`When faced with such problems as estimating the demand for forms of transport..., the economist is forced into a characteristics approach, whereby estimates are attempted of the demand for various characteristics of travel, the demand for...travel being synthesized from these' [LANCASTER 1971, p. 10].
I. INTRODUCTION
There is now an extensive and elaborate economics literature that bases the theory of consumer demand on the attributes of a good rather than on the good itself. The central figure in the contemporary development of this theory has been KELVIN LANCASTER [1966a; 1966b; 1971; 1979], although LANCASTER has credited incipient notions of the idea to MENGER [1871], CHAMBERLIN [1933], STIGLER [1945], BREMS [1951), THEIL [1952], HOUTHAKKER [1952], and more explicit versions to MAY [1954], HICKS [1956], BECKER [1965] and IRONMONGER [1972]. While acknowledging the priority of these earlier writers, LANCASTER assert that none of them attempted to integrate the `characteristics' approach into the central core of demand theory.
This paper explores and elaborates the unrecognized pioneer contribution of JULES DUPUIT to the characteristics-based theory of demand. Trained as a civil engineer, DUPUIT was the premier transport economist of the nineteenth century. As the opening epigram by LANCASTER indicates, and as we demonstrate below, problems in the area of transport economics invite the characteristics approach.
II. DUPUIT ON UTILITY, DEMAND, COMPETITION AND ENTREPRENEURSHIP
Although for centuries economists used the term utility loosely to describe either individual satisfactions or the general welfare, JULES DUPUIT [1844] was the first writer to link utility to demand in a formal, theoretic sense(n1). Correcting the errors in SAY's analysis of the utility of public works, DUPUIT developed the formal idea of marginal utility and identified it with demand(n2). He then combined this powerful insight with a serviceable theory of production costs to make numerous and important discoveries in microeconomic theory, not least of which were the efficiency and welfare implications of static monopoly theory, including the theories of price discrimination [EKELUND, 1970], marginal-cost pricing [EKELUND, 1968], excise taxation, spatial economics [EKELUND and SHIEH, 1986], and the comparative price-allocative effects of competition.
1. Utility, Product Quality, and the Competitive Process
While DUPUIT's substantial contributions to comparative statics establish him as an early and important neoclassical economist, they do not indicate his keen understanding of the dynamics of the competitive process in which the entrepreneur plays a crucial role. Our research shows that DUPUIT went beyond static analysis (which he knew to be a mere abstraction of market operations) to develop a view of competition that stressed interactions between gradations of product quality and nominal prices.
In DUPUIT's economic theory, demand presupposes utility. He never tired of stressing the point that `there is no utility other than that for which people are willing to pay' [DUPUIT, 1844, p. 49], and he argued that the maximum sacrifice in terms of money that individuals were willing to give was the best estimate of the utility of the object [DUPUIT, 1844, P. 39](n3). DUPUIT enriched his theory by treating goods as combinations of utility-producing characteristics rather than one-dimensional `lumps' of utility. The qualitative attributes that he deemed important (i.e., utility-producing) included product and service characteristics that increase consumers' utility in the conventional sense, and dimensions of a good that reduce disutility by lowering waiting time or transaction costs. Moreover, he envisioned markets as working to establish a full-price equilibrium wherein (in the case of transportation) waiting time is minimized through a menu of differentiated products (see EKELUND and SHIEH [1986; 1989]).
DUPUIT regarded quality dimensions as basic to the nature of a product and therefore, capable of many variations. These variations, moreover, are valued by consumers:
`(R)arely does a cost-reducing change in production not also change the quality of products; they become better or worse, larger or smaller, lighter or heavier, faster or slower, and so on...,[and] all these qualities have a value which can be measured by the calculation of utility' [DUPUIT, 1844, p. 54].
Clearly, utility envelopes both quantitative and qualitative aspects. DUPUIT in fact, characterized competition (in the dynamic sense) as the ongoing process of creating new combinations of utility-producing characteristics that will reward the creator with increased sales and profits.
