¹ Layer, Robert, “Wages, Earnings and Output in Four Cotton Textile Companies in New England: 1825–1860,” unpublished doctoral dissertation, Harvard University, 1952Google Scholar; McGouldrick, Paul, New England Textiles in the Nineteenth Century: Profits and Investment (Cambridge, 1965)Google Scholar; Davis, Lance E. and Stettler, H. Louis, “The New England Textile Industry, 1825–1860: Trends and Fluctuations,” in Conference on Research in Income and Wealth, Output, Employment, and Productivity in the United States After 1800, National Bureau of Economic Research, Studies in Income and Wealth, 30 (New York, 1966), pp. 213–38Google Scholar; Zevin, Robert B., The Growth of Manufacturing in Early Nineteenth-Century New England (New York, 1975)Google Scholar; David, Paul A., Technical Choice, Innovation, and Economic Growth: Essays on American and British Experience in the Nineteenth Century (New York, 1975)Google Scholar; Williamson, Jeffrey G., “Embodiment, Disembodiment, Learning by Doing and Constant Returns to Scale in the Nineteenth-Century Cotton Textiles,” this Journal, 32 (Sept. 1972), 699–705Google Scholar.

² See for example Abbot, Edith, Women in Industry: A Study in American Economic History (New York, 1910)Google Scholar; Josephson, Hannah, The Golden Threads, New England's Mill Girls and Magnates (New York, 1949)Google Scholar; Ware, Carolyn F., The Early New England Cotton Manufacturer: A Study in Industrial Beginnings (New York, 1931)Google Scholar. Ware, Norman, The Industrial Worker, 1840–1850 (Boston, 1924)Google Scholar.

³ David, Technical Choice, p. 122.

⁴ Davis and Stettler, “The New England” 227–29. Layer, “Wages, Earnings and Output,” 171–77. Paul F. McGouldrick, “Comment,” in Conference on Research in Income and Wealth, Output, Employment and Productivity in the U.S. After 1800, p. 240.

⁵ Gitelman, Howard M., “The Waltham System and the Coming of the Irish,” Labor History, 8, (Fall 1967), 227–53CrossRefGoogle Scholar. Nickless, Pamela J., “Changing Labor Productivity and the Utilization of Native Women Workers in the American Cotton Textile Industry: 1825–1860,” (Ph.D. diss., Purdue University, 1976), ch. 2.Google Scholar

⁶ David, Technical Choice, pp. 122–24. The easy availability of census data for female and male workers encourages this sort of analysis. In this case, however, the use of census data creates a serious problem. The firms David is analyzing are all integrated multi-mill firms, and the increase in the numbers of women in the industry may reflect the ongoing shift in production to integrated mills that employed larger numbers of women. The census data then may not be entirely appropriate and may bias the results.

⁷ Gitelman, “The Waltham System” 227–253. Nickless, “Changing Labor Productivity,” ch. 2.

⁸ Nickless, “Changing Labor Productivity,” ch. 2.

⁹ Ibid.

¹⁰ David, Technical Choice, pp. 169–72.

¹¹ Ibid., p. 165.

¹² Layer, Robert, Earnings of Cotton Mill Operatives, 1825–1914 (Cambridge, MA., 1955), p. 43.Google Scholar

¹³ Solow, Robert M., “Technical Change and the Aggregate Production Function,” Review of Economics and Statistics, 39 (Aug. 1957), 312–20CrossRefGoogle Scholar; see also Jorgenson, D. W. and Griliches, Z., “The Explanation of Productivity Changes,” Review of Economic Studies, 34 (July, 1967), 249–83.CrossRefGoogle Scholar

¹⁴ s₁ and s₂ could be calculated as follows.

(1)

where w₁L₁ = overseers' wage bill; w₂L₂ = operatives' wage bill. Substituting λ = w₁L₁/w₂L₂ (derived from the representative mills)

(2)

which is equal to

(3)

Since 1 − s₃ = s₁ + s₂; 1 − s₃ = (1 + λ)s₂ and therefore

(4)

Thus, estimates for s, and s2 were derived for each year.

¹⁵ David, Technical Choice, pp. 144–49. (David argues that monopsony existed in all labor markets.) Layer, “Wages, Earnings and Output,” ch. 5; McGouldrick, New England Textiles, p. 45; Ware, The Early New England, ch. 8.

¹⁶ Nickless, “Changing Labor Productivity,” ch. 2.

¹⁷ All of the following calculations use ΔL_t/L_t, as it appears in Table 2. The conversion to 1845 base is only for purposes of presentation.

