WPC 2 B3P Z Courier 10cpi#Xt4 PoG,XP#HP LaserJet IIIHPLASIII.PRSx  @,\,A]X@21<3L.X#|t#Xt4 PoG$XP#Courier 10cpiTms Rmn 12pt (Rmn-8) (25inOne!)PRSx  @ 0QX@2xxx,x  @87X@ *tC4,,Xt4 PoGXP2xxx,.x  `B7Xa) -pp2(#p a1DocumentgDocument 26cvpkCourier 10cpiTmsRmn 10pt (Pi) (K)Courier 10cpi Bolda8DocumentgDocument Style StyleXX` `  ` a4DocumentgDocument Style Style . a6DocumentgDocument Style Style GX  2kv,ta5DocumentgDocument Style Style }X(# a2DocumentgDocument Style Style<o   ?  A.  a7DocumentgDocument Style StyleyXX` ` (#` BibliogrphyBibliography:X (# 2 H   , a1Right ParRight-Aligned Paragraph Numbers:`S@ I.  X(# a2Right ParRight-Aligned Paragraph Numbers C @` A. ` ` (#` a3DocumentgDocument Style Style B b  ?  1.  a3Right ParRight-Aligned Paragraph Numbers L! ` ` @P 1. ` `  (# 2   ~ F a4Right ParRight-Aligned Paragraph Numbers Uj` `  @ a. ` (# a5Right ParRight-Aligned Paragraph Numbers _o` `  @h(1)  hh#(#h a6Right ParRight-Aligned Paragraph Numbersh` `  hh#@$(a) hh#((# a7Right ParRight-Aligned Paragraph NumberspfJ` `  hh#(@*i) (h-(# 2H !+a8Right ParRight-Aligned Paragraph NumbersyW"3!` `  hh#(-@p/a) -pp2(#p a1DocumentgDocument Style StyleXqq   l ^) I. ׃  Doc InitInitialize Document Style  0*0*  I. A. 1. a.(1)(a) i) a) I. 1. A. a.(1)(a) i) a)DocumentgTech InitInitialize Technical Style. k I. A. 1. a.(1)(a) i) a) 1 .1 .1 .1 .1 .1 .1 .1 Technical2JXa5TechnicalTechnical Document Style)WD (1) . a6TechnicalTechnical Document Style)D (a) . a2TechnicalTechnical Document Style<6  ?  A.   a3TechnicalTechnical Document Style9Wg  2  1.   2SGa4TechnicalTechnical Document Style8bv{ 2  a.   a1TechnicalTechnical Document StyleF!<  ?  I.   a7TechnicalTechnical Document Style(@D i) . a8TechnicalTechnical Document Style(D a) . 2,PleadingHeader for numbered pleading paperP@n   $] X X` hp x (#%'0*,.8135@8: 1994, p. 433.ē To this observation we would add the ILO work which contributed to thinking on a range of major international economic problems. Our objective is to sketch the ILO economists' perspectives in the light of growing interest by intellectual historians in the ideas of economists as one "recognizable professional cadre" in  ?L international agencies.LB T& ԍ A.W. Coats: "Introduction", in Coats (ed): Economists in international  tT' agencies: An exploratory study (New York, Praeger, 1986), p. 2. It is no longer satisfactory to acceptLg 0*((@@ the judgement expressed in 1933 by G.D.H. Cole and Margaret Cole, that "the ILO has done much useful work, though this has been rather in the field of promoting international discussions and spreading information about the various countries than in the  ? direct improvement of industrial legislation."mB T ԍ G.D.H. Cole and M. Cole: The intelligent man's review of Europe (1933)  tT quoted in J. Price: ILO: 50 years on (London, Fabian Society series 275, April  tT 1969), p. 1. For in our view the ILO economists did not merely promote discussion and disseminate information; they were active advocates of particular perspectives on the operation of international economic processes, and on the methods which might be used to stabilize and harness those processes for the purposes of economic and social reform. Indubitably, as Herbert Feis observed in 1927, the work of the ILO  ? was "primarily educational". B Tq ԍ H. Feis: "International labour legislation in the light of economic  tTN theory", in International Labour Review, Vol. 15, No. 4, 1927, p. 518. In fact, some of the contributions of the ILO economists in the 1920s and early 1930s advanced beyond a mere educational function. The contributions of the economists became engines of advocacy in the pursuit of a broad concept of economic and social justice. Justice was to be achieved through international economic policy coordination delivering high levels of employment and high living standards. The first exploratory study of economists in international agencies considered all the major agencies the United Nations, G.A.T.T., World Bank, Organisation for Economic Cooperation and Development and International Monetary Fund but the ILO was  ? conspicuously omitted.jB T! ԍ A.W. Coats (ed) op. cit. More recently, and relevant to our period of research, is a study of the ideas promoted by the League of Nations economists in the 1930s; N. de Marchi, with the collaboration of P. Dohlman: "League of Nations economists and the ideal of peaceful change in the decade of the  tT$ 'thirties", in C.D.W. Goodwin (ed): Economics and national security: A history  tT% of their interaction, (Durham, Duke University Press, 1991), pp. 143178. The main figures in our study are J.R. Bellerby, H. Fuss, P.W. Martin, J.H. Richardson, and E.J. Riches.  0*((@@ These economists analyzed the following issues in some detail during the course of their work for the ILO during the 1920s and 1930s: monetary policy and employment stabilization measures in the 1920s; the problem of "underconsumption" and wages policy in the 1920s and early 1930s; challenges to more orthodox policy responses to economic depressions, and a policy agenda for international economic reconstruction in the 1930s. We consider each of these topics seriatim and offer summary observations in the final section of the paper.  ?  Employment Stabilization and Monetary Reform in the 1920s From its inception in 1919, there was a grand vision underpinning the research focus of the International Labour Office which went beyond an interest in conditions of labour per se. The founding director of the office expressed this vision by linking social problems associated with conditions of labour internationally with general economic problems. Thus, it X"is no longer possible to separate them [social questions] from the whole economic question...Nowadays during employment crises, it is not merely relief or employment that is demanded but the whole system of organisation of industry, distribution of raw materials,  ? and foreign exchange is called to account".B TL ԍ A. Thomas: "The International Labour Organisation: Its origins,  tT) development and future", in International Labour Review, Vol. 1, No. 1, p. 19.  The advisory capacity of the Office (when labour matters were due for discussion at the Assembly of the League of Nations) was founded on a detailed knowledge of all subjects relating to international adjustment of industry and labour. So as to fully enlarge their understanding of such matters, the Office's researchers were led to consider broader economic and financial policies affecting industrial output, consumption, investment and business cycles on an international scale.T$50*((@@ԌFrom the standpoint of the ILO, while the Genoa Economic and Financial Conference of 1922 on monetary issues recommended measures of currency reform through "continuous cooperation among  ? central banks of issue" B T ԍ From Resolution 3 of the Report of the Financial Commission of the Genoa  tT Conference 1922.Ĕ international monetary reconstruction was not to be regarded as an end in itself; it was merely an instrumental goal. To be sure, from the early 1920s the problem of central bank cooperation over money, credit and currency policy preoccupied economists and policymakers, including those working  ?` for international agencies. ` 5B T% ԍ On this point and the following details in this paragraph we have drawn extensively on B. Eichengreen: "International policy coordination in historical perspective: A view from the interwar years", in W.H. Buiter and R.C. Marston  tT (eds) International economic policy coordination (Cambridge, Cambridge University  tTZ Press, 1985), pp. 139141.ģ In particular, Britain's return to the gold standard in 1925 signalled success (in the minds of some) in reconstructing the international monetary system in the interwar period, although the "correct" valuation of sterling remained a moot point. The League of Nations assisted nations experiencing hyperinflation in the early 1920s by providing stabilization loans in return for commitment to tight internal budgetary policies and a return to the gold standard. There was a distinctive line of thought on these events issuing from ILO publications a line which insisted that the goal of monetary reform should not take for granted the achievement and maintenance of a high, stable level of employment. A proximate goal of international economic policy should be price stability (through monetary reform) since price stability entailed keeping down the living costs of workers. The report of J.R. Bellerby for the International Association on Unemployment entitled "Control of Credit as a Remedy for Unemployment" (1923), was discussed  ?L approvingly in the ILR.l LB T& ԍ Bellerby was an economist on the staff of the ILO from 19211927.l Here the most pertinent remark onL 0*((@@ existing monetary institutions was that "to the extent that banks, by...regulating the issue of `money' of all kinds, can influence the movement of the price level, they are handling a most powerful  ? and sensitive instrument of industrial welfare".* mB T ԍ [Anonymous]: "Bank credit and unemployment", in International Labour  tT Review, Vol. 9, No. 1, 1924, p. 79. This article was probably written by Henri  tT Fuss, an economist and chief of the Unemployment Service of the ILO.* Quoting the views of W.C. Mitchell, a leading American economist, the article condemned widespread price fluctuations which "promoted speculation and reduced the efficiency of management and labour. We are poorer in goods, more quarrelsome in spirit, less ready to work because of  ?` these fluctuations".= ` B T ԍ Ibid: p. 80.