¹ United Nations, International Standard Industrial Classfication of All Economic Activities, Statistical Papers, Series M, No. 4, Rev. 2 (briefly, ISIC; New York, 1968). For the sake of convenience, public institutions and documents are referred to in the text directly by their brief titles.Google Scholar

² For example, Landes, David S., The Unbound Prometheus (Cambridge, 1969), p. 235;Google ScholarSchumpeter, Joseph A., Business Cycles (New York, 1939), pp. 397–98;Google ScholarSinger, Charles et al. , eds., A History of Technology (Oxford, 1954–1958), vol. 4, p. 274, vol. 5, p. 552.Google Scholar

³ A full account of the derivation of these estimates is available from the author on request, and will be published in due course in Fenoaltea, Stefano, Italian Industrial Production, 1861–1913: A Statistical Reconstruction (New York, forthcoming).Google Scholar This study identifies Italy with the area within the Kingdom's borders from 1871 to 1918. It makes no attempt to obtain regionally disaggregated estimates, even though individual time series are often constructed, or reconstructed, from place-specific data.

⁴ The calculation of real value added as output weighted by base-year value added is of course justified by its minimal data requirements rather than by its theoretical optimality; but it does not appear to be as poor a measure, next to the common alternatives, as is generally believed. See Fenoaltea, Stefano, “Real Value Added and the Measurement of Industrial Production,” Annals of Economic and Social Measurement, 5 (Winter 1976), 113–39.Google Scholar

⁵ Istituto centrale di statistica (briefly, Istat), Indagine statistica sullo sviluppo del reddiro nazionale dell' Italja dal 1861 al 1956. Annali di staristica, series 8, vol. 9 (briefly, Reddiro nazionale; Rome, 1957), PP. 210–11, 216–17, 219–20, and 224, taking the estimates within constant (present-day) borders;Google ScholarErcolani, Paolo, “Documentazione statistica di base,” and Vitali, Ornello, La stima del valore aggiunto a prezzi costanti per rami di attività,” in Lo sviluppo economico in Italia, ed. Fuà, Giorgio (Milan, 1969), vol. 3, pp. 383, 401–2, 463–78.Google Scholar

⁶ Istituto centrale di statistica, Sommario di starisriche storiche iraliane 1861–1955 (briefly, Sommario; Rome, 1958), pp. 133, 135.Google Scholar

⁷ Reddito nazionale, pp. 107–18. The account of the derivation of the gas series from 1895 to 1911 is somewhat ambiguous.Google Scholar

⁸ Fuà, Sviluppo economico, vol. 3, pp. 469–70. Before the appearance of electric power production, therefore, the Vitali series (col. 5) should vary in strict proportion to the Istat aggregate current-price series (col. 4). In the 1860s, however, it moves in strict proportion to the Istat current-price series for gas alone (col. 2).Google Scholar

⁹ Gerschenkron, Alexander, Economic Backwardness in Historical Perspective (Cambridge, Massachusetts, 1962), pp. 72–89, 367–421. The earlier indices by Dessirier, Golzio, and Tagliacame are discussed therein.Google Scholar

¹⁰ In the Sommario, pp. 133, 135, electricity production grows from 50 million kilowatt-hours in 1896 to 160 in 1900 and 1,150 in 1908, and gas production grows from 174 million cubic meters in 1896 to 194 in 1900 and 307 in 1908;Google Scholar in the Reddito nazionale, pp. 221–22, value added in 1900 equals 25 million tire in electricity production and 29 million lire in gas production.Google Scholar

¹¹ Since Gerschenkron's industry indices are combined with weights corresponding to each industry's value added in 1903, the weight here attributed to the utilities is obtained from the Reddito nazionale value added figures for 1903 as the ratio of the utilities' figure (96 million lire) to the sum of those for the utilities and the six industries in Gerschenkron's index (1,719 million lire).Google Scholar

¹² Fenoaltea, Stefano, “Public Policy and Italian Industrial Development, 1861–1913” (Ph.D. diss., Harvard University, 1967).Google Scholar The index appears in O'Brien, Patrick K., ed., Railways and the Economic Growth of Western Europe (London, 1982), pp. 55–56.Google Scholar Because of the relatively poor coverage of the manufacturing industry, the mining industry accounts for 6.4 percent of the Gerschenkron index in 1900 and 3.4 percent of it in 1913, while electricity and gas account for 3.9 percent of my index in 1900 and 11.1 percent of it in 1913. Since my index seems excessively sensitive to cyclical fluctuations, it probably overstates the growth rate actually achieved early in this century.

