics – a methodology which saw history as the testing ground for deductive economic hypotheses – a group of ancient historians, led by Finley, was showing that the inhabitants of the ancient world could not easily be forced into the mentality of modern capitalism. Division and conflict among ancient historians inevitably followed.Footnote 25 Keith Hopkins described Finley’s framework as something like an inoculation which protected the classical world from analysis by modern economic models:

[Finley and his followers] believe that it is impossible or at least unprofitable to use modern economic concepts in order to analyse a pre-industrial embedded economy. For them, the ancient economy was a cultural system, which was dominated by nonrational considerations of status and ritual and so was immune to cold rational analysis or reconstruction.Footnote 26

Finley immunized Greece and Rome using a distinct strain of Polanyi’s substantivism. As a side effect, scholarly positions took on distinctly ideological meanings. Many ancient historians found themselves on either side of a divide between substantivism and “formalism” – the use of formal theories and models in the study of ancient economies.Footnote 27

In this environment, Finley’s version of primitivism served as the rally flag for adherents or, conversely, as the main target of critics. Finley, however, was not nearly as ideological as Polanyi, but it is true that Finley’s politics were shaped during his time in New York by members of the exiled Frankfurt School of neo-Marxists.Footnote 28 Indeed, the idea of antiquity’s “fundamental break” with modernity was crucial to Karl Marx’s materialist historiography.Footnote 29 Finley drew upon this aspect of Marxism (through the lens of Weber’s revisions) to address a question of transcendent ideological significance: is capitalism inevitably conjoined with human society? The primitivism of The Ancient Economy, more than any other previous iteration of the idea, painted a vivid picture of a thriving society in the midst of a world which operated under different rules than the modern capitalistic order.Footnote 30 The implications were almost eschatological in scope: “we might now look to Rome to see how else things might have been – for example, it offers a reminder that capitalism is not the only way of organising a society – and perhaps also how else things might be.”Footnote 31

Finley’s ideas were compelling, convincing and powerful. Although primitivists – whether or not they self-identified as such – never fully dominated the field, their contributions to the discipline easily overcame their smaller numbers.Footnote 32 The counter-reformation was led by Hopkins. In the Journal of Roman Studies, Hopkins constructed a ‘taxes and trade’ model of the Roman economy, which showed how taxation stimulated and directed trade in the Roman world. The model was what Hopkins called a “wigwam argument” – an amalgamative approach which collected scattered pieces of empirical evidence, weak generalizations and theoretical models into an argument greater than the sum of its parts; each element was like the poles of a wigwam – structurally sound and circumscribing of the truth.Footnote 33 Some of Hopkins’ formulas were patently neoclassical, such as the one he employed to calculate the gross domestic product (GDP) – an estimate of the monetary value of all final-order or “consumer” goods – in a model year – within the territory of the Roman Empire. The same is true of Hopkins’ use of the equation of exchange (sometimes called the ‘Fisher equation’), which assumes an unbreakable mathematical connection between money supply, the speed at which money circulates, the total quantity of goods in the economy and the prices at which those goods are successfully exchanged.Footnote 34 The mathematics of the equation of exchange are predicated upon a certain verbal logic – that if the economic output of the economy stays the same while the amount of money increases, then prices must proportionally rise. Hopkins used hypothetical numbers in his equation to illustrate relationships rather than make predictions, yet he still felt the need to assure his assumedly skeptical readers of the purely speculative nature of the exercise. While Hopkins used neoclassical methods, his approach was not quite the same as the pure positive economics of Friedman (and some of the ancient economic historians who have since followed Hopkins) in which models are tested by data and either confirmed or falsified. Hopkins’ model needed merely to be plausible enough to compare with the fragmentary data; his was ultimately a qualitative if not ideal-typical enterprise. By deploying theory heuristically, Hopkins avoided the circularity of positivist economic analysis, where it is often unclear whether theory is testing history or if history is testing theory. Instead, ‘taxes and trade’ provided what might be thought of as ‘bounding’ knowledge about the structure and scale of economic activity in the Roman world.

