Arthur Cecil Pigou (; 18 November 1877 – 7 March 1959) was an English economist. As a teacher and builder of the School of Economics at the University of Cambridge, he trained and influenced many Cambridge economists who went on to take chairs of economics around the world. His work covered various fields of economics, particularly welfare economics, but also included Business cycle theory, unemployment, public finance, index numbers, and measurement of national output. His reputation was affected adversely by influential economic writers who used his work as the basis on which to define their own opposing views. He reluctantly served on several public committees, including the Cunliffe Committee and the 1919 Royal Commission on Income tax.
Early life and education
Pigou was born at Ryde on the Isle of Wight, the son of Clarence George Scott Pigou, an army officer, and his wife Nora Biddel Frances Sophia Lees, daughter of Sir John Lees, 3rd Baronet. He won a scholarship to Harrow School where he was in Newlands House and became the first modern head of school. The school's economics society is named The Pigou Society in his honour. In 1896 he was admitted to King's College, Cambridge, as a history scholar where he first read history under Oscar Browning. He won the Chancellor's Gold Medal for English Verse in 1899, and the Cobden (1901), Burney (1901), and Adam Smith prizes (1903), and made his mark in the Cambridge Union Society of which he became President in 1900. He came to economics through the study of philosophy and ethics under the Moral Science Tripos. He studied economics under Alfred Marshall, whom he later succeeded as professor of political economy. His first and unsuccessful attempt for a fellowship at King's was a thesis on "Browning as a Religious Teacher."
Pigou began lecturing on economics in 1901 and started giving the course on advanced economics to second year students on which was based the education of many Cambridge economists over the next thirty years. In his early days he lectured on a variety of subjects outside economics. He became a Fellow of King's College on his second attempt in March 1902, and was made Girdler's lecturer in the summer of 1904. He devoted himself to exploring the various departments of economic doctrine, and as a result published the works on which his worldwide reputation rests. His first work was more philosophical than his later work as he expanded the essay which had won him the Adam Smith prize in 1903 into Principles and Methods of Industrial Peace.
In 1908 Pigou was elected Professor of Political Economy at the University of Cambridge in succession to Alfred Marshall. He held the post until 1943.
Pigou's most enduring contribution was The Economics of Welfare, 1920, in which he introduced the concept of externality and the idea that externality problems could be corrected by the imposition of a Pigovian tax. The externality concept remains central to modern welfare economics and particularly to environmental economics. The Pigou Club, named in his honour is an association of modern economists who support the idea of a carbon tax to address the problem of climate change.
A neglected aspect of Pigou's work is his analysis of a range of labour-market phenomena studied by subsequent economists, including collective bargaining, wage rigidity, internal labour markets, segmented labour market, and human capital.
One of his early acts was to provide private financial support for John Maynard Keynes to work on probability theory. Pigou and Keynes had great affection and mutual regard for each other and their intellectual differences never put their personal friendship seriously in jeopardy.
Pigou was generally critical of Keynesian macroeconomics and developed the idea of the Pigou effect on real money balances to argue that the economy would be more self-stabilizing than Keynes proposed. In a couple of lectures delivered in 1949 he made a more favourable, though still critical evaluation of Keynes' work: "I should say... that in setting out and developing his fundamental conception, Keynes made a very important, original and valuable addition to the armoury of economic analysis". He later said that he had come with the passage of time to feel that he had failed earlier to appreciate some of the important things that Keynes was trying to say.
Keynes, in turn, was very critical of Pigou, mentioning Pigou at least 17 times in The General Theory of Employment, Interest and Money, usually disparagingly.
Pigou had strong principles, and these gave him some problems in World War I. He was a conscientious objector to military service when it required an obligation to destroy human life. He remained at Cambridge, but during the vacations was an ambulance driver at the front for the Friends' Ambulance Unit, and insisted on undertaking jobs of particular danger. Towards the end of the war he reluctantly accepted a post in the Board of Trade, but showed little aptitude for the work.
He was a reluctant member of the Cunliffe Committee on the Currency and Foreign Exchange (1918–1919), the Royal Commission on the Income Tax (1919–1920), and the Chamberlain Committee on the Currency and Bank of England Note Issues (1924–1925). The report of the last body was the prelude to the much criticised restoration of the gold standard at the old parity of exchange. Pigou was elected to the British Academy in 1925, but resigned in 1947. In later years he withdrew from national affairs and devoted himself to more academic economics and writing weighty letters to The Times on problems of the day. He was a foreign honorary member of the American Academy of Arts and Sciences, a foreign member of the Accademia dei Lincei, and an honorary resident of the International Economic Committee.
He loved mountains and climbing, and introduced climbing to many friends, such as Wilfrid Noyce and others, who became far greater climbers. An illness affecting his heart developed in the early 1930s, however, and this affected his vigour, curtailing his climbing and leaving him with phases of debility for the rest of his life. Pigou gave up his professor's chair in 1943, but remained a Fellow of King's College until his death. In his later years he gradually became more of a recluse, emerging occasionally from his rooms to give lectures or to take a walk.
- Browning as a Religious Teacher, 1901.
- The Riddle of the Tariff, 1903.
- "Monopoly and Consumers' Surplus", 1904, EJ.
- Principles and Methods of Industrial Peace, 1905.
- Protective & Preferential Import Duties. 1906.
- "Review of the Fifth Edition of Marshall's Principles of Economics", 1907, EJ.
- "Producers' and Consumers' Surplus", 1910, EJ.
- Wealth and Welfare, 1912.
- Unemployment, 1914.
- Some Aspects of the Housing Problem, Warburton Lecture, 1914.
- "The Value of Money." 1917, Quarterly Journal of Economics, 32( 1), pp. 38– 65.
- The Economics of Welfare, 1920.
- "Empty Economic Boxes: A reply", 1922, EJ.
- The Political Economy of War, 1922.
- "Exchange Value of Legal Tender Money", 1922, in: Essays in Applied Economics.
- Essays in Applied Economics, 1923.
- Industrial Fluctuations, 1927.
- "The Law of Diminishing and Increasing Cost", 1927, EJ.
- A Study in Public Finance, 1928.
- "An Analysis of Supply", 1928, EJ.
- The Theory of Unemployment, 1933.
- The Economics of Stationary States, 1935.
- "Mr. J.M. Keynes' General Theory of Employment ...," 1936, Economica, N.S. 3(10), pp. 115–132.
- "Real and Money Wage Rates in Relation to Unemployment", 1937, EJ.
- "Money Wages in Relation to Unemployment", 1938, EJ.
- Employment and Equilibrium, 1941.
- "The Classical Stationary State", 1943, EJ.
- Lapses from Full Employment, 1944.
- "Economic Progress in a Stable Environment", 1947, Economica.
- The Veil of Money, 1949. First-page chapter-preview links
- Keynes's General Theory: A retrospective view, 1951.
