Hostname: page-component-7c8c6479df-ws8qp Total loading time: 0 Render date: 2024-03-29T08:59:17.565Z Has data issue: false hasContentIssue false

Foundations, Statistics, and State-Building: Leonard P. Ayres, the Russell Sage Foundation, and U.S. Government Statistics in the First World War

Published online by Cambridge University Press:  13 December 2011

William J. Breen
Affiliation:
William J. Breen is a reader in history at La Trobe University, Australia

Abstract

This article examines the statistical work done by Leonard Porter Ayres for the Council of National Defense during the First World War. Director of statistics for the Russell Sage Foundation when war was declared, Ayres immediately volunteered his own and the foundation's statistical expertise. The article focuses on the first year of American intervention in the war and argues that Ayres's important statistical work evolved in three overlapping but distinct stages. The structure of the American state, however, confounded the wider ambitions of Ayres (just as it had those of his rival Edwin Gay) to centralize all government statistical data.

Type
Articles
Copyright
Copyright © The President and Fellows of Harvard College 1994

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Leonard P. Ayres, “Memorandum on Trip to Washington, Tuesday, March 20, 1917 (March 24, 1917),” subject files: World War I, 1917–25, box 3, folder “Memoranda, Mar. 1917–Dec. 1919,” Leonard Porter Ayres Papers, Library of Congress. The Division of Statistics at the Russell Sage Foundation consisted of three people and the Division of Education of seven people. Ayres directed both divisions. The work of both groups consisted of statistical investigations and reports, and “the combined staffs of the two divisions really constituted one well-trained statistical force made up of four men and six women.” Nine of these ten individuals transferred to Washington for war service with the Council of National Defense. See unheaded, ten-page memo dated 1 Dec. 1919, in folder. The quotation is on p. 1. See also Glenn, John M., Brandt, Lilian, and Andrews, F. Emerson, Russell Sage Foundation, 1907–1946, 2 vols. (New York, 1947), 1: 245–47Google Scholar.

2 Leonard P. Ayres, “First Annual Report, Division of Statistics, Council of National Defense. Now Statistics Branch, Executive Division, General Staff, War Department (April 6, 1918”) [46 pp.], 1–2, quotation on p. 2. Copy found in Record Group (RG) 62, Council of National Defense (CND) Records, 17–B1, Files and Records Division (Post-War), Administrative Files, Statistics Division, box 1087 (unmarked folder), Federal Records Center, Suitland, Md. This unpublished report is largely, but not completely, reproduced in the section on the Division of Statistics contained in the Second Annual Report of the Council of National Defense for the fiscal year ended June 30, 1918 (Washington, D.C., 1918), 200211Google Scholar; references are made to the unpublished report only when unavoidable. In fact, the Statistics Division was not the only branch of the Sage Foundation affected by the war. “Every department was affected in some degree, ranging from complete suspension of normal activities for thirty months to modifications of programs and curtailment or interruption of work under way because of temporary absences.” The foundation continued to pay full salaries until individuals were transferred to the federal payroll and then supplemented new salaries if they were lower than those paid by the foundation. Mary Van Kleeck, for example, the head of the foundation's Division of Industrial Studies, served initially in the Ordnance Department and then was appointed director of the newly created Women in Industry Service bureau of the Department of Labor. See Glenn, , Brandt, , and Andrews, , Russell Sage Foundation, 1: 245, 256–58Google Scholar.

3 Critchlow, Donald T., “Think Tanks, Antistatism, and Democracy: The Nonpartisan Ideal and Policy Research in the United States, 1913–1987,” in The State and Social Investigation in Britain and the United States, ed. Lacey, Michael J. and Furner, Mary O. (New York, 1993), 284–85Google Scholar.

4 Eakins, David W., “The Origins of Corporate Liberal Policy Research, 1916–1922: The Political-Economic Expert and the Decline of Public Debate,” in Building the Organizational Society: Essays on Assodational Activities in Modern America, ed. Israel, Jerry (New York, 1972), 179Google Scholar.