Both Hopkins’ model and his imaginative but controversial methodology provoked the ire of primitivists. He was criticized most famously and enduringly in the work of Richard Duncan-Jones throughout the 1980s and 90s.Footnote 35 Duncan-Jones overwhelmed his readers with tables of coin finds and carefully dissected hoard analyses – a strategy which directly attacked the empirical pole of the ‘taxes and trade’ wigwam. Hopkins, however, like Finley, had his own methodological inoculation: deductivism.

The logical connections [of Hopkin’s argument] cannot be proved or disproved by examples, or by counter-examples. The only proper disproof, I claim, is by counter-argument … some critics, true to the positivistic traditions of history, have simply taken for granted that showing errors of fact would be sufficient disproof of my propositions. I do not think that disproof is so simple.Footnote 36

Hopkin’s saw Duncan-Jones’ coin hoards as counter-examples, not counter-arguments; they did not damage the logical core of Hopkins’ deductive model and, hence, were ultimately impotent in fully discrediting it according to Hopkins.Footnote 37 Over the years, Hopkins became worn down by the constant need to defend both his ‘taxes and trade’ model, as well as the deductive approach behind it, from mischaracterizations and misunderstandings. Near the end of his life, Hopkins lamented that induction was “the dominant orthodoxy” while his approach was a “heresy.”Footnote 38 Hopkins was a rhetorical savant, and the “protestant atheist’s” use of religious language may reveal his evaluation of the nature of many of the arguments made against him.Footnote 39 Alternatively, Hopkins may have simply been expressing humility, or perhaps reflection, or even hyperbole.Footnote 40

Nevertheless, Hopkins’ claims about the epistemological structure of his model demands a level of methodological reflection and debate with which few ancient historians have engaged. Instead, there persists a certain amount of methodological schizophrenia in the field of Roman economic history: things have changed, but they have also stayed the same. Confusion persists, for example, over the differing epistemological statuses of theories, hypotheses and models as well as the role of empirical evidence within these categories of argumentation. The main historiographical dichotomies of ancient economic history – primitivism/modernism and substantivism/formalism – are muddled almost beyond usefulness. Primitivism and substantivism, for instance, are often used interchangeably.Footnote 41 Furthermore, ancient historians practice the age-old scholarly pastime of distinguishing ‘new’ work by segregating the ‘old’ work into extreme characterizations and subsequently arguing against these stereotypes.Footnote 42 Approaches which were once nuanced and complex are now straw men which show the ‘extremism’ which new research will supposedly avoid through a ‘third way’, compromise or moderation.Footnote 43 The primitivist/modernist and substantivist/formalist binaries have been unfortunately reduced into caricatures, as over time short-term efforts to create ‘third ways’ have cumulated in diluting these otherwise helpful historiographic categories.

In addition to the need for clarity in the field of Roman economic history is a need for reflection. Social-scientific theories, models and organizing concepts should be evaluated for “their usefulness, the extent to which they can offer a persuasive view of the evidence, rather than being automatically ruled invalid and unacceptable.”Footnote 44 Theoretical constructs should channel historians’ inquiries toward questions the empirical evidence does not ask of itself.Footnote 45 Pure induction is a myth, not a method. Generalization may be problematic, yet it is impossible to write history without it. One of the more eloquent commentaries upon methodology and the ancient economy is found in the editors’ introduction to Ancient Economies, Modern Methodologies:

the historian cannot simply withdraw from the realm of ideas to find safe and firm ground in the world of sources. We perceive the world through ideas and concepts. They determine what questions we ask of ancient material and how we organize the answers.Footnote 46

Inductivism may offer the promise of substance, even objectivity, but there is no self-evident framework for organizing or examining the facts.Footnote 47 Historians can and must debate the degree to which generalizations are useful; yet they cannot pretend to be pure inductivists, as it is impossible to write history without prior assumptions.Footnote 48 Hopkins rightly notes that our critical intelligence is prior to the sources.Footnote 49 It seems imprudent, therefore, to dismiss economic theory on ideological or political grounds alone.Footnote 50

