The neo-classical "revolution" involved the abandonment of the classical theory of value based on the centrality of labor or work in the production of wealth and its replacement with a preoccupation with the "utility" gained from the consumption of wealth. At the same time, the analysis of both production and consumption were reformulated in "marginalist" and mathematical terms that made possible a whole new assortment of analytical, mathematical treatments. These changes involved not merely a shift in theoretical focus but a profound narrowing of the scope of what was increasingly called "economics" as opposed to "political economy."
Alfred Marshall (1842-1924), perhaps the greatest popularizer of the shift from classical to neoclassical theory, is closely associated with the substitution of the term "economics" for "political economy". In a book written with his wife Mary Paley and published in 1879 they wrote:
"The nation used to be called 'the body Politic.' So long as this phrase was in common use, men thought of the interests of the whole nation when they used the world 'political'; and then 'political economy' served well enough as a name for the science. But now 'political interests' generally mean the interests of only some part or parts of the nation; so that it seems best to drop the name 'Political Economy,' and to speak simply of Economic Science, or more shortly, Economics." [emphasis in original]They go on to argue that:
"Economics is a science because it collects, arranges, and reasons about one particular class of facts. A science brings together a great number of similar facts and finds that they are special cases of some great Uniformity which exists in nature. It describes this Uniformity in a simple and definte statement, or Law." A Law of Science states that a certain result will be produced whenever a certain set of causes are present." [emphasis in original]
Heilbronner, in his book The Worldly Philosophers presents this shift as one from grand world views and attempts "to illuminate the whole avenue down which society was marching" to professoral specialization and the "elucidation" of the details of the workings of the economy. This shift took place, he argues, within a world of growth, rising real wages and shortening working hours, a world therefore of "hope and promise."
Unfortunately, like the work of the new economists that Marshall and Heilbronner are describing, his account leaves out the dark, bloodstained core of that growth --the class struggle over work-- a terrain of worker strikes and sabotage, of capitalist counter-violence using private goons and public police repression. That terrain of endemic conflict between workers and capitalists spread with the growth of the system. On the one hand worker self-organization took the form of trade unions and political movements and on the other hand, capitalist efforts to cope with these challenges involved everything from co-optation at home to the imperialism of colonialism abroad.
While Heilbronner claims that rising standards of living and shortening working hours made it possible for economists to ignore the big questions those changes were not the automatic byproduct of capitalist growth but victories won on the fields of industrial and political battle. Moreover, business in defeat at home was able to make such concessions because of its own victories abroad in the bloody conquest and exploitation of the workers in Europe's colonies. In short, the "big questions" remained far from settled and in terms of the numbers of people involved and the scope of conflict grew larger every year.
The uncomfortable truth is that the only response of most economists to the violent terrain of work and production was to construct arguments aimed at demonstrating the futility of workers struggles that threatened the stability of capitalist accumulation. The increasingly elegant "science" of utility and profit maximation was put to use to shore up the system and limit the goals of those who would change it. In short, the "marginalist revolution" involved a flight from the threat of revolution to a refuge of marginal change. The abandoned terrain would be taken up by others, by scientific managers (like Frederick Taylor) and goons in the factories and by industrial sociologists and psychologists in the academy.
Unlike some, Marshall paid considerable attention to workers' demands for higher wages and less work. He was well aware of the pervasive poverty of most workers and deplored the conditions in which so many lived. Marshall read revolutionaries like Marx and utopians like More and Morris but rejected their calls for radical change as impractical and even counterproductive. He looked, instead, for solutions where standards of living might be improved within capitalism.
For Marshall wages were one part of a flow of net income (after deducting depreciation) from which interest, rent and taxes were also deducted. Leaving aside rent and taxes as being determined exogenously, Marshall focused on what was left: wages and profits (interest). Centering his attention on these very class-based categories of income - and the conflicts between them - might have led Marshall to a theory of zero-sum games. But instead, using the kind of "marginalist" thinking that came to be characteristic of neoclassical thinkers, Marshall examined the conditions under which marginal increases in wages might reduce profits and those under which they would not. The key was the relationship between marginal increases in wages and marginal increases in productivity. In Chapter 11 on wages of The Economics of Industry we find:
A rise in the Time-wages of any trade tends to diminish profits. but if the wages that are paid for work vary according to its efficiency - if Task-wages are unaltered - the share of the produce of industry that is left for others [the capitalists] will be the same whether Time-wages are high or low. It is only where the rise in time wages is not accompanied by a corresponding increase in efficiency, and therefore Task-wages rise, that the change is injurious to capital."