2. Demand Discovery and the Role of the Entrepreneur
Within this dynamic view of competition, DUPUIT conceived the entrepreneur's role as `demand discovery'. In other words, the entrepreneur must estimate the utility that consumers assign to goods or services, and devise pricing schemes to get them to pay in proportion to the utility they receive. Since each good or service is potentially comprised of a number of utility dimensions, the entrepreneur's function is more subtle then it first appears. It requires choosing many attributes beyond the mere physical dimensions of the product. Market segmentation, quality variation, and all other kinds of product differentiation are choice variables in the entrepreneur's decision nexus.
DUPUIT was aware that such entrepreneurial talent is rare, sui generis, but he argued that it could be enhanced:
`It takes a special talent to estimate from certain specific clues the importance of the service rendered, to distinguish these attributes clearly, and classify them methodically in a schedule of prices that relates price to this importance, a talent which lies primarily in certain natural faculties, but which can, like any other gift, be considerably augmented by study, observation, theory, and experience' [DUPUIT, 1849, p. 162].
The entrepreneurial skill that competition invokes is a consequence of product uncertainty. CANTILLON [1755] told us long ago that the entrepreneur does not know the demand function for each (monolithic) product he/she sells. DUPUIT went even farther by claiming that the entrepreneur doesn't even know the product itself! He must discover it. DUPUIT was the first economist to explicitly recognize that the entrepreneur manipulates prices and product characteristics, experimenting in order to find profitable combinations of elements that make up a (sophisticated) product. Almost as though he anticipated the tendency of neoclassical theory to slight entrepreneurship, DUPUIT [1849, p. 137] underscored the burden of uncertainty on economics:
`If the law of consumption was perfectly known, if sales policy...was set and invariable, these calculations would become perfectly determinate problems, solvable by the simplest arithmetic'.
We can now appreciate more fully what DUPUIT meant by the seemingly innocuous phrase, `sales policy'. The realization of entrepreneurial profits demands that the entrepreneur employ imagination and skill in the face of ignorance. DUPUIT was optimistic about the ability of economics to assist:
`Ignorance of the law of consumption is w obstacle to the rational calculation of prices. The substitution of probability in place of certainty unquestionably makes the problem more difficult, but it also lends new charm to the solution. On the one hand, all business calculations turn on conjectures about the law of consumption; and, on the other, on the means of making consumers pay for the utility of the products....in the producer's uncertain world the solution depends both on his skill in estimating the needs of the consumers, and on his imagination in devising a method of making them pay as much as possible' [DEPUIT, 1849, pp. 139-37].
DUPUIT noted that may forms of product differentiation may appear to be trivial, but he allowed the consumer wide latitude in making consumption choices. From the standpoint of economic theory, the central issue is whether or not product differentiation creates, in the subjective judgment of the consumer, an increase in utility (or a decrease in disutility). DUPUIT was aware of different demand elasticities across consumers, and its bearing on the problem:
`The same merchandise, disguised in different stores under various forms, is often sold at very different prices to the rich, the well-to-do, and the poor. There is the fine, the very-fine, the extra-fine, and the super-fine, which, although drawn from the same barrel, present no real difference other than a better label land a higher price. Why? Because the same things has a very different utility value for the consumer. If the goods were sold only at an average price, all those who attached less utility than this price would not buy, and thus incur a loss; and the seller would lose because many of his customers would be paying for only a very small part of the utility they receive' [DUPUIT, 1853, p. 177].
The important issue for DUPUIT was the effect of product differentiation upon total sales. He took it as self-evident that consumers' total utility would rise if the number of goods or services purchased increased(n4).
In practice, the entrepreneur's role is even more subtle than suggested by these passages, because all exchange involves transactions cost --e.g., relative convenience, location, waiting, and so son. DUPUIT claimed that the entrepreneur must account for these factors in setting nominal prices. This idea comes through most forcefully in his analysis of transportation issues. He advised that in certain instances nominal rates might be lowered to offset such customer costs as slow-moving trains, inconvenient departures and arrivals, passenger congestion, and the high opportunity costs of alternative modes of transport [DUPUIT, 1849, pp. 159ff; 1854a, p. 343]. Thus the responsibility of devising a full-scale `products policy' is two-pronged: in determining the set of prices that maximize utility the entrepreneur must ascertain the optimal (and variable) physical characteristics of a good; and he must discover and accommodate customers' opportunity costs.