How then should Roman economic historians account for the otherness of the various cultural contexts for economic activity in the Roman world? One approach is to subvert it to supposedly universalizing concepts. The Roman Market Economy, the title of a recent book by one of the most influential economic historians of the late twentieth century, for example, seems to beg the question.Footnote 51 The book’s provocative title is no doubt meant to summarize the author’s characterization of production and distribution in the Roman economy as driven by market forces in both qualitative and quantitative terms.Footnote 52 The Roman ‘market’ would therefore tend toward equilibrium and prices would reflect supply and demand.Footnote 53 Characterizations such as these, in which the economic forces of markets, integration and money are dominant, justify the use of neoclassical and new institutional economics: “ancient economies clearly differed from modern ones, but the principles of economics still hold true, and economics can bring clarity to the analysis of how resources were allocated in the ancient world.”Footnote 54 Such characterizations, however, are odious to scholars sympathetic to substantivist approaches derived from economic anthropology, cultural history, classical philology and economic sociology.Footnote 55

Roman economic historians are rapidly becoming more literate in social-scientific approaches – a change which brings new opportunities but also some challenges. Paleoclimatology and other ‘hard’ science is also becoming increasingly relevant if not essential for understanding the Roman world.Footnote 56 New methodologies now confront Roman historians on a regular basis – an exciting but also potentially overwhelming situation.Footnote 57 Navigating the nuances of, for example, post-Keynesianism or economic anthropology is a burdensome dilemma for Roman historians who have already acquired background philological and historical knowledge and, in many cases, theory common to history-writing. In the case of economics, some practical helps have appeared such as Donald Jones’ Economic Theory and the Ancient Mediterranean or Morris’ now aged but still apropos “Hard Surfaces” chapter in the edited volume Money, Labour and Land.Footnote 58 Morris’ chapter, which contrasts the approaches to model-building in neoclassical economics versus those of economic sociology, performs an invaluable service to the field by unpacking concepts otherwise foreign to many ancient historians. The expanding methodological menu has also brought new and more sophisticated challenges to the formalist enterprise broadly. Naïve substantivism (why study theory which is inapplicable if not alien to antiquity?) has been replaced by theoretically literate critiques of formalism and the use of economic theory broadly. Recent debates have taken place upon a fitting battleground: new institutional economics. The ideas of new institutionalist and Nobel laureate Douglass North –both a critic and a proponent of neoclassical economics – wield tremendous influence over the use of economic theory in the field of Roman economic history. Assessing the presence of new institutionalism in Roman history is not only valuable for understanding the present state of the field, but it is also possible to signal trajectories for future debates on the place for economic theory in the study of the Roman economy.

New Institutional Economics and Beyond

New institutional economics enjoys something like ‘third way’ status in the field of ancient economic history because the new institutional framework seems to enable (primitive) pre-industrial economies to be studied using (formal) economic theory. Roman economic historians friendly to new institutional economics tend to use the version articulated by North in which economic analysis accounts for ‘institutional frameworks’ – the rules, laws, political bodies and social norms which play a key role in controlling, guiding or influencing the actions and interactions of individual actors.Footnote 59 ‘Institutions’ are typically defined as the practical mechanisms by which economic exchanges take place in societies – e.g. legal systems, property rights and money are all ‘institutions’. Dennis Kehoe argues that thinking in terms of institutional frameworks helps historians categorize and subsequently analyze a broad range of activities, laws and customs in the Roman world, both ‘formal’ and ‘informal’. Kehoe provides a succinct of explanation of the difference between formal and informal institutions in his book Law and the Rural Economy in the Roman Empire:

The formal institutions include the rules that the Roman state established to define property rights, legal rules relating to farm tenancy and other aspects of land tenure, as well as the courts and other institutions through which individuals could defend their rights. The formal institutions also include policies of the Roman government, such as how it defined and defended property rights, how it mediated between competing claims by various classes of landowners and tenants, and, ultimately, whose property rights it tended to foster. But institutional environment also embraces the informal institutions that helped determine how formal institutions worked, that is, the social values and practices that helped establish the ‘rules of the game’ for the rural economy in the Roman Empire.Footnote 60