By the time Marshall published his Principles of Economics in 1890, eleven years later, he could not only explain the linkage between wages, profits and productivity growth more precisely but put this "new view" in historical perspective:
"If the efficiency of industry were increased, and more things were produced, higher wages would be paid at once by drawing more rapidly on stocks already in hand. It might be necessary to be a little carful about the stocks of some kinds of raw produce which could not be replenished very quickly. but with few exceptions the increased supplies would come in so soon that the stores need never run low. Therefore, the younger economists do not speak of wages as limited by capital. But they say that every increase of capital raises wages, because it increases the productiveness of industry; it increases the competition of the capitalist for the aid of labour, and thus lowers the rate of interest and increases the part of the total produce which capital is compelled to resign to labour."
The emergence of a preoccupation by economists with "utility" or "happiness" can be found in the writings of two groups: a number of Italian writers of the 18th Century and the better known British "utilitarians" who built on the work of Jeremy Bentham.
The Italian work is less known in the English speaking world for the usual reasons. First, little has been translated into English and second, the direct lineage of the neoclassicals from Bentham is well known and repeatedly cited in the literature. Nevertheless, the early work of Galiani, Beccaria and Verri should be recognized and you can examine what little is available in English, i.e., Cesar Beccaria's work on crime and punishment in which he sets out what was to become a famous formula. That formula was that public policy should be directed to achieving the "greatest happiness of the greatest number." Beccaria's reasoning, for determining a pattern of punishment that would lead to such a result, also foreshadowed the logic of "utilitarianism": "Pleasure and pain," he wrote, are the only springs of action in beings endowed with sensibility." The proper calculation of the pain appropriate to the punishment of crimes aimed at achieving pleasure, therefore, could lead to the minimization of the latter and the maximization of social happiness, or what would later be called "social welfare."
This nod of recognition to the Italians is partly necessary because it seems that the better known work of Jeremy Bentham (1748-1832), English philosopher, economist and jurist was influenced by his reading of Beccaria. A child prodigy, Bentham entered the University of Oxford at age 12, studied law and, after being admitted to the bar, wrote on issues of legal reform and fought for their implementation. In this work he drew, at least in part, from Beccaria's earlier work. Bentham himself acknowledged that he had read an English translation (1767) of Beccaria's Crimes and Punishments and his writing shows that he took over several of the basic perspectives.
Bentham's first major work An Introduction to the Principles of Morals and Legislation, was published in 1780. In the first chapter on the "principle of utility" and in the fourth chapter on measuring pain and pleasure --the only ones I want you to read-- you can see both his debt to Beccaria and his development of the basic concepts of pain and pleasure in terms of the concept of "utility" and the creation of an "hedonic" or "felicific" calculus for measuring pain and pleasure and evaluating the balance involved in any action. As Bentham notes in his Preface, the text was written (like Beccaria's) to serve "as an introduction to a plan of a penal code." [Bentham's work on "punishment" would culminate in his (in)famous plan for the "panopticon" prison in which the inside of prisoners' cells could be constantly viewed by (unseen) guards.]
In these texts Bentham gives us a definition of "utility" that will continue to be used, more or less, right down to the present (in microeconomic textbooks): "By utility is meant that property in any object, whereby it tends to produce benefit, advantage, pleasure, good, or happiness [ . . . ]." As far as economics goes, the technical use of this concept has involved its reformulation in ever more elaborate mathematical terms.
Cournot was very conscious of being an innovator in the application of mathematics the the subject of political economy. So much so that he felt the need to justify his actions. In preface to his Mathematical Principles of Wealth (1838), he comments on the novelty of his project:
This is a plan likely, I confess, to draw on me at the outset the condemnation of theorists of repute. With one accord they have set themselves against he use of mathematical forms, and it will doubtless be difficult to overcome today a prejudice which thinkers, like Smith and other more modern writers, have contributed to strengthen.This said, Cournot is also quite explicit about the limits of his project. He has not, he explains in his "Preface", "set out to make a complete and dogmatic treatise on Political Economy; I have put aside questions, to which mathematical analysis cannot apply, and those which seem to me entirely cleared up already." In Chapter IV of his essay - that I take up below - he reiterates this choice:
[. . .]
most authors who have devoted themselves to political economy seem also to have had a wrong idea of the nature of the applications of mathematical analysis to the theory of wealth. They imagined that the use of symbols and formulas could only lead to numerical calculations, and as it was clearly perceived that the subject was not suited to such a numberical determination of values by means of theory alone, the conclusion was drawn that the mathematical apparatus, if not liable to lead to erroneous results, was at least idle and pedantic. But those skilled in mathematical analysis know that its object is not simply to calculate numbers, but that it is also employed to find the relations between magnitudes which cannot be expressed in numbers and between functions whose law is not capable of algebraic expression.