= Instead of inveighing against private employer organizations for irresponsible employment and wage setting practices, the message of ILO research in the early 1920s was that economic instability and unemployment were caused by the misguided policies of monetary authorities. The ILO economists often in collaboration with colleagues in the Economic and Financial Section of the League of Nations, undertook several crosscountry studies of monetary institutions and economic fluctuations, which were designed primarily to provide empirical  ? evidence in support of this view.0 mB T ԍ See, for example: ILO Studies and Reports Series C No. 8, Unemployment  tT 192023 (Geneva, 1924) especially part III, pp. 135154 and No. 9 Unemployment  tT in its national and international aspects (Geneva, 1924), pp.6669.0 Consistent with a fundamental tenet of economic theory in the 1920s, the ILO economists presented a simple monetary theory of economic instability, giving prominence to the link between the money supply and prices. The "orthodoxy" accepted at the ILO was mostly British going by the high frequency of citations to J.M. Keynes and R.G. Hawtrey, although the Swedish monetary theorist G. Cassel is also cited. Important theoretical differences between these economists were mostly set aside to underscore the essentially monetary aspects of international business cycles whichx 0*((@@ were thought to cause significantly high and persistent unemployment. Keynes had stressed (in a more sophisticated treatment than could be gleaned from ILO publications) that money balances held by individuals are an abode of purchasing power; they are instruments held for the utility that can be derived from the  ? goods they purchase.B T4 ԍ J.M. Keynes: A tract on monetary reform (London, MacMillan, 1923), pp.  tT 6162.Ď ILO publications always sought to associate the idea of money as an "instrument of payment" in exchange and international trade with prices and ultimately with conditions in labour markets. So we have, for instance, Fuss expressing a conviction that a "rise in the general price level indicates that the supply of instruments of payment is increasing more rapidly than the quantity of goods offered for sale; a fall in the price  ? level indicates a relative scarcity of these means of payment."5B T ԍ H. Fuss: "Money and unemployment", in International Labour Review, Vol.  tT 16, No. 5, 1927, p. 603.ġ Accepting empirical work published in the ILR demonstrating a close positive relation between movements of prices and employment in the U.S.A. 19031925, Fuss argued that an increase or decrease of unemployment is to be explained predominantly by changes in the ratio of money (the instrument of payment) and goods produced by labour. Therefore, the policies of monetary authorities in different nations should attempt "to ensure that the supply of the instruments of payment shall increase at a rate corresponding  ? exactly to the expansion of production."B T ԍ Fuss: ibid, pp. 604605. An early version of this position is  tT adumbrated in Fuss: "Unemployment in 1925", in International Labour Review, Vol. 14, No. 2, 1926, pp. 228231. The empirical research he referred to was completed by eminent American economist, Irving Fisher: "A statistical relation  tT# between unemployment and price changes", in International Labour Review, Vol. 13,  tT# No. 6, 1926, pp.787792. While there was some published comment on nonĩmonetary causes of unemployment in ILO work up to 1927, these always seemed to be mere obiter dicta. Instead, the focus was on causes of / 0*((@@ unemployment emanating from monetary disturbances. In this, it rightly drew attention to the functions and performance of the international economy as a whole in the long run; the macroeconomic causes of unemployment were patently more important in this context. Short run factors resulted in what Bellerby called "minor  ? cycles" in unemployment in specific countries.sB T4 ԍ J.R. Bellerby: Monetary stability (London, MacMillan, 1925), p. 6.s These included changes in: wage setting legislation; migration; domestic and international competition between firms; frictional unemployment; seasonal factors; tariff protection, taxes and technological processes. Also structural maladjustments "between the quantities of each commodity produced and the quantities actually needed to satisfy the infinite variety of changing human wants"  ? may result in transitory unemployment at the local level.RmB T  ԍ Fuss: "Money and unemployment", p. 602.R During the first half of the 1920s, measures designed to alleviate short run fluctuations in unemployment such as unemployment insurance, labour exchanges, vocational guidance and public relief work were  ? researched in some detail.w5B Tb ԍ See ILO Studies and Reports Series C, No. 10, the unsigned survey  tT? article "Employment and unemployment", in International Labour Review, Vol. 7, No. 23, 1923, pp. 298320, and K. Pribram: "The unification of social  tT insurance", in International Labour Review, Vol. 11, No. 3, pp. 303317.w However, the preeminent question of preventing unemployment by securing sound, coordinated international responses to international monetary disturbances took pride of place; it was a principal subject of ILO research. The work of Bellerby in particular brought together "an impressive collection of opinions from many sources"; he was intimately involved in "the almost revolutionary improvement in our understanding of the mechanism of money and credit and of the analysis of the trade cycle" (together with its labour market 0*((@@  ? impact) in the 1920s.mB TX ԍ These expressions of praise issued from J.M. Keynes in March 1924. See  tT5 J.M. Keynes: The collected writings of John Maynard Keynes, Vol. XI (London,  tT MacMillan, 1983), p. 419 and 419n. He was tireless in insisting on "the urgency of regulating the rate and range of price movements"  ?X through monetary policy reform so as to reduce unemployment.XB T ԍ Bellerby: "The controlling factor in trade cycles", in Economic Journal,  tT Vol. 33, September 1923, p. 305.Ī Again, this position accepted without question that price  ? fluctuations were paralleled by unwanted employment fluctuations.jB T ԍ E.g., ILO: Unemployment 192023, Studies and Reports Series C, No. 8,  tT 1924, p. 8 and p. 144.Ğ In principle, the reduction of unemployment as a problem of international economic policy turned on addressing general price level movements or price inflation and deflation in the first instance. Coordinated monetary policies between nations were a  ? prerequisite for success.N B T ԍ Ibid, p. 138 et passim.N By comparison, general national programmes designed to stabilize employment and alleviate unemployment had an important, though subsidiary, place; these could be much more diverse and multifaceted at the microlevel depending on a nation's institutional arrangements. Here too, "coordinated, organic action" was required, necessitating a "concerted plan of stabilization" between various components of  ? domestic economic policy (employment and financial components).| B T ԍ J.R. Bellerby: Stabilization of employment in the United States, ILO  tT Studies and Reports Series C, No. 11, 1926, p. 100.Ĺ The first issue to be encountered in practice was which institution at the national level should be responsible for policy implementation aimed at price stability. Expansions and contractions of credit led to price movements; private banks were responsible in one sense for regulating the level of money and credit in the economy. That is, "the banks...are the agency 0*((@@ through which the `money' factor in production is controlled; and if the monetary system fails to give satisfaction, it is through this agency that reform may be expected". Since unemployment at historically high levels often persisted in these conditions, an independent monetary policy operated by private banks was not in  ? the best interests of society.mB T4 ԍ [Anonymous]: "Bank credit and unemployment", pp. 79, 80. See also J.R.  tT Bellerby: Control of credit as a remedy for unemployment (London, P.S. King and  tT Son, 1923). Following recommendations of the Genoa Conference, ILO economists were adamant that central banks must be established in all countries. Central banks would play an active role in operating monetary policy and, at one remove, an employment policy based on the fundamental tenet of price stability. But how were central banks to behave? Under a simple gold standard operating efficiently internationally, central banks intervened in money markets by reinforcing the impact on national money markets of balance of payments conditions. A central bank losing gold reserves (perhaps gold outflows following a balance of payments deficit) would tighten the money supply and vice versa. Countries following this rule "reinforce one another's efforts to  ? restore external balance", currency values and price stability.B Ty ԍ Eichengreen: "International policy coordination in historical  tTV perspective", p. 139.