Kehoe is one of many Roman economic historians who have found in North’s new institutionalism a body of both abstract concepts and practical models seemingly tailor-made to overcome the unique methodological problems that have hitherto plagued the historiography of the Roman economy.Footnote 61 Perhaps more than any other school of economics, new institutionalism appears best equipped to handle the persistent problem of limited empirical evidence; at the same time, the approach seems to reveal new (although often hypothetical) ‘evidence’ about the costs and incentives created by ‘institutions’ and the response these elicited in economic actors.

The practical power of new institutionalism is in its use of ‘transaction costs’ to analyze and explain economic activity, whether such activity occurs in individual transactions or in a wider social context. Transaction costs are the costs associated with making an exchange – costs, for example, of defining and protecting property, negotiating contracts and completing transactions – all of which are often unexamined in the ‘clean’ models of neoclassical economics.Footnote 62 The ‘frictionless’ world of neoclassical models, North thoughtfully observes, “appears to beg all of the interesting questions.”Footnote 63 Neoclassical models, for example, assume perfect information – that prices, quality and value (utility) are thoroughly known by individuals. By contrast, new institutional economics factors (or attempts to factor) the cost of searching for goods, price comparisons, gathering information and information asymmetry. Markets in neoclassical models are usually perfectly or ‘purely’ competitive, in that no individual or firm is powerful enough to affect the equilibrium price – the price at which supply and demand are in balance. New institutional economics, however, acknowledges that there are costs involved when parties agree to exchange, such as the need to draw up contracts. States or other enforcement agencies are also important for policing contracts and perhaps providing legal systems; the costs of these must be factored into the ‘institutional environment’. Finally, neoclassical models usually assume rule-enforcement – that property rights, contracts and the rule of law will be upheld and respected without deviation – as a given. A framework of transaction costs, however, allows the models of new institutional economics comparative flexibility against those of neoclassical economics. Identifying and quantifying transaction costs in the Roman economy, however, is difficult. Only occasionally do ancient sources seem to mention transaction costs, such as when Tacitus claims that slave traders raised their prices after Nero required traders to pay the tax on slave purchases.Footnote 64 Furthermore, to what extent is it even possible to assign cardinal values to these costs? One might, for example, look at taxes paid as a way to calculate the costs of the judicial system, but not all taxes go exclusively to the provision of justice and rarely (if ever) has any justice system given equal treatment to all taxpayers.

New institutional economics also offers a different approach to rationality. Neoclassical economics tends to model rationality after homo economicus’ instinctive disposition toward profit-maximization and omniscience.Footnote 65 The use of this ‘perfect’ rationality for analyzing non-capitalist societies outraged Polanyi as well as Finley and his followers, and for good reason, as “neither the drives and psychological profile attributed to homo economicus, nor the assumption of the effective operation of price mechanisms, can be applied across the spectrum of ancient societies without creating a gross travesty of reality.”Footnote 66 New institutional economics’ understanding of transaction costs problematizes the rationality of homo economicus, as the presence of these costs in exchanges – costs which are often implied or even hidden during the decision-making process – leaves little room for a purely ‘rational’ actor to exist. Instead, new institutional economics uses the more nuanced ‘bounded rationality’ to theorize and analyze human decision-making under conditions of limited information, limited ability to assess value and the limited amount of time individuals have to make decisions. North posits that

institutions are a creation of human beings. They evolve and are altered by human beings; hence our theory must begin with the individual. At the same time, the constraints that institutions impose on individual choices are pervasive.Footnote 67