5 Lagemann, Ellen Condliffe, The Politics of Knowledge: The Carnegie Corporation, Philanthropy, and Public Policy (Middletown, Conn., 1989)Google Scholar; Critchlow, Donald T., The Brookings Institution, 1916–1952: Expertise and the Public Interest in a Democratic Society (DeKalb, III., 1985)Google Scholar, chap. 1.

6 For a brief statement of the theoretical debate, see Critchlow, The Brookings Institution, especially 6–8. See also Dubofsky, Melvyn, The State and Labor in Modern America (Chapel Hill, N.C., 1994)Google Scholar, introduction and conclusion; Block, Fred, “The Ruling Class Does Not Rule: Notes on the Marxist Theory of the State,” Socialist Revolution 33 (May-June 1977): 627Google Scholar, reprinted in Block, , Revising State Theory: Essays in Politics and Postindustrialism (Philadelphia, Pa., 1987), 5168Google Scholar; Skocpol, Theda, “Political Response to Capitalist Crisis: Neo-Marxist Theories of the State and the Case of the New Deal,” Politics and Society 10 (1980): 155201CrossRefGoogle Scholar.

7 Jordan, John M., Machine-Age Ideology: Social Engineering and American Liberalism, 1911–1939 (Chapel Hill, N.C., 1994), 93Google Scholar; see especially chap. 4, which focuses specifically on the First World War.

8 On the early work of the foundation, see Glenn, , Brandt, , and Andrews, , Russell Sage Foundation, vol. 1Google Scholar. See also the recent work of Guy Alchon on the role of Mary Van Kleeck, who headed the Women's Work—Industrial Studies Division: Alchon, Guy, “Mary Van Kleeck and Social-Economic Planning,” Journal of Policy History 3, no. 1 (1991): 123CrossRefGoogle Scholar; and Alchon, , “Mary Van Kleeck and Scientific Management,” in A Mental Revolution: Scientific Management since Taylor, ed. Nelson, Daniel (Columbus, Ohio, 1992), chap. 5Google Scholar.

9 Bremner, Robert H., From the Depths: The Discovery of Poverty in the United States (New York, 1956), 155–56Google Scholar.

10 On the early work of the Statistics Division, see Glenn, , Brandt, , and Andrews, , Russell Sage Foundation, 1: 171–76.Google Scholar

11 See Who's Who in America (various editions).

12 On this issue, see especially McClymer, John F., War and Warfare: Social Engineering in America, 1890–1925 (Westport, Conn., 1980), chap. 6Google Scholar; see 158 for specific reference to Ayres.

13 Clarkson, Grosvenor B., Industrial America in the World War: The Strategy behind the Line, 1917–1918 (Boston, Mass., 1923), 205Google Scholar.

14 Heaton, Herbert, A Scholar in Action: Edwin F. Gay (Cambridge, Mass., 1952), 126CrossRefGoogle Scholar; DeWeerd, Harvey A., President Wilson Fights His War: World War I and the American Intervention (New York, 1968), 174Google Scholar.

15 Alchon, Guy, The Invisible Hand of Planning: Capitalism, Social Science, and the State in the 1920s (Princeton, N.J., 1985), 27Google Scholar.

16 Cuff, Robert D., “Creating Control Systems: Edwin F. Gay and the Central Bureau of Planning and Statistics, 1917–1919,” Business History Review 63 (Autumn 1989): 588613CrossRefGoogle Scholar; quotations on 613 and 611. Drawing on Beniger, James R., The Control Revolution: Technological and Economic Origins of the Information Society (Cambridge, Mass., 1986)Google Scholar, Cuff locates Gay's efforts to develop adequate wartime statistics in the context of an underlying “control revolution” that “had already begun to transform key areas of American life before the war”; see 589–90.

17 Cuff, “Creating Control Systems,” quotations on 590 and 603.

18 See, for example, Alchon, The Invisible Hand of Planning, 27. Alchon, while emphasizing 1918 and the role of Dean Gay, does recognize the work of Leonard P. Ayres in 1917 as “one possible exception” to his bleak picture of haphazard and inadequate statistical machinery available to policymakers in the first year of the wartime mobilization. However, he does not follow up on this suggestion and brushes Ayres's contribution aside as “an anomaly.”