- Essays in Economics, 1952.
1The economics of A.C. Pigou is generally regarded as a body of policies, particularly ‘Pigouvian’ taxes and subsidies, designed to maximize economic welfare. This view is embedded in leading economics textbooks, the basis of the early socialization of economists and the principal artifacts in which ideas are institutionally certified as economic truths. Older books identify presumptively Pigouvian policies by attaching his name to them. Newer texts retain the policies but drop his name, indicating the extent to which the received view is woven into the conventional wisdom of economics.1 One notable result of this understanding of Pigou’s economics is the “Pigou Club” founded by Gregory Mankiw. Numerous economists have been inducted as members, among them Gary Becker, Robert Frank, Paul Krugman, Nouriel Roubini, and Lawrence Summers.2 Pigou’s place in the disciplinary consciousness of contemporary economics is nicely documented in a remark by William D. Nordhaus in a recent review essay on the economics of energy use. Taxes on negative externalities, Nordhaus observes, are “sometimes called Pigouvian taxes after their first important advocate, English economist Alfred [sic] Pigou” (Nordhaus, 2011, 30). Pigou’s name was of course not Alfred. Moreover, it is a mistake to suppose that he advocated taxes on externalities.
2 In this essay, we contest the conventional view of Pigou’s thought. By way of a preamble, some elementary distinctions may be useful. In Pigou’s time, factories in Britain were located in densely populated cities. Smoke from their coal-burning furnaces polluted the ambient air, imposing a cost on urban dwellers. These facts pose a problem of economic welfare. As one method of addressing this externality, Pigou, following Henry Sidgwick, considered the possibility that the state might tax factory owners, linking the size of the tax to the costs resulting from emissions. This is a welfare policy. In examining the policy, Pigou argued that the social costs of factory smoke exceeded its private costs. In order to arrive at a precise account of this discrepancy, he employed the marginal social cost curve, which lies above the marginal private cost curve, the vertical distance between the two representing the incidental cost of factory emissions. This is an economic analysis of a welfare policy. Although Pigou assessed welfare programs of his time, his judgments on policies were invariably prima facie, guarded, and hedged and qualified by a detailed consideration of a formidable array of restrictive and contingent variables: the significance of the problem at stake; the conditions under which it could be addressed given existing circumstances; the analytical tools, data, and competent personnel—economists, civil servants, and political leaders—available for handling the problem; and the potentially damaging consequences, both economic and extra-economic, of adopting policy proposals. In other words, there are no definitively Pigouvian welfare measures. Pigou was a logician of policy analysis, not a proponent of specific economic policies. Nor was he the architect of a general system of policy analysis—for the compelling reason, as we show in the ensuing, that he believed such a system is impossible.
3 In this essay, we pursue two lines of investigation. In Part I, we show that Pigou developed what he called a “machinery” of thought, favoring, as he often did, the mechanical metaphors of his teacher Alfred Marshall. The Pigouvian machinery constituted a sketch of a theory of economic policy analysis, a blueprint for an analytical theory of economic welfare that specified the conditions an analysis of economic policy must satisfy in order to qualify as scientifically legitimate. Employing Pigou’s metaphor, he was an artificer of the tools that comprise the economist’s toolbox and an artisan who employed these tools to explore the conditions under which specific economic policies can be expected to succeed in solving concrete problems of economic welfare.
4 In Part II, we analyze the main inference that Pigou drew from his theory: the historicity of policy analysis, an important aspect of his thought that has generally been neglected—or occasionally denied—in the secondary literature. The realistic possibilities of Pigouvian policy analysis are determined by the historical conditions under which the analysis is performed. Two respects in which this is the case can be distinguished. The range of possible economic policies, while not infinite, is very large. Pigou’s selection of policies for analysis was a function of the priorities of his political culture. He took up and dropped economic policies as candidates for analysis depending upon their salience in the agendas of parliament, the government, and the leading political parties in Britain. He also held that the validity of policy analysis is tied to current political and economic conditions and limited by changes in their complexion. As a result, the generalizations of economic science—which Pigou, following the Victorian positivism of his predecessors, called “laws”—that hold true today may prove to be false tomorrow.
5 In targeting, rejecting, or discarding policies for analysis, Pigou was opportunistic. His analytical choices were decided chiefly by his sense of the shifting distribution and balance of power that would determine which policies were likely to be enacted and which would disappear from the public agenda. His conception of the validity of policy analysis was based on the same considerations. As a result, Pigouvian economic generalizations were empirically fragile propositions, vulnerable to shifts in the polity and the economy. As these shifts occur, it is impossible for the economist, embedded in history, to make scientifically reliable estimates concerning their scope and implications for his work. An economic analysis—whether published in books or journals or presented before governmental commissions or committees—is an historical artifact, constrained by the same contingencies that define all historical phenomena. This is the ultimate reason why Pigouvian policy analyses are restricted to a case-by-case method of investigation. Because of the historicity of the conditions for their validity, reliable inferences from current to hypothetical future cases are out of the question. Here we part company with T.W. Hutchison, who maintains that Pigou, “though he came to economics from history, did not share Marshall’s intense concern with the historical dimension of economic processes, which, in fact, largely faded from the Cambridge scene with Marshall’s departure” (1981, 65).3
1. The Theory of Economic Policy Analysis
6Pigou’s theory of policy analysis is defined by several premises. Some he stated explicitly and stressed repeatedly (Pigou, 1906, 1-3; 1908; 1935). Others are tacit assumptions. He arrived at these principles in early manhood and adhered to them with remarkable consistency. There is a strong sense in which his conception of what the aims of economic analysis should be and what it means to be an economist had their source in the reformist social thought of late Victorian progressives and Edwardian New Liberals that became prominent in the early years of the twentieth century. In many respects, Pigou remained a man of the Edwardian era into old age.4
7First. economists do not make economic policy, which lies in the province of politics. Economic policy is a responsibility of political leaders, whom Pigou, playfully borrowing language from Plato’s Republic, characterized as “philosopher kings” (Pigou, 1939b, 220). Economists analyze effects of alternative policies, spelling out their consequences for the size, distribution, and stability of the national dividend. However, the choice among these alternatives is not a legitimate question for economics, which is a purely “positive” science. It can substantiate no judgments about what measures should be taken or what ought to be done, because its conclusions are restricted to propositions that can be established by logic and empirical investigation. In Weberian language, economic policy analysis is value-neutral. It cannot validate norms and is limited to clarifying the substance of alternative economic solutions to problems of policy and examining the consequences of acting on them (Pigou, 1906, 1-2).
8Thus economic theory is not the sole basis for economic policy. Nor should economic welfare be conflated with human welfare generally. Under ideal conditions, economists generate well-confirmed generalizations bearing on policy problems for the consideration of statesmen, who treat these generalizations as pertinent data. Together with other data essential to the achievement of extra-economic ends such as ethical, eudaemonistic, aesthetic, and political values, “philosopher kings build up policies directed to the common good” (Pigou, 1939b, 220).
