The Matter of the Market
The point at issue here is the integration of markets in the Roman Empire. Was there a set of unconnected markets? Or did different markets all interreact? Here, too, diametrically opposed opinions stand side by side, sometimes without impinging on or engaging with one another.
THE ENORMOUS CONGLOMERATION OF INTERDEPENDENT MARKETS: A RED HERRING
The broadest generalizations being the simplest ideas to handle, the statement that Rome enjoyed ‘an economic system that was an enormous conglomeration of interdependent markets’ has for some time been much discussed. According to E. Lo Cascio, it is a controversial notion. Peter Bang has argued against it; Leopold Migeotte quotes it as an indispensable reference point for his diachronic study of the economies of Greek cities from the archaic period to Roman times, but comes to the conclusion that it is untenable. Yves Roman, on the other hand, in his conclusion to the same collection, seems to take it for granted, a view that both Richard Hingley and Annalisa Marzano, despite some reservations, come close to sharing.
The statement derives from an article by Peter Temin, who says with laudable clarity at the beginning: ‘Finley declared that “ancient society did not have an economic system which was an enormous conglomeration of interdependent markets”’, and ends by saying: ‘Finley was wrong; ancient Rome had an economic system that was an enormous conglomeration of interdependent markets’.
The same clarity appears to mark Temin’s approach. He begins by equating Finley and Polanyi, the former having implicitly based his criticism of Rostovtzeff on the latter. Polanyi’s three categories of exchanges (redistribution, reciprocity, and market) are examined in turn: those that have no relevance to the market are inapplicable to Temin’s chosen set of examples. He notes the existence of coinage, prices, interest rates, contracts, a concern for profitability, and several intelligent economic behaviours. In his view, there were no completely autarkic agricultural enterprises; those who ran them were aware of market prices and took account of them in their economic forecasts. Even in Rome, the grain dole represented a small minority of transactions, all the rest being under the control of private traders and subject to market forces. State transport was covered by contracts. Prices were therefore central to transactions, varying according to supply and demand and regulating markets which were economically connected. So here we have the Roman Empire endowed with an economy made up of markets, all of them interconnected, though some more closely than others, depending on the region.
A great deal could be said about Temin’s examples, his choice of them, their accuracy, and the way he interprets them. His conclusions about the agricultural-produce market are highly debatable, as has been pointed out by Peter Bang, who has also remarked that the exorbitant interest rates cited as market effects were actually the hugely disproportionate ones demanded by the politically powerful.3 It should be added that, on three separate occasions, Ulpian makes the very point that any evaluation of valid interest rates must acknowledge that they are largely determined by the customs of a province or region;4 and, before Ulpian, Gaius had already stressed that, from place to place, there were great differences in the price of money.5 On the central point of grain for Rome, Temin ignores the long-standing estimates of total deliveries of tax grain: only about a third was doled out free; the rest was put on the market by the state, at least partly in Rome, and the price of it could be set politically.6 Nor should it be forgotten that exports of grain from Egypt to anywhere other than Rome were first forbidden, then controlled.7 As for contracts with shipowners, an addition to Diocletian’s edict found in the as the crow flies is given as an approximation for the distance from Lusitania to Rome; the Po Basin is linked to Rome by rivers, and so on.
- 3 Bang (2006: 54-5, n. 8; 2008: 30-2).
- 4 Dig. XXVI. 7, 7, 10; XXVII. 4, 3, 1; XXX. 39, 1.
- 5 Dig. XIII. 4, 3, which I quote on p. 184, n. 33. For a thorough discussion, see Andreau (2010: 161-4).
- 6 Van Berchem (1939). See more recent references in Virlouvet (2003: 67-72); Erdkamp (2005: 244-57, 326 (256: ‘Free trade in the empire’s capital operated in the margins of a system that was characterised by public supply channels’)); De Romanis (2007: 223), arrives at a figure of 35-40 million modii, substantially higher than Van Berchem’s 27 million. Bang (2008: 177, n. 14) summarizes the issue.
- 7 Casson (1980: 23-5); De Romanis (2003a: 703-5).
Aezani fragments makes it plain that they did not abide by the general rules, at that time at any rate.
One could comment on other details; and one might wonder whether the broader Roman economy can be fitted into categories as Procrustean as those of Polanyi. My purpose here, however, is first to inspect the structure and the basis of Temin’s text, useful as it is as a starting point for a critical examination of the whole matter. Three points, in my view, must be made.