Some of the points underscored in the foregoing discussion, and [DUPUIT's] adeptness on these issue, are illustrated in a technical treatise in which he confronted the problem of economic efficiency in the distribution of municipal water supplies. The problem was exacerbated by the fact that water meters were technologically primitive at the time, making optimal user charges difficult to determine. DUPUIT considered two alternatives: (1) the intermittent system, whereby citizens could draw water from a reservoir only at a scheduled time of day; and (2) the continuous system, whereby customers could access a continuous flow by turning a faucet. The former system involved very high transaction and enforcement costs, whereas the latter involved high metering costs.
DUPUIT concluded that the problem with the continuous system was that it utilized an inefficient pricing scheme which failed to acknowledge the different opportunity costs of different users. It was common practice to charge customers a flat rate per unit of water. This uniform-rate policy encouraged misallocation, and in some cases, a serious mismatch between demands and available supplies. In the case of Toulouse, for example, DUPUIT calculated that nearly half the municipal water supply was wasted, thereby requiring taxpayer subsidies to maintain production. His proposed solution relied not only on the technical expertise of the engineer. It captured, in microcosm, the essence of the entrepreneurial function. `Water...should be sold by differential subscription', DUPUIT [1865, p. 201] wrote, `at a price proportional to the utility which the subscriber derives from it(n5). Despite the known difficulties of estimating consumers' opportunity costs, DUPUIT outlined a rate design tailored to different water products. He asserted that `the amount which a landlord will pay for a subscription will depend upon the number and wealth of his tenants, the nature of their work, and the difficulty that they have of obtaining water' [DUPUIT, 1865, p. 202]. Therefore he endorsed the English practice of basing subscription rates on house rents. He also hypothesized that classes of houses could be established like classes of travelers on railroad trains, and user charges could be set accordingly. Most importantly, he showed that `free' wafer derived from the town square in most French municipalities was not free at all, because users must pay time and transaction costs to obtain it and are also frequently called upon to subsidize the municipal waterworks. In DUPUIT's view, both consumer welfare and economic efficiency could be enhanced by entrepreneurial manipulation of the price system, through demand discovery and differential subscription prices.
3. Demand Discovery and Product Quality: Theatre Tickets
In another concrete example, DUPUIT discussed how a successful impresario devises product and pricing strategies to maximize consumer utility. Again anticipating later developments, he argued that theatre tickets are not homogeneous goods -- like any and all products, they yield specific quality characteristics or combinations of such characteristics:
`A single price would not fill a theatre and may frequently produce only mediocre revenues. Thus the entrepreneur would suffer a pecuniary loss and the public a loss of utility. Divisions Of the theatre and of prices almost always increase receipts as well as the number of spectators. This is readily understandable if the divisions merely separate the seats where one can see and hear well from those where one cannot. But observation of how most theatres are actually divided leads us to conclude that this is one of the least considerations in pricing the tickets; that the entrepreneurs know how to adapt their prices to all the whims of the spectators, those that go to see, those that go to be seen, and those that go for every other reason. Customers are made to pay according to the sacrifice they are prepared to make to satisfy their whims, and not according to the show which they enjoy [DUPUIT, 1849, p. 142].
DUPUIT invented the marginal utility = demand curve in part to explain the implications of such pricing techniques as contained in the theatre example. Figure 1 reproduces the graphical apparatus he used to express the welfare implications of market exchange. The curve labelled FH is both a demand curve and a marginal-utility curve.
At a uniform (average) ticket price, P[sub a, the total utility produced by the performance is the area under the demand curve up to quantity Q[sub a] (area OFGQ[sub a]). For convenience, assume the marginal cost of tickets equals zero. In the spirit of DUPUIT's example, suppose the entrepreneur develops -- through hunch, observation, or experimentation -- two additional theatre products: box seats and balcony seats. By devising a scheme of differentiated products and prices, the entrepreneur thus attempts to segment both the high end and the low end of the demand curve, and to sell in both regions. If he is successful, the total utility of theatre-goers would rise from OFGQ[sub a] to OFG"Q[sub bal], for a net gain in utility of Q[sub a]GG"Q[sub bal]. This affirms DUPUIT's original thesis, reiterated later by ROBINSON [1933, p. 206], that price discrimination is welfare enhancing if and only if output is expanded beyond the level produced by a monopolist imposing a single price [EKELUND, 1970].
