¹ Landes, David S., “What Do Bosses Really Do?” Journal of Economic History 46 (Sept. 1986): 585–624CrossRefGoogle Scholar; Marglin, Stephen, “What Do Bosses Do? The Origins and Functions of Hierarchy in Capitalist Production,” Review of Radical Political Economics 6 (Summer 1974): 33–60CrossRefGoogle Scholar.

² Marglin, “What Do Bosses Do?” 84.

³ Landes, “What Do Bosses Really Do?” 610.

⁴ According to Beatrice and Sidney Webb, until 1850 union organization in cotton spinning was impermanent and its influence on wages and working conditions was limited. Webb Trade Union Collection, A Cotton Spinners' History, 1700–1896, vol. 34, Library of the London School of Economics and Political Science, London, England. For a recent history of trade union organization in spinning emphasizing its continuity, see Fowler, Alan and Wyke, Terry, The Barefoot Aristocrats: A History of the Amalgamated Association of Operative Cotton Spinners (Littleborough, England, 1987)Google Scholar.

⁵ Hobsbawm, E. J., Labouring Men (London, 1968), 347–50Google Scholar.

⁶ Lazonick, William, “Industrial Relations and Technical Change: The Case of the Self-Acting Mule,” Cambridge Journal of Economics 3 (Sept. 1979): 231–62Google Scholar; Lazonick, , “Production Relations, Labor Productivity, and Choice of Technique: British and U.S. Cotton Spinning,” Journal of Economic History 41 (Sept. 1981): 491–516CrossRefGoogle Scholar; Lazonick, , Competitive Advantage on the Shop Floor (Cambridge, Mass., 1990)Google Scholar.

⁷ Lazonick, William, “Theory and History in Marxian Economics,” in The Future of Economic History, ed. Field, Alexander J. (Boston, Mass., 1986), 279Google Scholar.

⁸ In 1815 and 1841 the median textile firm in Manchester employed between 351 and 400 workers. The average fine-spinning firm in Lancashire employed 191 workers. Lloyd-Jones, Roger and Le Roux, A. A., “The Size of Firms in the Cotton Industry: Manchester, 1815–41,” Economic History Review 33 (Feb. 1980): 76Google Scholar. Gatrell, V. A. C., “Labour, Power, and the Size of Firms in Lancashire Cotton in the Second Quarter of the Nineteenth Century,” Economic History Review 30 (Feb. 1977): 127CrossRefGoogle Scholar. The average firm spun coarse counts. Despite its distinctiveness, in an early study of M&K, G. W. Daniels argued that it was a “typical cotton spinning firm from that stage of the Industrial Revolution when the cotton industry was beginning its rise to prominence.” In his recent study of capital formation, Philip Richardson offered a more cautious view, but concluded that, despite its size and specialty, M&K was not unique. Daniels, G. W., “The Early Records of a Great Manchester Cotton Spinning Firm,” Economic Journal 25 (1915): 175CrossRefGoogle Scholar; Richardson, Philip, “The Structure of Capital During the Industrial Revolution Revisited: Two Case Studies from the Cotton Textile Industry,” Economic History Review 42 (Nov. 1989): 502CrossRefGoogle Scholar.

⁹ The evidence is from a comparison of the employment records of nine Manchester factories for 1818. H.L. 1818 (90) XCVI, Evidence on the Cotton Factories, appendix.

¹⁰ Kirby, R. G. and Musson, A. E., The Voice of the People: John Doherty, Trade Unionist, Radical and Factory Reformer (Manchester, England, 1975), 13–14, 52Google Scholar; Freifeld, Mary, “Technological Change and the ‘Self-Acting’ Mule: A Study of Skill and the Sexual Division of Labour,” Social History 10 (May 1986): 319–43CrossRefGoogle Scholar.

¹¹ The archives of the firm have been widely consulted. See, among others, Daniels, G. W., The Early English Cotton Industry (Manchester, England, 1920)Google Scholar; Smith, Roland, “Manchester as a Centre for the Manufacture and Merchandising of Cotton Goods, 1820–30,” University of Birmingham Historical Journal 4 (1953–1954): 47–65Google Scholar; Collier, Francis, The Family of the Working Classes in the Cotton Industry, 1784–1833 (Manchester, England, 1964)Google Scholar. The most thorough account of the firm's activities is Lee, C. H., A Cotton Enterprise, 1795–1840: A History of M'Connel and Kennedy, Fine Cotton Spinners (Manchester, England, 1972)Google Scholar. Despite the breadth of the secondary literature, the sources on production and industrial relations at the firm remain untapped.