19 DeWeerd, President Wilson Fights His War, 174.

20 Ayres is well known for his statistical work with the General Staff in 1918 and for the publication, at the end of the war, of his The War with Germany: A Statistical Summary (Washington, D.C., 1919)Google Scholar. However, his role in 1917, before he joined the General Staff, has been almost completely ignored.

21 For information on Gifford, see the three-page biographical outline headed “Mr. Walter S. Gifford” in RG 62, CND Records, 3-A5, Secretary's Office, Correspondence, box 171. See also Gifford, Walter S., “The Function of Statistics in the Telephone Business,” in Business Statistics, ed. Copeland, Melvin T. (Cambridge, Mass., 1917), 684—96Google Scholar.

22 L. P. Ayres to Mrs. Milan C. Ayres, 17 April 1917, Ayres Papers, box 2, file “Family Correspondence, 1902–1930.” Walter S. Gifford, the 31-year-old director of the Council of National Defense, was well aware of the potential value of statistics. A Harvard graduate in the class of 1905, he had joined the American Telephone & Telegraph Company in 1908 and became its chief statistician. He had given guest lectures on business statistics at the Graduate School of Business Administration at Harvard University and had read papers before the American Statistical Association and the American Economic Association. A fellow of the American Statistical Association and chairman of its Standing Committee on Business Statistics, he had also been active in the “preparedness” campaign and had served as Supervising Director of the Committee on Industrial Preparedness of the Naval Consulting Board. See “Mr. Walter S. Gifford,” RG 62, CND Records, 3-A5.

23 Ayres to Mrs. Milan C. Ayres, 17 April 1917.

24 In drawing up the organization chart for the Council of National Defense, Ayres cooperated with Henry P. Seidemann of the Institute for Government Research. They tried, unsuccessfully, to promote a reorganization of the council's administrative structure, because they believed the current form was very unwieldy and inefficient. See L. P. Ayres and H. P. Seidemann to W. S. Gifford, 4 May 1917 (with accompanying organization charts), RG 62, CND Records, 17-A2, Files and Records Division (Post-War), Correspondence, box 1045, and the photographs in this article.

25 Ayres, “First Annual Report, Division of Statistics,” 3, 30—31.

26 Ayres to Mrs. Ayres, 17 April 1917.

27 Division of Statistics, “Weekly Report (June 9, 1917),” 3–4. See also “Weekly Report (May 12, 1917),” 1–2. The installation of vertical filing systems in offices was a relatively new development in the prewar era and involved quite complicated decisions about different methods of organization. See Yates, JoAnne, Control through Communication: The Rise of System in American Management (Baltimore, Md., 1989), especially 5663Google Scholar.

28 Manuals were a quite recent office innovation; see Yates, Control through Communication, 71–72.

29 First Annual Report of the Council of National Defense, for the fiscal year ended June 30, 1917 (Washington, D.C., 1917), 5556Google Scholar. In addition, the group established a system of security passes to control access to the CND building for the approximately five hundred individuals involved and drew up a detailed list of the people, including the various volunteers, who were actually working as part of the council organization.

30 See Coffman, Edward M., The War to End All Wars: The American Military Experience in World War I (New York, 1968), 2728Google Scholar.

31 Second Annual Report of the Council of National Defense, 200.

32 A useful brief overview of the organization and supply problems facing the War Department in 1917 is contained in DeWeerd, President Wilson Fights His War, 220–25. See also the excellent essay detailing the problems in the artillery program of the Ordnance Bureau by Beaver, Daniel R., “The Problem of American Military Supply, 1890–1920,” in War, Business, and American Society: Historical Perspectives on the Military-Industrial Complex, ed. Cooling, Benjamin F. (Port Washington, N.Y., 1977), 7392Google Scholar.