DUPUIT's theatre example exposes two important ideas. The first and most obvious is that DUPUIT developed a sophisticated view of the entrepreneur's role in competitive markets. The second, more recondite point, is that product differentiation which correspondingly lowers customer transaction costs may help to prevent the dissipation of utility -- a point which DUPUIT understood but only partially developed.
As to the first point, DUPUIT's writings make it indelibly clear that price differentiation of all kinds (quantity augmenting as well as nonaugmenting) requires active participation by entrepreneurs in the development of product-quality characteristics. The theatre manager must know how to divide floor space so as to create utility-producing characteristics, which he must then successfully identify and sell. The entrepreneur treats products as neither monolithic nor homogeneous. Each item for sale offers a continuum of utility-producing characteristics(n6). The subtle, complex, but vital role of the entrepreneur, therefore, is to discover: (a) the technological possibilities for characteristics production; (b) the utility values of these characteristics in various combinations; and (c) appropriate market, or pricing, strategies that will maximize (b), given (a). DUPUIT believed that the competitive marketplace provided the largest opportunity set for achieving these multiple objectives, and he persistently defended competition in theory and practice.
The second point, which DUPUIT only imperfectly understood, is the fact that product-quality variations introduced by entrepreneurs may actually reduce aggregate transaction costs (i.e., lost utility) over the single-price alternative. Consider DUPUIT's theatre example once again. At price P[sub a] in Figure 1, the total quantity of tickets sold is Q[sub a]. But a more subtle issue is whether these tickets are homogeneous, or whether their sale, en bloc, generates transaction costs to potential buyers. DUPUIT held that they were not homogeneous goods, regardless of whether or not they were price-differentiated.
Those theatre patrons who are willing to pay P[sub box] lose utility, on average, if Q[sub a] tickets are distributed randomly and retracing is not possible. If retracing is allowed, ticket scalping can be expected, but some of the value of the gain (P[sub a] P[sub box] G'K') will be dissipated in transaction costs(n7). Conversely, if Q[sub a] tickets are sold on a first-come-first-served queue, the utility to consumers who want the better seats (P[sub a] P[sub box] G'K') may be dissipated through waiting time, a problem that DUPUIT [1849, p. 24] acknowledged in connection with passenger traffic on the railroads(n8). The key point here is that price differentiation linked in this manner to quality differences reduces the magnitude of non-appropriable utility losses --a conclusion DUPUIT implicitly understood, but did not elaborate.
It should be obvious that DUPUIT'S analysis is at variance with orthodox theory based on the assumption of homogeneous products. In Figure 1, for example, the horizontal axis does not and cannot represent the number of homogeneous theatre tickets sold, because the tickets are not homogeneous. This fact must be recognized whether or not entrepreneurs identify and sell perceived quality differences. An intriguing aspect of DUPUIT's analysis is his interpretation of products as stores of services which have qualitative differences of both a subjective and an objective nature. Later economists echoed this theme without acknowledging any input from DUPUIT, making it difficult to ascertain the full range of his influence(n9).
A fresh look at DUPUIT's analysis suggests several embellishments on LANCASTER's model. For example, DUPUIT laid the early groundwork for integrating consumer behavior and entrepreneurial stratagems. Clearly, the consumer's consumption-possibilities set would not exist without the active market participation of the entrepreneur. It is up to the entrepreneur to discover combinations of attributes and characteristics that consumers value, and to price them so as to maximize profits. In other words, entrepreneurial competition takes the form of price and product strategy in the fullest sense. The task that confronts each entrepreneur is to design new products and to match these new products with nominal-price vectors that simultaneously maximize consumers' utility and entrepreneurs' profits.
Another embellishment on LANCASTER's model is suggested by DUPUIT's (implicit) treatment of transaction costs. DUPUIT apparently recognized that the alternative forms (or characteristics-combinations) that products can take under the direction of able entrepreneurs include various alternative marketing strategies which produce utility not by augmenting the physical attributes of goods but by lowering the transaction costs of exchange. This is a very modem view, which LANCASTER did not sufficiently elaborate in his narrow attempt to reconstruct consumption theory.
