1. See, for example, Richard Rubinson, “The World Economy and the Distribution of Income within States: A Cross-National Study,” American Sociological Review 41 (Aug. 1976):638-59.

2. Samir Amin, Accumulation on a World Scale, vol. 1 (New York: Monthly Review Press, 1974); Celso Furtado, Economic Development of Latin America (Cambridge: Cambridge University Press, 1972); and Norman Girvan, “The Development of Dependency Economics in the Caribbean and Latin America,” Social and Economic Studies 22 (1973):l-33.

3. John C. H. Fei, Gustav Ranis, and Shirley W. Y. Kuo, Growth with Equity: The Taiwan Case (Oxford: Oxford University Press, 1979).

4. See, for example, the country studies included in Trade and Employment in Developing Countries, vol. 1, Individual Studies, edited by Anne O. Krueger, Hal B. Lary, Terry Monson, and Narongchai Akrasanee (Chicago: University of Chicago Press for the National Bureau of Economic Research, 1981).

5. Christian Morrisson, “Domestic Income Distribution and the Structure of Foreign Trade,” mimeo, 1985.

6. Morrisson's independent variables included per capita income, rate of secondary school enrollment, and share of exports in GDP as well as the export structure variables cited above. The question inevitably arises as to whether his equations omitted important determinants of the level of inequality.

7. Two interesting and ambitious attempts along these lines are Irma Adelman and Sherman Robinson, Income Distribution Policy in Developing Countries (Stanford, Calif.: Stanford University Press, 1978); and Gian S. Sahota and Carlos A. Rocca, Income Distribution: Theory, Modeling, and Case Study of Brazil (Ames: Iowa State University Press, 1985).

8. Shane J. Hunt, “Growth and Guano in Nineteenth-Century Peru,” in Latin American Economies: Growth and the Export Sector, 1880-1930, edited by Roberto Cortés Conde and Shane J. Hunt (New York: Holmes and Meier, 1985), 265.

9. Rosemary Thorp and Geoffrey Bertram, Peru, 1890-1977 (New York: Columbia University Press, 1978), 321.

10. These figures were taken from the Banco Central de Reserva del Perú, Memoria 1985 (p. 114). They differ somewhat from those of the World Bank in World Tables: The Third Edition.

11. See, for example, Shail Jain, Size Distribution of Income: A Compilation of Data (Washington, D.C.: World Bank, 1975).

12. Santiago Macario, “Protectionism and Industrialization in Latin America,” Economic Bulletin for Latin America 9, no. 1 (Mar. 1964):75, t. 5.

13. Large Latin American countries are defined here as those having over five million people in 1960.

14. Based on figures taken from World Bank, World Tables, 1980 and World Tables: The Third Edition, vol. 1, Economic Data, 1984.

15. Daniel M. Schydlowsky, “The Tragedy of Lost Opportunity in Peru,” in Latin American Political Economy: Financial Crisis and Political Change, edited by Jonathan Hartlyn and Samuel A. Morley (Boulder, Colo.: Westview, 1986), 220.

16. Carlos Alberto Boloña, “Tariff Policies in Peru, 1880–1980,” Ph.D. diss., St. Anthony's College, Oxford, 1981, 134.

17. Ibid., 232.

18. For example, see Gary Fields, “Employment, Income Distribution, and Economic Growth in Seven Small Open Economies,” Economic Journal 94, no. 1 (Mar. 1984): 74–83.

19. World Bank, World Development Report 1986, 198.

20. See Thorp and Bertram, Peru, 1890-1977, 4. For a detailed treatment of trade and economic change in the nineteenth century, see also Heraclio Bonilla, Un siglo a la deriva (Lima: Instituto de Estudios Peruanos, 1980).

21. As noted by Thorp and Bertram, Peru, 1890-1977, p. 353, n. 3.

22. Jonathan V. Levin, The Export Economies: Their Pattern of Development in Historical Perspective (Cambridge, Mass.: Harvard University Press, 1960), 112–20.

23. Hunt, “Growth and Guano,” 257 Other important analyses of the impacts of the guano episode are Juan Maiguashca, “A Reinterpretation of the Guano Age, 1840–1880,” Ph.D. diss., University of Oxford, 1967; and Heraclio Bonilla, Guano y burguesía en el Perú (Lima: Instituto de Estudios Peruanos, 1974).

24. Hunt, “Growth and Guano,” 258.

25. Ibid., 270.

26. Ibid., 274.

27. Ibid., 275.

28. Ibid., 277.

29. Ibid.

30. Ibid., 304.

31. Ibid., 286.

32. Thorp and Bertram, Peru, 1890-1977, 23–25.

33. Ibid., 38–40.

34. Carlos Alberto Boloña, “Tariff Policies in Peru,” 70.

35. As of the mid-1890s, Peru's labor force numbered some 1.53 million, interpolating between the number from the 1876 census and a 1908 figure of 1.71 million based on Hunt's estimate of growth at 1.15 percent per year between 1908 and 1940. See Shane Hunt, Real Wages and Economic Growth in Peru, 1900-1940, Discussion Paper no. 25 (Boston: Boston University, Center for Latin American Development Studies, 1977), 13.

