The research reported in this paper was financed by a grant from the National Science Foundation (GS 1837) and the University of Wisconsin Graduate School. It has benefited from the helpful comments of Russell J. Cheetham, Colin Clark, Jack Dowie, John McCarty, Alan Powell, Henry Rosovsky, R. P. Sinha, Joseph A. Swanson, Kozo Yamamura, our colleagues at the University of Wisconsin, and the participants at the Cliometrics Conference held at Madison, Wisconsin (April 29-May 1,1971).

¹ See, for example: Jerry Green, “The Effect of the Iron Tariff, in the United States 1847–1859: The Estimation of a General Equilibrium System with Non-Traded Goods;” Peter Passell and Maria Schmundt, “Pre-Civil War Land Policy and the Growth of Manufacturing;” and Clayne Pope, “The Effect of the Ante-Bellum Tariff on Income Distribution,” all presented to the Cliometric Conference, Madison, Wisconsin (1970).

² Much of the following paragraph is taken from the excellent summary in Hayami, Yujiro and Yamada, Saburo, “Agricultural Productivity at the Beginning of Industrialization,” in Ohkawa, K., Johnston, Bruce F., and Kaneda, H., editors, Agriculture and Economic Growth: Japan's Experience (Tokyo: University of Tokyo Press, 1969), pp. 105–135.Google Scholar

³ Rosovsky, Henry, “Rumbles in the Ricefields: Professor Nakamura vs. the Official Statistics,” The Journal of Asian Studies, XXVII (Feb. 1968), 360.Google Scholar

⁴ Nakamura, James I., “Growth of Japanese Agriculture, 1875–1920,” in Lockwood, W. W., editor, The State and Economic Enterprise in. Japan (Princeton: Princeton University Press, 1965),Google Scholar ch. vi. See also Nakamura, James I., Agricultural Production and the Economic Enterprise in Japan (Princeton: Princeton University Press, 1966).Google Scholar

⁵ Ibid., (1966), p. 12.

⁶ Harry T. Oshima, “Meiji Fiscal Policy and Agricultural Progress,” in Lockwood, editor, The State and Economic Enterprise in Japan, p. 356.

⁷ The most recent exchange appears in Rosovsky, “Rumbles in the Ricefields,” Nakamura, James I., “The Nakamura versus the LTES Estimates of the Growth Rate of Agricultural Production,” Keizai Kenkyu, XIX (Oct. 1968), 358–61Google Scholar, and Hayami and Yamada, “Agricultural Productivity.” In the present paper, we do not have space to review and evaluate the debate. The interested reader may wish to consult Kelley, Allen C., Williamson, Jeffrey G., and Cheetham, Russell J., Economic Dualism in Theory and History (Chicago: University of Chicago Press, 1972Google Scholar), ch. v.

⁸ Ohkawa and Rosovsky acknowledge the direction of the bias, but not its magnitude. “We are quite prepared to concede that there may have been some underestimation in the early government statistics, but we do not accept either Nakamura's assumptions or the highly inflated output figures which result …. Our reading of the evidence (indicates that) official yield figures perhaps contain a slight downward bias (we would suggest in the magnitude of 10–12%), but they are far closer to the mark than the alternative proposed by Nakamura.” Ohkawa and Rosovsky also note that the high growth rates are not “impossible” as often implied. Goldsmith's calculations on Tsarist Russia between 1860 and 1913 showed 1.75–2.0 percent per year; post-war growth in the ECAFE region was above 2 percent. K. Ohkawa and Henry Rosovksy, “A Century of Japanese Growth,” in Lockwood, editor, The State and Economic Enterprise in Japan, ch. ii.

Nakamura is pessimistic regarding a satisfactory resolution of the controversy: “… it is not likely that the final word on the controversy over the growth rate of agricultural production will ever be written due to data deficiency.” James I. Nakamura, “The Nakamura versus the LTES Estimates,” pp. 358–61.

⁹ Nakamura, Agricultural Production and the Economic Development of Japan, pp. 143–49 and 170–72.