¹⁸ McGouldrick, New England Textiles, pp. 139–40.

¹⁹ The output index and the productivity results are remarkably insensitive to the choice of weights. Average price weights for 1836–40 and 1856–60 produce similar results. An unweighted index also does not alter the conclusions.

²⁰ For a more detailed discussion of the properties of Divisia indices, see Richter, Marcel K., “Invariance Axioms and Economic Indexes,”. Econometrica, 34 (Oct. 1966), 739–55.CrossRefGoogle Scholar

²¹ Layer, “Wages, Earnings and Productivity,” Appendix C; McGouldrick, New England Textiles, pp. 157–59.

²² McGouldrick, New England Textiles, chs. 3 and 7.

²³ Ibid., pp. 69–71.

²⁴ It was necessary to use Layer's figures since no other total capital figures for these firms were available. The figures underestimate Nashua's capital and it was feared that the movement in capital input might be overstated, but when Nashua's figures were revised upwards to the level of the Hamilton Company, the resulting capital-input index was changed by one percentage point in only five years.

²⁵ The deflator used is the one McGouldrick uses to deflate investment—the Snyder-Tucker index of general prices. This index is inappropriate to investment in several ways, but it has the advantage that it is less sensitive to movements in material and semi-finished goods prices than other indices. As McGouldrick points out, most new investment was for new construction, which had a high degree of labor cost. Labor costs varied less than commodity prices and a general price index includes labor costs and shows less variation than the Warren-Pearson wholesale price index (New England Textiles, pp. 162–63, 292.)

The Snyder-Tucker general price index was also compared with Dorothy Brady's price index of machine shop products. Machine shop products included were steam engines, water wheels, lathes, and looms. The indices appeared very similar and produced similar results in deflating the investment series. Brady's series is not a continuous series and could not be used in place of the Snyder-Tucker index even though it is conceptually more appropriate (Dorothy S. Brady, “Price Deflators for Final Product Estimates,” in Conference on Research in Income and Wealth, Output, Employment and Productivity in the United States After 1800, pp. 91–115.)

It has been suggested that it would be more appropriate to combine the Snyder-Tucker index and the Brady index with data on interest rates and assumptions about depreciation rates to attempt to measure the price of capital services. This could then be used to derive a measure of real services per deflated unit of capital stock, thus eliminating the assumption of constant proportionality between the capital-stock index and the capital-service index. The suggestion was rejected because of the difficulty, of assessing the possible direction of bias. The measure used in the text would appear to understate capital-service flow over time, given the possibilities of improved machine quality and increasing speeds of production.

²⁶ Davis and Stettler, “The New England Textile,” pp. 552–53.

²⁷ McGouldrick, New England Textiles, 144–48.

²⁸ Ibid., 38–40.

²⁹ Ibid., Appendix C; Davis, Lance E., “The New England Cotton Textiles Mills and the Capital Markets: A Study in Industrial Borrowing, 1840–1860,” Essays in American Economic History, ed. Coats, A. W. and Robertson, R. M. (London, 1969)Google Scholar.

³⁰ McGouldrick, New England Textiles, p. 41. Gibb, George S., The Saco-Lowell Shops and Textile-Machine Building in New England, 1813–1849 (Cambridge, 1950), p. 80CrossRefGoogle Scholar.

³¹ Brito, D. L. and Williamson, Jeffrey G., “Skilled Labor and Nineteenth-Century Anglo-American Managerial Behavior,” Explorations in Economic History, 10 (Spring 1973), 235–51.CrossRefGoogle Scholar

³² Jeremy, David J., “Innovation in the American Textile Industry During the Early Nineteenth Century,” Technology and Culture, 14 (Jan. 1973), 40–76CrossRefGoogle Scholar.

³³ Crucial to this analysis is the assumption that wages accurately reflect marginal products; if this is not the case, then the analysis of both labor productivity and total factor productivity is questionable. Evidence from the firms' payrolls and census data analyzed by Howard M. Gitelman in “The Waltham System,” (227–53) indicates that the Irish were paid the same wages as their native counterparts. Data analyzed by the author indicate that women and men operatives were paid substantially the same wages. Thus it appears that wage discrimination was not an important factor in these firms. The possible non-existence of wage discrimination should not be interpreted as a denial of ethnic or sex discrimination. Both existed in the Waltham system, but took more subtle forms than simple wage discrimination.

³⁴ Robinson, Harriet Hanson, Loom and Spindle (Boston, 1898)Google Scholar is especially illuminating on this point.

³⁵ Braverman, Harry, Labor and Monopoly Capital, (New York, 1974)Google Scholar, especially ch. 3.