¹³ Mitchell, Brain R., European Historical Statistics, 1750–1970 (London, 1975), pp. 355–59;CrossRefGoogle ScholarReddito nazionale, pp. 98–99, 218; Fuà, Sviluppo economico, vol. 3, pp. 401–2, 468–69.Google Scholar

¹⁴ Feinstein, C. H., National Income, Expenditure and Output of the United Kingdom, 1855–1965, Studies in the National Income and Expenditure of the United Kingdom, vol. 6 (Cambridge, 1972), pp. 207–11, 111–12.Google Scholar

¹⁵ Feinstein, National Income, p. 210;Google ScholarHoffman, Walther G., British Industry, 1700–1950, trans. Henderson, W. O. and Chaloner, W. H. (New York, 1955), table facing p. 330;Google ScholarLewis, W. Arthur, Growth and Fluctuations, 1870–1913 (London, 1978), pp. 246–58.Google Scholar

¹⁶ Hoffman, British Industry, pp. 15–19, 289–90, table facing p. 330; Lewis, Growth and Fluctuations, pp. 252, 260 ff. Electricity does not enter the Hoffman index until after World War 1.Google Scholar

¹⁷ Lomax, K. S., “Production and Productivity Movements in the United Kingdom Since 1900,” Journal of the Royal Statistical Society, Series A, 122 (1959), 185–210.CrossRefGoogle Scholar

¹⁸ Markovitch, Tihomir J., “L'industrie française de 1789 à 1964—Sources et méthodes,” Cahiers de l' Institut de Science Economique Appliquée, series AF, no. 4 (1965), 213 ff.Google Scholar

¹⁹ Crouzet, François, “Essai de construction d'un indice annuel de la production industrielle française au XIX^e siècle,” Annales (Economies, Sociétés, Civilisations), 25 (1970), 56–99.CrossRefGoogle Scholar Annual estimates of industrial production were also presented in Lévy-Leboyer, Maurice, “La croissance économique en France au XIXe siècle. Résultats préliminaires,” Annales (Economies, Sociétés, Civilisations), 23 (1968), 788–807. The very succinct description of their derivation does not indicate whether they include series for the utilities industries.CrossRefGoogle Scholar

²⁰ Hoffman, Walther G., Das Wachstum der Deutschen Wirtschaft seit der Mitte des 19. Jahrhunderts (Berlin, 1965), pp. 63, 390–93.Google Scholar

²¹ Hoffman, Wachstum der Wirtschaft, p. 387.Google Scholar

²² Goldsmith, Raymond W., “The Economic Growth of Tsarist Russia, 1860–1913,” Economic Development and Cultural Change, 9 (1961), 454–70.CrossRefGoogle Scholar

²³ U.S. Office of Management and Budget, Standard Industrial Classification Manual (briefly, SIC; Washington, D.C., 1972).Google Scholar

²⁴ Gould, Jacob M., Output and Productivity in the Electric and Gas Utilities, 1899–1942 (New York, 1946);Google ScholarFabricant, Solomon, The Output of Manufacturing Industries, 1899–1937 (New York, 1940), pp. 334, 648, 657;Google ScholarFrickey, Edwin, Production in the United States, 1860–1914 (Cambridge, Massachusetts, 1947);Google ScholarGallman, Robert E., “Commodity Output, 1839–1899,” in National Bureau of Economic Research—Conference on Research in Income and Wealth, Trends in the American Economy in the Nineteenth Century, Studies in Income and Wealth, vol. 24 (Princeton, 1960), pp. 13–67;Google ScholarDay, Edmund E. and Thomas, Woodlief, The Growth of Manufactures, 1899 to 1923, Census Monograph 8 (Washington, D.C., 1928), p. 112.Google Scholar Gould's output estimates for the gas and electric utilities were revised and extended to 1890 by Ulmer, Melville J., Capital in Transportation, Communications, and Public Utilities: Its Formation and Financing (Princeton, 1960), pp. 455–99.Google Scholar