[. . .]
I propose to show in this essay that the solution of the general questions which arise form the theory of wealth, depends essentially not on elementary algebra, ubt on that branch of analysis which comprises arbitrary functions, which are merely restricted to satisfying certain conditions. As only very simple conditions will be considered, the first principles of the differential and integral calculus suffice for understanding this little treatise."
[. . .] we shall not accompany most speculative writers back to the cradle of the human race; we shall undertake to explain neither the origin of property nor that of exchange or division of labor. All this doubltless belongs to the history of mankind, but it has no influence on a theory which could only become applicable at a very advanced stte of civilization, at a period when (to use the language of mathematicians) the influence of the intial conditions is entirely gone."In short, he passes from the sweeping historical and ethical vision of Adam Smith's Wealth of Nations and Theory of Moral Sentiments or of Karl Marx's Capital to a much narrower focus. While his choices as to what to set aside are clear enough, his logic is questionable - while the intial conditions of mathematical models may cease to have relevance to the evolution of its variables, the continual reproduction of a particular kind of society envitably involves the continual reproduction of those characteristics which constituted it at the outset. Ignoring the circumstances of that consitution, therefore, may well involve ignoring central and vital aspects of the on-going social relations that need to be analyzed.
Within the context of our one-semester survey, we will look at only one of the tightly defined issues that Cournot takes up: market demand.
As we have seen, observers and theorists of markets had been discussing the "demand" for commodities in various ways for years. But none that we have examined formulated their ideas in terms of a demand schedule or a demand function. Cournot did, and relatively unknown as he was, it would be decades before it was done again.
In Chapter four of his work Mathematical Principles of Wealth (1838), Cournot begins his discussion of demand with the "single axiom" that "each one seeks to derive the greatest possible value from his goods or his labour." Later, in the 1870s such as statement about "value" would be interpreted in terms of utility; but in 1838 Cournot meant something simpler and more direct: each would try to get the most for his money. His notion of demand was intuitive: if things were cheaper, people would buy more of them, at least up to a point: "The cheaper an article is, the greater ordinarily is the demand for it."
This intuition he proposes should replace the usual formulation of his time, in which price is imagined to be in "direct ratio" to the quantity of goods demanded. That is to say instead of thinking about the relationship between price and demand in terms of p = f(D) where demand is the independent variable, he proposes D = f(p) where it is price that determines the quantity demanded. This, of course, is the contemporary conception of the functional relationship and it is usually thought that dD/dp < 0.
The first thing that strikes the modern eye, is that Cournot is not clearly distinguishing between the quantity demanded at a given price and demand, i.e., the whole schedule, or function, of quantities demanded at various prices. For if we interpret the statement that "the price of goods is in . . . the direct ratio of the quantity demanded" as an assertion that if "demand" (the whole function) rises (or shifts to the right), then (assuming an upward sloping supply curve) price will also rise, then Cournot's critique fails. Consistent with his way of looking at this issue is the way he equates demand with sales: "The sales or the demand (for to us these two words are synonymous, and we do not see for what reason theory need take acount of any demand which does not result in a sale) the sales or the demand generally, we say, increases when the price decreases." Even though Cournot goes on to develop the demand curve, we see again that in this discussion he has no sense of the demand curve as a set of possible quantities demanded encountering a supply curve as a set of possible quantities supplied and the level of "sales" being determined by the point of encounter and the dynamic of equilibrium and "clearing" in the marketplace.
Nevertheless, Cournot does set out the functional form of demand still used today: D = F(p) where D = demand (or annual sales) and p = the price of the commodity in question. It is this function that he calls the law of demand
Arguing that an attempt to derive such a curve from a theory that would include habits and customs, use-values, income distributions and moral causes would be impossible (exactly the later derivation from utility), he argues that one might more reasonably seek to know the actual form of such curves through the study of observations of what happens in the market.
One might, he says, draw up "a table of the corresponding values of D and p; after which by the well-known methods of interpolation or by graphic processes, an empiric formula or a curve can be made to represent the function in question." What he has in mind here is the kind of work that was later undertaken in econometric studies that, for instance, used statistical methods such as regression analysis to find curves that best "fit", or could represent, a set of observations. Given "the difficulty of obtaining observations of sufficient number and accuracy", however, he was dubious of the possibilities of such studies and went on to argue that the formalism of mathematical specification can still produce useful results.