9Second. It follows that economists engaged in policy analysis cannot be political partisans—advocates of political doctrines, programs, or ideologies (Pigou, 1906, 1-2). Economists investigate policy proposals supported by revolutionaries, reformers, conservatives, reactionaries, and revanchists—actors who span the entire political spectrum. They also consider policies enacted by diverse forms of government in different national and historical settings. Economic reasoning often provides fodder for political partisans, who employ it not as a basis for arriving at truths, but as a “kind of brickbat useful on occasions for inflicting injury on their opponents” (Pigou, 1935, 9). This prospect may tempt younger economists to tailor their results to the requirements of political programs in the hope of positioning themselves “near the centre of action.” However, Pigou held that it was an “intellectual crime” to succumb to such a temptation. In his biblical metaphor, economists who compromise their scientific integrity in this fashion sell their “birthright in the household of truth for a mess of political pottage” (Pigou, 1935, 10).
10 Pigou’s austere distinction between economic science and economic policy did not, in his view, commit him to a scientific asceticism, requiring him to write only for professional economists. He was also, in the parlance of our time, a public intellectual who wrote for non-specialist readers.5 However, when he entered the public domain, he did not abandon or compromise his commitment to scientific economics in favor of policy advocacy. The differences between his more strictly academic and popular works were determined by their level of technical sophistication. In writing specialized treatises and academic articles for students of economics and in publishing for the general reader, he employed the same mode of analysis. In the latter case, he simplified or elided complexities of technique that might be confusing, or even unintelligible, to the non-specialist. From the standpoint of logic, however, there were no differences between his contributions to these spheres of discourse.
11Philip Noel-Baker, Pigou’s friend of many years and recipient of the Nobel Prize for peace in 1959, recalled that he “carefully refrained from membership or other affiliation to any political party” (Noel-Baker, 1979). Above the fray of party struggles, he was able to draw on the works of a diverse group of British social reformers: from New Liberals such as Charles Masterman and Seebohm Rowntree to Fabians such as Beatrice and Sidney Webb and more conservative philanthropists such as Charles Booth, Helen Bosanquet, and Octavia Hill. Like the Webbs, he saw considerable value in a national minimum for housing and other necessities of life. He also found merit in the position on housing for the poor taken by Hill, who strenuously opposed both state provision and management of shelter (Pigou, 1914, 36, 47). Although he believed that Rowntree’s reforms at the family’s chocolate factory promised to bear fruit—his metaphorical criterion, drawn from Francis Bacon, both for the success of economic policy and the value of economic science—he did not take the same view of Rowntree’s stance on the minimum wage. This aspect of Pigou’s thought is a mark of his intellectual circumspection and conservatism. The policies that he regarded as legitimate candidates for economic analysis were not logical possibilities that he conceptualized in his rooms at King’s College, Cambridge, but actual programs that had been proposed or implemented by polities, parties, and interest groups, many in his own time.
12Third. Policy analysis requires economists to undertake ambitious ethnographic research on work places and working classes. The result should be a keen understanding of “economic life in its full concreteness, a moving, breathing process among actual men and women in their factories and their homes.” Pigou maintained that unlike Marshall, most economists had only a remote acquaintance with the economic realities confronted by the majority of the population. Their knowledge of economics was derived from the printed page, not first-hand experience. Left deserted on an island, Pigou claimed, Marshall could redesign “the great majority of important machines now in use.” Although this estimation of his master’s skills in industrial design may have been excessive, Pigou had a point. Unlike Marshall, who had done extensive field research in his youth, investigating the conditions of work and life in many industries, most economists lived “cloistered” academic lives comparable to his own. Thus Pigou advocated a method of research that he himself had “notoriously failed” to pursue: “it is up to the economist, when he is young and his mind is plastic, to take whatever opportunities he can find for direct acquaintance with the life of men and women in factory and field; to understand machinery; to see for himself at first hand how businesses are organised and run” (Pigou, 1935, 11-12).
13Fourth. Success in social reform depends on the sophistication of methods of economic analysis. Paramount in Pigou’s thinking on this point was the modest accuracy of quantitative estimates at the disposal of economists. Notwithstanding improvements in statistical techniques during his lifetime, he habitually lamented various sources of methodological and quantitative negligence on the part of economists that limited the powers of policy analysis. They seemed unwilling to tackle the problem of omitted variables. They often conflated correlation and causation. And they had failed to conduct careful quantitative analyses of short-period economics, disequilibrium processes, and cumulative causation (Pigou, 1908, 31; 1910a, 984-85; 1935, 21-22). However, even in a perfected regime of statistical refinement, serious methodological difficulties would remain. Some variables, for example, resist quantification. In his memorandum for the Royal Commission on the Poor Laws and Relief of Distress, Pigou addressed one such variable: the competence of the agency that administers relief to the poor: “on the really debatable question whether carefully administered out-relief would be better or worse in this respect [reducing pauperism] than carefully administered in-relief, refinements of statistical reasoning can throw no light” (Pigou, 1910a, 986). Because of this insurmountable problem, economics “must almost always speak with an uncertain voice” (Pigou, 1908, 31).
14Fifth. There are no definitively Pigouvian techniques. Pigou treated the selection of methods of economic analysis as a practical and provisional matter, not a question of doctrine. It is sensible to employ a technique if it proves useful in discovering relations of economic cause and effect. The more extensive the range of economic problems to which a technique can be successfully applied, the greater its analytical power. However, there are two reasons why no great confidence can be placed in the durability of a technique. Economic phenomena and the problems they pose are subject to change, which may spell an end to the efficacy of techniques successfully employed before the change. In that case, alternative methods will be needed. In addition, new techniques may achieve better results—more precise or comprehensive knowledge of causal relations—than current methods. In both cases, there are good grounds to discard elements of the current analytical apparatus. Thus it seems neither hyperbolic nor anachronistic to suggest that Pigou was a methodological pragmatist: there is no set of economic techniques that occupy a uniquely privileged status because they invariably achieve the most powerful analyses, but only methods that prove to be successful under specific conditions that are subject to change.