33 Division of Statistics, “Weekly Report (July 7, 1917),” 1, and also “Weekly Reports” for 23 June 1917, 3 and 30 June 1917, 2. Graphs were not commonly used in business until around the turn of the century, and it was not until 1914 that graphs “came of age as a managerial tool.” Ayres's use of graphic charts was in line with the latest business practice. During 1917, Henry Laurence Gantt developed the famous “Gantt chart” to monitor progress toward scheduled war production goals. See Yates, Control through Communication, 85–91, quotation on 85.

34 Division of Statistics, “Weekly Report (July 14, 1917),” 4.

35 Ibid. (28 July 1917), 1.

36 Ibid. (4 Aug. 1917), 1.

37 Ayres, “A Report of the Division of Statistics, April 9–July 1, 1917 (November 16, 1917),” [6 pp.], 6, RG 62, CND Records, 17-B1, Files and Records Division (Post-War), Administrative File, box 1085, file “Reports-Stat. Div.” The use of graphs for forecasting was very innovative: apparently graphs for that purpose only “came into common use between 1914 and 1919.” See Yates, Control through Communication, 292n85.

38 Division of Statistics, “Weekly Report (August 25, 1917),” 3. This became a regular weekly report to the secretary of war consisting of “a small folder containing a series of charts and diagrams relating to different topics.” For a detailed list, see Division of Statistics, “Weekly Report (September 1, 1917),” 2.

39 Ayres, “First Annual Report, Division of Statistics,” 6.

40 Division of Statistics, “Weekly Report (October 13, 1917),” 1–2.

41 Second Annual Report of the Council of National Defense, 204.

42 Division of Statistics, “Weekly Report (October 20, 1917),” 1. Some idea of the magnitude of the task is suggested in the following comment: “Between April 6, 1917, and June 1, 1919, the War Department alone had entered into approximately 300,000 contracts, disbursements on these exceeding $14,500,000,000.” See Noggle, Burl, Into the Twenties: The United States from Armistice to Normalcy (Urbana, III., 1974), 63Google Scholar.

43 Ayres, “First Annual Report, Division of Statistics,” 10–11.

44 Second Annual Report of the Council of National Defense, 203. A memorandum from the chief of staff to the adjutant general dated 27 Sept. 1917 amplified Secretary Baker's instructions. The bulk of that memorandum is reproduced in Ayres, “First Annual Report, Division of Statistics,” 12–14.

45 On 17 August, the War Industries Board passed a resolution requesting the Director of the Council of National Defense to enlarge the statistical work of the Division of Statistics “so as to provide for the assembling of information valuable to the Priorities Committee, the Purchasing Commission, and other branches of the Council.” See Second Annual Report of the Council of National Defense, 201. Also on 17 August, Ayres was formally appointed statistician of the War Industries Board and, on 8 September, of the Priorities Committee. See Ayres, “First Annual Report, Division of Statistics,” 14.

46 There were “twenty or more statistical agencies” in the federal government when America entered the war. In addition, the various war mobilization bodies, such as the Council of National Defense and the War Industries Board, also established their own statistical divisions. See Mitchell, Wesley Clair, The Backward Art of Spending Money and Other Essays (New York, 1950), 45Google Scholar.

47 Division of Statistics, “Weekly Report (April 21, 1917),” 1, and “Weekly Report (September 1, 1917),” 1. For a joint recommendation from the Division of Statistics and the Institute for Government Research concerning the organization of the Council of National Defense, see Ayres and Seidemann to W. S. Gifford, 4 May 1917 (with two accompanying organization charts).

48 Division of Statistics, “Weekly Report (October 6, 1917),” 2. In his report for 8 September 1917, Ayres had referred to the establishment of a working arrangement with the Geological Survey whereby it “virtually becomes the Geological Section of the Division of Statistics.”

