A Companion to the History of Economic Thought - Chapter 16 pps

C H A P T E R S I X T E E N English Marginalism: Jevons, Marshall, and Pigou Peter Groenewegen 16.1 INTRODUCTION English marginalism embraces a group of economists, active in the late ni

C H A P T E R S I X T E E N English Marginalism: Jevons, Marshall,

and Pigou

Peter Groenewegen

16.1 INTRODUCTION

English marginalism embraces a group of economists, active in the late nine-teenth and early twentieth centuries, who used the marginalist method for analyzing economic questions relating, in the first instance, to the theories of value and distribution, and later to a wider range of economic questions in the theories of money, fluctuations, income determination, and growth Jevons, Marshall, and Pigou were major contributors to this “marginal revolution” in England, but not the only important pioneers Edgeworth and Wicksteed (both contemporaries of Jevons and Marshall) and Marshall’s Cambridge students, of whom Pigou was among the first, are also important The “marginal revolution” had a well-known international dimension (see Horwitz, ch 17, this volume; Walker, ch 18, this volume) and its impact continues to reverberate in economic theory

The concept of marginalism itself has received surprisingly little attention

in the history of economics literature The word was slow to creep into the lan-guage of English economics As Howey (1972) indicates, it was first used by J A Hobson (1914, pp 174–5, 331–2) in a derogatory manner The original Palgrave

Dictionary of Political Economy did not have an article on “marginalism.” How-ever, Edgeworth’s second article on “margin” in its final paragraphs described the widespread use by mathematical economists of “margins” in economics in both its theories of consumption and production, to which “Jevons’ exposition forms

an admirable introduction [while their] fullest and most accurate exposition

is to be found in Prof Marshall’s Princ of Econ.” (Edgeworth, 1894, p 691) The later Encyclopaedia of the Social Sciences (Seligman, 1948 [1930] ) likewise

contains no article specifically devoted to “marginalism.” The same applies to

the New Palgrave Dictionary (Eatwell et al., 1987), which contains articles on

“marginal and average cost pricing,” “marginal efficiency of capital,” “marginal productivity theory,” and “marginal utility of money” as well as an article on

“marginalist economics.” The last acknowledges “the almost total domination that marginalist economics has enjoyed for about a century” without igniting

the “spark of an ultimate truth” (Campus, 1987, pp 321–2) The second quoted

remark is a comment from Sraffa (1926, p 535), foreshadowing his prelude to a critique of the dominant marginalist method (Sraffa, 1960, esp p v)

This chapter is basically concerned with Jevons’s application of the marginal method as one of its early pioneers; with Marshall’s masterful exposition of

mar-ginalism, especially in his Principles of Economics; and with its elaboration by

Marshall’s official Cambridge successor, Pigou The final section briefly refers to three other leading English economists of the period – Edgeworth, Sidgwick, and Wicksteed – and in addition raises selected post-Pigovian aspects in the history

of English marginalism

16.2 WILLIAM STANLEY JEVONS (1835–82)

Jevons was born in Liverpool, the ninth child in a large, middle-class family

He studied chemistry and engineering at London University in the early 1850s without taking his degree In 1853 he went to Sydney (Australia) as assayer

of its new mint, staying there for six years On his return to London, he completed his degree He also started working on a new, mathematical theory

of political economy, on which he read a paper to Section F of the British Association for the Advancement of Science (Jevons, 1911a [1862] ) Unlike his

A Serious Fall in the Value of Gold (1863) and his subsequent book, The Coal

Question (1865), this raised little interest His major claim to intellectual fame,

Theory of Political Economy, was published in 1871 (2nd edn., 1879) It elaborated his mathematical theory of political economy, based predominantly on the pleasure/pain (utility/disutility) principle and employing the marginalist method