15 Pigou’s methodological pluralism is clear in the remarks on economic analysis in his Sydney Ball Lecture at Oxford in 1929. Here, he attempted to shed light on the concept of economic analysis by juxtaposing it to economic description. Descriptive economics provides a narrative account of economic affairs. Analysis is an investigation of their causes and effects. Description addresses the question of what is the case, analysis the question of how and why events transpired as they did. Description gives a phenomenological account of events as they are experienced and observed. Analysis penetrates beneath the surface to discover the underlying mechanisms that produce events. However, economic analysis is not a methodologically uniform process. “With different problems and with different sets of data, different detailed methods are appropriate” (Pigou, 1929, 2). Economic analysis is performed in a “workshop.” The methods suited to different problems are tools, comparable to those employed in the undertaking that was Pigou’s other consuming passion: Alpine mountaineering. Although every analysis requires tools, just as every mountaineering expedition requires ice axes, no given tool is inherently and invariably indispensable: “those of them that cannot be made to work in elucidating the problems of the real world must be scrapped,” for the same reason that an ice axe that does not prove serviceable should be repaired, redesigned, or replaced (Pigou, 1929, 8). In characterizing techniques of economic analysis, Pigou’s main concern was to stress their purely instrumental and transient value. Because the terrain of economics changes, the same holds for the scope and character of economic problems. As new problems engage economists and old problems are reconceptualized from new perspectives, new tools can be expected to supersede, piece by piece, older techniques. In this manner, the instrumentarium of the economist’s workshop is endlessly renovated.
16 Pigou’s methodological pluralism does not conflict with his legendary dedication—“idolatrous” in the view of some (Robinson, 1968, 91)—to Marshall’s economics. It was notoriously difficult to persuade him that the Marshallian engine of analysis was deficient in any respect. In addition, he believed that there “is much to be urged against the employment of novel terminology” and tools. However, he was ready to refine or innovate if the results promised to strengthen economic analysis (Pigou, 1910b, 358). He betrayed no hesitation in making a case for defects in consumers’ and producers’ surplus—tools Marshall had developed to assess changes in welfare—when externalities were present. Nor was he reluctant to propose new tools—the now famous curves of marginal social costs and benefits—when he was convinced that circumstances called for technical innovation (Pigou, 1903c; 1910b). When Piero Sraffa’s critique of Marshallian partial-equilibrium appeared in the Economic Journal (1926), Keynes wrote him that Pigou “feels he must, in light of it, reconsider his whole position” (in Potier, 1987, 20). Although Pigou did not follow Sraffa’s suggestion—to abandon the competitive market analytical framework for monopoly—he revised his analysis in order to refine its logical consistency (Pigou, 1927b; 1928a). Pigou’s flexibility in adopting new techniques is perhaps most evident in his employment of the IS-LM version of Keynes’s model in 1938 (Pigou, 1938; see also Pigou 1941), only two years after his scathing review of the General Theory (Pigou, 1936). Years later, Pigou made a case for the unprecedented explanatory power of Keynes’s macroeconomic formulation. The “kernel” of Keynes’s contribution to economics, Pigou argued, lay in the development of a “single formal scheme” by means of which all significant economic factors, real and monetary, could be methodically investigated (Pigou, 1950, 65).
17Pigou’s methodological pluralism should be understood in light of a basic principle of his philosophy of economic science: economics progresses not by wiping the slate clean and destroying or demeaning the work of others, but by building on the past. In this regard, he urged economists to follow the scientific etiquette of Einstein, who “did not, in announcing his discovery, insinuate, through carefully barbed sentences, that Newton and those who had hitherto followed his lead were a gang of bunglers” (Pigou, 1936, 115). On this point, there is a fundamental conflict between Pigou and the post-Treatise Keynes, who believed in the revolutionary promise of eradicating the past and beginning de novo—a strategy that Pigou regarded as bad form professionally, politically hazardous, and logically impossible.
18Sixth. In the analysis of policy, it is a mistake to concentrate on the state as the exclusive vehicle for implementing welfare programs. Successful policy is generally achieved by cooperation between the state and civil society. State actors may find it necessary to legislate minimum standards for housing, inspect homes for overcrowding, condemn uninhabitable buildings, “tackle the collective problem of beauty, of air and of light” through urban planning, and subsidize rents for families that do not earn a living wage (Pigou, 1914, 37-55). However, philanthropists may also improve the quality of life by managing friendship societies and brotherhoods that provide social security and housing (Pigou, 1914, 46-47; 1901, 254). In many cases, the welfare policy spheres of the state, civil society, and even charitable individuals cannot be clearly delineated. Because “it is impossible to say what any one of them ought to do, without knowing what the others are doing,” significant public-private collaboration may be not only desirable but necessary. In simple cases of destitution, the state may be most effective in allocating resources by a quasi-centralized system of distribution, such as the Poor Law. Allocation by voluntary agencies may be the best course in more complex settings requiring a careful discrimination among heterogeneous needs of individuals (Pigou, 1901, 259). Although Pigou himself was an agnostic, he believed that religious charities were well suited to assist the poor since they were decentralized and maintained close ties to their communities (Pigou, 1901, 261).
19 Employers were often particularly well-qualified to serve as effective partners with the state, and on several grounds. Because of the length of the working day on factory floors, employers were in a unique position to improve the lives of their workers by introducing better working conditions and hiring supervisors with strong organizational skills and high moral standards. Instances of best practices were not hard to find. In Bournville, England, the Cadbury family offered its younger workers gymnastics classes, subsidized education, and regular physical and dental examinations. Underperformers suffering from ill-health were placed on a strict regimen of nutrition, exercise, and rest. In Rochester, New York, the Canadian entrepreneur Franck Brownell maintained comfortable year-round temperatures in his camera factory. And in York, England, Joseph and Seebohm Rowntree hired a social secretary in their factory to improve working conditions, organize social activities, and reduce employee-management conflicts (Pigou, 1913a, 18-22). Although these “chivalrous” employers could not expect a full return on their philanthropic investments, there were undeniable economic benefits. They enjoyed reputational advantages as well as higher productivity—a result of reductions in labor strife, tardiness, absenteeism, and sloth as well as improvements in work discipline (Pigou, 1913a, 17-18).
2. The Historicity of Policy Analysis
20In view of the above premises, Pigou concluded that economists should abandon the illusion of a grand theory of policy analysis and reconcile themselves to a much less grandiose program of inquiry: a methodology of painstaking, case-by-case investigations, each of which is limited by the irreducible complexity and variability of historical circumstances.
Each particular case must be considered on its own merits in all the detail of its concrete circumstances. High-sounding generalizations on these matters [such as state intervention] are irrelevant fireworks. They may have a place in political perorations, but they have none in real life. Accumulation of evidence, the balancing of probabilities, judgment of men, by these alone practical problems in this region can be successfully attacked. (Pigou, 1935, 128)6
21 Pigou ruled out a universally valid theory of policy analysis due to the intractable diversity and flux of individual lives, local histories, and institutional spheres. Consider the problem of poverty, which he regarded as a daunting challenge because of immense differences in the lives of poor people as well as the factors responsible for their poverty (Pigou, 1901, 240). “The poor” did not constitute a homogeneous class defined by common conditions of existence. People could be transiently or chronically poor. The transiently poor were otherwise independent families who had experienced a “sudden misfortune” (Pigou, 1901, 247). Chronic poverty might be a result of numerous causes: old age; frailty, which could be physical, intellectual, or moral; an inability to master crises and a consequent dependence on indiscriminate handouts; or the accident of being born to chronically poor parents (Pigou, 1901, 241-45). Diversity in local conditions and variation in the competence of public agencies made the problem of welfare for the poor even more complex. It was utterly unrealistic to develop regulations for charitable organizations without investigating the local cultures where poor relief would be administered. If the rich generally provided liberal and indiscriminate grants to the poor, it would be pointless for a local charitable organization to introduce strict regulations governing poor relief; they would be ignored (Pigou, 1901, 257). Finally, even if a need for government intervention could be demonstrated, it did not follow that public assistance should be forthcoming. The judgment and competence of public officials were crucial desiderata. It was essential to consider “how far, in the particular country in which we are interested and the particular time that concerns us, the government is qualified to select the right form and degree of State action to carry it through effectively” (Pigou, 1935, 124). Pigou’s view of the futility of a systematic theory of policy analysis is nicely documented in his testimony on April 23, 1919, before the Sankey Commission on the coal industry. Here he contested the ideal of a uniform policy of nationalization.