49 See, for example, Division of Statistics, “Weekly Report (October 6, 1917),” 1–2. On the appointment of a liaison officer from the General Staff to the Division, see Division of Statistics, “Weekly Report (November 3, 1917),” 1. The initial cooperative arrangement with the General Staff in late September had merely consisted of a reciprocal agreement “by which the statistical division is to furnish copies of summaries which it compiles relating to raw materials and is to receive from the General Staff such information as may be needed from their files.” See Division of Statistics, “Weekly Report (September 29, 1917).” Within a month the General Staff had decided to move beyond that arrangement and to appoint a liaison officer. See also Ayres, “First Annual Report, Division of Statistics,” 14.

50 Division of Statistics, 'Weekly Report (December 22, 1917),” 1.

51 Ibid. (December 8, 1917), 4.

52 The Geological Survey assigned some twenty-five specialists to assist the Division of Statistics in the work of preparing these weekly reports. The cooperation of the Survey was essential. For some years, the Survey had received regular, highly confidential reports from mining companies all over the country on production. Even the Justice Department was denied access to these reports. The Division of Statistics, with the cooperation of the Geological Survey, contacted the companies involved and secured their permission to use essential production data reported to the Geological Survey for warrelated purposes. The Office of War Minerals Statistics later developed into the Joint Information Board on Minerals and Derivatives with representatives from a dozen departments and war agencies. A similar cooperative committee, the Joint Office on Chemicals Statistics, was also formed in December 1917 to gather information on the production of chemicals needed for explosives. See Second Annual Report of the Council of National Defense, 206–7.

53 Glenn, , Brandt, , and Andrews, , Russell Sage Foundation 1: 246Google Scholar.

54 For the distribution list, see Ayres, “First Annual Report, Division of Statistics,” 20; see also 18–19. Fourteen numbered copies of the weekly report were being produced in March 1918, including two copies for the flies. The relationship between these reports and the reports subsequently produced by Edwin Gay when he organized the Central Bureau of Planning and Statistics is not clear. Gay was aware of the work that Ayres was doing and, through Bernard Baruch, the chairman of the War Industries Board, received a copy of Ayres's “weekly confidential report.” See Cuff, “Creating Control Systems,” 606–8.

55 A Conversation between Mr. R. H. Williams, Assistant to the Director of the Council of National Defense and Chairman of the Advisory Committee on Purchase of Public Animals and Remount Service, and Mr. Grosvenor B. Clarkson, Secretary of the Council of National Defense, October 2, 1917. Subject: “The Purchase of Horses and Mules for the United States Army, together with a General Discussion of the Organization and Work of the Remount Division of the United States Army” [16-page typescript], 1, 4–5; RG 62, CND Records, 3-B1, Secretary's Office, box 173.

56 Second Annual Report of the Council of National Defense, 208.

57 Division of Statistics, “Weekly Report (August 25, 1917),” 3.

58 Second Annual Report of the Council of National Defense, 209. The members of the Tonnage Board of the Statistical Division were Ayres, Major Coward, and Dr. Field. In early January 1918, General Tasker H. Bliss took Coward with him to the Supreme War Council at Versailles to act as statistical officer, particularly responsible for data on shipping. The Bliss mission took with it over one hundred reports and tables prepared by the Division of Statistics “bearing on the problems of the transportation of troops and the progress made to date in our material preparations for war; the situation with respect to explosives and propellants; and American resources in raw materials.” See Division of Statistics, “Weekly Report (January 5, 1918),” 1. Shortly afterward, Dr. Field went to London to act as statistical officer for the American Section of the Inter-Allied Shipping Board. See Ayres, “First Annual Report, Division of Statistics,” 28–29. For the comment on the “very full data” supplied by the U.S. Shipping Board, see Division of Statistics, “Weekly Report (October 27, 1918),” 1. Four members of the Division of Statistics and five liaison officers worked on different aspects of the tonnage problem. See Division of Statistics, “Weekly Report (November 3, 1917),” 1. The first report was completed toward the end of November. The emphasis placed by Heaton and Cuff on the importance of the report to the Shipping Board made by Edwin Gay at the beginning of January 1918, which called for an additional 3.6 million deadweight tons of shipping in the first half of 1918, needs to be seen in the context of the considerable work previously done by the Division of Statistics. Cuff reports that Gay “drew on the statistics departments of the Council of National Defense and the War Department and on English data gathered by the U.S. Mission, recently returned from Paris.” See Cuff, “Creating Control Systems,” 596. Given the highly sensitive nature of the material, Gay may not have been given access to all the data available.