In 1862 he was elected Fellow of the Royal Society He was appointed Pro-fessor of Political Economy at London’s University College in 1876, a post from which he retired in 1880 to have more time for writing However, he died

early from a swimming accident in 1882 In 1875 Jevons published Money and

the Mechanism of Exchange , and in 1882 The State in Relation to Labour appeared posthumously Volumes of essays appeared in 1883 (Methods of Social Reform) and 1884 (Investigations in Currency and Finance) Although Jevons’s work as an

economist was wide-ranging (for good surveys, see Black, 1981; Peart, 1996), this essay concentrates on his marginalist treatment of value, exchange, labor, and capital

Jevons’s (1911a [1862] ) “Brief account of a general mathematical theory of value” was his first presentation of the marginalist method After arguing that “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, 1911a [1862], p 304)

which treated pleasure and utility as synonyms, Jevons focused attention on what he then called “the coefficient of utility,” described as a “generally dimin-ishing function of the whole quantity of the object consumed.” This, in his view, provided the basis for “the most important law of the whole theory” ( Jevons, 1911a [1862], p 307) Earlier (Jevons, 1911a [1862], p 306), the coefficient of utility had been defined as “the ratio between the last increment or infinitely small supply of the object, and the increment of pleasure which it occasions.” This incrementalist, or marginalist, perspective was then used to explain the application of labor, the extent of exchange and the rate of interest In the first, Jevons established that “labour will be exerted until a further increment will

be more painful than the increment of produce thereby obtained is pleasurable” (Jevons, 1911a [1862], p 307) The theory of exchange was then deduced “from the laws of utility” by postulating that equality of the “increment of utility lost and gained at the limits of the quantities exchanged” determined the extent of trade, and that such findings based on a two individuals–two commodities case were easy to extend to “any number of commodities” and therefore applic-able to “generalised trade [and] international trade” (Jevons, 1911a [1862],

pp 308–10) Finally, the 1862 paper showed the universal determination of the

rate of interest “by the ratio which a new increment of produce bears to the increment

of capital by which it was produced,” clearly foreshadowing a marginal productivity theory of interest

The Theory of Political Economy elaborated the 1862 propositions without

ex-tending the principles and methods by which they had been originally derived Hence it is difficult to treat Jevons’s development of a marginalist method and its application to utility as part of a “marginalist revolution” in 1871 that was tak-ing place simultaneously in two other European countries The eight chapters of

Theory of Political Economy reveal this formal similarity of contents with the 1862 paper After an introduction (ch 1), Jevons deals with the theory of pleasure, pain, and utility (chs 2 and 3), exchange (ch 4), labor, rent, and capital (chs 5–7), and concluding remarks (ch 8) The book displays its marginalist credentials explicitly without using the terminology Thus, halfway through the work, while treating the theory of labor, it proclaims, “as in the other questions of Economics, all depends upon the final increments” (Jevons, 1911b [1871], p 171), while the major elements of the subject are identified as “[u]tility, wealth, value, commod-ity, labour, land, capital” (Jevons, 1911b [1871], p 1) The book likewise reasserts

the mathematical quality of the subject, “our science must be mathematical, simply

because it deals with quantities” (Jevons, 1911b [1871], p 3), with the calculus as

the basic mathematical tool, assisted by some Euclidean geometry (the Theory is

extensively illustrated by diagrams) The problem of economics is defined (Jevons, 1911b [1871], p 37) as the maximization of pleasure (minimization of pain), as

is particularly clearly illustrated in Jevons’s treatment of resource allocation, including that of labor The quotation below neatly illustrates the marginalist nature of his approach to utility and his nonuse of words such as “margin” or

“marginalist” therein:

We shall seldom need to consider the degree of utility except as regards the last increment which has been consumed, or, which comes to the same thing, the next increment which is about to be consumed I shall therefore commonly use the

expression final degree of utility, as meaning the degree of utility of the last addition.