In applying such formalism, he argues that it is reasonable to assume - in a fully developed capitalist society with many consumers and many markets - that such a demand function F(p) would be continuous. Such continuity implies that we can talk about small "marginal" changes, and that opens the door to the calculus:
If the function F(p) is continuous, it will have the property common to all functions of this nature, and on which so many important applications of mathematical analysis are based : the variations of the demand will be sensibly proportional to the variations in price so long as these last are small fractions of the original price.
He can then proceed to discuss the shape of the demand function D = F(p) and to argue that it is most likely to be convex to the origin --its familiar shape-- over most likely and reasonable ranges. He discusses this in terms of the maximization of the value of total sales pF(p), i.e., the size of the market, at various prices and uses first and second order derivatives to identify maxima and minima. The only unfamiliar aspect of his representation of the function is the reversal of axis with price being on the horizontal and quantity on the vertical.
It is worth noting that along the way Cournot also recognizes and discusses variations in what would later be called the elasticity of demand - that is to say the degree of sensitivity of changes in demand to changes in price.
The demand might be in the inverse ratio of the price; ordinarily it increases or decreases in much more rapid proportion - an observation especially applicable to most manufactured products. On the contrary, at other times the variation of the demand is less rapid.Formally, he puts this insight this way:
Suppose that when the price becomes p+dp [where dp = a small change in p], the annual consumption as shown by statistics, such as customhouse records, becomes D-dD [where dD = a small change in D]. According as dD/dp < or > D/p the increase in price, dp, will increase or diminish the product pF(p), and consequently it will be known whether the two values p and p+dp fall above or below the value which makes the product under consideration a maximum."As Clarence Morrison has recently pointed out in the Summer 2003 issue of the Atlantic Economic Journal: if we divide dD/dp < or > D/p by D/p, we get the familiar formula for elasticity: (dD/D)/(dp/p) < or > 1. This discovery of elasticity is usually attributed to Alfred Marshall who gave it that name at a much later date.
Here we are clearly a long way from such preoccupations of the classicals, and their critics, as to whether working class "demand" is enough to keep them alive, or enough to prevent the economy from plunging into "glut" depression and falling wages.
Like Cournot, Jevons is aware that his proposal for a "general mathematical theory" is a departure from the general practices of political economists and called for some explanation and caveats. Early in his essay he notes: "It is not to be supposed, however, that because economy becomes mathematical in form, it will, therefore, become a matter of rigorous calculation. Its mathematical principles may become formal and certain, while its individual data remain as inexact as ever." Later, after spelling out his main formal propositions, Jevons adds: "Of course such equations as are here spoken of are merely theoretical. Such complicated laws as those of economy cannot be accurately traced in individual cases. [. . .]We must think under the forms of these laws in their theoretic perfection and complication; in practice we must be content with approximate and empirical laws." Beyond such methodological observations, in his final "Concluding Remarks" in his book Jevons denounces - with as much vigor as any passage from Karl Marx - "the noxious influence of authority", attacking the tendency of economists to turn any one set of ideas into dogmatic creed while eschewing any critical examination of those received ideas. In his day the creed he was questioning was the "orthodox Ricardian school", but his call for critical inquiry might well have targeted the narrow "neoclassical price theory" that would be built - in part - on his own ideas, or the later "neoclassical synthesis" of the post-WWII period, or the current neoliberal economist worship of the market.
In his "General Mathematical Theory" essay, Jevons situates himself firmly within the utilitarian tradition with two assertions: first, "A true theory of economy can only be attained by going back to the great springs of human action -- the feelings of pleasure and pain." (Jevons even attempts to recast Bentham's exploration of the dimensions of pleasure and pain in mathematical terms: "As several writers have previously remarked, feeling have two dimensions, intension and duration. A pleasure or a pain may be either weak or intense [. . .] it may also last a long or a short time. [. . .] Thus, if the duration of a feeling be represented by the abscissa of a curve, the intensity will be the ordinate, and the quantity of feeling will be the area.")
And then the other utilitarian assertion: "A second part of the theory proceeds from feelings to the useful objects or utilities by which pleasurable feeling is increased or pain removed. An object is useful when it either affects the senses pleasurably in the present moment, or when, by foresight, it is expected that it will do so at some future time."
But his insight around which he would craft his theory -and what makes him one of the fathers of neoclassical economics- was contained in the following passages:
"8. Amount of utility corresponds to amount of pleasure produced. But the continued uniform application of au useful object to the senses or the desires, will not commonly produce uniform amounts of pleasure. Every appetite or sense is more or less rapidly satiated. A certain quantity of an object received, a further quantity is indifferent to us, or may even excite disgust. Every successive application will commonly excite the feelings less intensely than the previous application. The utility of the last supply of an object, then, usually decreases in some proportion, or as some function of the whole quantity received. This variation theoretically existing even in the smallest quantities, we must recede to infinitesimals, and what we shall call the coefficient of utility, is the ratio between the last increment or infinitely small supply of the object, and the increment of pleasure which it occasions, both, of course, estimated in their appropriate units.