59 Ayres, “First Annual Report, Division of Statistics,” 29.

60 Baruch, Bernard M., American Industry in the War: A Report of the War Industries Board (March 1921), ed. Hippelheuser, Richard H. (New York, 1941), 4445Google Scholar.

61 Hatfield, H. R., “History of the Division of Planning and Statistics” [WIR], 11Google Scholar-page typescript, War Industries Hoard, vol. 1, pt. 1, no. 18, Bernard M. Baruch Papers, Princeton University Library, Princeton, N.J. Hugh S. Johnson, the army's representative on the War Industries Board, later commented that the army supply systein during the First World War “just didn't work.” He attributed this to “the tremendous tenacity of life of a Government bureau.” The army supply system was composed of “a cluster of jealous and ancient bureaus,” which, he believed, “fought the exterior control of the War Industries Board at every step….”r See Johnson, Hugh S., The Blue Eagle from Egg to Earth (Garden City, N.Y., 1935)Google Scholar, quoted in Baruch, American Industry in the War, 36–37.

62 Ayres, “First Annual Report, Division of Statistics,” 15. Because the information received was inadequate, it was decided “to depend for deliveries records upon summarized statements of kinds, and amounts, rather than attempt to secure any extensive itemized reports by orders and firms” (p. 15). In other words, informed guesswork would remain the basis for policy in this area. This problem continued during 1918. See Clarkson, Industrial America in the World War, 202; Hatfield, “History of the Division of Planning and Statistics,” 3–4. See also the statement by Edwin Gay concerning the Division of Planning and Statistics of the War Industries Board, in which he complained about “not getting really effective cooperation until toward the close of 1918.” Quoted in Cuff, “Creating Control Systems,” 603.

63 The bulletin listed the causes of delay in deliveries, reported cases of unused plant capacity, and sought to provide information useful “to those who are placing war orders.” About eighty copies of the bulletin were issued twice a month. See Ayres, “First Annual Report, Division of Statistics,” 16. Note that the Second Annual Report of the Council of National Defense refers to the bulletin as a daily publication (pp. 205, 210). The War Contracts Section also published a weekly labor report, based on the replies to the questionnaires, summarizing reports on labor conditions in different industries.

64 Baruch, American Industry in the War, 45.

65 In spite of various efforts made during 1918 to remedy the problems associated with the statistics of war contracts, no satisfactory solution evolved, although, like many other aspects of the industrial mobilization, a satisfactory resolution was within reach when the Armistice was signed. Reviewing the work of the Division of Planning and Statistics of the War Industries Board in 1919, the director commented on the difficulty of securing statistical information on 1) estimates (preferably monthly) comparing military supplies needed and anticipated supplies available; 2) contracts actually placed compared with anticipated needs of military supplies; and 3) the production and consumption of commodities, especially manufactured articles. Initially, the division had to rely on the 1914 Census of Manufactures, which was organized by industry rather than by commodity. The gathering of comprehensive statistics on production and consumption was a very slow process. See Hatfield, “History of the Division of Planning and Statistics,” 3–4, 10–11.

66 Division of Statistics, “Weekly Report (October 20, 1917),” 1–2. A supplementary report estimating the capacity of plants to produce “smokeless powder and T.N.T.” was also produced by the division. The information “was found to be of the first importance” and the division was asked by the War Industries Board to make a report on the production of chemicals essential to the manufacture of explosives, including “sulphuric acid, alkali, ammonia, nitric acid, and toluol.”

67 W. Randolph Burgess, London, to Leonard P. Ayres, 21 Nov. 1917 [8 pp.], 6–7; RG 62, CND Records, 2-A1, Director's Office: General Correspondence, box 36, file 141 (“Great Britain: Ministry of Munitions”). Burgess was office manager of the Division of Statistics and looked after the business details. He left Washington for London in late October and returned to the division shortly before Christmas.

