The theory of employment, common to orthodox economic think-ing in the nineteenth century, was based on a deceptively simple generalisation known as Say's Law of Markets. The formulation of the "Law" is generally attributed to the French economist Jean Baptiste Say, although John Stuart Mili later claimed part of the credit for his father James Mili. Succinctly expressed in the phrase, "supply creates its own demand", the law can be elaborated in the following series of propositions. The outgoings or costs of business laid out on purchase of raw materials, payment of wages to labour, on rent of land, interest upon capital, etc, are the earnings of workers, landowners and capitalists who furnish the factors of production. The income constituted by these earnings is the source of demand for the product of industry and commerce. Incomes are spent, paid in taxes or saved. What is spent goes directly to the purchase of goods. What is paid in taxes or saved is also spent on 1 Cmd. 6527, Employment Policy, March 1944. buying goods, but indirectly through the intermediary of govern-ments and of those who borrow for business and other purposes.1 The costs of business are thus recouped from the sale of the product; and the mere fact of producing a supply of goods automatically generates income enough to ensure a demand which will absorb out-put as a whole at a remunerative price. A particular business or trade might, it was allowed, be turning out at any time, more of its own product than the public at that time were prepared to buy. But if the goods had been produced and the bills met, then the costs had been paid out and the income generated. That income was spent on something by the earners who received it, by the govern-ments who took a part in taxes and by the borrowers of the remainder which had been saved; and it was up to those engaged in the declin-ing industry to leave that particular business, and betake themselves to the trades to which the demand had now been transferred. John Stuart Mili, writing in 1848, would admit in this case no over-production. "Production", he said, is not excessive, but merely ill-assorted . . . demand canfnot] be wanting when there is an additional supply; though it is perfectly possible that the demand may be for one thing and the supply may unfortunately consist of another.2 As long as there was production, there was employment; and the unemployed, if indeed there were any, should turn to making the products on which income was now being or would be spent. The Reverend Thomas Robert Malthus, better known as the author of An Essay on the Principle of Population—writing twenty years before during the unemployment and distress which followed the war against Napoleon—had argued that, for a continuing pros­perity, a balance should be preserved between the "productive" classes, who added to capital, and the "unproductive" who main-tained the rate of consumption. But expenditure on improvements of lasting value, in an age which drew these distinctions in an agri-cultural setting, though undeniably "productive" had the defect of being spasmodic. Once an improvement to a property was com-plete, the occupation of the labourers engaged on the work was at an end; and there could be no assurance that other improvements 1 Say: Treatise on Political Economy, Prinseps Edition, 1821, Vol. 1, p. 115. J. S. Mili: Principles of Political Economy, Book III, c. 14. Ricardo: "Principles of Political Economy and Taxation", Collected Works of David Ricardo, edited by Piero Sraffa, Vol. I, p. 290. 2 J. S. Mili: Principles of PoliticalEconomy, Book III, c. 14. would immediately be set in train sufficient to continue their employ­ment. An unproductive expenditure on consumption, on the other hand, even if wasteful, had the great advantage of satisfying demands which are strong, recurrent and always with us. A landlord with an eye to the futurę might once in a lifetime "invest"—that is engage labourers productively to build new barns or to drain or otherwise improve some part of his property. But he would always want footmen and gardeners. Their employment was secure—at least, as long as the landlord's rent roli (or his credit) might last. Without this spendthrift margin, so Malthus claimed, there would be gluts of unsold goods for which no purchasers could be found; and honest and otherwise industrious men would find themselves without work. To the suggestion that unemployment might be caused be-cause the recipients of income were accumulating too assiduously, the proponents of Say's Law had a ready answer. David Ricardo, probably the ablest among them, retorted shortly that "Mr. Malthus never appears to remember that to save is to spend, as surely as what he exclusively calls spending".1 Economists in the classical tradition (and many others), thinking generally in terms of real wealth, have been apt to regard income as the actual goods and sendces in which output consists—and not as the sum total of the money by which the value of that output is measured. Thomas Attwood, a Birmingham banker, speaking at a town meeting in 1829, expressed an opinion commonly held—at least among the unlearned—when he said that "human wants had no limit but that which is imposed by deficiency of money"; and that, by increasing the supply of money, "you are sure to increase the demand for commodities and labour to any extent you please".2 That central distinction between "saving", the passive act (or should one say "non-act"?) of "not spending", and investment, the creative act of adding to the sum total of capital goods, was not to be ex-plicitly recognised by the economists for another century. Saving, to Ricardo and to Malthus, was one form of expenditure like any other but on goods for productive purposes and not an unproduc-tive consumption—expenditure, as we should cali it now, on goods and services for capital account as distinguished from expenditure on current consumption. Malthus and those who thought with him, accepting the same premise as Ricardo, could find no reply to his 1 David Ricardo: "Notes on Malthus", Collected Works of David Ricardo, edited by Piero Sraffa, Vol. II, p. 449. 