III. CONCLUSION DUPUIT'S THEORY OF MARKET PROCESS
In the final analysis, the quest for entrepreneurial profits, which is the essence of competition, requires perspicacity, intuition, and inventiveness. It is much more complex than merely selling the same product at different prices to different groups of consumers by taking advantage of natural market divisions (a process DUPUIT clearly understood)(n10). It means treating products, in the conventional sense, as intermediate means to an end. It involves constantly tinkering with ideas which eventually take form in `attributes', or `characteristics', that alter the utility-producing nature of a product. Most importantly, it is a quest that, if successful, rewards both the entrepreneur (through higher profits) and his customers (through greater levels of utility).
The idea that the entrepreneur produces and sells a single product has never fit comfortably into economic analysis, yet it has stubbornly remained at the center of microeconomics despite several prominent attempts to dislodge it. At a time when microeconomic analysis was still in its infancy, DUPUIT advanced a theory of competition that emphasized market activity rather than market structure. This makes him one of the earliest anticipators of ideas which subsequently flowered under the aegis of CHAMBERLIN and LANCASTER. DUPUIT's first-hand experience with specific market behavior, especially the economics of transport and service enterprises, schooled him early on in the wider context of competition and entrepreneurship. He turned this experience to good advantage, advancing our understanding of both concepts by stretching the boundaries of received economic theory in two important directions, only one of which has been fully appreciated before now.
In the first instance, DUPUIT established a correct appreciation of the relationship between utility and demand. He put an end to SAY's confusion on the matter, thereby setting the course of value theory on the straight and narrow. For this, DUPUIT has received just acclaim, albeit belatedly. In the second instance he refined the meaning of product so that it embodied the many dimensions of utility, tangible and intangible, that comprise our motivation to purchase. As a consequence of this second achievement, which has not been recognized before now, DUPUIT elaborated concepts of consumer behavior and entrepreneurial behavior that were ahead of their time.
DUPUIT probably would not have regarded these two things as separate accomplishments. He understood the fact that what we now consider the standard, pace-theoretic notion of utility and demand is prerequisite to the elaboration of characteristics (instead of products) as the ultimate end of production and consumption. Much to his credit, DUPUIT did not end the story where he began it, with a simple demand or utility curve. He treated demand as an abstract notion embodying consumer preferences only until it is discovered by an entrepreneur, who then must be willing to bet on his judgment. To DUPUIT, therefore, the theory of consumer behavior was never a matter of unilateral action. Even though entrepreneurs are profit motivated, not utility motivated, they nevertheless play a vital role alongside consumers in effecting utility maximization.
DUPUIT's insightful theory of entrepreneurship and consumer behavior culminated an early French tradition that linked entrepreneurship to uncertainty. His predecessors, CANTILLON and SAY, isolated and analyzed the effects of price uncertainty in market behavior. DUPUIT absorbed these lessons and carried the analysis farther, by extending the notion of uncertainty to encompass product uncertainty as well as price uncertainty. He could not have accomplished this within the framework of a competitive model bound by the static restrictions of market equilibrium.
Unlike some recent models patterned after LANCASTER's analysis, DUPUIT's theory of entrepreneurship was not deterministic(n11). He did employ a deterministic theory of demand and utility as an approximation of market behavior, which served as a useful antecedent to his complex ideas on the role of the entrepreneur in the economic process(n12). But in its mature form DUPUIT's inquiry was guided by the practical considerations of a person in full touch with economic action as well as economic theory.
In France, DUPUIT's only serious contemporary rival on the subject of demand, AUGUSTIN COURNOT, turned economic theory down a different track. COURNOT eschewed utility considerations as `unscientific', and those who followed him attempted to redefine economics as though it were a branch of rational mechanics. Some neoclassical writers, like JEVONS, WALRAS and MARSHALL, rooted the theory of consumer behavior in the principle of utility, as DUPUIT had done, but nevertheless followed COURNOT by treating `homogeneous' products as the proper subjects of analysis in microeconomic theory. DUPUIT alone among nineteenth century economists saw products as bundles of utility-producing characteristics that comprise the objects of exchange. This highly modern perspective, however tardily recognized, should serve to awaken renewed interest in DUPUIT as an economic theorist.