68 Cuff, “Creating Control Systems,” 613. See also Beniger, The Control Revolution.

69 Second Annual Report of the Council of National Defense, 210. A list of the eight bulletins produced by the division, which were almost all weeklies, and the eighty-five special reports prepared, is included in Ayres, “First Annual Report, Division of Statistics,” 39–43. The list does not include “large numbers” of small reports and memoranda. When the Statistics Division of the CND became the Statistics Branch of the General Staff, its composition did not change greatly: during 1918 over one hundred officers were assigned to the Branch and “more than half came from professions, and most of the rest from the statistical and accounting offices of business firms. Only three have been Regular Army officers. Nearly all were young men, and most of them had received thorough technical training in college, universities, and graduate schools.” See Annual Report of General Peyton C. March, Chief of Staff, United States Army, 1919 (War Department, 30 June 1919), 444Google Scholar.

70 On general developments within the War Department in the winter of 1917–18, see Beaver, Daniel R., Newton D. Baker and the American War Effort, 1917–1918 (Lincoln, Nebr., 1966), chap. 4, esp. 9497Google Scholar.

71 Second Annual Report of the Council of National Defense, 210.

72 Hatfield, “History of the Division of Planning and Statistics,” 1–2. In late 1918, Ayres went to France as chief statistical officer of the American Commission to Negotiate Peace. He returned to Washington in April 1919 to work on statistical aspects of the mobilization for the War Department. One result of this was the short statistical history, The War with Germany. On 1 October 1919, Ayres left government service and returned to his work at the Russell Sage Foundation in New York. In 1920, he accepted a position with the Cleveland Trust Company.

73 Cuff, “Creating Control Systems,” 605–9. The division between Ayres and Gay in 1918 has overtones of the bitter debate during and after the Second World War between those who appeared to favor either greater civilian or greater military control over the wartime economy. See Polenberg, Richard, War and Society: The United States, 1941–1945 (Philadelphia, Pa., 1972), esp. chap. 8Google Scholar; Carton, Bruce, The War Lords of Washington (New York, 1948)Google Scholar.

74 Heaton, A Scholar in Action, 131—36, quotation on 135. See also Alchon, The Invisible Hand of Planning, 34—38; Hawley, Ellis W., “Economic Inquiry and the State in New Era America: Antistatist Corporatism and Positive Statism in Uneasy Coexistence,” in The State and Economic Knowledge: The American and British Experiences, ed. Furner, Mary O. and Supple, Barry (New York, 1990), 290Google Scholar. The postwar period witnessed both the collapse of central government statistical work and the beginning of the proliferation of private “think tanks” in Washington, some of which focused on statistics. See Critchlow, “Think Tanks, Antistatism, and Democracy,” in Lacey and Furner, eds., The State and Social Investigation, chap. 7.

75 Lang, Richard O., “Problems of Statistical Control—Military Aspects,” Journal of the American Statistical Association 36, no. 213 (March 1941): 1117Google Scholar; quotation on 13.

76 The quotation is from the diary of Secretary of War Henry Stimson in early October 1940. For this quotation and for the reference to the Lang article in the previous footnote, I am indebted to Robert Cuff, “War Mobilization, Social Learning, and State Building: 1917–1941,” in Lacey and Furner, eds., The State and Social Investigation, especially notes 30 and 49. Cuffs essay explores the ways in which the administrative experience of the First World War was absorbed in the interwar period and how the growth of the American state itself, especially in the New Deal era, limited the applicability of that earlier experience. The crisis of the New Deal had stimulated the development of more centralized statistical work in the federal bureaucracy; in the Second World War, unlike in the First, the American state quickly developed quite sophisticated statistical controls. In the Second World War, Ayres served as coordinator of statistics in the Office of the Undersecretary of War from 1940 to 1942 and then as a consultant with the War Manpower Commission from 1943 to 1945. He died in 1946.