(Jevons, 1911b [1871], p 51)

This “degree of utility” is defined as a diminishing function of the quantity of

the commodity held, so that “the degree of utility decreases as that quantity

in-creases” (Jevons, 1911b [1871], p 53) Optimum allocation in consumption then requires equalization of the final degree of utility of the commodities consumed (Jevons, 1911b [1871], p 61), exchange equilibrium the equality of the price and the utility ratios (Jevons, 1911b [1871], p 95), optimum duration of labor supply the equation of marginal pain (disutility) and marginal benefit (utility) of that labor (Jevons, 1911b [1871], p 173), and the optimum allocation of capital the equalization of interest with the “advantage [in terms of product] of the last increment of capital [employed]” (Jevons, 1911b [1871], p 256) Jevons (1911b [1871], pp 212–13) also explicitly admitted that versions of the classical theory of rent, particularly those of James Mill (1821) and McCulloch (1838), were pieces of economic analysis using the incremental method In this respect, Keynes (1972 [1936], p 109) neatly identified Jevons’s contribution as belonging “to the group

of economists whose school of thought dominated the subject for the half-century after the death of Mill in 1873.”

Jevons’s economic writings encompass more than the marginalist theory of value and distribution He made interesting contributions to monetary and busi-ness cycle theory, and to the economics of energy resource scarcity, which has a

very a contemporary feel about it The last issue was raised in his The Coal

Question, which analyzed the implications of predictable shortages in the “pres-ent great supplies of coal,” then “the material source of energy” for Great Britain, underpinning its great industrial strength that was so successfully built on steel (Jevons, 1906 [1865], pp 2–3) Jevons’s treatment raised geological as well

as economic aspects of the problem (impacts on costs, prices, and imports, economizing the use of coal, and finding substitutes for coal) His conclusion emphasized the “momentous choice” facing the policy-maker “between brief but true greatness” from using available coal resources at a rate dictated by current technology, or a “longer [period] of continued mediocrity” if scarce coal resources were tightly rationed and industry was thereby effectively starved of its major energy source (Jevons, 1906 [1865], p 460) Although marginalist methods could have been used to advantage in explaining this argument, Jevons did not do so Neither did Jevons’s writings on monetary and the associated business cycle

theory, posthumously collected by H S Foxwell as Investigations into Currency

and Finance (Jevons, 1884) This volume contains various versions of his famous

sunspot theory of the cycle, as well as his 1863 monograph, A Serious Fall in the

Value of Gold Essays on social reform (Jevons, 1883) reveal Jevons’s broad social concerns, ranging from the use (and abuse) of free public libraries and museums,

to that of employing married women in factories, cruelty to animals, and the drink question, demonstrating that the noted marginalist theorist was also a

wide-ranging social scientist The splendid essays on Cantillon and on the future

of political economy appended to his unfinished and posthumously published

Principles of Economics (Jevons, 1905) show his gifts as a historian of economics

16.3 ALFRED MARSHALL (1842–1924)

Born in Bermondsey (London) in 1842, Marshall was seven years younger than Jevons He was educated at the Merchant Taylor School, where he gained a taste for mathematics: he subsequently completed the Cambridge Mathematical Tripos in 1865 as “second wrangler” (second in the first-class honors list), thereby securing a Fellowship at St John’s College He then gradually switched to the moral sciences, concentrating on economics from the early 1870s His first

book, Economics of Industry (1879, 2nd edn 1881) was written jointly with his

wife (a former student, whom he had married in 1877) That same year he privately published material on pure theory (Marshall, 1975b [1879] ) After holding academic positions at Bristol and Oxford, in 1884 Marshall became Professor of Political Economy at Cambridge He retired in 1908, when his student Pigou

(discussed below) succeeded him Marshall’s major book, Principles of Economics,

appeared in 1890 (eighth, definitive, edition in 1920; reprinted in 1961) During

his retirement, he published supplementary volumes (Industry and Trade, 1919;

Money, Credit and Commerce, 1923) as partial substitutes for a second volume

of his Principles, which he never managed to complete.