36. High labor productivity does not necessarily imply high capital intensity because total factor productivity in the export items was probably somewhat above the economywide average due to high levels of rents or efficiency. But with labor productivity this far above the national average, it would be very unlikely that the export mix was labor-intensive in the sense of its labor-capital ratio.

37. Hunt, “Growth and Guano,” 266.

38. A detailed picture of the functioning of the sugar estates is presented in Peter Klaren, Formación de las haciendas azucareras y los orígenes del APRA, 2nd ed. (Lima: Instituto de Estudios Peruanos, 1976).

39. See, for example, Thorp and Bertram, Peru, 1890-1977, 41.

40. Ibid., p. 359, n. 1.

41. Ibid., p. 359, n. 2.

42. Ibid., 41.

43. Management of the family hacienda was traditionally left to the least-able son, whose brothers would move into politics and the liberal professions in order to contribute wherever possible to that key attribute of the successful hacienda, a nearly limitless access to credit. When a financial crisis in the last half of the 1870s was followed by the physical destruction of the War of the Pacific, perhaps only a third of the sugar loans were recovered via liquidation. Various financial institutions collapsed as a result, and losses to individuals were widespread. See A. Garland, Estudio económico: la industria azucarera en el Perú (1550-1895) (Lima: Imprenta del Estado, 1895), 12.

44. Thorp and Bertram, Peru, 1890-1977, 340–41.

45. Unless Garland's 1895 figure as cited in Thorp and Bertram (p. 48) was seriously overestimated relative to Hunt's data for 1912 and subsequent years. See Hunt, Real Wages, t. 8.

46. This estimate was made by using data from Thorp and Bertram, Peru, 1890-1977, pp. 177, 340–41, and by relating the average for 1940–44 and 1935–39 to that for 1901–3.

47. Hunt, Real Wages, t. 10.

48. Thorp and Bertram, Peru, 1890-1977, 62–63.

49. Ibid., 63.

50. Heraclio Bonilla, “The New Profile of Peruvian History,” LAR 16, no. 3 (1981):220.

51. Norman Long and Bryan Roberts, Miners, Peasants, and Entrepreneurs (New York: Cambridge University Press, 1984), 1.

52. Webb estimated for 1961 that the top 1 percent in the income hierarchy received about 25 percent of the national income; he assigned nearly 80 percent of all capital income and “almost all property income” to that group. See Richard C. Webb, Government Policy and the Distribution of Income in Peru, 1963-1973 (Cambridge, Mass.: Harvard University Press, 1977), 8. The 1981 population census, no doubt understating capital incomes rather seriously, produced a figure of about 15 percent (imprecise because of the open-ended top income category); the true figure was presumably well above this level.

53. In estimating the income share of the richest percentile in 1876, three of Hunt's categories are relevant: taxpayers, 1.04 percent of the labor force with 7.1 percent of national income, using his figures (for which I suggest some adjustments herein); government employees, 0.74 percent of the labor force with 4 percent of national income; and the capital income categories not allocable, with the information available, to any group of the labor force, amounting to 7.6 percent of national income (see table 3). If the taxpayers were the top 1 percent in income level and had all of the capital income, their share would be 14.7 percent. In fact, some other persons must have had incomes greater than those of some taxpayers, putting them in that top 1 percent, including some government workers and some professionals (who are not distinguished in Hunt's calculations) as well as others. Thus on the one hand, the income share, before inclusion of capital income, would exceed the 7.1 percent accruing to the taxpayers. On the other hand, some (although perhaps not much) capital income would go to lower groups. These two biases might offset each other, neither being likely to exceed 2 percent of national income.

54. It is true that for some other groups to which the jornalero wage was applied an underestimate may have resulted. See Hunt, “Growth and Guano,” 289–90.

55. In this estimate, I have adjusted farmer incomes down by 20 percent (assuming the same average income for farmers as for other groups in Hunt's category “farmers”), tripled taxpayer income, quadrupled rural and urban rent, and quintupled returns to agricultural capital. The national income estimate is now higher by 31.2 percent (see table 4, “High Inequality”). The 1 percent who were taxpayers would then be earning 16.2 percent of national income; unallocated capital income would be 23.6 percent of total income, so if one attributes to them even half of that capital income, their share would exceed 25 percent.

56. Underreporting is assumed to be one-third for taxpayers, one-half for rental income, and two-thirds for income from agricultural capital; and the downward adjustment of farmer incomes by 20 percent is retained as in the “high inequality” case. National income would then be about the same as Hunt's estimate: the taxpaying 1 percent would have been earning 10.8 percent of income, capital income would have been 16.0 percent, and the (somewhat differently constituted) top 1 percent would be estimated to have received about 25 percent of all income.