9. The coefficient of utility is, then, some generally diminishing function of the whole quantity of the object consumed. Here is the most important law of the whole theory."
Here are Cournot's infinitesimal changes, not in demand, but in utility, and they too are diminishing. Jevons' "coefficient of utility" is, of course, the derivative dU/dx of the function U = f(x) where U represents the utility gained from the consumption of commodity x. And dU/dx < 0, as x increases marginally, the utility gained decreases marginally.
Please note that implied in this vision is possibility of a precise calculation of amounts of utility. Jevons' utility function is not about "more or less" utility, but involves measuring precise amounts more or less. This requires an actual unit of measurement, a quantum, a util. Jevons theory involves "cardinality" and is framed in terms of "cardinal utility," i.e., his theory implies that we can say precisely how much more or less utility one derives from the consumption of a given unit of good x.
When he turns from this utility calculus to labor, market exchange and capital Jevons bypasses supply and demand to argue that we can understand all these things by investigating the evolving relation between marginal gains in utility and marginal losses. For example, in the case of labor he argues:
Thus labor will be exerted both in intensity and duration until a further increment will be more painful than the increment of produce thereby obtained is pleasurable. Here labor will stop, but up to this point it will always be accompanied by an excess of pleasure.While his logic is clear enough, his "excess of pleasure" is clearly of a peculiar sort. To speak of "a further increment of labor being more painful than the increment of produce thereby obtained" implies that the labor itself is "painful" or unpleasurable. And in the usual situation of capitalist society where the "increment of produce" that comes to the worker is mediated by the wage - paid after work has been carried on for some time - the only possible sense of saying, as he does, that up to the point of stoppage work will be "accompanied by an excess of pleasure" is that the anticipated future pleasure to be derived from the wage is greater than the immediate pain involved in working!
In the case of market exchange he argues that people exchange with a view to maximizing utility and therefore will engage in exchange if and only if they will gain utility by so doing. Exchange will take place he argues until "the increments of utility lost and gained at the limits of the quantities exchanged" are equal. This point can be calculated by knowing the exchangers respective utility functions and the possibilities of exchange. This theory he argues can be generalized from a two person - two commodity exchange to any number of exchanges whether domestic or foreign.
Finally, with respect to capital, he defines it as "consisting of all useful objects which, supplying a laborer's ordinary wants and desires, enable him to undertake works the results of which will be deferred for a greater or lesser space of time". In other words, in a peculiar characterization that in some ways harkens back to the classics, Jevons calls capital "nothing but the maintenance of the worker".
Nevertheless, he does understand capital as a separate factor of production whose return (interest) can be calculated as equal to its marginal productivity: "thre rate of interest is always determined by the ratio which a new increment of produce bears to the increment of capital by which it was produced." And the unavoidable diminishing marginal returns from the employment of capital, he argues, explain "the known fact that the interest of capital always tends to fall very rapidly as its amount increases, in proportion to the labor it supports."
In the years following 1862 Jevons refined his theory of utility and demand, finally publishing his Theory of Political Economy in 1871 that includes Chapter III on "The Theory of Utility." In that chapter, general mathematical formulations of the 1862 essay are given more precise functional and graphic forms. Like Cournot with respect to demand, Jevons assumes the utility function to be continuous and that marginal changes in utility are a function of marginal changes in the quantity of the relevant good consumed. So that if U = f(x) then Jevons calls the limit of Δu/Δx as x approaches zero du/dx, or "the degree of utility." The marginal utility of the final "small, or infinitely small, quantity [added] to the existing stock," Jevons calls "the final degree of utility." He now states the "general law" of diminishing marginal (final) utility as: "the degree of utility varies with the quantity of commodity, and ultimately decreases as that quantity increases."
Jevons is quite explicit in his debt to Bentham, and prefaces his formal treatment with a discussion of "the laws of human want", a most general and philosophical beginning. He favorably quotes Nassau Senior' "Law of Variety" that emphasizes how humans are not merely preoccupied with quantity but also seek qualitative variety (as Jevons would insist the marginal utility of each consumption falls with the increase in its consumption). It is of some interest that Jevons speaks of the "insatiability" of diversity. As neoclassical theory is formalized an assumption of insatiability will be made in which it is supposed that people always want more, that utility rises as the quantity consumed rises, even if the marginal utility of each consumer good diminishes.