¹⁰ Both Rosovsky and Ohkawa have been very careful to present important qualifications to their investment and capital stock estimates. See Ohkawa, K., The Growth Rate of the Japanese Economy Since 1878 (Tokyo: Kinokuniya Bookstore Co., 1957Google Scholar); Ohkawa, K., Sinohara, M., and Umemura, M., Estimates of Long Term Economic Statistics of Japan Since 1868 (Tokyo: Toyo Keizai Shinpo Sha, 1966Google Scholar); and Rosovsky, Henry, Capital Formation in Japan 1868–1940 (Glencoe: The Free Press, 1961Google Scholar). Another estimate of capital stock has been compiled in Fei, John C. and Ranis, Gustav, Development of the Labor Surplus Economy (Homewood: Richard D. Irwin, 1964), pp. 125–31Google Scholar and pp. 146–48. A crucial assumption in their calculation is that the capital stock depreciates at an annual rate of 15 to 20 percent. Their tests relating to the historical analysis of the labor reallocation process uses capital stock estimates based on the 20 percent rate. We feel that the historical record does not justify such high rates of capital consumption. We thus reject the Fei-Ranis estimates and their resulting empirical analysis. A critique of the Fei-Ranis analysis and estimates is provided by Reubens, E., “Capital-Labor Ratios in Theory and in. History: Comment,” American Economic Review, LIV (Dec. 1964), 1052–62Google Scholar; see also Minami, Royshin, “The Turning Point in the Japanese Economy,” Quarterly Journal of Economics, LXXXII (Aug. 1968), 397–98Google Scholar. We have also had the benefit of yet unpublished estimates of capital stock compiled by Colin Clark.

¹¹ Lewis, William Arthur, “Development with Unlimited Supplies of Labour,” Manchester School of Economics and Social Studies, XX (May 1954), 139–92CrossRefGoogle Scholar; Fei, and Ranis, , Development of the Labor Surplus Economy: Theory and Policy (Homewood, Illinois: Richard D. Irwin, 1964Google Scholar); and Jorgenson, D. W., “The Development of a Dual Economy,” Economic Journal, LXXI (June 1961), 309–34CrossRefGoogle Scholar.

¹² For Japan consult Kaneda, H., “Long-Term Changes in Food Consumption Patterns in Japan, 1878–1964,” Food Research Institute Studies, VIII, 1 (1968Google Scholar).

¹³ Kelley, Allen C., “Demand Patterns, Demographic Change and Economic Growth,” Quarterly Journal of Economics, LXXXIII (Feb. 1969), 110–26CrossRefGoogle Scholar.

¹⁴ Yujiro Hayami and V. Ruttan, “Factor Prices and Technical Change in Agricultural Development: The United States and Japan, 1880–1960,” Staff Paper 69–19, Department of Agricultural Economics, University of Minnesota .(July 1969); Watanabe, Tsunehiko, “Industrialization, Technological Progress, and Dual Structure,” in Klein, Lawrence and Ohkawa, K., Economic Growth, the Japanese Experience Since the Meiji Era (Homewood: Richard D. Irwin, 1968)Google Scholar, ch. iv; and Williamson, Jeffrey G., “Dimensions of Philippine Post-War Economic Progress,” Quarterly Journal of Economics, LXXXIII (Feb. 1969), 93–109.CrossRefGoogle Scholar

¹⁵ Nakamura, “Growth of Japanese Agriculture.”

¹⁶ Our decision to omit land from production is based on several considerations. First, we know of no satisfactory theory of land expansion which commands considerable empirical support. Assuming a fixed supply of land merely guarantees diminishing returns, an hypothesis which could be obtained by assumption. Second, while demand and savings could be affected by differential behavior of property income recipients, again the empirical literature to guide our specification is meagre. See Kelley, Allen C. and Williamson, Jeffrey G., “Household Savings Behavior in the Developing Economies: The Indonesian Case,” Economic Development and Cultural Change, XVI (April 1968), 385–403CrossRefGoogle Scholar. A much more detailed defense of our treatment of land is found in Kelley, Williamson and Cheetham, Economic Dualism in Theory and History, ch. ii.

¹⁷ Parks, Richard W., “Systems of Demand Equations: An Empirical Comparison of Alternative Functional Forms,” Econometrica, XXXVII (Oct. 1969), 629–50CrossRefGoogle Scholar; and Yoshihara, Kuniq, “Demand Functions: An Application to the Japanese Expenditure Pattern,” Econometrica, XXXVII (April 1969), 257–74CrossRefGoogle Scholar.