Ď However, extensive research by economic historians has since shown that this system did not work harmoniously in the 1920s something that ILO economists were well aware of at the time. It was clear to Bellerby and Fuss in particular that the gold standard did not work quickly enough to secure high employment levels; the standard was usually "effective" only with wage deflation, major downward price fluctuations and the creation of unemployment. One of the principal methods used by central banks to intervene in money markets, the discount rate, was rendered effective irrespective of the efficient operation of a gold standard. A central bank could vary the rate at which it would j0*((@@ lend to other banks by discounting their bills. For example, when private banks are obliged to pay more for replenishing their cash reserves, they must charge more for credit given to customers; interest rates will therefore tend to rise, thus restricting economic activity and dampening price inflation. Another more direct method of controlling the money supply, open market operations, involved timely buying and selling of government securities to the public. Central banks could thereby change the reserves held by private banks which were used as a basis for credit creation. These reserves were recognized by the ILO as the most important variable affecting the money supply and hence the  ? level of economic activity. B T< ԍ This point was first very forcefully made in 1923 by Keynes: A tract on  tT monetary reform, pp. 141 and 144, and it was repeated in ILO publications, e.g.: Studies and Reports Series C, No. 8, 1924, p. 151; [Anonymous] "Bank credit", p. 221; [Anonymous] "Unemployment in 1924", p. 90. See also J.R. Bellerby: "The  tTq monetary policy of the future", in Economic Journal, Vol. 34, June 1924, p. 184. Now it is at this point in their reasoning that the ILO economists made an innovative contribution both to contemporary theory and international economic policy. The problem for many central banks in operating monetary policy was the choice of which economic indicator to consider when targeting bank reserves or interest rates. If the quantitative indicators used by policymakers to regulate money and credit were inadequate, narrowly based or poorly designed, employment levels would be reduced unnecessarily. Missing in contemporary monetary policy deliberations were indices of production and employment. The manipulation of bank reserves and interest rates by central banks had to be forward looking or anticipatory and must not be attempted without reference to the progress of nonmonetary variables viz,: the level of output and employment. Indeed "the opportunness of the occasion [for interest rate manipulation] is best judged by the position of the employment index, taken in conjunction perhaps with  ? an index of physical production."B T& ԍ [Anonymous]: "Unemployment in 1924 and the beginning of 1925", in  tT' International Labour Review, Vol. 12, No. 2, 1925, p. 224.Ľ 0*((@@ԌIn the above vision, monetary policy based on the movement of gold reserves movement responding to international trade and payments flows would need to be replaced by a continuous system of credit regulation based on the observation of industrial conditions assessed by reference to a wide range of quantitative variables. While price stability was a prime objective, the major problem with a reliance solely on price indicators was that there were irregularities in the relationship between prices and employment: "a rise in prices does not always mean that trade is  ? active, and a decline does not always point to depression."M B T ԍ [Anonymous]: "Bank credit", p. 84.M It was crucial that the effects of price movements, that is, effects on production and employment, be understood and monitored in tandem with price indices. The link between unemployment and price level fluctuations was wellestablished by copious ILO case studies of experience in the Netherlands, Germany, France, Scandinavia, Czechoslovakia, Belgium and the U.S.A. In early ILO publications these studies were used to infer that a stable or slightly rising price level was consistent with high levels of employment: X"Although many other factors influenced the labour market by expanding or restricting the demand for labour, there were many striking coincidences between the varying periods at which unemployment began to increase in the different countries and the periods...at which the general price level ceases to rise and begins to fall. Moreover, in countries where the fall of price was most prolonged unemployment was most severe and persistent, while...unemployment decreased most rapidly in countries where the fall of prices was brief and was soon succeeded  ? by a rise or by stability."W mB T$ ԍ [Anonymous]: "Unemployment in 1924", p. 223.W  Data on employment and unemployment, if taken in conjunction with the movement of price indices, would therefore form appropriate(# 0*((@@ criteria for the guidance of monetary policy. Production indices would also help, although they were closely correlated with  ?X employment at least in the U.S.A. where statistics were reliable.XB T ԍ Bellerby: Control of credit, p. 90. Bellerby (ibid., pp. 8283) found correlation of +0.86 between employment and production indices for New York and New Jersey data, and reported a correlation coefficient of +0.89 between employment and pigiron production for the same area. See also Bellerby: Studies  tT and Reports Series C, No. 11, 1926, pp. 8499.ħ The use of employment and price level data would allow a "scientific" approach to the setting of the relevant discount rate and the achievement of stable income and employment: "if such principles were observed, there would seem reason to believe that, by a policy first of stimulation, then, when the moment arrived, of stabilization, industry might be maintained at a higher average  ? level of activity than has ever previously been achieved."  B T ԍ [Anonymous]: "Bank credit", p. 85, emphasis in original. In a private expression of opinion formed by his research at the ILO, Bellerby estimated that when the unemployment rate reached 2.5 to 3%, rising production was overtaken by price inflation. He recommended that monetary policy might be driven by a rule to halt economic expansion at that point; Bellerby: "The controlling factor in  tT trade cycles", p. 310. ILO publications did not discuss practical problems accompanying any attempt to synchronize national and world price level stability. An "international policy of price stabilization" was asserted as being desirable. Ostensibly, the interdependence of economies implied that no country (unless it was completely closed to international trade) could operate an independent monetary policy for very long. Given this interdependence, especially between major world economies, it was importunate that monetary authorities collaborate in following the new procedure which set the proximate objective of price level stability. Achievement of the latter would be made possible through a "scientific" combination of employment, production and price indices. The Genoa Conference recommended collaboration between central banks; ILO research had suggested the quantitative indicators over which such collaboration should be effected. The   0*((@@ reaction of central banks to quantitative indicators, including the problem of weighting each indicator would require policy guidelines. Bellerby's staunch internationalism led him to suggest the creation of a new international body to set the rules for controlling credit; thus "a small executive comprising the more powerful economic nations could, by action in their own countries, and recommendation to the remaining Members, direct with accuracy  ?4 and certainty the currency policy of the whole world."R!4B T ԍ Bellerby: "Controlling factor", p. 328.R The ILO would concentrate on the collection of data and the design of specific quantitative indicators relating to employment, unemployment, and output, capable of valid crossnational  ? comparison." mB T ԍ Bellerby: Control of credit, p. 90. Also Fuss: "Money and  tT unemployment", p. 617.Ē Economic investigations at the ILO in the 1920s up to about 1927, were not unique in holding the view that the monetary authorities can remedy high levels of unemployment indirectly by a macroeconomic policy which avoided price and wage deflation and promoted "confidence of the business world in the existing level of  ? prices."#B TV ԍ J.M. Keynes: 1923 quoted in [Anonymous]: "Bank credit and unemployment",  tT3 p. 91.Ċ Price level stabilization, however operationally defined and achieved by means other than reliance on the gold standard, was a prescription that ILO economists held in common with J.M. Keynes and several other luminaries at the time. There would have been general agreement among economic researchers at the ILO with Keynes's cautious comment that (at least) debate was required in the 1920s on "whether by a control of the monetary machine we cannot cure one of the greatest and most dangerous evils  ?L of the economic system" [ viz,: unemployment].$LGB T#& ԍ J.M. Keynes: "Discussion on monetary reform", in Economic Journal, Vol.  tT' 34, June 1924, p. 170.ğL $0*((@@ԌIn sum, the literature issuing from ILO research surveyed up to 1927 on the subject of monetary reform and employment have not been insipid, uninteresting collections of data and case studies. What published ILO research offered was a determinedly activistic, interventionist stance on this subject. Policy proposals were concerned with preventing unemployment by acting on macroeconomic causes of a monetary nature. We have identified five major policy recommendations at this level: (i) strengthen central banking institutions so as to make private banks more dependent on these institutions; (ii) the chief objective of monetary policy should be price level stability; (iii) price level stabilization is not meaningful in isolation and must be contemplated and achieved in the light of the ultimate goals of output and employment stabilization at high levels; (iv) formulate active, discretionary monetary policies using discount rates and open market operations. If they were to be used wisely, these policies required concerted action by central banks in the major world economies otherwise they could be inimical to employment; and (v) monetary policies may guide the macroeconomy successfully only if price, output and employment indices are used to inform discretionary actions.  ?!  "Underconsumption" and Wages Policy in the Late 1920s and Early  ?(# 1930s That the ILO economists had demonstrated how monetary policy was closely linked to unemployment, did not peremptorily imply a panacea which could be adopted with ease to deal with the malady. '$0*((@@ First, there were practical issues of timing monetary responses in anticipation of "instability" in prices and employment. Secondly, and more crucially, the difficulty of contriving a consensus both within and between nations on the precise meaning of price  ? stability could not be underestimated.Z%B T ԍ See Bellerby: Monetary stability, p. 146.Z Thirdly, there was also the vital change in thinking on the negative consequences of using the discount rate as a monetary tool which is associated with J.M. Keynes; ILO economists remained unaware of these consequences in the 1920s. Discount rates can be used to tighten monetary policy and contract credit; tighter credit would not just connote falling commodity prices, it also necessitated reductions in money wages and employment. As Keynes came to recognize in 1930, partly as a result of logical implications and partly from experience in the 1920s, there "is no way the Bank rate brings down prices except  ?h through the increase of unemployment."(&mhmB Te ԍ J.M. Keynes: "The MacMillan committee", in Activities 19291931:  tTB Rethinking employment and unemployment policies, in The collected writings of  tT  John Maynard Keynes, Vol. 20 (London, MacMillan, 1981), pp. 4849.( In short, successful price stabilization following a period of inflation depended for its success on the creation of unemployment. A policy induced fall in the price level was a cause of unemployment and, contrariwise, unemployment reduces national expenditure and therefore could be considered a contributor to falling prices. The objectives of price stabilization and high levels of employment may conflict. As if implicitly recognizing a conflict of objectives, increasingly in the 1920s and early 1930s ILO research shifted its emphasis from price stabilization toward the demand side of the economic system. Specifically, there seemed to be a wider search for preventative measures and remedies for unemployment which no longer relied heavily on monetary policy and price adjustments to effect a high and stable level of employment. ILO economist P.W. Martin accepted the goal of price stabilization, but argued trenchantly that employment increases did not follow automatically!&0*((@@ or very speedily when the level of prices was stabilized. A high level of employment required that "effective demand in the hands of the community" be adjusted so as to provide sufficient purchasing power to "buy all that industry can produce". In Martin's view it could not legitimately be presumed that there were adequate markets for all society's produce; instead a "scientific reinforcement" of buying power was required which was in principle compatible with the operation of a monetary policy having price stability as its  ?` objective.'` B T ԍ P.W. Martin: "Overproduction and underconsumption: A remedy", in  tT International Labour Review, Vol. 14, No. 1, 1926, pp. 47, 48, 51. In this context the strategic variable for policymakers was the level of wages since wages constituted a large component of a society's purchasing power. Countervailing government intervention in the face of cutthroat international competition was often needed so as to protect domestic wages and workers' living standards. Moreover, consequent upon a shortage of purchasing power, underconsumption prevailed; there followed general deflation and human misery. With slow labour mobility and the immobility of large items of capital, the reduction of wages enforced by international competition would not always generate economic readjustment or "rationalization" in a manner which was by any means harmonious or organized. The notion that large, discontinuous shifts in the competitive positions of different industries may be destructive of purchasing power in the countries negatively affected, received some support from research by an  ? independent economist on the pages of the ILR.(5B T ԍ E.g., H. Feis: "International labour legislation in the light of  tT economic theory", in International Labour Review, Vol. 15, No. 4, 1927, p. 516. Notwithstanding this support, ILO investigations usually favoured the progressive elimination of international trade barriers and the reform of trade policies in the direction of freer trade so as to increasex(0*((@@  ? employment.A)5B TX ԍ E.g., [Anonymous]: "Unemployment in 1925", p. 224, and see ILO:  tT5 Unemployment: Some international aspects 192028, Studies and Reports Series C, No. 13, 1929, p. 40 for a critique of "economic nationalism" and "excessive  tT customs protection".A Martin's work at least illustrated that a freer trade policy was encumbered by doubts and not just over practical problems to do with the international synchronization of trade policy deregulation; for free trade did not guarantee higher levels of employment especially if it was introduced at onefellswoop. Martin appeared to have in mind a policy which retained the benefits of freer trade yet moderated its worst effects. The policy implications of underconsumption were clear. Buying power may, from timetotime, need to be reinforced. Easy monetary policies could be used, implying a boost to wages and perhaps profits, or (and this is radical for the 1920s) governments may use deficit finance ("a government prints paper money and uses  ? it to meet its liabilities").K*B Te ԍ Martin: "Overproduction", p. 45.K There is a crude sense in which Martin offered protoKeynesian solutions to underconsumption, although no great leap of intellect was necessary to reach the conclusion that economic slumps could be countered by allowing more purchasing power to be released into the economy. The main message of Martin's contribution was that governments need not wait for automatic market forces to restore high levels of employment, even when prices have been stabilized. The study of underconsumption was linked to ILO research on wages policies, since a "sound" wages policy was one which did not result in general reductions in either workers' purchasing power or labour conditions. The problem of underconsumption suggested that governments might need a macrostance on wages that went beyond concern for minimum wage standards for the weakest groups of workers. On the latter issue, the ILO view was that minimum wage standards were aspects of an "international social policy"; social justice is secured by adopting the value judgement that "human j*0*((@@ labour must not be treated as a commodity, namely, an article of commerce whose price on the open market is determined solely by the  ?X forces of supply and demand."+XB T ԍ K. Pribram: "The regulation of minimum wages as an international  tT problem", in International Labour Review, Vol. 17, No. 3, 1928, p. 331. Having accepted this principle, minimum wage standards may differ depending on customs and habits of consumption in different nations. However, much more than minimum wage standards must be considered in the wage setting process if governments were to be concerned with the level of purchasing power in the community. Extensive research on wages by Bellerby and J.H. Richardson tended to undermine the wages policy implications of Martin's doctrine of underconsumption. Firstly, Bellerby's threepart survey of the "evolution of a wage adjustment system", consistent with ILO principles, concluded that "it would seem fully adequate to vest the responsibility for evolving the principles [of wage adjustment] in the employers' and workers' organizations" of any  ? country.,5B TY ԍ J.R. Bellerby: "The evolution of a wage adjustment system: III", in  tT6 International Labour Review, Vol. 16, No. 3, 1927, p. 330.Ŀ Beyond some minimum wage legislation, there seemed to be no role for government in machinery designed to set wages; governments could play a useful part in dispute conciliation and arbitration, but not in establishing some macrolevel wage share to maintain purchasing power. Moreover, no attempt had been made through the ILO to secure international action dealing with an "appropriate" level of wages or of the wage share in national income. By contrast, there were numerous efforts to coordinate international monetary policies. Richardson was openly critical of the universal application of underconsumptionist arguments because it necessitated raising real wages (or at least maintaining real wages) by government decree, irrespective of the ability of employers to pay; such a policy may give impetus to employers to increase productive efficiency but it may just as well result in higher unemployment. Richardson!,0*((@@ reported that in the U.S.A. application of the doctrine of high purchasing power (or high wages) ensured harmonious employeremployee relations. And high wages were "an important element in  ? the demand for commodities."-mB T ԍ J.H. Richardson: "Some aspects of recent wage movements and tendencies  tT in various countries", in International Labour Review, Vol. 17, No. 1, 1928, p.  tT 200. He discovered, as well, that high wages "may...by their physical and psychological effects result in  ? greater efficiency of workers.".B Ti ԍ J.H. Richardson: "The doctrine of high wages", in International Labour  tTF Review, Vol. 20, No. 6, 1929, p. 814.