Nationalisation in any sense cannot be judged on grounds of general principle. What would work well under one kind of Government would work badly under another: and what, under any given Government would work well for one industry would work badly for another. The desirability or otherwise of the nationalisation in any sense of any industry can only be determined after a detailed study of the characteristics of the industry in relation to the qualities of the country’s Governmental machinery. (in HMSO, 1919, 416)
22 Within the limits of a single essay, Pigou’s conception of the historicity of policy analysis can perhaps be best demonstrated by considering his writings on public finance. This sphere of his work offers the most thorough and comprehensive documentation of the substance and intentions of Pigou’s thinking—academic publications, pamphlets written for a general readership, letters to editors of magazines and newspapers, memberships on government commissions and testimony before such bodies, and archival sources.
23 The genesis of Pigou’s project of developing an analysis of public finance lies in the tariff reform controversy in Britain. The controversy began with a speech given by Joseph Chamberlain, Secretary for the Colonies in the Conservative government of Prime Minister in Birmingham. Arthur Balfour, on May 15, 1903. Chamberlain challenged the principle of free trade and its benefits for the British empire, a doctrine deeply ingrained in late Victorian political and economic thought and rhetoric if not always implemented in practice. In its various permutations, the fiscal reforms proposed by the Balfour government included elements of several policies: retaliatory tariffs against nations that dumped cheap goods into unprotected British markets, permanent protective tariffs that defended domestic producers against foreign competition, and preferential duties designed to secure the British empire against increasingly damaging American and German competition by creating a united trading bloc.7
24At the turn of the century, the British government was under immense pressure to increase revenues. The Boer War was the most expensive British military initiative since the Crimean War of 1853-56, imposing costs of £35,750,000 between April and July 1902 (The New York Times, 1901). In the five-year period beginning 1895-96, normal government expenditure rose by 40 percent (Daunton, 2001, 303-304). Central grants to local governments did not keep pace with their responsibilities, resulting in tax increases of 141 percent on building occupants between 1875 and 1900 (Packer, 2001, 55). The campaign against free trade had important ramifications for public finance. Advocates of tariff reform promised that preferential import duties would be an effective source of revenue for both local and central governments.
25As a public intellectual, Pigou proved to be a fierce critic of tariff reform, relentlessly exposing flaws in the reasoning of reformers in popular lectures, articles and pamphlets. As an academic economist, he subjected Chamberlain’s proposals to a “non polemical, but scientific” analysis—as he described the objective of his proposed book, Protective and Preferential Duties,to his publisher (Pigou to Macmillan, 8/25/1906, Macmillan Archive). Pigou acknowledged that protective tariffs could be helpful in sustaining infant industries during the early stages of national economic development. In more mature economies such as Britain, however, they would reduce the national dividend and damage economic welfare. Thus no generally valid laws of free trade were possible. The question of whether free trade or protection was more effective in promoting economic welfare depended on factors that not only varied from one economy to another, but were subject to change within a given economy as it developed, prospered, or declined. In making this case, Pigou briefly considered the revenue potential of tariffs, finding no general “a priori presumption either for or against the imposition of protective duties as a means to raising revenue” (1906, 32). There was no reason to believe that a randomly selected good subject to protective duties would either damage the national dividend or raise revenues more or less effectively than a randomly selected good subject to non-protective tariffs. In the real world, however, products were not taxed at random. Existing non-protective duties imposed on sugar and tea were based on assessments of their revenue potential. Chamberlain and his supporters selected products for protective tariffs based on the probability that they would reduce competition, not on their capacity to generate revenue through taxation. On Pigou’s analysis, such tariffs were likely to be ineffective sources of revenues: “for the more completely they protect the home producer, the smaller, of necessity, is the revenue they yield” (Pigou, 1906, 32).
Land Value Taxation
26Chamberlain’s proposals created a furious national debate among policy makers, economists, and the informed public. Supporters of the government formed the Tariff Reform League. The Unionist Liberals established the Unionist free Trade League, and the Liberals organized the Free Trade Union. Although the Tariff Reform League was funded more generously than its rivals, all three groups held rallies and distributed countless tendentious pamphlets and flyers (Irwin, 1994, 85). The Conservative Party was split, with cabinet members on both sides of the debate resigning. The public was confused, and in the landslide defeat of the Conservatives in the general election of 1906, the Liberals returned to power. Reducing dependence on indirect taxation, the Liberal government introduced the modern system of resource extraction via income taxes, adopting both differentiation and graduation. Land reform was also a significant policy on the Liberal agenda. David Lloyd George, the charismatic and tactically brilliant chancellor of the exchequer (1908-1915), included the taxation of land values—taxing landlords on the basis of the unimproved value of their land—as a key component of his “People’s Budget” of 1909.8
27Before World War I, ownership of land in both rural and urban Britain was concentrated in families of considerable wealth. A comprehensive national survey conducted in 1873 showed that in that year, 7000 people owned eighty percent of the land in the country (Horn, 2002, 104). The skewed distribution of land ownership had not changed markedly by 1900. In rural areas, agricultural workers lived in substandard cottages owned by landlords, reducing their bargaining power in wage negotiations and their mobility to pursue better opportunities elsewhere. Liberals argued that taxation of the unimproved—or socially created—value of land could fund many initiatives. Ownership of smallholdings by agricultural laborers could be increased. The housing-employment connection could be broken, ending the serf-like dependence that bound agricultural workers by holding them in thrall as vassals of the landlord. Finally, minimum wages for agricultural laborers could be introduced, giving them access to housing of higher quality. In urban areas, traditional ownership, lease, and tax regimes discouraged construction. Landowners leased large parcels to developers, retaining ownership rights to the site and claiming improvements as their own. Despite dramatic increases in land values in congested cities such as London, landowners contributed little to local government finances. Rates were imposed on occupants based on rental values that failed to distinguish the value of the land and the value of its improvement. The principle of land taxation was use value, which reduced owner incentives to improve or build (Packer, 2001, 28, 54-55). Between 1906 and 1915, the Liberal government considered various measures of land taxation, the most creative of which were suggested by Lloyd George and the unofficial land inquiries he commissioned. Under the umbrella of land reform and site-value taxation, other policies were also considered: smallholdings, a minimum wage, tenure security, and the quantity and quality of housing available to the poor.9
28Although discussions of land value taxation had begun in 1906 when Asquith was chancellor of the exchequer, it quickly became clear that Britain had no coherent method for evaluating landed property. How were improved and unimproved values to be distinguished? Perhaps the most difficult problem was how to replace archaic methods of evaluation. Outside London, 648 parishes valued property on the basis of outdated lists. Even more worrisome, the boundaries of parishes and counties did not coincide, a foundation for administrative chaos (Packer, 2001, 60). It was in this confused setting that Pigou published Policy of Land Taxation (1909), an amplification of his July 1907 article in the Edinburgh Review. He urged policy makers to consider his analysis carefully before introducing legislation (Pigou, 1909, 32).