²⁵ Nutter, G. Warren, Growth of Industrial Production in the Soviet Union (Princeton, 1962), pp.381–84.Google Scholar

²⁶ Kendrick, John W., Productivity Trends in the United States (Princeton, 1961), pp. 204–5, 565–79.Google Scholar

²⁷ U.S. Bureau of the Census, Historical Statistics of the United States, Colonial Times to 1970, Bicentennial Edition (briefly, U.S. Historical Statistics; Washington, D.C., 1975), pp. 427, 434;Google ScholarKuznets, Simon, National Product Since 1869 (New York, 1946), pp. 202, 213.Google Scholar The water consumption figures imply an average annual growth rate of about 5 percent; but they probably underestimate the industry's growth rate, since growing urbanization and rising health standards suggest that the water was collected progressively farther from the consuming centers (or more extensively treated) and distributed through progressively more elaborate networks. The asset value figures grow surprisingly slowly (from $221 million in 1880 to $275 million in 1912); the retrospective extrapolation, based on the apparent growth from 1900 to 1904, would seem to be defective. Ulmer, Capital in Transportation, p. 471, simply reproduces Kuznets's figures.Google Scholar

²⁸ For example, Direzione generale delle gabelle, Staristica delle imposre di fabbricazione (briefly, Imposte fabbricazione), 1911/1912, pp. 132–37.Google Scholar

²⁹ Ministero di agricoltura, industria, e commercio, Notizie statistiche sugli impianti elettrici esistenti in Italia alla fine del 1898 (briefly, Statistica eleurica 1898; Rome, 1901)Google Scholar and Statistica degli impianti elettrici attivati od ampliati in italia nel decennio 1899–1908 (briefly, Statistica elettrica 1908; Rome, 1911).Google Scholar

³⁰ Mortara, Giorgio, ed., Nel cinquantenario della Società Edison (Milan, 1934), vol. 2, pp. 103, 120, 142, 321, vol. 4, p. 160;Google ScholarUngaro, Mario, “L'industria elettrica italiana,” in Ministero per la Costituente. Commissione economica, Rapporto presentato all' Assemblea costituente. II. Industria, III. Appendice alla Relazione (Rome, 1946), p. 149.Google Scholar

³¹ Mortara, , Società Edison, vol. 2, p. 177. The output and utilization estimates for 1909–1913 yield capacity figures for those years.Google Scholar

³² To allow for its greater uncertainty, the latter measure is attributed half the weight of the former.Google Scholar

³³ Mortara, , Società Edison, vol. 2, p. 344;Google ScholarMilano, Comune di, Dati statistici a corredo del resoconro dell'amministrazione comunale (briefly, Statistica Milano), 1912, p. 487;Google ScholarAzienda elettrica municipale, Milano, Conto consuntivo 1916 (Milan, 1917), pp. 6–7, 24–34;Google ScholarTorino, Comune di, La vita amministrativa del Comune di Torino nel quinquennio 1903–08 (Turin, 1909), vol. 2, p. 250–51.Google Scholar Ancillary sources include Ernesto Cianci, Dianamica del prezzi delle merci in italia dal 1870 al 1929, Istituto centrale di statistica, Annali di statistica, series 6, vol. 20 (Rome, 1933), p. 307;Google ScholarCorpo delle miniere, Rivista del servizio minerario (briefly, Rivista mineraria), 1900, p. 368; Credito italiano, Notizie statistiche sulle principali Società italiane per azioni (subsequently Credito italiano, Socierà italiane per azioni. Notizie statistiche; briefly, Notizie S.p.A.), 1912, pp. 393–454;Google ScholarVismara, Carlo, “I salari degli degli operai dalle statistiche delle Cassa Nazionale Infortuni,” Rassegna della Previdenza Sociale, 8 (1921), 85.Google Scholar

³⁴ Average capacity utilization rates in the utilities and other industries were probably quite similar, judging from the American data for 1912 in the U.S. Historical Statistics, pp. 820, 824.Google Scholar The estimate of aggregate value added in Zamagni, Vera, Industrializzazione e squilibri regionali in Italia (Bologna, 1978), p. 246 is almost identical to the Istat figure, and appears to be similarly derived from output data that include the in-house production of other industries.Google Scholar