After a long quote from Banfield on the hierarchy of human wants, Jevons highlights the relativity of utility to the desirer. Things have no "intrinsic" utility, he notes, but only in so far as they relate to the desires they satisfy. Now while this implies that the same good may have quite different "utilities" to different people (an automobile provides transportation to one, status to another), Jevons is mainly concerned to pave the way for his discussion of declining marginal utility. So his discussion of the "variation of utility" mainly concerns the consumption of a single individual and he uses food to illustrate the point that the total utility (or what he calls the "intensity" or "degree" of utility) of each portion declines. (We might imagine an exception to this when the utility of food derives from display and the status it confers rather than from the nutrition or pleasue of its eating. Think of some wedding receptions or banquets in which not only the variety but the quantity of food is maximized to display the wealth of the sponsor and impress the invitees.)
In this further mathematical development of his theory that involves plotting utility curves in two-space, Jevons also recognizes that not only does marginal utility decline, but it may become negative and therefore he must allow for "disutility" or the curve passing through the horizontal axis into the lower right hand quadrant. In other words beyond a certain point (of intersedtion) consuming more causes a loss of utility. Jevons even went so far as to label such commodities "discommodities" --ones that caused "inconvenience or harm." So as we follow the utility function downward to the right we pass from "diminishing marginal utility" through the intersection point with the horizontal axis (the point of "inutility" in Jevons words) into the realm of "disutility." Assuming the curve continues to fall we would have to say we are in a realm of "growing marginal disutility." In the spirit of Jevons' own example of the utility of more and more food, we might imagine the case of consuming something like slices of green apples where eating a few might provide some marginal satisfaction/pleasure, though at a declining rate, at some point eating one more would do nothing for you and after that eating more would make you sick (give you pain, literally).
Jevons ends his chapter with some reflections on the time factor in utility and with the differences among actual utility, prospective utility and potential utility. The former discussion explains why a static analysis is sufficient; the latter is reminiscent of Bentham's efforts to grasp the complexity of utility.
This vision has parallels with Marx's analysis of the investment process in that he saw money expended as wages that hired labor power as part of capital (the other part being the money expended on the means of production). But the parallel continues when Jevons explicitly argues that the consumption of the worker, the consumption that sustains the worker during the production process in which new commodities are produced, is part of the capitalization process. The self-reproduction of the worker is part of the self-reproduction of capital. This is something that Marx ignored in his discussion of investment but grasped quite explicitly in his "reproduction schemes" that sketched the self-reproduction of capital as involving the self-reproduction of all of its components, including labor (or labor power in Marx's terms).
In the course of discussing such maintenance of labor, Jevons takes up the issue of "capital invested in education". In his discussion of this we can see both the extent and the limits of his understanding. He argues that feeding a child is mere consumption (for the pleasure of both the child and the parents) whereas feeding students who are going to school when they might be earning income amounts to a capital investment. The foregone income economists call the "opportunity cost" of schooling, but for Jevons the main point is that the money spent on schooling is money invested to bring about a greater future wealth. If we expand our vision from that of Jevon's individual, or family, to the economy as a whole, then it would seem obvious - as so many pro-natalist and educational programs testify - that all of the expenses of rearing children destined for the labor market and future work should fall, at least to some degree, into Jevons' category of "capitalisation." They are the expenses of producing "human capital" -in neoclassical jargon- or "labor power" in Marxist jargon. In both cases they are investments in future production and thus "capital" expenditures.
Be that as it may, Jevons was convinced that both capital and labor were necessary for production and that the capitalist society of his time was wracked by unnecessary conflict between the two. As to the causes of the conflict, he allocated blame to both sides, to both employers who are "too apt to resent and refuse every demand . . . as an infringement of [their] rights of judgement and management," and to labor that all too often ignored the ways in which their struggles harmed themselves, other workers and the economy as a whole. That said, Jevons' notion of how to reduce such conflicts was quite one-sided. The "greatest evil", he felt, lay with the workers and they must be educated to the point of understanding which kinds of struggles were legitimate and which were not.
In a fascinating lecture on "The Importance of Diffusing a Knowledge of Political Economy" that he gave to a group of school teachers, Jevons argued for the instruction of workers to begin at a very early age - "almost as soon, in fact, as a boy has acquired the power of reading with facility." The urgency for such education was two-fold, first it was aimed at reducing future strife such as strikes and violence, the second, the prospect of the working classes of England obtaining the right to vote raised the spectre of workers successfully imposing laws such as "minimum wage" or "living wage" laws that would violate the "natural laws that govern the relations of capital and labor" and disrupt the economy, undermine liberty and drive capitalist to invest abroad.