29In generating revenues, Pigou argued, the Liberal party had two options: to increase existing taxes on incomes, inheritance, commodities, and imported goods; or to identify untapped sources of revenue. In considering land taxes, he noted that they could be based on either the unimproved value of land or increments in its value . Once available, revenues could be used to adjust other national and local finances (Pigou, 1909, 5-6). In his view, unimproved land values formed the basis of a very effective and relatively equitable form of taxation, chiefly because they were a consequence of unearned spillover benefits—value generated by the efforts of neighbors, the government, or general economic progress (Pigou, 1909, 11). Thus taxes on unimproved land values would not damage productivity. Although they were unfair in targeting a specific economic group, they confiscated only socially generated, unearned incomes. From the standpoint of distributive justice, therefore, they were unobjectionable, especially in comparison with alternatives. In Pigou’s view, all taxes were unfair to some extent. Taxes on commodities such as tea or tobacco, for example, compelled consumers either to pay more in order to purchase the item or to forego consumption altogether. In either case, the consumer was burdened. He suggested that the distributive injustice of land value taxation could be mitigated by maintaining moderate and stable rates (Pigou, 1909, 12-13). If the cost of evaluating land proved to be low, land value taxes would qualify as an ideal source of revenue.10 They could also preempt new taxes that would be more burdensome. Revenues raised from land taxes could be used to relieve local rates on improvements to land and buildings, reduce migration from the countryside to congested cities, and improve urban sanitation as well as the availability and quality of housing. Because of inadequate data, Pigou was not prepared to draw confident conclusions about how funds might be allocated (Pigou, 1909, 31-32).
30Taxation of the incremental value of land was a much more complex matter. In general, Pigou was strongly in favor of imposing very high taxes on all genuine windfalls when two conditions were satisfied: they could be differentiated from other increments, and the costs incurred in taxing them were not prohibitive (Pigou, 1909, 22). However, it was not clear how these conditions could be employed to craft a workable policy. How could pure windfalls be distinguished from other increments, such as those caused by changes in the price level or interest rates? Should taxes be imposed only on increments, or should decrements be taken into account as well? Although Pigou could envision solutions to these problems, he regarded them as excessively complicated for policy making. Pragmatism dictated taxing extravagant windfalls—for example, land values that tripled over a fifteen year period. Although this simplification would generate little revenue, it would reduce administrative costs substantially.
The War and the National Debt
31On August 4, 1914, Britain declared war on Germany. World War I made food production a national security requirement, and taxing land quickly gave way to more pressing issues. Pigou’s theoretical priorities as an economist changed with the dramatic political and economic exigencies that the war imposed. The Asquith government had underestimated the scope and aggressiveness of Germany’s expansionist ambitions and misjudged the threat, both financial and military, posed by German war preparations. The British anticipated a short war. In the vision of Asquith’s strategic planners, British military commitments would not extend beyond a naval blockade of Germany, deployment of its small professional army to support much larger French forces in the West, and non-military assistance to allies. This vision was shattered in autumn. Following the September failure of the German army to achieve a decisive breakthrough to Paris in the Battle of the Marne, the inconclusive results of Allied counterattacks, the inability of either army to outflank the other, and shocking casualties in Flanders—especially in the first Battle of Ypres in October-November—a grim template was set for a conflict of indeterminate duration on the Western Front: industrialized carnage, massive casualty rates, immobility of forces as armies mounted grand offensives that produced gains measured in meters—taking, losing, and retaking the same devastated terrain—and the stalemate of trench warfare. Fighting a war with no end in sight, even as the Imperial General Staff promised victory with each successive new offensive, parliament passed legislation in the early months of 1916 mandating general conscription and imposing an enormous financial burden on the state. The government had increased its revenues by raising income tax rates, lowering the income threshold for taxation, and introducing excess profit duties. However, the expenses of the war coupled with a determination to maintain the sterling on the gold standard resulted in significant budgetary deficits (Horn, 2002, 26, 76-94).
32 In 1915, the socialist trade union leader Benjamin Tillett called for conscription of wealth to pay for the war, a position that the War Emergency Workers’ National Committee and the Trade Union Congress adopted following the introduction of military conscription. On this view, it was unconscionable, and also politically unwise, for the country to expect its men to risk their lives only to return home facing income insecurity and massive debt service to rentiers. The Labor Party adopted this position in its election campaigns of 1919 and 1923: outdoor relief to capitalists would have to end. Labor proposed to solve the budgetary problem with a one-time capital levy that would dramatically reduce the national debt and interest payments, lower income tax rates on work and saving, and diminish the concentration of wealth. Imposition of a capital levy during wartime was regarded as economically dangerous and politically impossible. It was seriously considered after the war but encountered numerous objections. How should capital be defined? What consequences would the levy entail for capital and credit markets, consumption, saving, and investment? In paying off the national debt, was it politically prudent to favor rentiers over the poor and the middle classes, whose income taxes had increased significantly during the war? Would a levy encourage tax avoidance and evasion? These difficulties seemed intractable, making a capital levy an unattractive fiscal policy even to its earlier supporters. When Labor came to power in 1924, the government formed the Committee on the National Debt and Taxation—the Colwyn Committee—with a view to ascertaining the most effective way to retire the war debt. In 1927, a year before Pigou published A Study in Public Finance (1928b), the levy was rejected as unsound on both political and economic grounds (Daunton, 2002, 49-74).
