Hostname: page-component-7bb8b95d7b-qxsvm Total loading time: 0 Render date: 2024-09-07T13:54:36.132Z Has data issue: false hasContentIssue false
  • English
  • Français

International resource flows and construction movements in the atlantic economy: the kuznets cycle in Italy, 1861–1913

Published online by Cambridge University Press:  03 March 2009

Stefano Fenoaltea
Stefano Fenoaltea
Affiliation:
School of Social Science, The Institute for Advanced Study, Princeton
Get access Rights & Permissions [Opens in a new window]

Abstract

Nineteenth-century Italy experienced the long swings in migration, capital flows, and construction characteristic of the international Kuznets cycle, but in an unusual combination: its external migration swing may have resembled Britain's, but its capital flows and construction swing resembled America's. Construction in Italy was finance-sensitive rather than population-sensitive, and reacted primarily to exogenous shifts in the supply of foreign capital. The Italian experience suggests that changes in perceived risk altered the relative supply of capital in Britain and abroad and thereby induced the opposite swings in construction and the swing in migration.

Type
Articles
Information
The Journal of Economic History , Volume 48 , Issue 3 , September 1988 , pp. 605 - 637
Copyright
Copyright © The Economic History Association 1988

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

The author is affiliated with the School of Social Science, The Institute for Advanced Study, Princeton, NJ 08540.

For their helpful comments on earlier versions of this paper he wishes to thank Moses Abramowitz, Lee Alston, Arthur Bloomfield, Charles Calomiris, Michael Edelstein, Robert Gallman, Stephen Golub, David Gordon, Knick Harley, Robinson Hollister, Peter Lindert, Ellen Magenheim, Joel Mokyr, Hugh Rockoff, Richard Sylla, and Jeffrey Williamson; the participants at the 1985 Cliometrics Conference, the Conference on “Bilancia dei pagamenti e sviluppo economico nel caso italiano dall'Unità a oggi” (Cortona, 1987), and the Conference on “El desarrollo económico en la Europa del Sur: Italia-España, 1860–1986” (Alcalà de Henares, 1987); and the participants at seminars at Cambridge, Columbia, Harvard, Illinois-Chicago, Indiana, North Carolina State, Oxford, Rutgers, the Triangle Economic History Workshop, and the Banca d'Italia. He is also grateful to the National Science Foundation for supporting his research on Italian industrial production (Grant No. SES-7913427), to the Exxon Research and Engineering Company for supporting his tenure at the Institute for Advanced Study, and to the Banca d'Italia and the Istituto centrale di statistica for the use of their research facilities. He alone, of course, is responsible for the views expressed here.

1 See, for example, Thomas, Brinley, Migration and Urban Development: A Reappraisal of British and American Long Cycles (London, 1972), p. 4.Google Scholar

2 Habakkuk, H. J., “Fluctuations in House-Building in Britain and the United States in the Nineteenth Century,” this Journal, 22 (06 1962), pp. 213, 227Google Scholar; O'Leary, P. J. and Lewis, W. Arthur, “Secular Swings in Production and Trade, 1870–1913,” The Manchester School of Economic and Social Studies, 23 (05 1955), p. 127Google Scholar; Saul, S. B., “House Building in England, 1890–1914,” Economic History Review, 2nd ser., 15 (08. 1962), p. 136.Google Scholar