57. See Albert Berry, “Foreign Trade and Income Distribution in Peru,” mimeo, 1986, t. 6.

58. See Instituto Nacional de Estadística, Censos Nacionales VIII de Población III de Vivienda, 12 de julio de 1981, vol. B, tomo 1 (Lima: Instituto Nacional de Estadística, 1984), p. 297.

59. See Hunt, Real Wages, 3.

60. Boloña's figures and an assumed growth of 2.5 percent over 1900–1904 imply an average of 4.1 percent over the thirty-year period. See Boloña, “Tariff Policies,” 70.

61. This estimate is based on the modern-sector employment figures presented by Hunt together with estimates of the total labor force based on his Table 3 and interpolations. See Hunt, Real Wages, t. 10, t. 3.

62. A comparison of incomes across such different settings is inevitably open to question. The behavior of modern-sector recruits suggests an income and welfare gap for a given level of skills, but its dimensions could be debated. See Hunt Real Wages, 21–22.

63. Ibid., t. 20.

64. Ibid., p. 28.

65. As of 1940 or so, wages in mining, while not high, were well above average labor earnings in the Peruvian economy as a whole (probably a little less than twice as high). Those in the sugar industry roughly equaled the average economy wide, although they were well above the blue-collar average and far above the norm for agriculture. Cotton wages were somewhat but not greatly below those of sugar. Overall, it would seem that these sectors exerted upward pressure within the labor markets where they operated. Their competitiveness was not based on paying less than going wages, although the sugar industry was notorious for its efforts to get cheap labor and its initial inability to attract Indian labor from the sierra. Use of the enganche notwithstanding, the sugar workers on the coast undoubtedly enjoyed higher levels of income in material terms than they would have obtained in the traditional sector. Whether living standards were better is less clear and partly a matter of definition. Still, it is of interest that many of those who escaped the clutches of their labor contractors chose not to return to the sierra but merely went to work for a different sugar hacienda. See Hunt, Real Wages, 21.

66. Thorp and Bertram, Peru, 1890-1977, 148.

67. The 1920s fell within the eleven-year dictatorship of Augusto Leguía, a time of grandiose economic plans, extravagant government spending, and occasional harsh repression justified in the name of economic progress. But employment in commodity production expanded little, and total employment expansion relied on the tertiary sectors, whereas its principal sources in the 1930s were to be cotton and mining.

68. Thorp and Bertram, Peru, 1890-1977, 147

69. These data refer to daily market wages and understate (sometimes greatly) the incomes of members of co-operatives after the agrarian reform.

70. Thorp and Bertram note a study by CONESTCAR (Convenio de Cooperación Técnica, Estadística y Cartográfica), Aspectos económicos an el cultivo del algodón en el Perú (Lima: Ministerio de Agricultura, 1965). It found that the wage bill made up only a third of total costs whereas in the 1930s, this ratio had been nearer 50 percent even though wages were at that time substantially lower (as were modern input costs in all probability). See Thorp and Bertram, Peru, 1890-1977, 402.

71. Figures on cotton employment are hard to interpret due to the mixture of paid workers (braceros), share-tenants (yanaconas), and others as well as the combination of seasonal and more permanent work.

72. A decline in the labor intensity of coastal agriculture is reflected in the 10 percent reduction in total agricultural employment over the intercensal period between 1961 and 1972.

73. Shane Hunt, “Distribution, Growth, and Economic Behavior in Peru,” in Government and Economic Development, edited by Gustav Ranis (New Haven, Conn.: Yale University Press, 1971), 408.

74. This interpretation is based on applying Webb's tax incidence ratios (from his Table 4.2) to his income distribution figures (his Table 2.1). See Webb, Government Policy, pp. 2, 47

75. Prior to this period, it was probably even lower. Boloña, whose ratio for the 1940s is a little higher than Hunt's, puts it at 8.6 percent for 1900–1909 and 78 percent for 1910–1919. See Boloña, “Tariff Policies,” 349.

76. This judgment was made by Hunt, using figures from Jeffrey Williamson, “Public Expenditure and Revenue: An International Comparison,” Manchester School 29, no. 1 (Jan. 1961). See Hunt, “Distribution, Growth, and Economic Behavior,” 391. Such comparisons are, of course, hazardous.

77. Hunt, “Distribution, Growth, and Economic Behavior,” 398.

78. Gillis et al. have noted that more open economies are easier to tax than less open ones. They also observed that countries having a relatively large production of oil and minerals, and hence a high share of resources in exports, possess considerably greater taxable capacity than other countries at similar levels of per capita income and openness. See Malcolm Gillis, Dwight Perkins, Michael Roemer, and Donald Snodgrass, Economics of Development (New York: Norton, 1985), 289–91. As of the late 1970s (to which their data refer), Peru appeared to have a lower than average tax ratio given its openness (measured by the ratio of exports to GDP) and its export structure. If this finding is and has been true of Peru, it would be helpful to know whether Peruvian openness has not had the generally expected positive effect on the tax share of GDP or whether other factors have outweighed that positive effect.