North’s bounded rationality seems to find itself somewhere between ‘methodological individualism’ – the idea that the study of society must begin with the study of individuals – and neoclassical ‘ontological individualism’ – the idea that only individuals exist. North here draws upon the idea advanced by fellow Nobel laureate Herbert Simon that human knowledge is “local” rather than “global.”Footnote 68 In practice, individuals rarely make “optimal” decisions; most often, they make “satisfactory” decisions instead.Footnote 69 Compared to perfect rationality, bounded rationality offers historians a better way to think about microeconomic factors and how economic activity might work within an embedded economy.Footnote 70

Is the wholesale adoption of new institutional economics the answer to the substantivist/formalist debate?Footnote 71 Evaluations are mixed. Historian of capitalism Francesco Boldizzoni argues that new institutionalism’s accommodation of embeddedness is a deception, and that new institutional economics is a mere rebranding of neoclassical economics, only with a more refined set of categories and a consistent although sometimes vague lexicon of jargon: “what [North] did was to extend the neoclassical explanatory model to the realm of social relations.”Footnote 72 North himself called his efforts a “modification” of neoclassical theory, as his approach retains most of neoclassical microeconomics as well as many basic neoclassical assumptions about scarcity and competitive markets.Footnote 73

Some historians believe that the Roman Empire is “a test case for Douglass North’s claim about the importance of institutions,” but how should economic historians perform such tests and would these tests differ substantially from the kind of testing performed in neoclassical analyses?Footnote 74 How meaningful is a term like ‘institution’, for example, which encompasses many facets of the process of exchange and even the wider political, legal, social and cultural context in which exchange takes place? Economist and economic historian Peter Temin’s use of new institutional concepts in his scholarship on the Roman world has led him to conclude that market systems were dominant in classical antiquity.Footnote 75 Bruce Frier and Kehoe, by contrast, use new institutional concepts in ways which are surprisingly compatible with and even at times reinforce a primitivist and substantivist view of the ancient economy: “ancient evidence can seldom if ever be used to test the hypotheses that are characteristic on the current research agenda in new institutional economics.”Footnote 76 Even Bang’s primitivistic “Roman bazaar” economy is rooted in a comparative analysis of institutions, albeit an analysis which emphasizes the dominance of predation and prevailing cultural norms. Therefore, while Boldizzoni’s critiques no doubt apply to many applications of new institutionalism, there are also noteworthy exceptions.

Roman economic historians’ use of new institutionalist concepts has now spanned multiple decades – enough time to afford some hindsight on the results. Despite its clear improvements over neoclassical analysis, some Roman historians remain unconvinced that new institutional economics offers a credible way forward. Is it a form of cliometric abuse, for example, to mine ancient sources for ‘statistics’ (and in the process discard cultural and social contexts) in order to quantify transaction costs and the wider ‘institutional environment’? Are analyses driven by new institutional concepts more likely to ignore historical problems which are not answered by appeals to quantitative data or statistics – especially questions of meaning? Roman historians have long understood, for example, the social and cultural stigmas held (at least superficially) by Roman elites against certain forms of trade and entrepreneurship.Footnote 77 Making money as an end in itself implied a lack of self-control – the willful indulgence in a never-ending cycle of insatiable greed, as the accumulation of money merely precipitated the need to exchange it to gain more.Footnote 78 In a recent book, Neil Coffee provides economic historians with a thorough survey of Roman writers’ frequent condemnation of ‘money-making’ for its own sake, especially through money-lending.Footnote 79 As Kehoe, Bang and others have shown, new institutionalism offers Roman historians some tools for productively exploring such issues, but new institutionalism’s vestigial neoclassicism limits the historical applicability of many of its concepts.