SUMMARY
The French engineer-economist, JULES DUPUIT (1804-1866) fashioned theories of consumer behavior and entrepreneurship that stressed the following p,propositions: (1) that goods are `bundles' of utility-generating characteristics, and (2) that the entrepreneur's job is the dual task of discovering which combinations of characteristics consumers want to buy, and then pricing these combinations so as to simultaneously maximize sellers' profits and consumer' utility. Unlike some recent models patterned after LANCASTER, DUPUIT's theory of entrepreneurship was nd deterministic. He perceived competition in the dynamic sense, and emphasized market processes rather than end-states.
Auburn University, Auburn, Alabama, U.S.A. For citation purposes, translations from French to English were made by authors, unless otherwise noted. We are grateful to RICHARD AULT and DAVID KASERMAN for comments on earlier drafts of this paper. They are, of course, blameles for any errors or emissions that remain.
(n1.) COURNOT [1838, p. 10], who preceded DUPUIT in the formal analysis of demand, found utility `ill-suited for the foundation of a scientific theory', and developed his demand curve on empirical-statistical grounds instead (see FRY and EKELUND [1971]).
(n2.) See EKELUND and HEBERT [1976] for a more detailed account of the context of DUPUIT's discovery of the marginal utility principle.
(n3.) DUPUIT [1844, p. 491 placed an important and unappreciated caveat on his money measure of utility. Political economy, he argued, must use the money measure, but he added that `political economy ... is not, in the final analysis, a rigorous measure of the ability of things to satisfy mankind's needs; it would be difficult to say whose hunger was the greater--that of the rich man, who would be willing to pay a million for a kilogram of bread, or that of the beggar, who, having nothing to give, would risk his life for it. But political economy, confining itself to questions of wealth, can measure the intensity of a desire only by its monetary expression. It bakes bread only for those who can buy it, and leaves to social economy the trouble of supplying it to those who cannot afford to give anything in exchange'.
(n4.) We will argue below that utility may increase in a multi-price, product differentiated scheme -- compared to the non-differentiated, single-price scheme -- even when total output is not expanded through product differentiation.
(n5.) Absent the technical apparatus developed later, DUPUIT fully understood Ramsay pricing and optimal departures therefrom. He emphasized the fact that the results of such pricing provide the acid test of investment projects. In the case of water, he noted: `Differential subscription could lead to a consumption of water greater than the quantity one can use. If such were the case, it would merely prove that the distribution is inadequate, that a mistake had been made in the initial planning, and that steps should tee taken to improve provision' DUPUIT, 1865. p. 201].
(n6.) This characteristics-based approach to demand for the theatre tickets is not an isolated example. DUPUIT's bridge, railroad, and highway pricing analyses also employed `convenience' and `time' as elements to be included in consideration of `full price' [DUPUIT, 1849, pp. 129 ff.]. The similarities between a characteristics-based approach to consumer demand and spatial competition (e.g., ARCHIBALD, EATON and LIPSEY [1986]) is also of special note in the context of DUPUIT's clear conceptualization of spatial pricing in transport markets (cf. EKELUND and SHIEH [1986]).
(n7.) Actually, P[sub a] P[sub box] G'K' is the lower limit to utility dissipation. With uncertainty or grossly imperfect information, the total loss could be as high as P[sub a] FG'K'.
(n8.) These kind of transaction costs are explicit in the recent work of BARZEL [1974] and SAH [1987]. In one of the transport problems analyzed by DUPUIT [1849, p. 24] he noted that `certain fast stains leaving at convenient hours carry only first-class passengers, others first- and second-class ones. Third-class travelers must put up with the slowly plodding train, and arrivals and departures at awkward times -- not because anyone wants to make the poor arrive less quickly or leave at times other than the rich; the campaign is not waged against the poor, but against the avarice of the rich, against the hurried man to whom the trip by stain offers great advantages which he would like not to pay for'.