3 Abramowitz, Moses, “The Nature and Significance of Kuznets Cycles,” Economic Development and Cultural Change, 9 (04. 1961), p. 248Google Scholar, and “The Passing of the Kuznets Cycle,” Economica, n. s., no. 140 (11 1968), p. 366Google Scholar; Cairncross, Alec, Home and Foreign Investment, 1870–1913: Studies in Capital Accumulation (Cambridge, 1953), pp. 195, 209Google Scholar, and “Economic Growth and Stagnation in the United Kingdom before the First World War,” in Gersovitz, Mark et al., eds., The Theory and Experience of Economic Development: Essays in Honor of Sir W. Arthur Lewis (London, 1982), p. 293Google Scholar; Kelley, Allen C., “Demographic Change and Economic Growth: Australia, 1861–1911,” Explorations in Entrepreneurial History, 2nd ser., 5 (Spring/Summer 1968), pp. 207–77Google Scholar; Thomas, Brinley, Migration and Economic Growth: A Study of Great Britain and the Atlantic Economy (Cambridge, 1954), pp. 174–75Google Scholar, and Migration and Urban Development, pp. 24–41Google Scholar; Williamson, Jeffrey G., American Growth and the Balance of Payments, 1820–1913: A Study of the Long Swing (Chapel Hill, 1964), p. 11Google Scholar; Kuznets, Simon S., Capital in the American Economy: Its Formation and Financing (Princeton, 1961), p. 327Google Scholarff. See also Lewis, J. Parry, “Growth and Inverse Cycles: A Two-Country Model,” Economic Journal, 74 (03 1964), pp. 109–18CrossRefGoogle Scholar. Some authors stress the role of primary product prices and the terms of terade in determining the long swing in migration; see the early analysis in Cairncross, , Home and Foreign Investment, pp. 189–195Google Scholar, the evidence in Lewis, W. Arthur, Growth and Fluctuations, 1870–1913 (London, 1978), p. 59Google Scholar, and especially Harley, C. Knick, “Transportation, the World Wheat Trade, and the Kuznets Cycle, 1850–1913,” Explorations in Economic History, 17 (07 1980), pp. 218–50.CrossRefGoogle Scholar

4 These series are derived in Fenoaltea, Stefano, “Public Works Construction in Italy, 1861–1913,” Rivista di storia economica, 3 (International issue, 1986), pp. 1–33Google Scholar, and “Construction in Italy, 1861–1913,” Rivista di storia economica, 4 (International issue, 1987), pp. 21–53Google Scholar; see also “Railway Construction in Italy, 1861–1913,” Rivista di storia economica, 1 (International issue, 1984), pp. 27–58.Google Scholar

5 While the spectrum of growth rates across cities is always very wide, the entire spectrum moves as the index does; see Fenoaltea, “Construction in Italy,” table 1.

6 Fenoaltea, Stefano, “Public Works Construction in Italy, 1861–1913,” Rivista di storia economica, 3 (International issue, 1986), pp. 26–27 and fn. 14.Google Scholar

7 Istituto centrale di statistica, Sommario di statistiche storiche italiane, 1861–1955 [henceforth Sommario] (Rome, 1957), p. 65.Google Scholar

8 Bloomfield, Arthur I., Patterns of Fluctuation in International Investment Before 1914 (Princeton, 1968), p. 32Google Scholar fn., asserts that “Italian emigration shows clear long-swing movements that tended to have a positive relation to the swings in the domestic variables but to be even more closely related to and influenced by the Kuznets cycle in the United States.” The domestic variables are railway track added and industrial production, as measured in my own early work, while emigration is measured by the series at hand (pp. 32, 49). In a similar vein, Lauricella, Francesco, “Emigrazione italiana di massa in Argentina e in Brasile e ciclo agricolo (1876–1896),” in Assante, Franca, ed., Il movimento migratorio italiano dall'Unità nazionale ai giorni nostri (Geneva, 1978), vol. 2, p. 372, plots an 11-year moving average of the ab'Fute changes in gross overseas migration; his series displays a long swing with high levels in the 1880s and 1896–1909, with an intervening trough in the early 1890s (and another very sharp one after 1910, as the war years get averaged in). His interpretation stresses the push of declining grain prices in the 1880s, and the pull of overseas labor markets in the later surge (pp. 379, 382–383).Google Scholar

9 Sommario, p. 5.Google Scholar

10 Coletti, Francesco, “Dell'emigrazione italiana,” in dei Lincei, R. Accademia, Cinquanra anni di storia italiana (Milan, 1911), vol. 3, pp. 10–11.Google Scholar