³⁵ Mortara, , Società Edison, vol. 3, pp. 19, 207, vol. 4, p. 141.Google Scholar

³⁶ Luzzatto, Gino, L'economia italiana dal 1861 al 1914 (Milan, 1963), vol. 1, pp. 239–68.Google Scholar

³⁷ Gerschenkron, Economic Backwardness, e.g., pp. 87, 363;Google ScholarConfalonieri, Antonio, Banca e industria in Italia (Milan, 1976), vol. 3, pp. 459–60.Google Scholar

³⁸ Mortara, , Società Edison, vol. 2, pp. 98–99;Google ScholarUngaro, “Industria elettrica,” pp. 148–50.Google Scholar

³⁹ Direzione generale della statistica, Statistica del Regno d' Italia. Industria mineraria 1865 (briefly, Statistica mineraria; Milan and Florence, 1868), pp. 50–51; Rivista mineraria, e.g., 1911, p. LII.Google Scholar The 1868 figure used by Istat is the aggregate output reported for 1865, reproduced in a later secondary source (Reddito nazionale, p. 111).

⁴⁰ Rivista mineraria, 1905, p. 162; 1908, pp. 202, 206; 1910, p. 133; 1912, p. CXC; 1913, p. CLXX.Google Scholar

⁴¹ For example, Imposte fabbricazione, 1911/1912, pp. 126–31, noting the growth of tax-exempt consumption in Naples after 1910.Google Scholar

⁴² Rivista mineraria, 1911, p. LII.Google Scholar

⁴³ See the references in note 33; also Rivista mineraria, 1911, p. LIX,Google Scholar and Enciclopedia italiana di scienze, lettere, ed arti (briefly, Enciclopedia italiana; Rome, 1929 ff.), vol. 16, p. 416.Google Scholar

⁴⁴ Sommario, p. 133. The Sommario tar series includes that obtained in other industries, and is therefore not comparable to the present one.Google Scholar

⁴⁵ Reddito nazionale, pp. 111–13; Cianci, Dinamica dei prezzi, p. 307.Google Scholar The slightly higher estimate (40 million lire) in Zamagni, Industrializzazione, p. 246 is obtained as the value of output times a ratio of value added to value calculated from the British 1907 census of production (55 percent); since that ratio was plausibly lower in Italy, where the cost of materials was inflated by the cost of transporting them from England, it can be taken to confirm the present figure.Google Scholar

⁴⁶ From 1861 to 1913 the ratio of coke consumed as fuel to coke produced in the distillation process declined from approximately 50 percent to less than 25 percent. Statistica mineraria, pp. 50–51; Enciclopedia italiana, vol. 16, pp. 416.Google Scholar

⁴⁷ Fuà, Sviluppo economico, vol. 3, pp. 440–41.Google Scholar

⁴⁸ Unione statistica delle città italiane, Annuario statistico delle città italiane (subsequently Istituto nazionale di urbanistica, Annuario delle città italiane; briefly, Annuario città), 1906, pp. 92–97, 1913–1914, pp. 230–33.Google Scholar The series on taxed consumption in the Sommario, p. 134, is a good measure of use for domestic lighting and heating; it shows an acceleration between the wars, and again after World War II. In a technical sense, gas was produced for heat rather than light ever since the adoption of the Welsbach incandescent gas mantle (patented 1885). This also allowed a change in the manufacturing process to one closer to that of the coking industry and accounted for a good part of the growth in fuel efficiency noted above. Enciclopedia italiana, vol. 16, p. 414; Singer et al., History of Technology, vol. 4, p. 274.Google Scholar

⁴⁹ Direzione generale della sanità pubblica, Inchiesta sulle acque potabili nei Comuni del Regno d' Italia al 31 dicembre 1903 (briefly, Acque potabili; Rome, 1906).Google Scholar

⁵⁰ Annuario città, 1934, part 2, pp. 476–535; also, e.g., 1906, pp. 83–84, 1913–1914, p. 229. Examples of other sources are Comune di Firenze, Annuario statistico del comune di Firenze (briefly, Annuario Firenze), 1907, p. 241; Statistica Milano, 1913, p. 500;Google Scholar Direzione generale della statistica, Notizie sulle condizioni demografiche, edilizie, ed amministrative di alcune grandi città italiane ed estere nel 1891 (Rome, 1891), pp. 18, 68.Google Scholar