As is the case with the agenda of neoclassical economics, new institutionalism prioritizes questions of causality over questions of meaning. Tacitus, for example, claims that Tiberius injected 100 million sesterces into the Roman economy via “banks” (mensae) in ad 33.Footnote 80 Recently, scholars seem much more interested in theorizing about how this injection may have influenced GDP than understanding the social and cultural meanings of, let alone the motivations behind, Tiberius’ intervention.Footnote 81 Collingwood summarized the insufficiency of mere causal analysis in matters of economic history:

No historian can claim to have shown that a certain sequence of events must have fallen out thus and not otherwise. The fall of a man’s income may lead him to retrenchment or to bankruptcy: which it does, depends certainly on what kind of man he is, but what kind of man he is can never be finally determined: he determines it himself in his own action as he goes on. He goes on to bankruptcy and we say he was an extravagant and thriftless man, but this does not explain why he chose that alternative, it is only a way of saying that he did choose it.Footnote 82

New institutional economics may have better concepts and more accommodating jargon than its neoclassical parent, but new institutionalism offers historians few if any tools for evaluating the meanings of historical actors’ choices, economic or otherwise. The potential for reorienting new institutional approaches toward identifying and understanding the sorts of ex ante attitudes, mentalities and ideologies which were ultimately articulated in ‘costs’ – economic costs to be sure, but also social and cultural costs – seems limited.

Another consequence of new institutionalism’s presence in the field of Roman economic history is a gradual but unmistakable modernist resurgence. Many Roman historians would now agree that the ‘institutions’ of the Roman Empire promoted, either by accident or design, markets and economic growth. Finley’s lingering influence has likely prevented excesses here; at least Roman historians have selected North’s new institutionalism (which focuses upon the state as the entity which sets the “rules of the game”), and not a more ideologically capitalist flavor.Footnote 83 Yet while Finley and his followers focused upon the state’s redistributive and predatory role, many new institutional studies reorient the state’s role toward raising or (more often) lowering transaction costs and thus facilitating market exchange and economic rationality. The triumph of this flavor of new institutionalist historiography is surely The Cambridge Economic History of the Greco-Roman World:

the creation of more peaceful and safer conditions translated to a marked decrease in transaction costs. The suppression of piracy in the final decades of the Republic, the diffusion of a ‘technology of measurement’ and of common metrological systems, and above all the creation of a unitary monetary zone and of common legal rules, especially in the field of commercial law, were all quite remarkable contributing factors in this reduction of transaction costs, in so far as they reduced uncertainty and improved access to information. The imperial state could define and enforce the fundamental ‘rules of the game,’ in particular exclusive property rights, not only in the Italian core but also in the provinces. The spread of the Roman notion (and practice) of private property was fostered by the increase in the number of urban communities of Roman or Latin status.Footnote 84

The most recent new institutional analyses give the Roman state an even more powerful and conscientious role as market facilitator:

both economic growth and intensification of trade … was not merely the unintended and unexpected result of the unification of the Mediterranean under Roman rule, but also the product of the role that the political organization of Rome purposely undertook in regulating market transactions … [this regulation] aimed at guaranteeing that the price in the forum rerum venalium would always be the market price, formed through the encounter between supply and demand: a market price conceived as a ‘fair price’.Footnote 85

In other words, the Roman state did not merely passively benefit markets; rather, the state designed the institutional environment for the purpose of growing markets. Such claims seem to vindicate those who fear that new institutionalism is merely a thinly veiled modernism. These skeptics believe that “cultural biases” are embedded in new institutional models and concepts, and these biases make it difficult for new institutionalists to distinguish the cultural elements which constrain their own thinking with what might be called “constants of nature.”Footnote 86 Both Boldizzoni and Jérôme Maucourant observe, for example, that despite North’s effort to incorporate non-market “substitutes” (not “alternatives”), North still premises these “substitutes” on capitalist concepts: contracts, property rights, economic efficiency and a pre-existent market mentality.Footnote 87 The rules adopted by institutions, North argues, “are derived from self-interest … with compliance costs in mind.”Footnote 88 If North’s new institutionalism assumes that “institutions,” even non-market ones, are magnetically drawn to market-like characteristics, then is it surprising that the adoption of new institutionalism by Roman historians has coincided with a distinctly modernist turn or, even more problematically, a teleology which idealizes (American) capitalism?Footnote 89