Marshall is sometimes bracketed with Jevons (whose Theory of Political Economy

he reviewed – Marshall, 1925 [1872] ), Menger, and Walras as a participant in the marginal revolution This is misleading, if only because his initial price analysis ignores utility considerations (Marshall, 1975a [1871] ) He was, however, a very significant pioneer of marginal analysis following Cournot (1963 [1838] ) and von Thünen (1966 [1826–63] ), from whose work he had benefited at an early stage in his economic studies Another differentiating factor from Jevons was that Marshall never showed strong hostility to his classical predecessors, instead incorporat-ing part of their work within his own system This is therefore appropriately described as neoclassical, a genuine merger of old and new doctrines Marshall’s

work, including his Principles, also combined theory with much factual matter

drawn from personal empirical observations and painstaking study of empirical and historical work Again unlike Jevons, Marshall strongly disliked display-ing mathematics in his theory – banishdisplay-ing his mathematical economics to an

appendix and burying his diagrams in the footnotes of his Principles Through

his tremendous influence as a teacher, he founded the Cambridge school of economics His economic principles based on supply and demand analysis

endured until well after his death, the Principles surviving as a major university

economics text until after World War II Useful overviews of his economics are O’Brien (1981) and Whitaker (1987); Groenewegen (1995) presents a full-scale biography, including detailed assessment of his economic work

As a trained mathematician, Marshall immediately took to the marginal method

in analyzing economic problems, to which study of Cournot’s and von Thünen’s

economics had introduced him His early economic writings, published with editorial introductions in Whitaker (1975), clearly illustrate this Examples are Marshall’s stress on the “balancing of advantages” in the theory of decision-making as applied to money (Whitaker, 1975, vol I, pp 166–7) and his use of incrementalist diagrams for explaining the theory of rent (Whitaker, 1975, vol I,

p 240) By the early 1870s, sophisticated incremental analysis guided Marshall’s thinking on tolls, monopoly, and growth and enabled him to sketch a marginal productivity theory of distribution (Whitaker, 1975, vol II, pp 281–3, 284–5, 309–

16, 323–5) This mathematical economics was completely hidden in the initial textbook presentation of the theory (Marshall and Marshall, 1879) but fully dis-played in the privately printed pure theory of international trade and domestic value (Marshall, 1975b [1879], vol II, pp 117–81, 186–236)

At a higher level, much of this theory was carefully reworked for Marshall’s

magnum opus, Principles of Economics (Marshall, 1961 [1890] ) This had been ori-ginally intended as a two-volume work, of which the second, never completed, volume, was at one stage to contain “foreign trade, money and banking, trade fluctuations, taxation, collectivism [and] aims for the future” (Groenewegen, 1995,

p 407; its chapter 10 presents a detailed discussion of the long haul of the

Prin-ciples from 1881 to 1922) The first volume became a classic text on value, produc-tion, and distribution From the second edition onward, its structure emphasized the “general relations of demand, supply and value” (book V), and their use for

a theory of “the distribution of the national income” or of wages, rent, interest, and profit (book VI) These followed two preliminary books, the second of which was definitional, and two books that provided the foundations for the theory of value in the theory of demand (book III, “Of wants and their satisfaction”) and of supply or production (book IV, “The agents of production, land, labour, capital

and organisation”) In addition, later editions of the Principles contained 12

appendices (amounting to over 13 percent of the text) and a concise set of

math-ematical notes The extent to which Marshall polished his Principles can be seen

in the variorum edition painstakingly prepared by his nephew, C W Guillebaud (Marshall, 1961 [1890] ), the edition used in what follows

The preface to the first edition (reprinted in all subsequent editions) signaled the specific marginalist intent of the work, and provided a concise statement

of Marshall’s views on the essential role of mathematics in the elucidation of economic principles The relevant paragraph can be quoted in full, since it also provides a clear acknowledgment of Marshall’s mentors in marginal analysis and

of how he saw Jevons’s role in his economics education:

Under the guidance of Cournot, and in a less degree of von Thünen, I was led to attach great importance to the fact that our observations of nature, in the moral as in the physical world, relate not so much to aggregate quantities, and that in particular the demand for a thing is a continuous function of which the “marginal”* increment

is, in stable equilibrium, balanced against the corresponding increment of its cost of production It is not easy to get a clear full view of continuity in this aspect without the aid either of mathematical symbols or of diagrams The use of the latter requires

no special knowledge, and they often express the conditions of economic life more

accurately, as well as more easily, than do mathematical symbols; and therefore they have been applied as supplementary illustrations in the footnotes of the present volume The argument in the text is never dependent on them; and they may be omitted; but experience seems to show that they give a firmer grasp of many impor-tant principles than can be got without their aid; and that there are many problems

of pure theory, which no one who has once learnt to use diagrams will willingly handle in any other way.

* The term “marginal” increment I borrowed from von Thünen’s Der Isolirte Staat,

1826–63, and is now commonly used by German economists When Jevons’ Theory appeared, I adopted his word “final”; but I have been gradually convinced that

“marginal” is the better.

(Marshall, 1961 [1890], p x and n.1)

The “balancing of advantages” underlying both Marshall’s theory of consump-tion (as, for example, done by “the primitive housewife” in book III, chapter V; Marshall 1961 [1890], p 117) and his theory of production (for example, book V, chapter VIII; Marshall, 1961 [1890], p 405) was shown to be particularly fruitful soil for the application of the marginal method It was also crucial for the theory

of distribution, where Marshall emphatically indicated that the marginal

pro-ductivity theory applied the principle that “we must go to the margin to study

the action of those forces which govern the value of the whole” (Marshall, 1961 [1890],

p 410) This applied with equal force to use of the margin in profit maximiza-tion (where the condimaximiza-tion of equating marginal cost with marginal revenue is implied following the manner of Cournot) and use of consumer surplus for solv-ing problems pertainsolv-ing to social welfare The book abounds with exhortations

to “study the margin of profitable expenditure” (Marshall, 1961 [1890], p 432), and

portrays the resource allocation decision facing housewife and businessman as fundamentally the same, being designed to distribute their resources so “that they have the same marginal utility [benefit] in each use” (Marshall, 1961 [1890],

pp 358–9)

The economics of the Principles, it needs to be stressed, is not confined to the

presentation, and solution, of static allocation problems in consumption and production as a way for highlighting the use of marginal method in expositing the theory of value and distribution Just as Marshall was aware of the dangers

in over-reliance on mathematics in economics, so he realized the complexity for the theorist of grasping the realities of economic life In this quest, he held up biology as the mecca for the economist This was a sign that the dynamics and the evolutionary processes by which the economic institutions of markets, firms, competition, and productive organization gradually altered and adapted, needed both induction (observation and study of facts) and deduction (logical, including mathematical analysis) The laws of economics were analogous to laws of the tides, not to the laws of physics, as represented by the law of gravitation They referred

to tendencies and not to precise truths (Marshall, 1961 [1890], book I, ch III, esp

pp 29, 32–33) This principle of Marshall’s economics is evident in his specific treatment of virtually every economic question It enabled him to reconcile com-petition and increasing returns, to introduce issues of family, education, arbitration, trades unions, and custom into his discussion of wages and labor; and to blend

well-established insights into economic behavior from his favorite classical mentors (Smith, Ricardo, and J S Mill) with the tools of modern, marginalist reasoning It created a spirit of eclectic tolerance in his views on scope and method, which makes present-day study of his great text still eminently worthwhile