⁵¹ For example, Notizie S.p.A., 1912, pp. 715–22.Google Scholar

⁵² Annuario città, 1934, part 2, pp. 476–535; Acque potabili, vol. 3, e.g., pp. 195, 218, 224; Direzione generale della statistica, Bilanci comunali del Regno d' Italia (briefly, Bilanci comunali), e.g., 1899, pp. 76–83.Google Scholar

⁵³ Notizie S.p.A., 1912, p. 715; 1918, p. 1397.Google Scholar

⁵⁴ This sample consists of the suitably dated aqueducts listed on every tenth page of the Acque potabili, vol. 1, starting with p. 6; the smaller sample is the subset of these for which the source supplies all the information needed to calculate output.Google Scholar

⁵⁵ The Acque potabili, vol. 1, includes aqueducts dated 1904. A comparison of the output figures obtained for these aqueducts to the estimated values for 1904 suggests that that survey was in fact substantially complete through that year.Google Scholar

⁵⁶ The source are those cited in note 50.Google Scholar

⁵⁷ These coefficients are derived from the data in the Annuario città, 1906, pp. 83–84. On the attribution of the Acque potabili figures to 1904, see note 55.Google Scholar

⁵⁸ The coefficient g is a constant (0.542 percent per year) estimated from the data for 1872 and 1901 in Direzione generale della statistica, Censimento della popolazione del Regno d' Italia al 10 febbraio 1901 (Rome, 1904). vol. 5. p. XXVIII.Google Scholar

⁵⁹ These population series are only indices, and not estimates, of the population actually served because the aggregate figures for 1904 include the numbers served by undated aqueducts; in addition, some of those served by the aqueducts built in 1901–1904 are lost as the annual figures are shifted forward beyond 1904.Google Scholar

⁶⁰ Annuario città, e.g., 1906, pp. 83–84; 1913–1914, p. 229.Google Scholar

⁶¹ Acque potabili, vol. 2, pp. 750–51, vol. 3, pp. 370–71.Google Scholar

⁶² Royal decree 619 of November 17, 1904 and law 835 of July 21, 1911.Google Scholar

⁶³ Annuario città, 1911–1912, p. 149.Google Scholar

⁶⁴ Acque potabili, vol. 1, pp. 186–87, 250–51; Annuario città, 1913–1914, p. 185; Annuario Firenze, 1911, p. 228; Statistica Milano, 1913, p. 500.Google Scholar

⁶⁵ Fuà, , Sviluppo economico, vol. 3, pp. 440–41;Google ScholarBilanci comunali, e.g., 1899, pp. XXXVIII–XXXIX; Direzione general di statistica, Annuario statistico italiano (briefly, Anuario statistico), 1913, p. 387.Google Scholar

⁶⁶ The Marcian aqueduct, apparently built in stages, is attributed 13,500 equivalent daily ton-kilometers from mid-1870, plus 5,400 from mid-1880 and again from mid-1887, 10,500 from mid-1890 and again 5,400 (for a total of 40,200) from mid-1907; the Serino aqueduct is attributed 28,500 equivalent daily ton-kilometers from mid-1885; and the Scillato aqueduct is attributed 14,100 equivalent daily ton-kilometers from mid-1895. Acque potabili, vol. 1, pp. 256–57, 282–83;Google Scholar Società dell' Acqua Pia Antica Marcia, Prospetto generate degli acquedotti (Rome, 1941).Google Scholar

⁶⁷ Reddito nazionale, passim, and Zamagni, Industrializzazione, p.246.Google Scholar

⁶⁸ Annuario statistico, 1913, pp. 268–69; Rivista mineraria, 1911, pp. LII–LIII.Google Scholar

⁶⁹ Annuario statistico, 1911, pp. 222–26; Sommario, p. 205;Google Scholar Ufficio del censimento, Censimento degli opifici e delle imprese industriali al 10 guigno 1911 (Rome, 1914), vol. 4, pp. 476–77. The present employment figures include those reported in category 8.21 and 7 percent in category ω.62.Google Scholar

⁷⁰ Setting 1861 = 100, in 1870 the Vitali series equals 300, and the present series 141; setting 1870 = 100, in 1900 the Vitali series equals 533, and the present series 523; setting 1900 = 100, in 1913 the Vitali series equals 649, and the present series 366. The vitali series also grows less smoothly than the present one; but a recalculation of his series suggests that its sharp dip in 1895 is due to clerical or typographical error.Google Scholar