11 Gould, J. D., “European Inter-Continental Emigration. The Road Home: Return Migration from the USA,” Journal of European Economic History, 9 (Spring 1980), pp. 88–90Google Scholar; see also Sori, Ercole, L'emigrazione italiana dall'Unità alla seconda guerra mondiale (Bologna, 1979), p. 55.Google Scholar

12 Gould, , “European Inter-Continental Emigration,” pp. 79, 84–85; Sommario, pp. 65, 67Google Scholar. The transatlantic return flow figures, compiled by the Commissariato dell'emigrazione, are not comparable to the outflow data; see Coletti, , “Dell'emigrazione,” pp. 14–15, 38, 74.Google Scholar

13 Bacci, Massimo Livi, “I fattori demografici dello sviluppo economico,” in Fuà, Giorgio, ed., Lo sviluppo economico in Italia (2nd edn, Milan, 1975), vol. 2, p. 21.Google Scholar

14 Gould, , “European Inter-Continental Emigration,” p. 82.Google Scholar

15 Giusti, Franco, “Bilanci demografici della popolazione italiana dal 1861 al 1961,” in Istituto centrale di statistica, Sviluppo della popolazione italiana dal 1861 al 1961. Annali di statistica, series 8, vol. 17 (Rome, 1965), p. 116.Google Scholar

16 Gould, , “European Inter-Continental Emigration,” pp. 79–87, investigates and explains the differences between the Livi Bacci and Giusti figures. While not unaware of the weaknesses of the latter, particularly as an annual series, he accepts Giusti's implication that “the Italian net population loss through emigration peaked in the 1890s” (p. 87).Google Scholar

17 As indicated in the Appendix, the impact of conditions in the receiving areas on Italy's net emigration is clearly apparent in the available figures for transoceanic repatriation after the turn of the century.

18 A long swing similarly appears in the real per capita stock of private buildings. That stock is indexed, with a short lag, by the estimates of the corresponding maintenance at 1911 prices; the maintenance figures for the six population benchmark years yield successive decadal increases in the per capita stock equal to –0.4, 1.4, 2.1, 0.8, and 5.1 percent. See Fenoaltea, , “Construction in Italy,” table 4 and pp. 34, 37–39.Google Scholar

19 Somogyi, Stefano, “Evoluzione della popolazione attraverso il tempo,” in Istituto centrale di statistica, Sviluppo della popolazione italiana, p. 55.Google Scholar

20 The dispersed rural population grew from 6.3 million in 1861 to 9.2 million in 1901 and 9.9 million in 1911; the agricultural population, from 13.5 million in 1861 to 17.3 million in 1901, declining to 17.0 million in 1911. See Fenoaltea, “Construction in Italy,” table 3.

21 Somogyi, Stefano, “Nuzialità,” in Istituto centrale di statistica, Sviluppo della popolazione italiana, p. 375. The geographic basis of the early figures is not clear: contrast Somogyi, “Nuzialità,” p. 320, and Sommario, pp. 4, 43. The rate figures are based on the population estimates which are challenged above; but for these purposes the errors should be minor.Google Scholar

22 Somogyi, , “Nuzialità,” pp. 322–24.Google Scholar

23 The correlation between the reported number of marriages (table 2, column 3) and the level of construction (table 1, column5) falls to +0.724 if it is calculated over the full period from 1861 to 1913, that is, without shedding the defective marriage figures for the early period.