In that lecture, and in a subsequent invited lecture given to "Trades Unionists' Political Association" in Manchester, Jevons spelled out in some detail which kinds of struggles he though legitimate for union actions and which were not. Among those actions he felt were legitimate were: the mutual aid activities of "friendly societies", the demands for shorter working hours, and demands for greater safety, santitation, etc., in working conditions. In the case of working hours, Jevons clearly thought that workers were perfectly justified in wanting to take part of the fruits of rising productivity in less work - his only caveat was that such reductions should be limited by the rise in productivity and not exceed it - which would have the effect of cutting into profits, capital investment and future accumulation.
Those actions that he felt were illegitimate were struggles for wage increases that exceeded productivity increases and worker resistance to the introduction of productivity raising technological changes, e.g., the introduction of new machinery. These two were clearly related. After all, new technology made possible the increases in productivity that could pay for higher wages and less work, so that he saw a clear contradiction between efforts to raise wages while resisting the intoduction of new machines. Why Jevons didn't seem to recognize how successful struggles over things like "the hours of labor, health, safety, comfort and moral conditions of the operative" might also raise costs more rapidly than productivity and thus undermine profits, is not at all clear.
However, Jevons' arguments against worker attempts to raise wages went beyond the issue of productivity. He also pointed out that because some workers were better than others in getting organized and in imposing wage increases, their very success had undesireable side effects. First, the success of some in the presense of the failure of others accentuated the wage hierarchy pitting workers against each other. Second, wage increases were often passed along to consumers through higher prices so that some workers gained at the expense of others. Third, any general success by workers in raising wages (beyond productivity increases) would result in capitalists closing factories at home and investing abroad where wages were lower and workers more controllable. "Capitalists", he warned, "will gradually withdraw their capital from home employment and invest it in the colonies, United States and foreign countries."
Such undesireable effects, Jevons thought, could only be avoided by workers restraining their struggles; but such restraint could only come from thorough understanding and such understanding could only come from early and thorough schooling of workers in political economy. (Other methods of education - such as "cheap treatises" and short stories - he found wanting because they reached few workers.)
Finally, Jevons evoked to workers the possibilities of saving and of their collective investment of savings into the formation of co-operative socities so as to become themselves capitalists. Of course, he also recognized that "there are many branches of trade, however, in which such great capitals are required that you can hardly be able to undertake them safely without the aid of capitalists." In such cases collaboration between capitalists and workers was, he felt, necessary and could be managed through Boards of Conciliation in which workers and capitalists could minimize misunderstanding and maximize cooperation.
The language that we use today in neoclassical is largely due to Marshall's sifting and evaluating the various ways of talking about the new insights. In the book The Economics of Industry written with his wife Mary Paley and published in 1879, they argue that the term "political economy" should be replaced by "economics":
"The nation used to be called 'the body Politic.' So long as this phrase was in common use, men thought of the interests of the whole nation when they used the world 'political'; and then 'political economy' served well enough as a name for the science. But now 'political interests' generally mean the interests of only some part or parts of the nation; so that it seems best to drop the name 'Political Economy,' and to speak simply of Economic Science, or more shortly, Economics." [emphasis in original]Marshall goes on to argue that:
"Economics is a science because it collects, arranges, and reasons about one particular class of facts. A science brings together a great number of similar facts and finds that they are special cases of some great Uniformity which exists in nature. It describes this Uniformity in a simple and definte statement, or Law." A Law of Science states that a certain result will be produced whenever a certain set of causes are present." [emphasis in original]
Beyond this re-titling of the field of study - which opened the way for later economists to lay claim to the same supposed "objectivity" "value-free" character as the natural sciences - Marshall recast many earlier terms. As you will see in the text and footnotes of the third chapter of the Third Book of his Principles of Economics (1890) some terms, such as Jevon's "final degree of utility" or "coefficient of utility" are set aside in favor of today's more familiar "marginal utility" and "declining marginal utility." Instead we will leap ahead to what I think of as the great political "cleansing" of neoclassical theory in the work of Pareto and Hicks.
The political problem was the following. If the utility individuals obtain from the consumption of goods can be precisely measured (say as amounting to so and so many 'utils'), then the loss to them can equally be measured and therefore the social impact of redistributing income or wealth can be measured. If you accept the utilitarian principle that the object of economic policy should be the greatest happiness of the greatest number, then income and wealth should be redistributed until the marginal utility of a receiver of income or wealth is equal to the marginal loss of the loser. In other words if the marginal utility of a dollar to a rich person is, say, one util and that to a poor person is 1000 utils, then the overall utility of society can be increased by redistributing that dollar from the rich person to the poor person. Any such redistribution, however, would strike at the heart of the income and wealth hierarchy upon which capitalism is based and which it reproduces daily. The economists we have examined created a theory that could be used to attack the existing hierarchy of capitalism.