Reaching his conclusion, Temin says that ‘market prices moved together’. There is, of course, no other way to demonstrate interconnection among markets that were far apart. Yet, on this matter, Temin offers not a shred of evidence, for the simple reason that there are no sources that could provide any. In 2004, when David Jacks scotched the doubts expressed by Allen and Unger about the integration of markets in Baltic and North Sea towns during the seventeenth and eighteenth centuries, he made the initial comment that it is not enough to note that prices differed. One must also take account of transport costs, or more generally of all transaction costs, between those towns that were most advantaged by their positions and the others. Next, Jacks collated the prices of wheat and rye from four or five towns over a period of three centuries, and from a few more for the eighteenth century. By calculating the variance factors mathematically, he drew up graphs showing the trend towards greater integration and the level of it in relation to later references. This enabled him to demonstrate that this integration, on the increase from 1500, had been disrupted about 1650, before increasing again in the first half of the eighteenth century and reaching a very high level in the second half. Any attempt at doing something comparable for the ancient world is bound to fail. In fact, as Bang aptly puts it, ‘Temin’s case, however, rests on little more than the widespread existence of prices and markets within the empire. This, in his eyes, constitutes sufficient evidence that the imperial economy was organized as an interconnected market system resembling Europe and the Americas in the eighteenth century. But this is to jump to conclusions.’
By managing only to show the existence of prices and the role of supply and demand in the Roman economy, while offering no evidence for any generalized integration, Temin has gone to much trouble to refute ideas that were never Finley’s. On the contrary, thirty years before, Finley had rather vehemently distanced himself from Polanyi on the very matter of price regulation by supply and demand. Not only did Finley see this as the way in which the
Athenian economy functioned, but the mechanism had been conceptualized: ‘Price variations according to supply and demand were a commonplace in Greek life in the fourth century bc.’ To this Finley added a footnote:
I should perhaps not have bothered with these seeming platitudes, were it not that Karl Polanyi... makes the strange remark that ‘the supply-demand-price mechanism escaped Aristotle. The distribution of food in the market allowed as yet but scant room to the play of that mechanism. . . . Not before the third century BC was the working of a supply-demand-price mechanism in international trade noticeable.’ How wrong that is will be evident from [different texts from Lysias, Demosthenes, and Pseudo-Demosthenes].
Clearly, what was true of Athens in the fourth century bc still held good in the Roman world. In Latin, to speak of the threat of shortages, the idea was expressed as ‘dearness’, caritas annonae. Calpurnius Piso, an annalist from the early years of the first century bc, tells us of Romulus, who, as a guest at the house of friends, was joshed one evening for having drunk very little wine: ‘If everyone did as you do, wine would be cheaper.’ To which he replied, ‘On the contrary, it would be dear if everyone drank his fill, because I have drunk my fill.’ There could be no more compact exposition of the law of supply and demand, no more concise way of saying that it applies to goods in short supply and that, on that evening, the degree of marginal utility of wine was lower for Romulus than it was for the average consumer.
No one now doubts that economic life in Roman cities was monetized. It is clear that their inhabitants felt they were living in a milieu in which the economy was primarily (but not exclusively) based on prices that were the outcome of an equilibrium between supply and demand, but with variations dependent essentially on things affecting supply, such as good or bad harvests, hoarding, or weather that was favourable or unfavourable for shipping. Discussions about how to define the proportion of goods not subject to this law, because of free distributions or for other reasons, the role of autarky in agricultural enterprises, and how accurate landowners were in their forecasts, do not invalidate the principle, to which Temin’s examples add nothing new.
My final comment goes beyond all this. Finley was not the originator of the expression ‘an enormous conglomeration of interdependent markets’. It is, in fact, a quotation from the great book by Lord Roll, A History of Economic Thought, which went through many editions between 1938 and 1992. Finley asked why the Ancients, in particular Aristotle, had written no work marking off economic thinking from matters such as domestic management or moral conduct. By way of seeking an answer, and needing a definition of the object of modern economic science, he borrowed one from Roll, making plain that it struck him as being broadly acceptable to economists. It ran as follows: ‘If, then, we regard the economic system as an enormous conglomeration of interdependent markets, the central problem of economic inquiry becomes the explanation of the exchange process, or, more particularly, the explanation of the formation of price.’ From this, Finley concludes: ‘But what if a society was not organized for the satisfaction of its material wants by “an enormous conglomeration of interdependent markets”? It would then not be possible to discover or formulate laws (“statistical uniformities” if one prefers) of economic behaviour, without which a concept of the “economy” is unlikely to develop, economic analysis impossible.’ In Finley’s view, the lack of a conglomeration of interdependent markets in Antiquity explains the lack of a separate line of economic thinking; and, conversely, he saw the lack of an idea of an economy in the modern sense as a sign that there was no great unified market.