24 The relation between marriages and household formation in Italy is in any case moot, as it was common practice to marry before emigrating, leaving the bride at home; see Gould, , “European Inter-Continental Emigration,” p. 71Google Scholar. The emigration-related marriages would not result in the formation of a new household in Italy unless and until the migrant returned home to stay; they could nonetheless be correlated with domestic construction because the latter was itself correlated with overseas construction and transoceanic migration. Marriages were also related to return migration, and the 1908 marriage peak is attributed precisely to the large number of repatriants from the New World in that year; see Sori, , L'emigrazione italiana, pp. 189, 195.Google Scholar

25 See Williamson, , American Growth, p. 11Thomas, , Migration and Urban Development, pp. 77 ff., 124Google Scholar; Edelstein, Michael, Overseas Investment in the Age of High Imperialism: The United Kingdom, 1850–1914 (New York, 1982Google Scholar). Williamson sees the long swing in capital flows as induced by, and in turn permitting, that in infrastructure growth; Thomas, in this second book, suggests that the actions of the Bank of England generated the long swing in capital flows and thereby the common construction cycle in the recipients of British capital and European migrants. The British capital export series in table 4, column 1 is from Edelstein, , Overseas Investment, pp. 313–14.Google Scholar

26 This association also is noted in Bloomfield, , Patterns of Fluctuation, p. 32.Google Scholar

27 Istituto centrale di statistica, Indagine statistica sullo sviluppo del reddito nazionale dell'Italia dal 1861 al 1956. Annali di staistica, series 8, vol. 9 [henceforth Reddito nazionale] (Rome, 1957), p. 253.Google Scholar

28 Gould, J. D., “European Inter-Continental Emigration. The Road Home: Return Migration from the USA,” Journal of European Economic History, 9, (Spring 1980), pp. 159–177, 253–59, whence the following discussion is drawn.Google Scholar

29 Stringher, Bonaldo, “Gli scambi con l'estero e la bilancia dei pagamenti italiana,” in dci Lincei, R. Accademia, Cinquanta anni di storia italiana, vol. 3, pp. 120–22Google Scholar. Stringher was then head of the Banca d'Italia. His estimate of earnings from tourists is under 500 million lire per year (Stringher, Bonaldo, “Gli scambi con l'estero e la bilancia dei pagamenti italiana,” in dci Lincei, R. Accademia, Cinquanta anni di storia italiana, vol. 3, pp. 122–25)Google Scholar, again below the figure entering the balance of payments series. An estimate of remittances similar to Stringher's is provided in Coletti, , “Dell'emigrazione,” pp. 238–44.Google Scholar

30 Reddito nazionale, p. 175Google Scholar. Balletta, Francesco, “Emigrazione italiana, cicli economici e rimesse (1876–1976),” in Ro'Fi, Gianfausto, ed., Un secolo di emigrazione italiana, 1876–1976 (Rome, 1978), p. 81, provides a graph illustrating annual gross emigration and deflated remittances (as per the official balance of payments series; p. 66). Until 1940 these move very much together, confirming the indication in the Reddito nazionale. In later years, when the series were obtained independently, their movements appear uncorrelated.Google Scholar

31 Coletti, , “Dell'7;emigrazione,” pp. 240–42; Stringher, “Gli scambi,” p. 121; Gould, “European Inter-Continental Emigration,” pp. 55–56. These considerations further suggest that the balance of payments series'7; allocation of total remittances between temporary and permanent migrants is grossly biased toward the former, particularly if “temporary” is identified with “seasonal.”Google Scholar

32 Coletti, , “Dell'7;emigrazione,” p. 80Google Scholar. These figures include those temporarily abroad as well as long-term migrants, and presumably include migrants‘ children even if born abroad. One notes the deceleration in the growth of this stock in the 1890s, and its acceleration in the 1900s; but these totals are not precisely dated, as they are collations from a variety of non-contemporaneous sources. See also Direzione, generale della statistica, Annuario statistico italiano, 1912, pp. 31–32.Google Scholar

33 Williamson, , American Growth.Google Scholar

34 In the absence of evidence sufficient fully to identify the relevant demand and supply curves, the analysis here seeks to identify only the dominant autonomous shifts. For example, an increase in price and quantity together establishes that demand increased; supply may have moved in either direction, but if it increased at all its shift was less than that in demand.