2 Thomas Attwood: Distress and the State of the Country, pp. 19-20. I owe the reference to Professor E. W. Fetter. criticism: and Say's Law was successfully enthroned as the orthodox dogma. The recurrence after 1870 of long-continued depression, following the twenty years or so of comparative prosperity in the mid-Victorian age, stimulated further reflection. The possibility of a generał under-consumption was revived both by liberał thinkers and by the social revolutionaries. J. A. Hobson, in a series of booLs_bćgiPJiing with the Physiology of Indus try,1 set himself to promrjte the heretical doc-trine that unemployment was caused by a supposed periodic (or chronić) insufficiency in the income distributed to consumers. The productive capacity of mechanical industry, he argued, was too great. Too large a share of income was being accumulated as an addition to the capital stock; and too little was left to furnish'the means of purchasing the resulting output. The conseąuences were glut and unemployment among the producers of the consumption goods which were being left unsold. This was followed inevitably by un­employment among the producers of capital goods, whose markets had disappeared in the depression of the consumer goods trades. But the "underconsumptionists" of the late nineteenth and early twentieth centuries, like the Malthusians before them, did not suffi-ciently distinguish between the act of saving—considered as the with-holding of money income from expenditure—and the act of invest-ment understood to be the actual laying down of plant or other capital eąuipment and the purchase of goods for stock. Saving again appears in the underconsumptionist literaturę, as in the writings of the orthodox, as synonymous with an output of capital goods. Saving is postponed consumption—i.e. the production of futurę goods, plant, machinery and raw materials in their several stages, instead of commodities for immediate consumption.2 The heretics laid themselves open to the obvious ripostę. If capital-ists, by saving, were actually responsible for producing goods— capital goods, it is true, and not consumption goods; but goods nevertheless—then income would be generated by the production of those goods. That income had to be disposed of; and it would pro-vide for employment, whether spent on the production of consump­tion goods, or saved in the production at that time of goods for futurę consumption. And if capitalists, faced with the lalling returns which all agreed to be the result of excessive investment, should 1 Written jointly with A. F. Mummery and published in 1889. See also Eco­nomic Journal, 1953, pp. 933-4. 2 Hobson: Evolution of Modern Capitalism (The Walter Scott Pub. Co., 1906), Chap. XI, p. 296. stop accumulating—should stop spending, that is, on capital goods— they would then surely spend on consumption goods the income they no longer saved. "Spending" and "saving" were the alternative means of disposing of income. If income was not "spent", it had to be "saved"—and vice versa. What else could be done with it? There could, of course, be unemployment in this or that trade as a result of a mistaken anticipation of demand. But unemployment from that cause could continue only when, by trade union restrictions or by disproportionately high wages, labour was held stagnant in trades depressed because demand had failed for that particular product. Unemployment unhappily did persist. One Royal Commission on the Depression in Trade reported in 1895 on the possible causes of unemployment; another Royal Commission issued to enquire into the Poor Laws and the provision for the destitute. Mr. W. H. (now Lord) Beveridge contributed his study of unemployment in 1908. He recommended payment to the unemployed of benefit financed from a scheme of State insurance, and a system of employ­ment exchanges to facilitate the movement of labour. A minority of the Poor Law Commission in 1909 made the radical proposal "that Government contracts for structural work should be concen-trated upon times of bad trade". This was criticised by one well-known economist on the orthodox ground that the "Government, by the very fact of borrowing for this expenditure, is withdrawing from the investment market savings which would otherwise be applied for the creation of capital". But "the whole point", as another eminent economist, Mr. D. H. (now Professor Sir Dennis) Robertson, so justly replied, was "that in times of depression savings are not otherwise so applied".1 The keystone had been unseated in the arch. If it could not be shown that income saved must be spent on capital goods, either by the saver or by a borrower, then it could not be assumed that the whole of the income generated in production was necessarily spent on buying the product, either as capital or as consumption goods. That income saved is spent just as surely as any other direct outlay, but on capital goods instead of consumption goods, is fundamental to Say's Law. Once Professor Robertson's point is admitted, that 1 Robertson: Industrial Fluctuations, p. 253 note (London, 1915); and his Presi-dential Address before Section F of the British Association for the Advancement of Science, 1947. Professor Robertson on this