(n9.) STIGLER [1964, p. 45] argues that profits are reduced when firms ignore differences among buyers, even in the case of so-called homogeneous products, and that such disregard imposes `an excise tax' upon buyers, but one that is not collected by the seller. DUPUIT also believed that incentive structures within a dynamic competition process would maximize welfare (the sum of producer and consumer surplus). Like STIGLER [1964, p. 60], he represented product quantities grahically as functions of nominal price and depicted qualify changes as a rightward shift in the demand curve [DUPUIT, 1844, pp. 63 - 64].
(n10.) In the case of utility maximization of a bridge, for example, DUPUIT [1849, p. 140] suggested that lower rates could be charged to the working class, who were readily distinguishable by their dress (short-waist coats were worn exclusively by the French working class through the mid-nineteenth century).
(n11.) Several contemporary writers (e.g., PHILIPS [1983], SRINAGESH and BRADBURD [1989]) correctly credit DUPUIT with insights into the static theory of monopoly, specially his association of price discrimination with quality. Moreover, BEARD and EKELUND [1991] demonstrate that DUPUIT's conjectures (as interpreted by the writers mentioned above) are formally correct. Although these ideas comprise advances in static monopoly theory, DUPUIT did not confine his analysis to statics. He described product quality variations combined with nominal price policies as the essence of perpetual entrepreneurial activity. Even if monopoly existed in the short run, entrepreneurial ativity of the kind described in this paper would insure a competitive outcome. DUPUIT [1954b, pp. 850 - 851: 1854a, p. 344] repeatedly championed the competitive approach as the key to optimal resource allocation, on issues ranging from toll systems to international trade. His concept of competition, as embodied in entrepreneurial activity, is basically dynamic.
(n12.) Despite his extensive discussion of the entrepreneur and his role of `demand discovery' in the market process, DUPUIT constructed his forma models using a single product uncomplicated by the subtleties of various utility-producing `characteristics'. However, this practice is consistent with his view of the role of economic theory, which he regarded as a starting point for the practicioner. In political economy', DUPUIT [1853, p. 191] wrote, `defective data is what usually bars a compete solution; but this inconvenience only nukes it more necessary to know the rules and general principles that are at the base of every solution. They alone can fill in our gaps of knowledge, indicate what is missing and, consequently, furnish the wherewithal to search and find a solution if possible, or to provide one if it is not. As in geometry.... political economy must draw its adroitness sad precision in practice from the analytical rigor of science, because the data available are often incomplete and uncertain'.
REFERENCES
ARCHIBALD, G. C.; EATON, B. C. and LIPSEY, R. G.: `Address Models of Value Theory', in: STIGLITZ, J. E. and MATHEWSON, G. F. (eds.), New Developments in the Analysis of Market Structure, Cambridge, MA: M.I.T. Press, 1986, pp. 3-47.
BARZEL, Y.: `A Theory of Rationing by Waiting', Journal of Law & Economics, Vol. 17 (1974), pp. 73-95.
BEARD, T.R. and EKELUND, R. B., JR.: `Quality Choice and Price Discrimination: A Note on Dupuit's Conjecture', Southern Economic Journal, Vol. 57 (1991), forthcoming.
BECKER, G. S.: `A Theory of the Allocation of Time', Economic Journal, Vol. 75 (1965), pp. 493-517.
BREMS, H.: Product Equilibrium Under Monopolistic Competition, Cambridge, MA: Harvard University Press, 1951.
CANTILLON, R.: Essai sur la nature du commerce en general [1755], H. HIGGS (trans.), London: Royal Economic Society, 1931.
CHAMBERLIN, E. H.: The Theory of Monopolistic Competition, Cambridge, MA: Harvard University Press, 1933.
COURNOT, A. A.: Recherches sur les principes mathematiques de la theorie des richesses [1838], centenary edition, introduction and notes by GEORGES LUTFALLA, Paris: M. Riviere et Cie., 1938.
DUPUIT, J.: `De la mesure de l'utilite des travaux publics' [1844], in: DE BERNARDI, M. (ed.), De l'utilite et de sa mesure, Torino: La Riforma Sociale, 1933, pp. 29-65.
DUPUIT, J.: `De l'influence des peages sur l'utilitee des voies de communication' [1849], in: DE BERNARD, M. (ed.), De l'utilite et de sa mesure, Torino: La Riforma Sociale, 1933, pp. 97-162.