Marshall’s mixed methodology is even more strikingly visible in his Industry

and Trade, his last major completed book It provided a careful, comparative, study of industrial techniques and business organization in the major industri-alized countries of the early twentieth century, Great Britain, Germany, and the United States It is an exercise in realistic economics, with theory mixed in at all the appropriate places The book combines statics and dynamics, history and contemporary analysis, and examines competition as essentially a monopolistic phenomenon involving a finite and often rather small number of large firms, thereby rejecting the artificial construct of perfect competition so beloved by much contemporary theory A great deal of it remains a rich source for the economic history of the late Victorian and Edwardian eras, because it reflected

the insights gained from Marshall’s wanderjahre in factories (see Groenewegen,

1995, ch 8, esp pp 208–14) Its contents continue to be a source of inspiration to contemporary investigators of industrial organization It shows to perfection the two sides of Marshall’s skills in combining fact with theory

Marshall’s last book, Money, Credit and Commerce (Marshall, 1923) is little more

than a pastiche of early work, and as such remains of interest It brings together

in one book the pure theory of international trade and reflections on monetary problems and business fluctuations presented to Royal Commissions and other government inquiries It indicates what could have been, had less time been

spent on perfecting the Principles (see Groenewegen, 1995, ch 19) These later

vol-umes also provide a clear link with what was to happen in Cambridge economics during the decades after his death – the “imperfect competition revolution” in the theory of the firm as business organization, and the “Keynesian revolution”

in the theories of money, inflation, employment, and output

As was the case with Jevons, Marshall’s work in economics also covered social reform and practical policy matters In later life, he claimed that his personal concern with the problem of poverty for sections of the working class, and with members of what he termed “the residuum,” had initially sparked his interest in economic studies as a practical means of helping the lowest orders of society to gain access to the resources needed in order to enjoy a fruitful life Some of these

concerns are visible in the pages of the Principles, especially in the final chapter of its last editions Specific policy contributions were edited by Keynes as Official

Papers of Alfred Marshall These reprinted his evidence to Royal Commissions and

to other government inquiries, relating to monetary questions, to “depressions of trade and industry,” to taxation policy, and to international trade A supplement

to Marshall’s Official Papers (Marshall, 1996) provides easier access to Marshall’s

views on education as given to a Committee of Inquiry in 1880, substantial extracts from work for the Labour Commission (of which he was a member) that are directly attributable to him, as well as more monetary evidence and material on the fiscal question which (probably mistakenly) has been attributed to him The economics of these volumes is not easily summarized (see Groenewegen,

1995, ch 11) However, they more clearly reveal Marshall, the well-rounded economist and social scientist, competent to deal with a wide variety of policy issues not only in money, banking, the cycle, and trade policy to which Marshall (1926) drew attention, but also to education and, especially, to labor relations

in all its manifold aspects Like Jevons, Marshall was no narrow theorist of marginalist economics but a student of the subject as a whole, anxious to grapple with its various problems, human, social, and theoretical, using every conceivable method that assisted in the task

16.4 ARTHUR CECIL PIGOU (1877–1959)

Pigou was born in 1877, in Ryde, on the Isle of Wight He was Head of School at Harrow and won a scholarship for King’s College, Cambridge There he gained first-class honors initially in the History Tripos, and subsequently in the Moral Sciences, of which economics was then still a part It made him a Fellow of King’s, the Cambridge college with which Keynes was also associated He began

to teach economics in 1903, and succeeded Marshall as Cambridge Professor of Economics in 1908 Pigou did not retire until 1943 During World War I, he was

a conscientious objector but worked close to the front as part of an ambulance team during the long university summer vacations His wartime experiences may have turned him into the recluse that he became from the early 1920s (Pigou,

1952 [1939]; Johnson, 1978 [1960]; Collard, 1981, pp 105–10)

Pigou’s economics is now largely remembered for two things First, and negat-ively, he is remembered for his quarrel with Keynes over the theory of employ-ment, caused by Keynes’s devastating critique of Pigou’s own work on the subject