occasion disguised his early self as "another Cambridge economist" and, with characteristic excess of modesty, gave his audience no reference to the source from which he drew his quotation! Economic Journal, Sept. 1947. See also Robertson: Banking Policy and the Price Level, first published in 1926. savings can be altogether withheld from the market, it can no longer be claimed, as John Stuart Mili had put it, that "whoever brings additional commodities to the market brings an additional power to purchase . . . [and] also an additional power to consume". There could, it seemed, be a generał overproduction of commodities if, for example, savers and other holders of accumulated wealth should stand off the market and refrain from investing their funds in stock, equipment and other real capital. The income generated by pro­duction would not then be wholly returned as the purchase price of the product. One cause of unemployment could thus be the inability of business to sell goods at a profit, because of a generał failure of demand for the output of industry as a whole. The fuli implications of this pregnant observation were not realised at the time by a gener-ation brought up in the orthodox tradition, and it was another ten years before the fruit was borne. The Liberał Party in 1929, led by Mr. Lloyd George, had issued an election manifesto entitled "We Can Conquer Unemployment". Based on a detailed study by the Liberał Industrial Enquiry,1 the manifesto pledged the Liberał Party to an ambitious programme of public works. The proposals were defended by Mr. J. M. (the late Lord) Keynes, and Mr. H. D. (the late Sir Hubert) Henderson. In a pamphlet entitled "Can Lloyd George Do It?" Keynes and Hender­son put it forward, at the same time, the more generał and much more significant idea (significant, at least, among the professional economists of the day) that generally speaking, the indirect employment which schemes of capital expenditure (during depressions) would entail is far larger than the direct employment. This fact is one of the strongest arguments for pressing forward with such schemes; for it means that the greater part of the em­ployment they would provide would be spread far and wide over the industries of the country. Expenditure on works on a sufficiently large scalę could, in fact, stimulate a revival of trade. The fact that many workpeople who are now unemployed would be receiving wages instead of unemployment pay would mean an increase in effective purchasing power which would give a generał stimulus to trade. Moreover, the greater trade activity would make for further trade activity; for the forces of prosperity, like those of trade depression, work with a cumulative effect.2 1 Britain's Industrial Futurę, Benn, London. s Keynes and Henderson: Can Lloyd George Do It?, London, 1929, p. 25. See also Harrod: Life of John Maynard Keynes, pp. 295,346,413,422,424,441 and 447. B The Liberałs were not in fact returned at the election of 1929, nor did these ideas commend themselves to the leaders of the two larger parties. Compelled to find millions to meet the increasing debt of the Unemployment Insurance Fund, Tory and Labour Chancellors of the Exchequer alike clung firmly to national economy and Iow rates of interest, retrenchment and cheap money, as the only prac-ticable defences against and remedies for depression and persistent unemployment. The Chancellor in 1929, when opening his budget, reported that the results of expenditure on public works had been disappointing, in fact so meagre as to lend considerable colour to the orthodox Treasury doctrine which has steadfastly held that whatever might be the political and social advantages, very little additional employment and no per-manent additional employment can, in fact and as a generał rule, be created by state borrowing and sta te expenditure.1 The (May) Committee on National Expenditure, reporting two years later at a time when unemployment had taken a decided turn for the worse, were unable to endorse a policy of great expenditure on capital works, irrespective of their economic value, as a means of providing work. It is too expensive. On the usual calculations of 4,000 man years of labour, direct or indirect for £1,000,000 expenditure it costs £250 to keep a man in work for one year. The saving on maintenance of the man is usually put at about £60 a year. Thus, proceeding on this basis, unless the man's work is worth at least £190 to the nation when it is done, the nation loses economically by carrying it out.2 Ideas about unemployment were growing, nevertheless, both in the academic community and without. Lord Keynes' General Theory of Interest, Employment and Money, appearing in 1936, had been preceded and was followed by much discussion, often acrimonious but always informative. Men's minds, towards the end of the second world war, began to turn once more to the old problem of recurrent and chronić unemployment. Economists, almost unanim-ously, were now ready to recommend that in the event of a depres­sion, Government should undertake to assist with a programme of public works, financed if necessary at the cost of deficit in the budget. This much was common ground within the Government and without. It was accepted on both sides of the Atlantic. The American Govern-ment of the day, no less than the British, would have been ready to 1 Mr. (now Sir) Winston Churchill, 227 H.C. 54. 2 Report of the Committee on National Expenditure, para. 358, p. 139 (Cmd. 3920, 1931). incur a deficit on its budget in order to prevent the re-emergence of unemployment. The long dominion of Say's Law was over, and the way was elear for a plan, the plan against unemployment.