New institutionalist studies (again, with some exceptions), tend to underappreciate the pervasive problems of violence and predation in the Roman world.Footnote 90 Scheidel rightly characterizes the Roman imperial economy as “a product of organized violence and coercive integration.”Footnote 91 Some applications of new institutionalism retroject the modern notion of property rights, with its clear division between public and private, onto Roman ideas and practices of property ownership. In fact, property ownership in the Roman world often came with public obligations. Moreover, coercion and violence were common mechanisms by which the elite shareholders in the empire’s social hierarchy grew their share of resources at the expense of others’ (including fellow elites’) so-called ‘property rights’:

Monopolies normally abuse their power to charge a monopoly rent. Interpretations, therefore, that emphasize the benign institutional conditions produced for economic life by the Roman state, need to develop an explanation why the imperial government did not abuse its monopolistic position.Footnote 92

Behind senators’ terror of the rise of soldier emperors of the third century ad, for example, was concern that the coercive power apparatus, which had hitherto benefited the landholding elite, would be turned upon them. In fact, the subsequent course of events justified these fears. Many Roman historians, including myself, are the passive beneficiaries of exploitative systems – an embedding context which easily blinds us to fully appreciating the violence and coercion undergirding the societies we study.

Sociologist Mark Granovetter forcefully argues that research on markets must study “the actual patterns of personal relations by which economic transactions are carried out.”Footnote 93 In the case of the Roman economy, the role of coercion, predation and violence must be accounted for. Unfortunately, the networks of interpersonal connections, whether violent or voluntary, of pre-modern economies are impossible to quantify and exploit via formal economic modeling; even new institutional economics, despite being an improved version of neoclassical economics, offers limited avenues for circumventing the problems raised by embeddedness (of both observers and subjects). Cultural historians and economic anthropologists, most of them skeptical of economic theory, are unlikely to find the approaches offered by new institutional economics more useful than the tools they already know and use. Still, new institutionalism provides Roman economic historians with novel concepts, even if new institutionalism as a whole clearly cannot pull the field out of well-worn dichotomies. Despite its promise, the agenda of new institutionalism has thus far not materialized as a true ‘third way’ between either primitivism and modernism or substantivism and formalism. I do not believe that new institutionalism should be discarded entirely, but historians would be wise to use care and caution.


All historical applications of formal economic models require justification – not merely within their own closed system of logic, but in a wider historiographical context which includes serious and thoughtful substantivist critiques of the formalist enterprise more generally and especially of applied economic theory. New institutional economics may not be the final solution, but are there other ways Roman economic historians might use economic theories to better understand economic choices as well as the embedding contexts which channeled such choices? Although I share many of the substantivists’ concerns about new institutional economics, I wonder: can some new institutional concepts (e.g. transaction costs and bounded rationality) be redeemed in order to help historians understand the Roman monetary system? History and economics, despite fundamental differences embedded in each discipline, can meaningfully and symbiotically intersect. While I hesitate to suggest that the two disciplines are entirely reconcilable, economics offers Roman historians valuable and helpful organizing concepts, so long as these concepts are used within an agenda of historical understanding.

3 White (Reference White2014), xiii.

4 Morley (Reference Morley1999), 57.

6 Keynes (Reference Keynes1936), 378.

8 Friedman (Reference Friedman1953), 9.

9 E.g. Boldizzoni (Reference Boldizzoni2011).

10 Bücher (Reference Bücher1917). A long list of Rodbertus’ relevant works is found in Schiavone (Reference Schiavone2000), 227 n. 2.

14 Rostovtzeff (Reference Rostovtzeff1936), 250, 252.

16 Polanyi (Reference Polanyi1944), 60.

19 Polanyi (Reference Polanyi1944), 57.

20 Bang (Reference Bang2007), 7. Another thoughtful revision of Finley’s primitivism is in Hitchner (Reference Hitchner, Manning and Morris2007).