(Pigou, 1933) This is particularly visible in his hostile review of The General

Theory (Pigou, 1936), only partly retracted in a subsequent appraisal (Pigou, 1950) His analytic contribution to the real balance effect, sometimes called the Pigou effect, greatly assisted the “neoclassical synthesis” between Keynes’s theory and marginalist equilibrium economics, of Clower and Patinkin, in the 1950s More positively, and more importantly in the long run, was his path-breaking work on welfare economics This expanded upon various hints on the subject left in Marshall’s economics Pigou’s pioneering and highly original welfare economics

was first published in Wealth and Welfare (Pigou, 1912) Its contents in turn spawned three major theoretical works in the 1920s: the Economics of Welfare (Pigou, 1920b), Industrial Fluctuations (Pigou, 1927), and A Study of Public Finance (Pigou,

1928) These books not only reveal his skills as a major marginalist economist but

as a student of Marshall in the fullest sense of the word Pigou (1920b, p vii) expressed the view that economics provides the “instruments for the bettering

of human life [by restraining] the misery and squalor that surround us, the injurious luxury of some wealthy families, the terrible uncertainty [from unem-ployment and business cycles] overshadowing many families of the poor [which constitute social] evils too plain to be ignored.” This explicitly Marshallian credo

was elaborated in his Wealth and Welfare, and more fully detailed in the three

publications of the 1920s

The essential concept in Pigou’s broad perception of welfare was the national income or dividend However, the marginal method tended to be employed in analyzing the welfare implications of changes in national income Much of Pigou’s welfare analysis was in fact conducted through balancing the advantages of various small changes in output in different industries, thereby equating costs and benefits at the margin Business fluctuations of course also dealt with varia-tions in national income, but of a different order Public finance was largely a redistributive exercise channeling parts of national income to disadvantaged sections of society either by public expenditure or by tax and borrowing policies, hence maximizing welfare through bringing marginal utilities of money incomes closer to equality Moreover, expenditure and tax policies also affected size of national income and output, either favorably or unfavorably

An aspect of this last problem had a distinct, albeit limited, Marshallian

ped-igree In his analysis of increasing and diminishing returns in the Principles,

Marshall (1961 [1890], vol I, pp 467–70) had tentatively suggested that welfare (in terms of consumer surplus) could be increased if industries operating under increasing returns were encouraged to expand their output by means of a gov-ernment bounty, while taxes could be levied to lower the output of diminishing returns industries Social welfare gains arose from the cost and price consequences inherent in such changes in output (Marshall later thought that Pigou had carried the tax/bounty analysis too far, partly because Pigou was attempting

to treat the dynamic issues underlying increasing returns by the using tools specifically designed for static analysis.) Pigou’s emphasis on measuring changes

in national dividend as the index of welfare, thereby making his welfare analysis essentially an analysis of social product, was also innovative Consequently, even

if inadvertently, Pigou became an important and innovative pioneer in national income analysis well before regular official estimates of national income had become available

The welfare implications of the national dividend had a production as well as

a distributive aspect for Pigou The size of the national income clearly influenced national welfare Per capita growth of the national income therefore also enhanced national welfare, and Pigou did much to clear up potential ambiguities

in growth measurement This part of the welfare problem was a direct legacy

of the British classical economics tradition from Smith to John Stuart Mill, with its focus on the growth of the wealth of nations Such a legacy had been strongly

preserved in the production analysis of Marshall’s Principles One ambiguity

in growth measurement was that raised by the presence of externalities, a

con-cept derived from Sidgwick’s Principles of Political Economy (Sidgwick, 1887),

but greatly refined by Pigou His definition of such externalities has become classic:

the essence of the matter is that one person A, in the course of rendering some service, for which payment is made, to a second person B, incidentally also renders services or disservices to other persons C, D, and E, of such a sort that technical considerations prevent payment being exacted from the benefited parties or com-pensation being enforced on behalf of the injured parties (Pigou, 1920b, p 159)