¹⁸ We emphasize traditional since the new economic historian has almost exclusively utilized comparative static analysis of formal general equilibrium models.

¹⁹ Kelley, Williamson, and Cheetham, Economic Dualism in Theory and History, ch. iii.

²⁰ Kuznets, S., “Quantitative Aspects of the Economic Growth of Nations: II. Industrial Distribution of National Product and Labor Force,” Economic Development and Cultural Change, V (July 1957Google Scholar), suppl.; and Chenery, Hollis B., “Patterns of Industrial Growth,” American Economic Review, L (Sept. 1960), 624–54Google Scholar.

²¹ This theme is also developed by Nakamura. “The Japanese experience, however, differed from that of the Western economies in many important respects, at least in degree where not in kind. Growth was assured within twenty years after the Meiji Restoration…. Her development from a feudal to an industrial economy of the 193O's was crowded into a shorter span of years.” (Emphasis ours.) Nakamura, Agricultural Production and the Economic Development of Japan, p. 21.

²² Chenery, Hollis B. and Taylor, Lance, “Development Patterns: Among Countries and Over Time,” The Review of Economics and Statistics, L (Nov. 1968), 393–403Google Scholar.

²³ The countries included in the average are Burma, Cambodia, Thailand, Taiwan, South Korea, Philippines, Malaya, and Japan. Ibid., p. 414.

²⁴ Ibid., and Kuznets, S., Modern Economic Growth (New Haven: Yale University Press, 1966)Google Scholar.

²⁵ Ohkawa, K. and Rosovsky, Henry, “The Role of Agriculture in Modern Japanese Economic Development,” Economic Development ana Cultural Change, II (Oct. 1960), 43CrossRefGoogle Scholar.

²⁶ Ohkawa, The Growth Rate of the Japanese Economy Since 1878; and Taeuber, Irene, The Population of Japan (Princeton: Princeton University Press, 1958Google Scholar).

²⁷ Kelley, Williamson, and Cheetham, “Biased Technological Progress and Labor Force Growth in a Dualistic Economy,” Quarterly Journal of Economics (forthcoming); and Kelley, Williamson and Cheetham, Economic Dualism in Theory and History, ch. vi.

²⁸ Note that n = n₁u + n₂ (l − u). Taking n = .009, u = .3, and the dualistic assumption that n₂ > n₁, and in particular, n₂/n₁ = 1.5, the resulting values of n₁ and n₂ are .0067 and .01, respectively.

²⁹ While this experiment could be carried out by resimulating the model, a simpler method which yields roughly equivalent results involves the use of structural elasticities already estimated for the Southeast Asian prototype. See Kelley, Williamson, and Cheetham, Economic Dualism in Theory and History, chs. v and vi.

³⁰ Ohkawa, The Growth Rate of the Japanese Economy Since 1878, pp. 26–7. Nakamura argues that the Ohkawa labor force series overstate the percentage of the work force in the agricultural sector. (See Nakamura, Agricultural Production, p. 147.) For 1885 two separate estimates by Nakamura yield 71.1 and 72.1 percent. For the same period, the Ohkawa estimates yield 79.2 percent.

³¹ In Kuznets' framework, MEG is characterized by four conditions: (1) the application of modern scientific thought and technology to industry, (2) a sustained and rapid increase in real product per capita, usually accompanied by high rates of population growth, (3) rapid rates of transformation of the industrial structure (changing sectoral output, labor force and entrepreneurial distributions), and (4) the emergence of or expansion in international contacts. Kuznets, S., Six Lectures on Economic Growth (Glencoe: The Free Press, 1959Google Scholar), Lecture 1.

³² While marked economic, political, and social change occurred in Japan beginning with the Meiji Restoration of 1868, these events established the necessary, but not the sufficient conditions, for MEG. Namely, the Restoration is a period when MEG becomes a national objective, not a revealed occurrence as identified by Kuznets' empirically verifiable criteria. Rosovsky thus characterizes the period 1868–1885 as Japan's “transition” to modern economic growth. During this period the banking system was developed, fiscal and monetary reform undertaken, social barriers to progress attenuated, political power in the central government consolidated, and so forth. See Rosovsky, Henry, “Japan's Transition to Modern Economic Growth, 1868–1885,” in Rosovsky, Henry, editor, Industrialization in Two Systems: Essays in Honor of Alexander Gerschenkron (New York: Wiley, 1966), pp. 91–139Google Scholar. See also Ohkawa and Rosovsky, “A Century of Japanese Growth,” pp. 47–66.