35 Fratianni, Michele and Spinelli, Franco, “Currency Competition, Fiscal Policy and the Money Supply Process in Italy from Unification to World War I,” Journal of European Economic History, 14 (Winter 1985), pp. 473–99.Google Scholar

36 Mundell, Robert A., International Economics (New York, 1968), chap. 18.Google Scholar

37 This series appears in the Sommario, p. 166.Google Scholar

38 For the reasons given above, the path of capital imports (up to a trend correction) seems best represented by that of the merchandise deficit; for simplicity, the present discussion ignores the differences between them.

39 With the nominal exchange at par, the equilibrating variations in the real exchange rate occur through changes in the price level; and since the prices of tradable goods are constrained by their international values, the variations will be concentrated in the prices of nontradable goods and services. The evidence on the latter provided by the wage and rent series in Fenoaltea, “Public Works Construction,” table 4, “Construction in Italy,” table 2, points to a strong swing in step with that in construction and the trade deficit; the rent swing would presumably have been even stronger but for the small parallel swing in the real per-capita stock of buildings noted in fn. 18.

40 While capital imports and the exchange rate were both higher in the mid-1870s than in the neighboring years, the trough in capital imports in 1871 was also associated with a temporarily high exchange rate. That appears to be the one clear case in which the exchange and capital flows were dominated by the market for goods rather than by that for financial assets, as exports surged in response to the Franco-Prussian war; see Stringher, , “Gli scambi,” p. 16.Google Scholar

41 This discussion assumes that capital is rationed by the interest rate. A more complete view would recognize that access to capital is rationed directly as well as by price; the evidence on crowding out thus provides information additional to that provided by the interest rate.

42 Italian con'F rates are reported in Bianchi, Bruno, “Appendice statistica: il rendimento del con'Fidato dal 1862 al 1946,” in Vicarelli, Fausto, ed., Capitale industriale e capitale finanziario (Bologna, 1979), pp. 156–58Google Scholar; they are here extended to 1861 with the aid of the comparable series in Pozzo, Mario Da and Felloni, Giuseppe, La borsa valori di Genova nel secolo XIX (Turin, 1964), p. 469.Google Scholar

43 This discussion ignores the war-induced blip in capital flows in 1871; see fn. 40.

44 The change in price expectations is plausible not only because prices stopped falling and started rising, but because that change could be associated with the gold discoveries at the end of the nineteenth century. Price expectations during the inflation of the early 1870s are more difficult to pin down, as the price level may then have been expected to revert to its “natural” level; but some evidence that the real interest rate moved inversely to capital imports and construction before the turn of the century as well as after it is provided by the Genoa stock price index calculated in Pozzo, Da and Felloni, , La borsa valori, pp. 499–508. That index, equal to 34 in 1861, exhibits peaks in 1863 (46), 1873 (93), and 1887 (98), and troughs in 1867 (33), 1877 (72), and 1896(44, the terminal value of the series).Google Scholar

45 See Fratianni, and Spinelli, , “Currency Competition,” pp. 490–91; Fenoaltea, “Railway Construction,” table 2, “Public Works Construction,” table 3.Google Scholar

46 See Fenoaltea, Stefano, “Decollo, ciclo, e intervento dello Stato,” in Caracciolo, Alberto, ed., La formazione dell'7;Italia industriale (Bari, 1969), pp. 95–114Google Scholar, and “Riflessioni sull'7;esperienza industriale italiana dal Risorgimento alla prima guerra mondiale,” in Toniolo, Gianni, ed., Lo sviluppo economico italiano 1861–1940 (Bari, 1973), pp. 121–56Google Scholar; Warglien, Massimo, “Nota sull'7;investimento industriale in macchinari e altre attrezzature meccaniche: Italia, 1881–1913,” Rivista di storia economica, n.s., 2 (02 1985), pp. 125–46Google Scholar; Romeo, Rosario, Risorgimento e capitalismo (Bari, 1959)Google Scholar, and Breve storia della grande industria in Italia (Bologna, 1961)Google Scholar. Romeo thought that investment was constrained by domestic saving, and oddly attributed the industrial expansion from the 1880s to a putative decline in infrastructure investment. For a brief presentation and evaluation of the Romeo hypothesis, see Fenoaltea, Stefano, “Italy,” in O'Brien, Patrick K., ed., Railways and the Economic Development of Western Europe (London, 1983), pp. 54–60.Google Scholar