33 Pigou held that it was critical to analyze the problems posed by financing the war, which had never been systematically investigated. Shortly after war was declared, he delivered two public lectures at Cambridge on the cost and financing of the war (see Pigou, 1916a, 5). They were followed by a series of works that considered the merits of large increases in the income tax to pay for the war (Pigou, 1916a, 1916b, 1916c, 1916d, 1918). Pigou understood that the principle guiding peacetime taxation—extracting equal fractions of wellbeing from everyone—could not apply during the war, when both the weak and strong were required to make equal sacrifices of life and limb. It was just to expect generous monetary contributions from noncombatants. Extreme progressive income tax rates that would leave noncombatants—rich or poor—with equal after-tax incomes were morally justifiable but politically unimaginable. However, tax rates for citizens exempted from military service on grounds of health, age, gender, or conscientious objection could be amply increased. Because Pigou expected Britain’s peacetime tax structure to be reinstated after the war, he rejected arguments that a wartime increase in income tax rates would reduce incentives to work and save.
On the contrary, heavy temporary taxes now are likely, on the whole, to hinder production much less than the heavy continuing taxes which we shall otherwise be compelled to impose for an indefinite period to provide the money for war loan interest. In many economic problems considerations about what is just and considerations about what is best for production join in opposite direction. In this problem they point in the same direction. (Pigou, 1916c; see also Pigou, 1916d)
34 As war costs mounted, the British government increasingly relied on loans to finance its deepening budget deficits. On Pigou’s analysis, Britain had two options for settling its £6-6.5 billion post-war internal debt—he did not believe that the British external debt was a matter of great consequence (Pigou, 1920a, 8). It could employ a conventional strategy of raising annual taxes to finance interest charges on the debt and establish a sinking fund that would retire the principal over several decades. Or it could impose a one-time capital levy on real and financial assets that would quickly pay off much of the debt. In his analysis of this choice, the relative advantages and burdens of the capital levy championed by Labor leaders had a prominent place. Although steep, the capital levy would diminish subsequent tax rates. It would also render a sinking fund unnecessary, and interest payments would be minimal. Pigou’s immediate post-war analysis suggested that this option, if properly crafted, was the preferable course: a carefully structured capital levy complemented by taxes on windfall wartime profits and moderately progressive income tax rates. Unlike the first option, a capital levy would also reduce disincentives to work and save over the coming decades (Pigou, 1920a, 17).11
35 When enthusiasm for a capital levy waned in the mid-1920s, Pigou’s thinking on the two options for repaying the national debt also changed. In his testimony before the Colwyn Committee, he no longer regarded a large capital levy as feasible, and for several reasons. It would encounter powerful organized resistance from various quarters. It would encourage fears that reliance on a capital levy might become a routine tactic in the fiscal armory of the state and exacerbate existing pessimism caused by post-war British industrial depression. Finally, his initial assessment of the advantages of a capital levy assumed that future revenues from income, super, and death taxes would not be significantly diminished. However, new research, including Josiah Stamp’s work on the economic effects of a capital levy, persuaded him that this assumption was mistaken: “I have not, until within last year, realised how very large a proportion of a levy at steeply progressive rates—and the rates would have to be steeply progressive—would be used in repairing ravages in the future revenue consequent upon the levy itself” (in HMSO, 1927, 436-37).12
Peacetime Public Finance
36Pigou’s thinking on public finance was powerfully affected by the transformation in British financial planning occasioned by the scope and duration of World War I. The war convinced him that financing government expenditures was a critical field of economic analysis. In revolutionizing the fiscal challenges of the great powers, the war had transformed state extraction of resources so radically that “the problems it presents are different, not merely in degree, but in kind from what they were before the war” (Pigou, 1920a, 7). If Britain could count on a post-Versailles era of peace—comparable to the relative pacification of Europe following the Congress of Vienna—Pigou would have been prepared to take a benign view of debt repayment as a matter of no great urgency. However, he held a much darker vision of the future of Britain. He was, as he put it, “obsessed” by the prospect that “there might quite well be a war in 20 years.” In the event of another general European war in two decades, a substantial outstanding debt would reduce the credit rating of the state, making it difficult to borrow at low interest rates. Pigou, therefore, suggested that the existing sinking fund be doubled to £100 million annually. Although this policy would entail higher tax rates and diminish productivity, the security of the country required severe measures (in HMSO, 1927, 444).
37 It would be necessary to finance the sinking fund through general taxes in peace-time. Sound fiscal planning required that debt be undertaken only to cover extraordinary expenditures. Recurrent borrowing would increase debt levels and interest obligations, damaging the legitimacy of the state and, in the extreme case, threatening eventual bankruptcy. By financing standard operations of the state through tax revenues on various sources of income, expenditure, saving, and inheritance, it would be possible to escape these consequences (Pigou, 1920c, 589). When Pigou gave evidence before the Colwyn Commission, its members were interested only in certain aspects of these taxes. Neither their questions nor Pigou’s answers addressed problems of peacetime public finance systematically. However, it is clear that in his appearances before the Commission, Pigou drew heavily on Part IV of the first edition of The Economics of Welfare (1920c).13 Part IV—“Public Finance”—was based in part on the analyses of tariffs, land value taxation, and wartime taxes that he had developed in 1903-1918. He wrote its treatment of income taxes after his membership on the Royal Commission on the Income Tax, also chaired by Lord Colwyn (Frederick Henry Smith), in 1919-20.
38The brief of the Royal Commission was to examine the British income tax system and recommend measures to improve its fairness, working on the assumption that tax revenues would be maintained at their immediate post-war levels. The Commission began meetings on May 7, 1919, and issued its report on March 11, 1920. In the interim, it held fifty sessions and examined 187 witnesses, representing the government, a broad spectrum of industries and occupations, and various other organizations. As the signatories of the report noted: the “evidence we have heard and the information put at our disposal have been voluminous and fairly exhaustive, for we believe that we have been in touch with most of the important sources of knowledge and of informed criticism on the subject with which we were called upon to deal.”14 Pigou’s service on the Commission gave him a more sophisticated understanding of the British tax system as well as an appreciation of the principles of peacetime taxation. By the time he wrote The Economics of Welfare, he was prepared to devote some 100 pages to public finance (Pigou, 1920c, 587-688).
39 Pigou argued that a sound system of taxation is grounded in two principles. First, its impact on employment, consumption, saving, and investment would vary across income groups. The poor, living near subsistence and with little or no savings or other assets, would very likely meet their tax obligations by increased effort—the value of their marginal dollar increases markedly in comparison with the disutility of additional labor. The rich, on the other hand, had several options. They could reduce their consumption, draw on their savings, or liquidate assets. Second, the chief desideratum of a tax is its impact on the national dividend. The consequences of a tax for the national dividend depended on several factors: expected revenue; administrative costs; the perception on the part of taxpayers that the distribution of tax burdens was fair; and predictability, conceived as minimal interference with the ability of taxpayers to manage their economic affairs in an instrumentally rational fashion (Pigou, 1920c, 593-99).