¹² This account is based on Lee, History of M'Connel and Kennedy, 10–90.

¹³ Total capital figures are net of credit outstanding. Richardson, “Structure of Capital,” 499.

¹⁴ M&K Inventory Book, John Rylands Library, University of Manchester, Manchester, England; Lee, History of M'Connel and Kennedy, 162.

¹⁵ For an elaboration, see Lazonick, William, Business Organization and the Myth of the Market Economy (New York, 1991)Google Scholar.

¹⁶ Hughes, J. R. T., Fluctuations in Trade, Industry and Finance (Oxford, England, 1960), 90Google Scholar; Matthews, R. C. O., A Study in Trade Cycle History: Economic Fluctuations in Great Britain, 1833–42 (Cambridge, England, 1954), 138–39Google Scholar.

¹⁷ “[T]he maximisation of production and also of productive capacity represented the appropriate business strategy.” Lee, C. H., “The Cotton Textile Industry,” in The Dynamics of Victorian Business, ed. Church, Roy (London, 1980), 170–71Google Scholar. With industry-wide prices falling, less efficient firms were forced to exit. Matthews, Trade Cycle, 129–31.

¹⁸ For evidence of the failure of wage cuts to match the fall in margins, see the testimony of employers in Parliamentary Papers (P.P.) 1833 (690) VI, Select Committee on Manufacture, Commerce, and Shipping, evidence of G. Smith, W. R. Greg, and J. Milne.

¹⁹ Lee, History of M'Connel and Kennedy, 137–43.

²⁰ Pollard, Sidney, “Capital Accounting in the Industrial Revolution,” Journal of Economic History 24 (Sept. 1964): 299–315CrossRefGoogle Scholar; John S. Lyons, “Vertical Integration in the British Cotton Industry, 1825–1850: A Revision,” ibid. 45 (June 1985): 419–25.

²¹ For different approaches to the problem, see Lazear, Edward, “Salaries and Piece Rates,” Journal of Business 59 (No. 3, 1986): 405–31CrossRefGoogle Scholar; Edwards, Richard, Contested Terrain (New York, 1979), 98–99Google Scholar.

²² This was clearly the intention of the firm. H.L. 1819 (24) CX, Minutes of Evidence on Children Employed in Cotton Manufactories, 237, 342.

²³ A comprehensive treatment of the early strikes and union formation is found in Fowler and Wyke, Barefoot Aristocrats, 14–36.

²⁴ M&K Letterbooks, 4 Nov. 1811, 23 Oct. 1813, 17 Sept. 1818.

²⁵ Sutcliffe, John, A Treatise on Canals and Reservoirs (Rochdale, England, 1816), 36Google Scholar.

²⁶ Freifeld, “Technological Change,” 335.

²⁷ “Certain large mills in 1818, like M'Connel and Kennedy, employed women and young boys on small mules.” Hall, Roger G., “Tyranny, Work and Politics: The 1818 Strike Wave in the English Cotton District,” International Review of Social History 34 (1980): 450Google Scholar. Hall cites statements of employers, overseers, and workers before the commissions on factory conditions. See also Lazonick, “Industrial Relations,” 235.

²⁸ Freifeld, “Technological Change,” 334.

²⁹ Kirby and Musson, Voice of the People, 370; McConnel, J. W., A Century of Fine Spinning (Manchester, England, 1906)Google Scholar; see also, Collier, Family Economy, 17.

³⁰ Lee, History of M'Connel and Kennedy, 114; H.L. 1818 (90) XCVI, appendix 6. The increase in the proportion of female workers was not the result of a decline in the number of male mechanics or engineers. Although after the turn of the century the firm stopped selling spinning mules, it retained mechanics and engineers to build and maintain its own increasing complement of machinery. In 1818 the firm employed about fourteen mechanics, and by the 1830s, about twenty; see M&K Deeds and Documents.