47 Pedone, Antonio, “Il bilancio dello Stato e lo sviluppo economico italiano: 1861–1963,” Rassegna economica, 31 (03–04 1967), pp. 285–341.Google Scholar

48 On stock price movements see fn. 44.

49 Had the swings in construction been set in motion by population growth or household formation, they would have put upward pressure on interest rates and crowded out other domestic borrowing and capital formation. As noted in fn. 34, the evidence reviewed here reveals only the dominant shifts in the underlying schedules; it thus supports the exact range of hypotheses advanced in the penultimate paragraph of the section on construction and population growth.

50 An autonomous increase in the domestic demand for bonds (supply of capital) would similarly put downward pressure on domestic interest rates and lead to increased domestic borrowing and investment across the board; but it would lead as well to capital exports or reduced imports, and thus to a negative correlation of capital imports with domestic capital formation.

51 Bloomfield, , Patterns of Fluctuation, pp. 8, 47Google Scholar; Fishlow, Albert, “Lessons from the Past: Capital Markets during the 19th Century and the Interwar Period.” International Organization, 39 (Summer 1985), pp. 388–89, 393–95.CrossRefGoogle Scholar

52 Since the capital-flow variables do not perform better if lagged, one may surmise that the interest rate affected the decision to build, while quantitative restrictions influenced construction as the project proceeded. The nominal con'F rate is of course a proxy both for other nominal (mortgage) rates, and for the relevant real rates.

53 The British con'F yield is taken as reported in Harley, C. Knick, “The Interest Rate and Prices in Britain, 1873–1913: A Study of the Gibson Paradox,” Explorations in Economic History, 14 (01 1977), p. 87CrossRefGoogle Scholar, for 1873 ff., and extended to 1861 in proportion to the comparable series in Homer, Sidney, A History of Interest Rates (New Brunswick, 1963), p. 196.Google Scholar

54 Edelstein, , Overseas InvestmentGoogle Scholar, argues that the swing in British capital exports resulted from the interaction of a relatively steady rate of accumulation by British savers and an autonomous long swing in domestic investment opportunities (tied to internal demographic and productivity movements) in Britain (p. 232). If British capital exports were thus essentially a residual, they should be high when low domestic investment made for low interest rates; but that seems to be the opposite of what actually occurred, at least from the 1880s. The correlation of British capital exports and the British con'F yield from 1880 to 1913 is +0.680, and the cyclical paths of the real and nominai rates seem to have been quite similar (Harley, , “The Interest Rate,” p. 85)Google Scholar. See also Richardson, H. W., “British Emigration and Overseas Investment, 1870–1914,” Economic History Review, 2nd ser., 25 (02 1972), pp. 102, 106.Google Scholar

55 If the British yield is risk-free, and the only risk attached to the Italian yield is a probability p of permanent default, then the British yield will equal (1 – p) times the Italian yield, and the relative yield premium calculated here is exactly the perceived probability of default p. In actuality, of course, the perceived risks include those of currency depreciation and changes in tax treatment as well as of the simple suspension of interest payments.

56 In light of this, Fratianni and Spinelli'7;s claim that “net capital inflows … were responding to a high interest-rate differential” (“Currency Competition,” p. 483) and their contempt for contemporary economists'7; stress on investors'7; attitudes (Ibid.) both seem unwarranted.