ĭ Nonetheless, it was important to recognize that the U.S. economy was a relatively closed one; any inherent bias in the system toward underconsumption could be addressed by applying a doctrine of high wages as a route to economic progress and high employment without being concerned with external balance of payments effects, or the effects of international competition. In this case it was noteworthy that American manufacturers preferred higher wages rather than price reductions when distributing the benefits of productivity improvements. Governments had no role whatsoever in determining this preference; there was no macrowages policy requiring such an  ? outcome./jB T ԍ Richardson: "Some aspects of recent wage movements", p. 201 and "The  tT doctrine of high wages", pp. 812813.ĥ Per contra, in Europe we find much greater dependence on internationally tradeable goods (imports and exports), so raising or retaining high real wages come what may, would immediately downgrade a country's competitive position in international commerce. Therefore the European approach was understandably that "[e]conomic rapprochement and the reduction of trade barriers as foreshadowed by the World Economic Conference of 1927 will facilitate the permanent improvement of real wage  ?  standards."g0 B T& ԍ Richardson: "Some aspects of recent wage movements", p. 201.g Nothing could have been further from Martin's position; Richardson was far more discriminating and less sanguineL| 00*((@@ about the possibility of reducing unemployment simply by boosting workers' purchasing power. He was right to insist that underconsumptionist arguments depend for their validity on the  ? relative closure of economies from foreign competition.\1B T ԍ Richardson: "The doctrine of high wages", p. 815.\ The only exception to this rule might be an international slump in economic activity where widespread underconsumption prevailed, and, as we shall see, these were precisely the conditions in which underconsumptionist arguments came into vogue in the early 1930s.  ?  Challenges to Traditional Policy Responses to Economic Depression ILO economists came to understand the seriousness of the world economic crisis in 192930 probably more quickly than the majority of experts in the fields of economics and politics. As early as August 1930 Bellerby and Isles were pleading for a quite radical programme of international reflation, beginning with initiatives from the world's largest economies. In Great Britain, they maintained, for X"any remedy to be fully effective, [it] must accomplish two things. In the first place, it must raise the average level of prices into its due relation to wages and other costs. Secondly, it must do this primarily through the raising of `unsheltered' prices [for internationally tradeable commodities produced by unprotected industries]. In essence, this signifies that the ideal remedy would be one which would raise the world  ?x level of prices."c25xmB Tu" ԍ J.R. Bellerby and K.S. Isles: "Wages policy and the gold standard in  tTR# Great Britain", in International Labour Review, Vol. 22, No. 2, 1930, p. 153. (By the time this article was published Bellerby had departed the ILO for an academic  tT$ position at the University of Liverpool).c  In order to effect reflation, they recommended a conference of representatives from all central banks so as to formulate an j20*((@@ agreement to ease credit internationally. Perhaps this proposal was rather fanciful given the problems involved in generating an international consensus and implementing any agreement. Martin was happy to concur: "No single country can attempt such a programme by itself. If it did its easier credit conditions and its high prices relatively to other countries would be likely to have disastrous effects. In particular, the fact that goods could be obtained more cheaply elsewhere would cause its exports to decline and its imports to increase, and would...leave it exposed to a fluctuating  ? foreign exchange."3m B T ԍ P.W. Martin: "Finance and industry: the international significance of  tT the MacMillan report", in International Labour Review, Vol. 24, No. 4, 1931, pp.  tT 369370. Just as in the 1920s, ILO economists took a stance identical to J.M. Keynes who, from an early stage, also advocated international cooperation to achieve reflation: there should, he wrote, be cooperation between "leading central banks to raise prices to a parity with the international level of money  ?h costs of production.",4mhB T ԍ J.M. Keynes: Submission to the MacMillan Committee 192931, quoted in  tT M.J. Flanders: International monetary economics 18701960: Between classical and  tT new classical (Cambridge, Cambridge University Press, 1989), p. 116., According to Flanders, Keynes "cherished" this approach to recovery from economic depression "until the end  ? of his life".@52B T ԍ Flanders: ibid.@ In addition, it is striking that Bellerby, Isles, Martin and Keynes left no room for doubt on the possibility of successful collaboration between central banks across countries; perhaps their outlook on the extent, likely duration and profundity of the economic decline placed a very high premium on the expression of cynicism or scepticism. The immediate prospect for real wages following worldwide reflation were likely to be negative. Real wages, not money wages, would initially decline but this was an unintended consequence of action taken for other reasons, namely, to raise the prices of internationally tradeable goods relative to producers' costs andx50*((@@  ? nontradeables.6B TX ԍ Doubtless, research at the ILO illustrating the disastrous effects of falling world agricultural commodity prices on agricultural wages would have reinforced the ILO economists' favourable attitude to reflation. See L.E. Matthaei: "Some effects of the agricultural depression on agricultural labour",  tT in International Labour Review, Vol. 23, No. 4, 1931, pp. 455465.ı Certainly, the reflation proposal was underwritten by a presumption that money wages were rigid downwards in all major world economies (unless governments intervened to reduce them directly) a presumption that was also consistent with J.M. Keynes's views and those of the British MacMillan Committee on Finance and Industry which reported in 1931. Keynes's conviction that money wages tended to be rigid downwards became clear in his  ?4 testimony to the MacMillan Committee.a74B TQ ԍ See Flanders: ibid, pp. 89 note 1 and pp. 8796.a The prevailing aggregate level of real wages was the byproduct of inflexible wages and prices in sectors of economies sheltered from international trade and competition. Simply acting to reduce all money wages by government intervention in labour markets would "intensify the slump. A fall in [money] wages would imply at first a fall in consumption; and only after prices had been adjusted downwards, through the failure of markets, would there be a tendency for trade  ? to recover and expand."l82B TV ԍ Bellerby and Isles: "Wages policy and the gold standard", p. 152.l Raising prices relative to producers' cost of production should imply that rentiers and those on fixed interest incomes would also lose proportionately, but profits would expand, especially in internationally tradeable goods industries, and unemployment would fall. Reflation would initially reduce real wages although it would not be possible to predict the real wage reductions required for each industry or country. Research at the ILO, reported by Richardson in 1930, indicated that a theoretically perfect comparison of real wages between countries was not easy80*((@@  ? given different consumption patterns.J95B TX ԍ Nevertheless, Richardson produced some of the earliest estimates available of relative levels of real wages between countries: "International  tT comparisons of real wages", in Journal of the Royal Statistical Society, Vol. 93,  tT No. 3, 1930, pp. 398423.J By the early 1930s it is clear that just as for the 1920s, ILO economic research viewed the price level as a prime target of international monetary policy. The objective of restoring high levels of employment remained vitally important, but at one remove, following in the train as it were, of price level stabilization. Unpalatable alternatives included large money wage reductions and deflation, international beggarthyneighbour tariff restrictions and increased taxation leading to full scale, centrally planned government dictated recovery programmes. The ILO economists were committed to the ideal of beneficent international collaboration through monetary and trade policy reform as the means to widespread economic prosperity. In this they shared a perspective with their  ?< colleagues at the League of Nations.Y:<B T ԍ See de Marchi: "League of Nations economists".Y The universal appearance of mass unemployment by 1931 called for international action, not inwardlooking protectionist policies. In retrospect, the exceedingly early prescription provided by Bellerby, Isles and Martin for reflation on an international scale was unavailing; it was offered well before its time, although it is not obvious that Martin fully shared the BellerbyIsles view that real wages might need to fall to stimulate recovery in the early stages. The 1931 Report of the MacMillan Committee (hereafter: Report) was discussed at length in the ILR by ILO economists Martin and E.J. Riches. Martin found in the Report some support for his unrepentantly underconsumptionist position. The Report acknowledged that money wage reductions were no remedy for unemployment. According to Martin's interpretations, the Report proposed that purchasing power be maintained. The problem was insufficient demand not overproduction. The Report "aims at j:0*((@@ putting sufficient purchasing power into the hands of consumers to keep prices and production on an even keel." The Report was "an authoritative indication of the direction that informed opinion is taking in the light of presentday conditions: viz, that the way of world depression is not to reduce standards but increase activity.  ? It marked a revolution in economic thought."];B T4 ԍ Martin: "Finance and industry", pp. 366, 370, 373.] As we saw above, Richardson's work at the ILO had already anticipated some key recommendations of the MacMillan Committee Report, although his views did not have as much authority. As well, the main monetary policy proposals emphasizing active, discretionary monetary management by central banks had long been the centrepiece of ILO research since the early 1920s. All this should not be surprising. The ILO economists were closer to J.M. Keynes on many important aspects of economic theory and especially in their international economic policy recommendations in the 1920s, and Keynes was a member of the MacMillan Committee as well as a major witness before it. It would be interesting to know how much the ILO economists were directly influenced by Keynes, but the importance of their contributions whether developed independently, or otherwise, should not be dismissed as just a derivative of Keynes's work. Everything published by ILO economists from 19301933 raised a plenitude of questions about the sufficiency of a key recommendation of the MacMillan Committee to reform the international monetary system in the interests of achieving price stability. ILO research had always deprecated a singleminded concentration on this objective. Furthermore, it is difficult to find an enthusiastic defence of the gold standard in ILO work during this period. Reliance on monetary reform may have been a necessary condition for economic progress, but it may not ensure high levels of employment. The gold standard, as analyzed by the Gold Delegation of the Finance Committee of the League of Nations, would need modifications so that gold movements in response toT$m;0*((@@ changes in nations' net balance of external payments should not be prevented (by governments) from having an influence on domestic  ?X price levels and thence foreign exchange rates.<XB T ԍ League of Nations: Report of the gold delegation (Geneva, League of  tT Nations, 1932), p. 24.ě With easy limpidity, Martin demonstrated that the buildup of gold reserves by France and the U.S.A. indicated that their central banks were  ? not playing by the rules of the gold standard game.W=5B T ԍ Martin: "Finance and industry", pp. 361364.W To be sure, the buildup of gold in these countries was partly due to tariffs and investment policies which prevented the free flow of goods, services and capital internationally, but it was also due to inflexible monetary policies: X"An efflux or an influx of gold is not permitted to have its full effect upon the monetary system of the country concerned...In this way, contrary to the whole theory of the `automatic' gold standard, one country may lose gold without any marked effect upon its economy, while another may accumulate gold rapidly and at the same time have its  ? price level rapidly decline.">B T* ԍ P.W. Martin and E.J. Riches: "The social consequences of a return to  tT gold", in International Labour Review, Vol. 27, No. 1, 1933, p. 7.  There are parallels here with Keynes's message in A Treatise on Money: "the great evil of the moment" urged Keynes, resided in the "unwillingness of the central banks of the world to allow the  ?p market rate of interest to fall fast enough." ?mpGB TG ԍ J.M. Keynes: A treatise on money: Volume II (London, MacMillan, 1930)  tT$! in The Collected Writings of John Maynard Keynes, Vol. 6 (London, MacMillan,  tT! 1973), p. 185. See also p. 344.  The immediate solution was a world policy and response programme which overrode national constraints. Martin and Riches recognized the complications: short term, speculative capital movements and factors specific to the crisis circumstances in some countries war reparations and trade restrictions also weakened theL| ?0*((@@ possibility of an immediate international response strategy. And a response relying solely on monetary policies was, by 1933, now considered to lack potency. In a remarkable, prescient passage, Martin and Riches refer to what Keynes was also to recognize as a serious problem: X"In particular, experience would not appear to warrant the belief that these [monetary] instruments alone are capable of restoring the volume of buying once depression has set in. It is one thing to make credit cheap by lowering the discount rate, but it is quite another at such a time to find sound and willing borrowers ready to  ? take it up."o@ B T< ԍ Martin and Riches: "The social consequences", p. 16, emphasis added.o  Here we have a plain statement of the wellknown "Keynesian" case of investment being interestinsensitive; the propensity to save outweighs the propensity to borrow and invest regardless of the effects of monetary policy on the prevailing rate of interest. If such a situation obtained in practice, as Martin and Riches believed in 1933, then the level of production in any economy would  ? never be consistent with a high level of employment.nAUmB T ԍ Martin and Riches were generalizing from the experience of Britain and the U.S.A. in particular. It is not a foregone conclusion that the level of interest rates compatible with higher private investment, production and employment would have to be the same in different economies. This point is eloquently argued by leading Swedish economist G. Cassel: "World economic reconstruction: A criticism of the economic resolution adopted by the  tT International Labour Conference", in International Labour Review, Vol. 26, No.  tT 5, 1932, pp. 646647.n Monetary policy on its own would not suffice. The earliest hints of this argument may be located in Martin's comments on the MacMillan Committee's Report two years earlier: since the operation of monetary policy on the demand side of economies is "slow, it would be necessary to forsee the depression some months ahead; and in the event of action being delayed or not taken with sufficient energy,  A0*((@@  ? it is questionable whether such measures would prove sufficient."sB5B TX ԍ P.W. Martin: "Finance and industry: The MacMillan report as a basis for  tT5 international action", in International Labour Review, Vol. 24, No. 6, 1931, p.  tT 692, and also Martin: "World economic reconstruction", in International Labour  tT Review, Vol. 26, No. 2, 1932, p. 205.s In a period of great intellectual ferment among economists over the means of extricating the world economy from the economic crisis, ILO economic research setup a challenge to existing, more orthodox proposals. It was a challenge with positive, concrete suggestions for international economic reconstruction to at least directly ameliorate the social consequences of economic depression. We turn next to consider this positive agenda.  ?  Toward Economic Reconstruction in the 1930s From the first session of the ILO Conference in 1919, it was deemed desirable that public works should be planned so as to counteract economic fluctuations. Public authorities should reserve works wherever possible for periods of rising unemployment. The 1931 survey of the theory and practice of public works distinguished three types of unemployment: (i) unemployment of a "more or less permanent" character, (ii) cyclical unemployment, and (iii) seasonal unemployment. Category (ii) was identified as being relevant to public works activity, provided work could be varied upwards in cyclical downturns and vice versa. However, these three categories of unemployment became less relevant to circumstances in 193133. The idea of "advance planning of public works...to further the regularisation of the demand for labour over a period of years" may have been appropriate for economic cycles of a normally experienced amplitude; public works could then act as a "temporary palliative" of unemployment. By contrast the economic crisis of the 1930s presented a special case of mass unemployment which was severe and prolonged. In these conditions, "[s]ooner or later a point is reached at which there are no more public works on!B0*((@@  ? the programmes of the authorities concerned."CB TX ԍ ILO: Unemployment and public works, Studies and Reports Series C, No.  tT5 15 (Geneva, 1931), pp. 4, 5, 25, 26, and also see p. 32.Ŀ Some important points in favour of public works were made in the 1931 ILO investigation. First, employment was preferable to relief work of any kind. Secondly, public works have positive downstream effects on the demand for goods, services and labour in other private industries, but these were difficult to estimate quantitatively. Thirdly, public works expenditure raises the purchasing power of the community as a whole. The survey is defensive about a warning in orthodox economic literature (attributed to the influence of British economists R.G. Hawtrey, T.E. Gregory and evident in the British Treasury view) that attempting to increase employment through public works was futile because it merely diverts labour and capital from works which would otherwise have been undertaken by private industry. Financing public works presented difficulties as well. In a prolonged depression, government surpluses usually eroded quickly into deficits and loan finance may not easily be sourced. In short, public works could not necessarily be brought forward in these  ? conditions.BD5B T ԍ Ibid: pp. 24, 30.B An essential feature of policy recommendations on public works in the 1920s as outlined in the 1931 ILO survey, was the planning and coordination of work at the national level on a relatively small scale. As generalizations these recommendations were unexceptionable; public works would serve as a response to temporary economic downturns. The sixteenth session of the ILO Conference in April 1932 adopted a resolution marking a paradigmatic shift in thinking on policy responses to the Great Depression: X"That Government delegates...should as soon as possible be given instructions to draw up a list of big!D0*((@@ international works for economic equipment and for national works on a large scale calculated to encourage the general development of the economic situation of the countries concerned, to make the necessary financial arrangements therefor and to have them put in hand  ? without delay." (emphasis added)EB T4 ԍ The full text of the resolutions adopted by the Sixteenth Session of the  tT ILO Conference is printed in International Labour Review, Vol. 26, No. 2, 1932, pp. 220221. The Conference passed the resolution on public works 80 votes to  tT 1; see A. Alcock: History of the International Labour Organisation (London,  tTi MacMillan, 1971), p. 104.