³³ Ohkawa and Rosovsky divide their examinations of long term change in the Japanese economy into three periods: period I, from the Meiji Restoration to World War I; period II, from World War I to World War II; and period III, from World War II to the present. Justifying the demarcation at World War I, they note, “In spite of the sustained character of growth in Japan, the relative position of major sectors in the economy changed considerably…. World War I marked a distinct structural change especially pronounced in the relations between agriculture and industry.” Ohkawa and Rosovsky, “The Role of Agriculture,” p. 43.

In a later study, these authors consider the period 1886–1905 as the initial phase of modern economic growth. From 1906–1930 a different economic structure is created which sets the pattern for the second phase of modern economic growth. Ohkawa and Rosovsky, “A Century of Japanese Economic Growth,” ch. ii.

³⁴ It should be noted that had we used the recently revised non-primary LTES series (not yet available), then the correspondence would have been much closer with the LTES and Nakamura figures.

³⁵ Fei and Ranis have, found significant capital shallowing over the entire period 1888–1916 in the industrial sector. See Fei, John C. H. and Ranis, Gustav, “Innovation, Capital Accumulation and Economic Development,”. American Economic Review, LIII (June 1963Google Scholar); and Fei and Ranis, Development of the Labor Surplus Economy. For reasons discussed in the ensuing paragraphs, we cannot accept their estimates and conclusions.

³⁶ While the possible understatement of agricultural capital stocks and investment due to the exclusion of land improvement must qualify this observation, the conclusion is not likely to be overturned. On measurement biases, see Nakamura, Agricultural Production and the Economic Development of Japan, 1873–1922, p. 172; Ohkawa, K., book review, Economic Development and Cultural Change, XII (Oct. 1963), 100;Google Scholar and Rosovsky, Capital Formation in Japan 1868–1940, p. 6.

³⁷ From 1920–1942 Ohkawa finds a decrease in both the economy-wide and in the industrial sector's labor share. See Ohkawa, K., “Changes in National Income Distribution by Factor Share in Japan,” in Marchal, J. and Ducros, B., editors, The Distribution of National Income (New York: St. Martin's Press, 1968), pp. 177–96CrossRefGoogle Scholar.

³⁸ Ohkawa and Rosovsky, “A Century of Japanese Economic Growth,” pp. 77–79.

³⁹ Rosovsky, Capital Formation in Japan 1868–1940; Rosovsky, “Rumbles in the Ricefields,” p. 360; Ohkawa and Rosovsky, “The Role of Agriculture in Modern Japanese Economic Development.”

⁴⁰ Nakamura, Agricultural Production and the Economic Development of Japan, 1873–1922, p. 151 and pp. 155–69.

⁴¹ Rosovsky, “Rumbles in the Ricefields,” p. 360.

⁴² Nakamura, Agricultural Production and the Economic Development of Japan, 1873–1922, p. 167.

⁴³ All of the comparisons utilize net output. These measures were also employed by Ohkawa and Rosovsky in their study of agricultural productivity. See Ohkawa and Rosovsky, “The Role of Agriculture.” In contrast, Hayami's examination of fertilizer demand in Japanese agriculture employs gross value produced per factor. See Hayami, Yujiro, “Demand for Fertilizer in the Course of Japanese Agricultural Development,” Journal of Farm Economics, XLVI (Nov. 1964Google Scholar).

⁴⁴ The best study is by Watanabe who finds technical change to be labor saving in manufacturing, 1904–1933. See Watanabe, Tsuenehiko, “Economic Aspects of Dualism in the Industrial Development of Japan,” Economic Development and Cultural Change, XIV (April 1965), 293–312CrossRefGoogle Scholar. Agriculture over the period 1887–1936 does not appear to reveal a significant bias. These are, as Watanabe later notes, highly tentative estimates. See Watanabe, “Industrialization, Technological Progress, and Dual Structure.” The impossibility of simultaneously estimating the elasticity of substitution and the bias in technical change [P. A. Diamond and D. McFadden, “Identification of the Elasticity of Substitution and the Bias of Technical Change: An Impossibility Theorem,” Working Paper No. 62, University of California (mimeo., March 1965)], weaknesses in existing data (output, capital excluding land improvements), and the necessity of making limiting assumptions all plague his empirical study.