⁷¹ See note 8. While the Istat series for the water supply industry also grows rather faster than the present one, its inclusion would have reduced the growth of the Vitali series over this period from threefold to twofold.Google Scholar

⁷² As noted above, the gas production which Istat appears to have taken as a complete figure for 1868 was in fact a less than complete figure for 1865. The data in the Statistica mineraria, p. LI suggest that current-price value added in 1865 was in excess of 6 million lire, or twice the Istat estimate for that year. The Enciclopedia italiana, vol. 16, pp. 413–14 indicates that the Italian gas utilities experienced their initial rapid growth in the 1830s and 1840s.Google Scholar

⁷³ Fuà, Sviluppo economico, vol. 3, p. 470;Google ScholarSommario, pp. 133, 135, noting the rapid growth of tax-exempt gas consumption and of metallurgical coke production. The recalculation of Vitali's series suggests that his estimates for the water supply industry also grow excessively fast; since they are obtained as a percentage of those for electricity and gas together, however, this discrepancy is itself the result of the inappropriate relative weights assigned to those industries.

⁷⁴ In addition, as noted above, various elements of the local water works are assumed to grow as other such elements, or as the population served.Google Scholar

⁷⁵ Data on the length of the gas distribution networks may be found in the Annuario città, e.g., 1906, pp. 92–96, 1913–1914, pp. 230–32. These suggest that the average length per unit of output in large cities was significantly less than in smaller ones, and declining over time; however, the national average may have been stabilized by the extension of the industry to progressively smaller towns.Google Scholar

⁷⁶ Derry, T. K. and Williams, Trevor I., A Short History of Technology (Oxford, 1961), pp. 420–24.Google Scholar

⁷⁷ Derry and Williams, Short History, pp. 510–14. From 1861 to 1913, the aqueduct sector grew ninefold. The local water works grew elevenfold if the allowance for public wells and cisterns is included, and fourteenfold if it is excluded. The gas industry grew fourteenfold.Google Scholar

⁷⁸ This refinement would be warranted if old aqueducts and public wells and cisterns appeared to be as significant in Britain and Germany as in Italy. In general, it may be noted, the problem is that of estimating an unknown series from a sample of known series; as in a regression, there is no need to impose equiproportionate weights on the components of that sample, or to restrict the sample to the other members of the same industry group.Google Scholar

⁷⁹ The 1907 census value added weights underlying the Feinstein series equal £5.6 million for electricity and £17.3 million for gas. The present figures allow water supply its corresponding weight (£9.1 million). The lower figure assumes water supply grew from 1855 through 1907 to 1913 at an average annual rate of 3.74 percent, against 4.36 percent for gas; the higher figure assumes that it grew in strict proportion to gas.Google Scholar See Feinstein, National Income, p. 210;Google Scholar Board of Trade, Census of Production. Final Report on the First Census of Production of the United Kingdom (1907) (briefly, Census of Production; London, 1912), p. 831.Google Scholar

⁸⁰ The 1907 weights underlying the Hoffman series are reported only in relative terms (47.3 percent for electricity, and 52.7 percent for gas). The present figures allow the water supply industry a 1907 weight equal to 30.6 percent of the sum of electricity and gas, as suggested by the German employment data (23,200 in electricity, 39,100 in gas, and 10,300 in water supply) and the British value added per worker data (£247 in electricity, £207 in gas, and £411 in water supply). The lower figure assumes that water supply grew from 1868 through 1907 to 1913 at an average annual rate of 6.11 percent, against 7.13 percent for gas; the higher figure assumes that it grew in strict proportion to gas. If the relative weights of electricity and gas are also calculated from the German employment and British value added per worker data, the estimates of the utilities' growth from 1868 to 1913 are further reduced (to thirty-three-or thirty-eightfold). Curiously, Hoffman's weights for electricity and gas do not correspond to the employment figures which they are said to reflect. Hoffman, Wachstum der Wirtschaft, pp. 197, 336, 387–88; Reichsamt, Statistisches, Statistisches Jahrbuch für das Deutsche Reich, 1909, pp. 79, 82; Census of Production, p. 831.Google Scholar