Ĵ  The necessity for "planning on a world basis" had been presaged in a report on the 1931 World Social Economic Congress by an ILO  ?` official.F` B T} ԍ G.A. Johnston: "Social economic planning", in International Labour  tTZ Review, Vol. 25, No. 1, 1932, p. 58.Ĩ In the same year the Deputy Director of the ILO considered the time ripe for "international solutions to the  ? world's difficulties".G B TB ԍ H.B. Butler: "The social effects of economic depression in North  tT America", in International Labour Review, Vol. 23, No. 3, 1931, p. 323. So by 1932 the ILO advocated international planning of public works in a manner which moved decisively beyond ILO studies and Conference resolutions in the 1920s resolutions focussing on encouraging smaller, nationallybased countercyclical public works schemes. The analytical groundwork for the historic 1932 resolution on an internationally based, large scale public works policy, was completed by ILO economists Martin and Riches. Other ILO investigatory work concentrated on the direct social consequences of the economic crisis, for instance, on employment problems in particular countries or for particular social groups, and on  ? methods of ameliorating these consequences at the local level.Hg B T$ ԍ For example, H. Fuss and D.C. Tait: "Unemployment benefits and measures  tTp% for occupying the unemployed in Great Britain", in International Labour Review, Vol. 27, No. 5, 1933, pp. 595619; M. Thibert: "The economic depression and the  tT' employment of women: I and II", in International Labour Review, Vol. 27, No. 4, 1933, pp. 443470 and No. 5, pp. 620630; [Anonymous]: "The problem of the'G0*((  tT unemployment of young persons", in International Labour Review, Vol. 27, No. 4, 1933, pp. 499523; D. C. Tait: "Unemployment among young people in Great  tT  Britain", in International Labour Review, Vol. 31, No. 2, 1935, pp. 166189; and  tT H. Fuss: "Unemployment and employment among women", in International Labour  tT Review, Vol. 31, No. 4, 1935, pp. 463497.Į xH0*((@@ Martin and Riches were acutely aware that any international policy response to mass unemployment must be made "as in every other sphere of public policy", in the light of "the probable effect on social conditions, on the life and work of the great mass of the people, which in turn must act as the final standard of sound  ? policy."sIxB T ԍ Martin and Riches: "The social consequences of a return to gold", p. 21.s Theirs was to be a macroeconomic perspective linking international monetary policy coordination with large scale public works programmes. They envisaged a sequence of responses beginning in the larger industrialized economies which were creditor (surplus) nations in international payments terms. But was there a logically necessary connection between monetary policy reform and public works programmes? And could the ILO economists demonstrate that their proposed policy responses were implementable? Martin and Riches elaborated on their positive agenda for economic reconstruction first by presupposing that price stability achieved through an apposite monetary policy, on its own would not suffice to maintain a high level of employment. As we found in the previous section, they insisted that expansionary monetary policy may not be able to stimulate private investment demand so vital to rejuvenating employment. In the first place, expansionary monetary policies must be aimed at financing public works "directly or indirectly [via downstream activities]...by what is to all intents  ? and purposes new money."[JB Tu" ԍ Martin: "World economic reconstruction", p. 202.[ Secondly, since capital and raw materials supplies were available, and cheap capital was not readily invested by private industry, then the only feasible alternative was for large scale government directed public works. Public works would mobilize otherwise idle bank balances. In thexJ0*((@@ 1931 ILO study of public works, it was recognized that expansionary monetary policies at that time would release "a superabundance of money seeking investment" as the rate of interest declined. That higher private investment expenditure would not result from lower interest rates in some economies was not stressed in this 1931 study at all. Nevertheless, evidence reported in the previous section, especially from Martin's research, indicated low confidence amongst private investors and thus a reduced likelihood of cheaper credit being quickly or fully utilized. In the event, there was no justification in practice for public works expenditure acting as a deterrent to private investment expenditure; employment would not simply be diverted from socalled more productive private employment to public sector employment. If, indeed, the private demand for credit began to increase as interest rates declined, this would not be problematic provided an expansionary credit policy was maintained. For as Keynes explained in 1929: X"In so far as the greater volume of employment and business means that wage earners carry in their pockets increased purchasing power in bank notes and that business men keep at their banks an increased purchasing power in bank deposits, there can be an increased volume of credit of an entirely innocent and noninflationary  ? description."KB T  ԍ J.M. Keynes: "Letter to the Times" (7th August 1929), quoted in extenso  tT in ILO: Studies and Reports Series C, No. 15 (Geneva, 1931), p. 27.  Martin was definitely thinking along these lines when addressing the 1932 ILO resolution on public works. In his estimation, the principal advantage of public works expenditure was that it leads to "an increase both in the quantity, and in the velocity, of currency and credit used to buy goods; so that, with this method, a relatively small volume of public works tends to result in a  ?! relatively large increase in the volume of buying."[L!5B T& ԍ Martin: "World economic reconstruction", p. 202.[! L0*((@@ԌProvided central banks in surplus countries pursued expansionary monetary policies and provided public authorities carefully planned public works schemes, then public works activity in one country would have demonstration effects for others and lead to a revival of employment internationally. Moreover, the revival would be enduring if public works ceased to be regarded as temporary measures; ongoing public works schemes may be required for a long time to maintain a level of purchasing power consistent with high employment. Martin promised that the ILO would assist authorities in ILO member countries with planning, given that the Office had collected much detailed information on suitable  ? schemes.KM B T< ԍ Martin: ibid, pp. 202203.K In 1931 it seemed reasonable to conclude that "public  ? works would not be put in hand for long [periods] of time".rNmB T  ԍ ILO: Studies and Reports Series C, No. 15, 1931, p. 32, emphasis added.r Now, in 1933, inspired by the ILO Conference resolution, the position of ILO researchers was that carefully planned public works not only helped revive economic activity, they also had a long term place in economic development by maintaining investment and purchasing power at high levels. On the question of financing public works there were several answers offered in ILO investigations. Underwriting these answers is a strong dirigiste tendency: investment expenditure in disparate public infrastructural and social projects needed active government planning and direction because private enterprise was too slow to act or less reliable. Financing, therefore, should not pose a problem because the strength and extent of government planning offered greater security. Of first importance was the activation of idle cash balances existing levels of credit could be fully employed by public authorities on major investment projects. Surely a reduction in interest rates to a sufficiently low level would have stimulated private investment without the need for!!N0*((@@  ? public works?RO5B TX ԍ This question was frequently put by orthodox economists. On orthodox doctrine in the interwar period see G.C. Peden: "The `Treasury view' on public  tT works and employment in the interwar period", in Economic History Review, Vol.  tT 37, No. 2, 1984, pp. 167181.R Perhaps so, but the egregious time lags involved, so frequently referred to in the work of Martin and Riches, were unacceptable to them. It is curious that Martin and Riches minimized the administrative and institutional obstacles in the way of planning and activating public works, for the time lags here may  ? also have been unacceptably long.PB T1 ԍ Institutional time lags in public works planning were highlighted in recent research by M. Weir and T. Skocpol: "State structures and the possibilities for `Keynesian' responses to the Great Depression in Sweden,  tT Britain and the United States", in P.B. Evans et.al. (eds) Bringing the state  tTf back in (Cambridge, Cambridge University Press, 1985), pp. 107163.