⁴⁵ Ohkawa and Rosovsky, “The Role of Agriculture,” p. 46; Johnston, B. F. and Cownie, J., “The Seed-Fertilizer Revolution and Labor Force Absorption,” American Economic Review, LIX (Sept. 1969), 569;Google ScholarJohnston, B. F. and Tolley, G. S., “Strategy for Agriculture in Development,” Journal of Farm Economics, XLVII (May 1965), 370;Google Scholar Shujiro Sawada, “Innovation in Japanese Agriculture, 1880–1935,” in Lockwood, editor, The State and Economic Enterprise in Japan, p. 335; and Nakamura, Agricultural Production and the Economic Development of Japan, 1837–1922, pp. 141–42 and 149–50.

⁴⁶ Solow, Robert, “Technical Change and the Aggregate Production Function,” Review of Economics and Statistics, XXXIX (Aug. 1957);Google ScholarBruton, Henry, “Productivity Growth in Latin America,” American Economic Review, LVII (Dec. 1967);Google Scholar A. C. Harberger and M. Selowsky, “Key Factors in the Economic Growth of Chile: An Analysis of Sources of Past Growth and Prospects for 1965–1970,” Conference on the Next Decade of Latin American Economic Development, Cornell University, April 20–22, 1966; Samuel Bowles, “Sources of Growth in the Greek Economy, 1951–61,” Memorandum No. 27, Project for Quantitative Research in Economic Development, Harvard University (mimeo., Nov. 1966); T. C. Liu, “The Tempo of Economic Development of the China Mainland, 1945–65,” An Economic Profile of Mainland China, The Joint Economic Committee, Congress of the United States (Feb. 1967); and Williamson, “Dimensions of Postwar Philippine Economic Progress.”

⁴⁷ Our percentage reference is to existing rates. For example, a 3 percent rise in growth rates may be from 3.0 to 3.1 in response to a savings parameter change from .312 to .315.

⁴⁸ A. R. Tussing, “The Labor Force in Meiji Economic Growth: A Quantitative Study of Yamanashi Prefecture,” in K. Ohkawa, B. F. Johnston, and H. Kaneda, editors, Agriculture and Economic Growth, p. 220.

⁴⁹ Fei and Ranis, Development of the Labor Surplus Economy, p. 132.

⁵⁰ Most recently Colin Clark has reconstructed the capital stock series using the same procedure as Fei and Ranis (perpetual inventory) but employing different depreciation parameters. Clark's figures show capital deepening throughout the entire period under consideration. We are indebted to Professor Clark for permitting us to utilize his estimates before publication.

⁵¹ Fei and Ranis, Development of the Labor Surplus Economy, pp. 263–64.

⁵² See Tables 6, 7. Minami also performed an econometric test for “turning points” in Japanese history. See Minami, “The Turning Point in the Japanese Economy,” pp. 380–402. The results are difficult to interpret given the debate over the real wage series and aggregation problems.

⁵³ Williamson, Jeffrey G., “Regional Inequality and the Process of National Development: A Description of the Patterns,” Economic Development and Cultural Change, XIII (July 1965Google Scholar), Supplement.

⁵⁴ Kelley, Williamson, and Cheetham, “Biased Technological Progress.”

⁵⁵ Ohkawa and Rosovsky, “The Role of Agriculture”; Ohkawa and Rosovsky, “A Century of Japanese Economic Growth”; Yujiro Hayami, and Saburo Yamada, “Technological Progress in Agriculture,” in Lawrence Klein and K. Ohkawa, editors, Economic Growth, the Japanese Experience Since the Meiji Era; and Johnston and Cownie, “The Seed-Fertilizer Revolution.”

⁵⁶ Kelley, Williamson, and Cheetham, Economic Dualism in Theory and History, chs. v, vi, and ix.