³¹ M&K Inventory Book.

³² Lazonick, “Industrial Relations,” 235.

³³ The count of number of yarn is the number of hanks per 1 lb. of yarn. A hank is a single stand of cotton 840 yards long. Thus, 1 lb. of no. 150 consists of 126,000 (150 × 840) yards of yarn.

³⁴ No correction was made for inflation because the price index showed little change between the two years. Cost estimates for 1811 from Kennedy's statement reproduced in Baines, History of Cotton, 353; Wages: M&K Cash Ledger, Price Index: Williamson, Jeffrey G., Did British Capitalism Breed Inequality? (Boston, Mass., 1985), 212Google Scholar.

³⁵ The equation waste = a + b(count) was estimated by ordinary least squares, using rates of change of monthly observations between 1810 and 1818 (N = 60). The regression was significant (F = 7.05) and had an R = 0.108. Regressions were also run for levels and logs of waste and count, and the results did not alter significantly. Source: M&K Yarn Output Book.

³⁶ If multiple end breakages were not pieced up, then the mule would have come to a complete stop. This “downtime” would have lowered productivity. The output growth in Table 2 is thus net of downtime.

³⁷ Statement before the factory commissioners cited in Cohen, Isaac, American Management and British Labor: A Comparative Study of the Cotton Spinning Industry (New York, 1990), 64Google Scholar.

³⁸ Lazonick, “Industrial Relations,” 236. Lazonick's evidence is from the testimony of workers and overlookers in the Reports of the Factory Commissions of 1833. See also, Pollard, Sidney, The Genesis of Modern Management (Harmondsworth, England, 1965), 213–25Google Scholar.

³⁹ Pinchbeck, Ivy, Women Workers and the Industrial Revolution, 1750–1850 (London, 1930), 186Google Scholar; see also, Cohen, American Management, 63.

⁴⁰ P.P. 1833 (450) XX, First Report of the Commissioners on the Employment of Children, D1, 52–53; P.P. 1831–32 (706) XV, Report from the Committee to Regulate the Labour of Children, 440. The evidence on overlookers is not strictly for M&K, but refers to the overall situation in Lancashire. An indicator of the scarcity of overlookers is the increase in their real wages over the period. See Huberman, Michael, “Vertical Integration in Lancashire: A Comment,“ Journal of Economic History 50 (Sept. 1990): 688–689CrossRefGoogle Scholar and Huberman, , “How Did Labor Markets Work in Lancashire? More Evidence on Prices and Quantities in Cotton Spinning, 1822–52,” Explorations in Economic History 28 (Jan. 1991): 5–6CrossRefGoogle Scholar.

⁴¹ Wastage was about 6 percent of unit costs. (Waste comprised 20 percent of cotton costs, which in turn represented about 30 percent of all costs in 1811.) The prices of Sea Island cotton fluctuated wildly in the period. For purposes of calculation, the change in price was based on the low and high prices of Sea Island cotton in Table 1 and was assumed to have doubled.

⁴² Fitton, R. S., The Arkwrights: Spinners of Fortune (Manchester, England, 1989), 165Google Scholar. Fitton's evidence is taken from H.L. 1819 (24) CX, 436.

⁴³ H.L. 1818 (90) XCVI, 177. The high turnover at M&K was typical of the period. See Fitton's analysis of employment records from the Factory Commissions of 1818 in The Arkwrights, 152.

⁴⁴ M&K Inventory Book.

⁴⁵ Ellison, Thomas, The Cotton Trade of Great Britain (London, 1886), 32Google Scholar.

⁴⁶ H.L. 1818 (90) XCVI, appendix 6. M&K Deeds and Documents.

⁴⁷ Turner, H. A., Trade Union Growth, Structure and Policy: A Comparative Study of the Cotton Unions in England (Toronto, Ont., 1962), 114, 128Google Scholar; Marianna Valverde, “Giving the Female a Domestic Turn: The Social, Legal and Moral Regulation of Women's Work in British Cotton Mills, 1820–1850,” Journal of Social History (June 1988): 619–34; Chapman, S. J., The Lancashire Cotton Industry: A Study in Economic Development (Manchester, England, 1904), 59Google Scholar. Lazonick, “Industrial Relations,” 236; Cohen, Isaac, “Workers' Control in the Cotton Industry: A Comparative Study of British and American Mule Spinning,” Labor History 26 (Winter 1985): 72CrossRefGoogle Scholar. For a survey of different views on the demise of women spinners, see Freifeld, “Technological Change.” Freifeld is concerned with the transition from common-mule to self-actor spinning. Her argument is that there was a failure to transfer skills across generations of women. The argument is not applicable to M&K because of the delay in the introduction of self-actors in fine spinning.