21 Mises (Reference Mises1949), 286–91.

23 Dale (Reference Dale2010), 141.

24 Garnsey and Saller (Reference Garnsey and Saller1987), 48.

27 Cartledge (Reference Cartledge and Ormrod1998), 114–5.

28 Tompkins (Reference Tompkins and Harris2013), 18–9.

31 Morley (Reference Morley2000)(b), 207. Emphasis added. See also Morley (Reference Morley2004), 39.

33 Hopkins (Reference Hopkins1978), 19–20.

34 See 108ff.

35 Duncan-Jones (Reference Duncan-Jones1990), 30–47; Morley (Reference Morley and de Callataÿ2014). See also Garnsey and Saller (Reference Garnsey and Saller1987), 49–53; Goldsmith (Reference Goldsmith1984), 273. On rhetorical style, see Morley (Reference Morley, Bang, Ikeguchi and Ziche2006a), 33–5.

36 Hopkins (Reference Hopkins, Scheidel and von Reden2002), 210–11. See also Hopkins (Reference Hopkins1980), 103.

37 Morley (Reference Morley2004), 24–6; Manning and Morris (Reference Manning and Morris2007), 38.

39 See his obituary in The Telegraph (2004).

40 Manning and Morris (Reference Manning and Morris2007), 15. Contra Schiavone (Reference Schiavone2000), 176.

41 As observed by Saller (Reference Saller, von Reden and Scheidel2002), 252–5; Morley (Reference Morley2004), 44.

42 Katsari (Reference Katsari2011), 4.

44 Morley (Reference Morley2004), 44–5.

47 Mises (Reference Mises and Reisman1933), 107–8.

50 Bang (Reference Bang1998); Katsari (Reference Katsari2011), 34.

52 Some, however, maintain that Temin’s work has “made the issue effectively closed” and “puts to rest Finley’s claim”; see Grantham (Reference Grantham, Erdkamp and Verboven2015), 59.

53 Temin (Reference Temin2013), 6; Temin (Reference Temin2001), 170. Contra Bang (Reference Bang2007), 25 n. 59; Haley (Reference Haley2003), 11–2.

60 Kehoe (Reference Kehoe2007), 29–30.

61 See, for example, the authors in recent volumes: McGinn (Reference McGinn2012); Kehoe, Ratzan and Yiftach (Reference Kehoe, Ratzan and Yiftach2015b).

63 North (Reference North1981), 5.

64 Tac. Ann. 13.31. Koops (Reference Koops, du Plessis, Ando and Tuori2016), 612.

65 Madra (Reference Madra2017), 116.

66 Davies (Reference Davies, Manning and Morris2007), 130. See also Bresson (Reference Bresson2015), 13.

67 North (Reference North1990), 5. See also 86ff.

68 Simon (Reference Simon1957); Simon (Reference Simon1955); Metcalfe (Reference Metcalfe2010), 61–3.

70 See for example Kehoe (Reference Kehoe2007), 38–9; Bannon (Reference Bannon2009), 30–6.

71 Scheidel (Reference Scheidel2005), 35.

72 Boldizzoni (Reference Boldizzoni2011), 18.

77 Morley (Reference Morley2007), 83–5.

78 Schaps (Reference Schaps2003), 179–93.

80 Tac. Ann. 6.17.

81 See 89ff.

83 As opposed to that of, say, Oliver Williamson, who assumes the primacy of the market: “in the beginning there were markets,” Williamson (Reference Williamson1975), 20. Rather than the state, Williamson focused upon “firms” which correct information asymmetries and promote the smoother operation of markets. See Williamson (Reference Williamson1985); Williamson (Reference Williamson1981).

86 Boldizzoni (Reference Boldizzoni2011), 23.

87 Maucourant (Reference Maucourant2012), 196–201; Boldizzoni (Reference Boldizzoni2011), 21–3. The specific object of their critique is North (Reference North1977).

88 North (Reference North1990), 48.

89 Maucourant (Reference Maucourant2012), 197.

92 Bang (Reference Bang2009), 205. See also von Reden (Reference Von Reden2010), 13–5; Bang (Reference Bang2008), 97–103; Weber (Reference Weber1968), 164.

93 Granovetter (Reference Granovetter1985), 504.