40 Pigou invariably considered alternative modes of taxation in light of their effects on the national dividend. A windfall tax—levied on unforeseen and unearned increases in property values—was efficient and relatively equitable. In a peacetime economy, however, it might not be feasible to identify windfalls. Property values could vary due to factors independent of windfalls, including changes in the price level and interest rates. As a pragmatic solution to this problem, Pigou suggested that windfall taxes be limited to very large incremental changes in property values (Pigou, 1920c, 601-608). Because of their efficiency, taxes on the unimproved public value of land compared favorably with windfall taxes. As noted above, they singled out landowners invidiously and thus were defective on grounds of equity. Nevertheless, states were compelled to raise revenue, and the public value of land was a most attractive source, at least within reasonable limits (Pigou, 1920c, 609-615).
41 Taxes on expenditures could take different forms, targeting specific commodities or all consumer goods and services. Uniformity was not a requirement of taxes on specific products. Although taxes on articles of mass consumption were objectionable because they were regressive, their administrative costs were low. Because they were imposed at differential rates, they changed the relative volume of the consumption of goods, the marginal net products of resources invested in their production, and the national dividend—which would increase or decrease depending on whether the product taxed was produced under conditions of diminishing or increasing returns (Pigou, 1920c, 618-21). A uniform and modest ad valorem tax on all consumer goods and services—essentially a “general unregulated tax on expenditure”—would have little effect on the national dividend (Pigou, 1920c, 616). Although it might marginally reduce saving and on-the-job-diligence, it left the relative magnitude of the marginal net product of resources intact. The drawbacks of an ad valorem tax were based on other considerations: it was blind to product quality, consumer income, and family size. In addition, the difficulty and cost of collecting revenues from shopkeepers and providers of professional services were enormous. Such a tax was “an open invitation to fraud,” prevention of which would require “an army of inspectors” (Pigou, 1920c, 617-18).
42 A progressive income tax that exempted saving would eliminate disincentives to save and invest. Because it was derived from an income tax, it could exempt consumption for basic needs, differentiate between income levels, and accommodate family size (Pigou, 1920c, 625-27). Because it exempted saving, it affected the national dividend only by dampening incentives to work. Pigou considered the possibility that intellectual workers or the propertied class might seek refuge from such a tax in other countries. Even workers without means to emigrate might relax their efforts (Pigou, 1920c, 627). However, he discounted the weight of these possibilities. Relocating was inconvenient and costly, and countries that host expatriates also had income tax systems. Moreover, the rich preferred to live in their own countries due to the emulative social advantages conferred by wealth. Domestic entrepreneurs were not likely to work less energetically because of a tax on expenditures. They were motivated by a passion to succeed, which should not be conflated with profitability. Finally, the poor would respond to the tax by attempting to increase their hours of employment (Pigou, 1920c, 627-29).
43 The disadvantages of an income tax that did not exempt saving varied with several factors: the percentage of total income collected in taxes, the normal saving rate of the economy, and motivations for saving. If savers accumulated a certain sum for the education of their children or set aside their post-consumption income, taxes would not markedly reduce saving. However, insignificant changes in saving and capital might be magnified over time, translating into a lower national dividend in the long term (Pigou, 1920c, 631-32). In addition, the choice between an income tax that does and one that does not penalize saving depended on equity as well as efficiency. Even if income taxes did not exempt saving, they could differentiate earned and unearned income. A moderate tax on unearned income limited to the affluent would improve the fairness of the tax system. However, it might entail a cost in efficiency if it reduced the national dividend. Beyond certain thresholds, earned incomes could be progressively taxed and equal incomes earned by people of differing circumstances taxed differently. Equity dictated that bachelors be taxed more heavily than heads of families, with relief to family men diminishing as family income increased. Pigou devoted little attention to flat taxes, believing that there was a general consensus supporting non-uniform taxation based on ability to pay (Pigou, 1920c, 632-637).
44 For a variety of reasons, Pigou believed that property taxes were more effective than taxes on unearned income: they could be defined more easily and precisely; they did not penalize business incomes; and unlike income taxes, they captured capital gains. Nevertheless, property taxes were objectionable in several respects. Because income from property was volatile, taxing it in lean years could impose an undue burden on property owners. If property ownership promised potential future income, property values might change. However, it was unfair to tax potential, as opposed to actual, income. It might also be difficult to graduate taxes on property. More importantly, property taxes, unlike income taxes, were not amenable to a simple assessment and were more easily evaded. Property could be taxed annually or over longer intervals. In Pigou’s view, the latter was a more productive fiscal tool than annual taxes on property or unearned income. Taxpayers discounted the burden of taxation at longer intervals since it occurred in a remote future. One variant of such a tax, death duties, had the advantage of a remarkably heavy discount, since it was charged at an unknown future date after the death of the taxpayer. Taxes on inheritance could not be expected to discourage saving or reduce the national dividend significantly. They fell chiefly on the very rich, who would continue to accumulate notwithstanding taxes because of their love of the “power and prestige that riches confer” (Pigou, 1920c, 640-42).
45 Compared with contemporary economics, Pigou’s theory of economic policy analysis reserved a decidedly modest and narrowly circumscribed role for economists in human affairs. Political leaders, not economists, make economic policy. Economists who indulge in normative pronouncements on economic objectives transgress the boundaries of science. In advocating economic policies, they speak only as informed citizens with no special scientific qualifications. Otherwise Pigouvian theory marks them as incompetent—confused about the proper scope of their work—or as charlatans. Economists are restricted to the analysis of economic causes and effects—only one among many variables that statesmen consider in arriving at economic policies. Moreover, considerable circumspection is required in the statement of economic generalizations. Because of the complexity of economic problems and the resistance of basic economic factors to formalization and statistical analysis, generalizations have a limited scope that cannot be specified with confidence. Thus the necessity of Pigou’s case-by-case methodology. Finally, economic analysis is constrained by its historicity. Because economic generalizations and the techniques used to produce them are historical artifacts, their value for policy varies with contingent changes in circumstances that cannot be predicted.
46 As the foregoing account demonstrates, the trajectory of Pigou’s work in public finance from 1903 through the mid-1920s was tied to changes in the political and economic conditions of Britain. The political exigencies of 1903 were not those of 1914. Policies that were politically possible in one year might have no prospects for success in the next. Because of changes in the economic priorities of the state or civil society, policies once regarded as critical were replaced by new priorities. What Pigou wrote, when he wrote it, the theses he defended, and the arguments he employed mapped these changes. A Study in Public Finance incorporated the results of the analyses discussed above, following his case-by-case mode of investigation. In this book, the specifics of cases and the historical conditions that formed them were elided. However, our reconstruction of his research program in public finance documents the extent to which it was anchored in the politics and economics of his time. In Pigou’s view, attempts on the part of economists to achieve more ambitious aims in the domain of economic policy were scientifically unsound, economically dangerous, and damaging to the legitimacy of economics as a science.
We are grateful to the British Library and the Syndics of Cambridge University Library for permission to quote unpublished material and to the chief editor and referees of Œconomia for advice and criticism.