And the theory was used for precisely such purposes. It was appropriated by socialists to argue for the kind of redistribution of wealth that John Hobson had called for in his writings against crisis and against imperialism. The logic was different, this appropriation of theory was not concerned with such things as aggregate demand or the need for foreign markets but the conclusion was the same: an attack on the wealthy and a demand for the redistribution of income and wealth downward.
Such an appropriation of the theory, at the turn of the century amidst conflicts between labor and capital at home and over imperial adventures abroad, a time that saw the emergence of the Second International, the first Russian Revolution of 1905-07 and the Mexican Revolution of 1910 led, not surprisingly, to a revision of the theory that did away first, with the cardinality upon which such arguments were based and secondly, with utility itself and all of its philosophical concerns with the general happiness of humankind.
In first chapter of his book Value and Capital (1939) John Hicks traces the intellectual groundwork for this fundamental shift. He reminds us of Pareto's appropriation of Edgeworth's concept of three dimensional utility surfaces and two dimensional indifference curves that permit an elegant analysis of consumer choice on the basis of a fairly rigorous set of assumptions about infinite divisibility, continuity and convexity. Assuming positive utility from all goods, movement along an indifference curve must involve offsetting loses for gains and the negative slope of the curves will be the ratio of marginal utilities of the goods involved. The result is a "map" of an infinite number of indifference curves sloping downward toward either axis and convex to the origin.
When the possibilities open to a consumer (given by income and the relative prices of the commodities) are combined with with this map of indifference curves the process of utility maximization will lead the consumer to choose that combination of goods that gives the highest maximization income allows and that combination will be represented by a position of tangency between the consumer's "budget possibilities" and some indifference curve.
The exciting thing about this, from Hick's point of view, is that Pareto's formulation allows us to ignore utility all together, including the political sticky issue of cardinality (though he doesn't put his argument in political terms). Although the indifference curves were originally derived from a three dimensional space that included utility, he points out, they need not be. All that is required for their existence is a rank ordering of consumer preferences among various baskets of goods. This formulation "only tells us," he writes, that the individual prefers one particular collection of goods to another particular collection; it does not tell us, as the utility surface purports to do, how much the first collection is preferred to the second. Cardinality has been escape and "ordinality" substituded for it. This move, Hicks argues, is a "conclusion of wide methodological importance." Having cleansed the theory of utility, Hicks then sets out on what he calls a "purge" of all concepts derived from it, including marginal utility, dimenishing marginal utility and ratios of marginal utilities. In their place he will argue for preferences, marginal rates of subsitution and diminishing marginal rates of substitution.
Hicks is quite explicit about his own desire to set aside the remnants of utilitarian philosophy in marginalist theory, though he does not tell us why he desires this escape - other than an invocation of Occam's Razor and the principle that the fewer assumptions the better.
But the political advantages are clear enough. With utility and cardinality banished from the theory there is no longer any grounds for arguing for such politically touchy programs as income redistribution. You might suspect that the transference of income from the rich to the poor might raise general social welfare, but there is nothing in the theory to support this. In fact, by removing not only utility but ANY common factor in the determination of preference the revised theory essentially makes all interpersonal comparisons impossible including estimates of relative losses and gains. If you redistribute income from the rich to the poor, the rich persons' budget possibilities will contract, presumably moving them to lower indifference curves, while those of the poor expand, presumably moving them to higher indifferenc curves, but no quantitative comparison of the two changes is possible. All that can be said is that the new situation will be less prefered by the rich and more prefered by the poor.
In chapter two of his book, Hicks continues his "purging" of utility by demonstrating how the downward sloping demand curve can be derived from the preference theory he delineated in the previous chapter. The chapter is interesting methodologically not so much from the specifics of the derivation - which should be familiar to you from a course in microeconomics - but from the process by which he justifies his derivation. At virtually each step of the way he shows which assumptions have to be made in order to obtain the desired end results. In the process you can see a theory being constructed to justify a desired conclustion. This is no exercise in deduction from first principles, but a crafting of theory for a specific end. In the proces there is an amusing twist in chapter one when Hicks trys to find a justification for his assumptions that goes beyond reference to experience, i.e., by making another assumption (that kinks can be neglected), yet that further assumption is justified only by reference to experience.