DUPUIT, J.: `De l'utilite et de sa mesure. De l'utilite publique' [1853], in: DE BERNARDI, M. (ed.), De l'utilite' et de sa mesure, Torino: La Riforma Sociale, 1933, pp. 163-191.
DUPUIT, J.: `Peage', in: Dictionnaire de l'economie politique, Vol. 2, Paris: Guillaumin, 1854a.
DUPUIT, J.: `Voies de communication', in: Dictionnaire de l'economie politique, Vol. 2, Paris: Guillaumin, 1854b.
DUPUIT, J.: `Du mode de distribution des eaux aux particuliers. Du prix de vente' [1865], in: DE BERNARDI, M. (ed.), De l'utilite et sa mesure, Torino: La Riforma Sociale, 1933, pp. 193-203.
EKELUND, R. B. JR.: `Jules Dupuit and the Early Theory of Marginal Cost Pricing', Journal of Political Economy, Vol. 76 (1968), pp. 462-471.
EKELUND, R. B. JR.: `Price Discrimination and Product Differentiation in Economic Theory: An Early Analysis', Quarterly Journal of Economics, Vol. 84 (1970), pp. 268-278.
EKELUND, R. B. JR., and HEBERT, R. F.: `Dupuit and Marginal Utility: Context of the Discovery', History of Political Economy, Vol. 8 (1976), pp. 266-273.
EKELUND, R. B. JR. and SHIEH, Y. N.: `Dupuit, Spatial Economics and Optimal Resource Allocation: A French Tradition', Economica, Vol. 53 (1986), pp. 483-496.
EKELUND, R. B. JR. and SHIEH, Y. N.: `Full Price Competition and Dupuit's. Defense of the Long-and-Short-Haul "Discrimination"', Journal of Regulatory Economics, Vol. 1 (1989), pp. 359-372.
FRY, C. L. and EKELUND, R. B. JR.: `Cournot's Demand Theory: A Reassessment', History of Political Economy, Vol. 3 (1971), pp. 190-197.
HICKS, J. R.: A Revision of Demand Theory, Oxford: Clarendon Press, 1956.
HOUTHAKKER, H. S.: `Compensated Changes in Quantities and Qualities Consumed', Review of Economic Studies, Vol. 19 (1952), pp. 155-164.
IRONMONGER, D.S.: New Commodities and Consumer Behavior, Cambridge: Cambridge University Press, 1972.
LANCASTER, K. J.: `A New Approach to Consumer Theory', Journal of Political Economy, Vol. 74 (1966a), pp. 132-57.
LANCASTER. K.J.: `Change and Innovation in the Technology of Consumption', American Economic Review, Papers and Proceeding, Vol. 61 (1966b), pp. 14-23.
LANCASTER, K. J.: Consumer Demand: A New Approach, New York: Columbia University Press, 1971.
LANCASTER K. J.: Variety, Equity, and Efficiency, New York: Columbia University Press, 1979.
MAY, K. O.: `Intransitivity, Utility and the Aggregation of Preference Patterns', Econometrica. Vol. 22 (1954), pp. 1-13.
MENGER, C.: Principles of Economics [1871], DINGWALL, J. and HOSELITZ, B. F. (trans.), New York: New York University Press, 1981.
PHILIPS, L.: The Economics of Price Discrimination. Cambridge: Cambridge University Press, 1983.
ROBINSON, J.: The Economics of Imperfect Competition, London: Macmillan, 1933.
SAH, R. K.: `Queues, Rations, and Market: Comparison of Outcomes for the Poor and the Rich', American Economic Review, Vol. 77 (1987), pp. 69-77.
SRINAGESH, P. and BRADBURD, R. M.: Quality Distortion by a Discriminating Monopolist, American Economic Review, Vol. 79 (1989), pp. 96-105.
STIGLER, G. J.: `The Cost of Subsistence', Journal of Farm Economics, Vol. 27 (1945), pp. 303-314.
STIGLER, G. J.: `A Theory of Oligopoly', Journal of Political Economy, Vol. 72 (1964), pp. 44-61.
THEIL, H. H.: `Qualities, Prices and Budget Enquiries', Review of Economic Studies, Vol. 19 (1952), pp. 129-147.