Ľ By 1933 ILO economists had clearly become impatient about private sector responsiveness to expansionary monetary policy. In some countries loan financed public expenditure would be needed to activate public works. Deficit finance should not be disparaged because public works could be selffinancing in time. First, the cost of social transfer payments and relief works would decline. Secondly, increased tax revenue could be expected as national  ?< income and employment expanded pari passu.dQ< B TV ԍ Martin and Riches: "The social consequences", pp. 17, 18.d Budgetary orthodoxy in the 1920s and early 1930s emphasized balanced government budgets; direct government employment creation was only possible if surpluses were available. For example, a representative of this orthodox view criticized Martin's research for suggesting deficit financing for public works: it was "essential" according to orthodox doctrine, "that government enterprise for stimulating recovery should be entirely separate from the ordinary budget. The budget must be balanced, for without a balanced budget there will  ? be no confidence."kR/ B T& ԍ Cassel: "World economic reconstruction", p. 646, emphasis added.k It was from this line of thought that the ILO economists were trying to escape. They were on common ground with" R0*((@@ J.M. Keynes, who had not merely made a leap of faith in advocating deficit finance, for there were sound logical reasons for expansionary public works. As early as 1931, the ILO survey on public works recognized the "multiplier" effects on all incomes generated by public works, which were sufficient to increase tax revenue. At that time there was concern that public works could generate a significant increase in foreign imports of raw  ?4 materials.fS4B T ԍ ILO: Studies and Reports Series C, No. 15, 1931, pp. 2528.f In 1933 this concern was allayed given the objective of expanding public works internationally; in the event, all nations should benefit from an increase in international trade stimulated thereby. ILO economic research in the 1930s understood that public works would need promotion on a vast scale in order to reduce mass unemployment; they understood, pace orthodox economists, that public works would need to become a normal part of government expenditure. In 1936 an economic advisor to the ILO, L.L. Lorwin, reflected on the achievements of the ILO in the 1930s and saw public works as one of the "large measures" advocated for economic recovery, equal in priority with monetary policy reform and the  ?D expansion of international trade.TDmB TA ԍ L.L. Lorwin: "The ILO and world economic policy", in International  tT Labour Review, Vol. 33, No. 4, 1936, pp. 462463.ĵ Public works now meant any government activity designed for industrial development including housing and health care. The ILO survey on "Public Works Policy" in 1935 showed that it did not need the formal conceptual framework of Keynes's General Theory of Employment, Interest, and Money (1936) for governments to vigorously promote public works. A reader of the ILO survey would nonetheless observe a disparate set of uncoordinated works schemes, rather than an internationally  ? integrated approach as advocated by ILO economists.UB T& ԍ ILO: Public works policy, Studies and Reports Series C, No. 19 (Geneva,  tT& 1935).ď When an#GU0*((@@ unsigned article entitled "Public Works as a Factor in Economic Stabilization" appeared in the ILR, it reported favourably on a new ILO initiative; the creation, in 1937, of an International Public Works Committee which would collect data, analyse and share information between member countries on public works worldwide. This article was based on Resolutions adopted by the 1937 ILO Conference; it took for granted what a few years earlier would have been heretical, viz. the "objective of national planning of public works is to maintain purchasing power"; that the issue of public works "embraces public expenditure in general"; that public expenditure should promote developments in the capital structure (human and physical) and be accompanied, where necessary, by an accommodating monetary policy and, initially, by unbalanced  ?< budgets.aV5<B T ԍ [Anonymous]: "Public works as a factor in economic stabilization", in  tTq International Labour Review, Vol. 38, No. 6, 1938, pp. 731, 732, 734. The full text of the ILO Conference resolutions in respect of the International Public  tT Works Committee is reprinted at pp. 754759.a After surveying public works schemes in a wide range of ILO member countries, the 1938 article observed that the 1937 ILO Conference favoured international agreements enabling public works in different countries to be developed on "parallel lines, so that an expansion of credit and an increase in the volume of public works may be as far as possible simultaneous in all countries". Furthermore, the International Public Works Committee would have the far reaching responsibility of promoting international public works, that is to say, of planning and then assisting in implementing and assessing schemes where works in some debtor  ?  countries would be financed by stronger creditor countries.DW B Tu" ԍ Ibid: pp. 744, 751.D Here was a responsibility portentous of the link, to be so welldeveloped in the 1950s, between foreign investment and foreign aid flows and public works in recipient (debtor) countries. Later, this was to become a principal responsibility of the World Bank. $jW0*((@@Ԍ ? ԙ Summary and Conclusions From the early 1920s until the years of economic recovery in the 1930s, economic research at the ILO was always innovatory, always challenging the presuppositions of traditional doctrine on matters of monetary policy, the economics of price stability and employment stabilisation, wages as a component of aggregate demand or purchasing power, responses to the economic crisis in the 1930s and the role of public works and public expenditure. The ILO researchers constantly kept in view the significance of these issues in an international economy context. An early intention to relate the traditional economic goal of price stability to the labour market and to the level of aggregate employment in particular, generated policy prescriptions favouring active government intervention in, and guidance of, monetary affairs in the major world economies. To inform monetary policies and assist international collaboration between central banks, ILO researchers argued for the construction of reliable price, output and employment indices. The crucial role of demand in the international economy was highlighted in ILO research in the late 1920s when Martin first raised the problem of underconsumption. Bellerby and Richardson did not fully share Martin's perspective, but they carefully considered the level of wages as a major determinant of purchasing power while at the same time connecting wages to international competitiveness. By the late 1920s the ILO researchers were, without exception, disseminating the idea that governments had to be mindful of the role of demand because the restoration of employment to high levels could not be guaranteed by a singleminded concentration on the price stability objective. In the incipient stages of the World Depression ILO researchers were quick to offer reflationary solutions well before they were regarded as fashionable and implementable. Perhaps their closeness to data on unemployment and industrial production which was flowing into the ILO from many different countries, made them'%W0*((@@ more aware of the gravity of the situation and therefore more open to new ideas? Sound evidence was presented by Martin and Riches against reestablishing a gold standard regime in the 1930s; the deflationary impact of the gold standard for some countries was regarded as too severe and the unemployment consequences too damaging to make this a socially viable option; certainly it was not a policy likely to generate sufficient international cooperation to effect economic recovery. There was a special value attached to international policy coordination in respect of money, trade and public works in ILO research although successful coordination was always going to be difficult given the variety of policymaking institutions in different countries. The ideal of workable international collaboration reached a high point in programmatic ILO studies of public works. Public works were increasingly viewed as part of general government outlays, sometimes deliberate deficitfinanced outlays made for longterm social and industrial development rather than for temporary unemployment relief. What emerges from our account is a significant intellectual linkage previously unrecognised: there were striking similarities between J.M. Keynes's ideas as they evolved from the early 1920s to the late 1930s, and economic thought and policy proposals emanating from ILO researchers. This is not to assert that ILO economists were slavish followers of Keynes (the question of influence in either direction can only be speculated upon), only that their views were evocative of Keynes's. We therefore conclude that the ILO economists like Keynes, were in the vanguard of a stream of economic thought which regularly questioned orthodox economic theory and policy in the interwar period and which, by the 1950s (if not the late 1930s) had been fully vindicated.