⁴⁸ Quoted in Lazonick, “Industrial Relations,” 236; Lazonick's explanation of the diffusion of the new technology is that at the point of installation the length of the mule was a variable and the strength of the worker was fixed. Lazonick, Competitive Advantage, 84.

⁴⁹ Between 1818 and 1838 the proportion of young males 13 to 20 at M&K rose from 14.0 to 24.3 percent. Source: M&K Deeds and Documents.

⁵⁰ With the introduction of longer mules, spinners preferred to hire boys as piecers because they occasionally had to push the heavy carriages. Cohen, American Management, 67. The number of adolescent males, at the expense of females, rose after the introduction of long mules. The hiring of more male piecers made it unlikely that after the introduction of long mules women acquired experience in spinning, thereby dampening employers' incentives to hire them.

⁵¹ On the decline of family work groups in spinning, see Anderson, Michael, “Sociological History and the Working Class Family: Smelser Revisited,” Social History 3 (Oct. 1970): 317–34Google Scholar. For M&K, evidence on family relations is from H.L. 1818 (90) XCVI, appendix 6; and Shuttleworth, John, “Vital Statistics of Piecers and Spinners Employed in the Fine Spinning Mills of Manchester,” Journal of the Royal Statistical Society 5 (1842): 273Google Scholar.

⁵² Lazonick, “Theory and History,” 278.

⁵³ For evidence of maltreatment, P.P. 1833 (450) XX, D1, 688; P.P. 1833 (519) XXI, Second Report of the Commissioners of the Employment of Children, D2, 191, 194; Cohen, American Management, 64. Shuttleworths survey recorded the percentage of spinners who inflicted punishment. At M&K 63 percent of spinners maltreated their piecers, but the average of nineteen spinning firms surveyed was 48 percent. Shuttleworth, “Vital Statistics,” 273.

⁵⁴ Catling, H., The Spinning Mule (Newton Abbott, England, 1970), 149Google Scholar.

⁵⁵ On short time, see M&K Letterbooks, 28 Feb. and 30 March 1826.

⁵⁶ An examination of local newspapers and other contemporary accounts confirmed that there were no protracted stoppages at M&K or at neighborhood firms.

⁵⁷ For a review of the period's spinning manuals, see Chapman, Stanley D., “The Textile Industries,” in Where Did We Go Wrong? Industrial Performance, Education and the Economy in Victorian Britain, ed. Roderick, Gordon and Stephens, Michael (Sussex, England, 1981), 125–38Google Scholar. For different analytical treatments about the gains from restricting output, see Lazonick, Competitive Advantage, appendix; Gibbons, Robert, “Piece Rate Incentive Schemes,” Journal of Labor Economics 5 (1987): 413–29CrossRefGoogle Scholar.

⁵⁸ The author of a popular spinning manual wrote, “Having been engaged in the different departments of the cotton business for many years, I have found a general deficiency in theoretical knowledge amongst practical persons filling important situations in cotton and other factories.” Scott, Robert, Scott's Practical Cotton Spinner and Manufacturer, 3d ed. (London, 1851), ivGoogle Scholar.

⁵⁹ The Tukey test of multiple comparisons was used to isolate the cell means that were statistically different from each other. The test was conducted using the means from a two-factor analysis of variance of years and trimesters. For 1822–26, output of all mules in April, July, and September were grouped in cells and the variation analyzed between these cells.

⁶⁰ On wage cuts in 1816, see Kirby and Musson, Voice of the People, 18; Hall, “Tyranny, Work and Politics,” 451.

⁶¹ Proceedings of the Manchester Chamber of Commerce, Feb. 1826.

⁶² There were twenty firms in the Association. For a detailed study of the 1829 dispute, see Kirby and Musson, Voice of the People, 59–85.