57 The corresponding F statistic equals just 0.58.

58 Harley, , “The Interest Rate,” p. 85, suggests that the real con'F rate in Britain rose from the mid-1870s to the late 1880s, in opposition to the movement of the nominal rate; the negative coefficient on the British yield in equation 5 is thus not particularly troublesome. In the Italian case, deflationary price expectations in the 1890s may have temporarily raised the real rate above the nominal rate.Google Scholar

59 The value of the F statistic comparing the sum of squared residuals in equations 6 and 9 together to that in equation 3 is just 0.25. To return to an earlier topic, the regressions in equations 1 through 9 were repeated with the addition of the number of marriages (Table 2, column 3). Whether introduced contemporaneously or with a one-year lag, this variable never comes close to having a significantly positive impact on construction; even in the later subperiod, when the marriage data seem generally accurate, its t statistic varies from –0.1 to –1.6.

60 The sharp blip in the yield of the con'Fidato in 1894 similarly seems tied to the nervousness caused by major bank failures in Italy. Italy'7;s exchange rate may also be a measure of country risk, considered lower when the exchange was at par than when it was floating below par; since the first suspension of convertibility was no doubt that of greatest symbolic significance, it stands to reason that the exchange rate seems to matter more in the earlier decades.

61 Edelstein, , Overseas InvestmentGoogle Scholar, unpublished appendix 7. I am grateful to the author for making this material available to me. “The extreme discredit which overtook foreign investment after the … crises of the early nineties” is noted, for example, by Habakkuk, H. J., “Fluctuations in House-Building,” p. 226Google Scholar. The distinction between changes in attitudes and changes in objective conditions at that time is stressed by Lewis, , Growth and Fluctuations, p. 180: “There was a large bandwagon effect in foreign lending… With lending to the USA, Australia and Argentina out of fashion, all other countries would find themselves deprived of loans beyond any point that objective economic analysis of their own economic 'Futions could justify.”Google Scholar

62 See Pedone, , “Il bilancio dello Stato,” p. 333, and fn. 44. Perhaps not coincidentally, the decline in construction over the later 1870s was concentrated in railways and private buildings.Google Scholar

63 Thomas, , Migration and Urban Development, pp. 20–58.Google Scholar

64 Easterlin, Richard A., Population, Labor Force, and Long Swings in Economic Growth: The American Experience (New York, 1968), p. 31.Google Scholar

65 Feinstein, Charles 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. 192, T88–89.Google Scholar

66 Giusti, , “Bilanci demografici,” p. 116.Google Scholar

67 The seasonal pattern of departures is documented in Coletti, , “Dell'7;emigrazione,” p. 70.Google Scholar

68 Sommario, p. 65.Google Scholar

69 The correction factors are 1.30 in 1872–81 (somewhat higher than in later years, as suggested by Sori, , L'7;emigrazione italiana, p. 54), 1.25 in 1882–1901 (as suggested by Gould'7;s figures cited in the text; the discrepancy includes people leaving without passports, and people leaving for Old World destinations but then moving overseas), and 0.92 in 1902–1911 (again from Gould'7;s figures; the discrepancy reflects people obtaining passports but not departing).Google Scholar

70 Sori, , L'7;emigrazione italiana, p. 337ff.Google Scholar

71 See, for example, the spectacular rise in return migration from the United States in 1908, Sommario, p. 67; and Argentina in 1890, Sori, L'7;emigrazione italiana, p. 56.Google Scholar

72 Gould, , “European Inter-Continental Emigration.” p. 87, asserts that the repatriant ratio was “probably higher in the '7;80s than in the '7;90s.” Since he gives no reason as to why this should be expected a priori, the most likely basis for his claim is simply the pattern thrown up by Giusti'7;s estimates.Google Scholar

73 The present calculations accept Giusti'7;s estimate of the year-end population in 1901, even though it is no more reliable than his corresponding figure for 1911; given the lack of data on repatriants, and the discontinuity in the gross emigration series, its bias is not easy to evaluate.

74 Sommario, p. 65; Coletti, , “Dell'7;emigrazione,” p. 74.Google Scholar