This choice of Finley’s was unfortunate. Roll’s sentence is in no way an attempt at a general definition of the object of modern economic science. It belongs in an introductory passage from the chapters on the transition from classical economic theory, about 1870, to neo-classical theory, known as marginalism. It is only the third point in a list of the principles underlying this new way of thinking that would make it possible to conceive of the economy in mathematical terms. Roll spells out these principles, or rather assumptions, as follows:
 In regard to the goods and services which individuals require directly for the satisfaction of their wants, the general purchase-and-sale character of individual behaviour is easy to recognize;
 But even the transactions of the productive process are seen to resolve themselves into the purchase and sale of raw materials, capitals goods, money capital, and labour;
 If, then, we regard the economic system as an enormous conglomeration of
In the final third of the nineteenth century, neoclassical theorists were not trying to describe an economic reality determined by certain historical circumstances. Their aim, as Roll puts it two pages later, was to ‘develop a theory of value which is independent of any specific social order’. It applies both to Robinson Crusoe and to Wall Street, to prehistoric people or to nineteenth- century Europeans. The geographical space in which a market functions is of no account for an abstract theory of the ways in which human beings choose between their differing needs. Moreover, in Roll’s view, it was not the extension of the market, but the creation of manufacturing, that gave rise to the conditions that shifted the accent from exchange to production and thus brought about the emergence of the ideas of William Petty, Quesnay, and Adam Smith.
Read in context, Roll’s sentence about ‘an enormous conglomeration of interdependent markets’, with its suggestion of rather monstrous dimensions, clearly does not mean some huge spatial spread of markets, but the enormous number of goods that satisfy the needs of humans as consumers and producers, and their constant ordering of these needs. The markets that are interdependent are those of different goods and services of various kinds, not those located in different places.
Finley took from Roll’s words a meaning they did not have; and Temin rushed into the breach opened by the mistake, dragging after him all those who draw on his work. The ‘enormous conglomeration’ is not the right door through which to enter the debate on whether markets were integrated in the Roman Empire.
-  Hingley (2005: 106); Bang (2006; 2008: 30-2); Lo Cascio (2007c: 622 (more critical in 2009:208)); Marzano (2007: 7); Migeotte (2008); Roman (2008).
-  Temin (2001). A second article (Kessler and Temin 2008), which aims to demonstrate aninverse correlation between distances from Rome and the price of wheat, is unfortunately ruledout of court at the very outset by its preposterous geography: the distance from Madrid to Rome
-  Arnaud (2008: 131-2): the words praeter onera fiscalia quae formam suam optinent(‘except for cargoes of tax goods, which observe their own rules’) were added to the price of allloads bound for Rome. The state also had the option of requisitioning shipowners: Dig. XIV. 2,10, si prior nauta publice retentus...
-  Jacks explains this by the simple formula: P1 = P2 + t (P1 being the price at point 1, P2 theprice at point 2, and t the transaction cost). Cf. Bang (2008: 154).
-  Bang (2006: 54).
-  Finley (1970: 13-14, and n. 45).
-  See Corbier (2005: 258-60), for an excellent analysis; and, whatever interpretation is putupon the legislator’s intention, the texts quoted by Lo Cascio (2006: 230-1).
-  Quoted by Aulus Gellius, NA XI. 14, 2. See also Nicolet (1988: 167-8), for citations fromother corroborating sources.
-  Veyne (2001: 139, n. 22 (1979a)) says that, having read the Digest in the light of MarcBloch’s argument that mentions of monetary exchanges in documents from the Middle Agesoften mask the bartering of things, he concluded: ‘The Digest gives the impression of acompletely monetized society and I had to abandon my initial scepticism.’
-  Nicolet (1988: 176), comes to the same finding though working from different sources. Seealso the examples quoted by De Ruyt (1983: 357).
-  Roll (1992: 338). 2 Finley (1985: 22 (1975b: 22)).
-  18 Restated in the chapter ‘Further Thoughts’ (Finley 1985: 176-9).
-  19 The standard definition given in handbooks is by Lionel Robbins: ‘Economics is a sciencewhich studies human behaviour as a relationship between ends and scarce means which have
-  alternative uses.’ But this is a definition that could not help explain how Aristotle’s analysis waslimited by the historical context.
-  Roll (1992: 338).
-  Le Masne (2002) says of Carl Menger: ‘Economics, having a theoretical object (humannature in relation to its needs), became a theoretical science... It does not deal with relationsbetween a man and other men in production and consumption, but with the isolated individual,the economic “Robinson Crusoe”... The neoclassical theorists defined laws which they saw ashaving universal validity, based on the assumed characteristics of human nature. The laws applyboth to Robinson Crusoe’s behaviour and to behaviours in Wall Street.’
-  Finley (1985: 22) noted that ‘the word “market” is used abstractly’, but he took no accountof it.
-  Roll (1992: 83-4).
-  Let us imagine, for instance, that, because the high price of petrol will increase the cost ofoutings by car, I decide not to go for a drive but to buy a kilo of cherries instead of half a kilo. I amjudging that the marginal utility of a small extra amount of cherries is greater for me than onemore outing by car would be. It is clear in that case that there is interdependence between the oilmarket and the cherry market.
-  I am grateful to Helene Hetier, agregee de sciences economiques et sociales, for kindlyreading the preceding pages and checking that they contain no flagrant falsehoods. Anotheranalysis of Roll’s sentence has been made by Lerouxel (2012); quite independently of mine, itreaches the same conclusion.