⁶³ M&K Letterbooks, 17 Oct. 1829.

⁶⁴ Applying efficiency wage theory, if all firms paid by the list and fixed rates of pay, wages would have cleared the market and unemployment would have resulted. The stick of unemployment was an added incentive for spinners to work hard. Akerlof, George A. and Yellen, Janet L., eds., Efficiency Wage Models of the Labor Market (Cambridge, England, 1986)CrossRefGoogle Scholar.

⁶⁵ The absolute mean changes in percent were: earnings, 9.22; piece rates, 14,45; output, 18.43. For calculations, see Table 2 in Huberman, Michael, “The Economic Origins of Paternalism: Lancashire Cotton-Spinning in the First Half of the Nineteenth Century, “ Social History 12 (May 1987): 189–90CrossRefGoogle Scholar.

⁶⁶ Calculations based on annual price and output changes as proxies for piece rate and employment changes. See Huberman, “How Did Labor Markets Work, “ Table 6. Spinners in Manchester had higher earnings than fine spinners in regions, like Oldham, that did not pay by a list. Wood, G. H., The History of Wages in the Cotton Trade During the Past 100 Years (London, 1910), 25, 58Google Scholar.

⁶⁷ M&K Letterbooks, 28 Oct. 1830.

⁶⁸ Manchester Guardian, 6 Feb. 1830; Kirby and Musson, Voice of the People, 109. Initially, small firms were reluctant to abide by the list because they could not compete with larger spinners and their longer mules. However, the vast majority of fine spinners in Manchester were large, and by the early 1830s they succeeded in imposing the list on all firms. Fowler and Wyke, Barefoot Aristocrats, 30. On the survival of the list in the 1830s, see Ure, Andrew, The Philosophy of Manufacturers (London, 1835), 348Google Scholar.

⁶⁹ For a detailed discussion of the causes of short-hour working and for references, see Huberman, “Economic Origins,” 188–90.

⁷⁰ For the views of workers of different ages in support of short-time, see P.P. 1849 (1017) XXII, Reports of the Factory Inspectors, 207–17.

⁷¹ Only aggregate data are available for the 1830s. The data have two drawbacks: it is not possible to distinguish between short and long mules at the firm, and the presence of short time that can only be estimated imparts a downward bias in the calculation of the number of hanks per spindle.

⁷² The average output per spindle in the period between 1810 and 1817 was 5.61, and for 1834 to 1840, (excluding 1837 when short time was worked almost continuously), 6.38, or an increase of about 14 percent. Taking the two best years in each of the two periods gives an increase in output per spindle from 6.01 to 7.02, or about 17 percent. It is difficult to compare fluctuations in productivity among the periods because of the type of evidence available for the 1830s and the amount of short-hour working in that period.

⁷³ Labor productivity estimates confirm the trend in output per spindle. In 1816 the number of hanks per worker per week was 490.6, in 1825, 506.5, and in 1838, 608.1. These years were selected because of the availability of data. Output was corrected by Kennedy's index. The average count in 1816 was no. 165; 1825, no. 115; 1838, no. 150. The actual output values are 1816 = (79,020 spindles × 5.76 hanks per spindle)/1,020 workers; 1825 = (124,848 spindles × 7.81 hanks per spindle)/1,540 workers; 1838 = (124,848 spindles × 7.55 hanks per spindle)/1,550 workers. See Tables 2, 3, and 6 and M&K Inventory Book, Deeds and Documents.

⁷⁴ On the development of the lists, see Chapman, S. J., “The Regulation of Wages by Lists in the Spinning Industry,” Economic Journal 9 (1899): 80–89CrossRefGoogle Scholar; British Association for the Advancement of Science, Regulation of Wages by Means of Lists in the Cotton Industry (Manchester, England, 1887)Google Scholar.

⁷⁵ Lee, History of M'Connel and Kennedy, 152–53.

⁷⁶ Jewkes, J. and Gray, E., Wages and Labour in the Lancashire Cotton Spinning Industry (Manchester, England, 1935), 35–53Google Scholar.

⁷⁷ Lazonick, “Production Relations,” 503–4.

⁷⁸ See for example, Clay, H., Report on the Position of the British Cotton Industry (London, 1931)Google Scholar.