¹ The development of the Meiji financial system is treated in more detail in H. T. Patrick, “Japan,” in a forthcoming book edited by Rondo Cameron, Banking in the Early Stages of Industrialization.

² In 1868, taxes were only about 10 per cent of central government expenditures, and in 1872 only 40 per cent; once the land tax was imposed in 1873, taxes tended to cover expenditures until 1877. See Emi, Koichi, Government Fiscal Activity and Economic Growth in Japan, 1868–1960 (Tokyo: Kinokuniya, 1963), pp. 111–13.Google Scholar

³ See Crawcour, Sydney, “Japan's Economy on the Eve of Modernization,” in Lockwood, W. W., ed., The State and Economic Enterprise in Modern Japan (Princeton University Press, 1965)Google Scholar. If James Nakamura's estimates (“Agricultural Production in Japan, 1878–1922,” in ibid.) are correct, the daimyo and samurai share in total agricultural output would be reduced to some 20 to 30 per cent and Restoration government tax share to about 12 per cent. By 1878–82, the bondinterest income of the former ruling class relative to agricultural output was only 2.9 per cent (or 1.5 per cent, using the Nakamura estimate).

⁴ “Within less than ten years 67 per cent of the bonds had been sold by the original owners, most at a low price.” Kimura, Motokazu, “Fiscal Policy and Industrialization in Japan, 1868–95,” in Berrill, Kenneth, ed., Economic Development with Special References to East Asia (London: Macmillan, 1964), p. 280Google Scholar.

⁵ For useful descriptions in English of the Meiji financial system and policies, see Matsukata, Masayoshi, Report on the Adoption of the Gold Standard in Japan (Tokyo, 1899)Google Scholar; “Banking in Modern Japan,” Fuji Bank Bulletin, Vol. XI, No. 4 (Dec. 1961; special issue)Google Scholar; and Takaki, Masayoshi, The History of Japanese Paper Currency, 1868–1890 (Baltimore: Johns Hopkins Press, 1903)Google Scholar. For greater detail and analysis, see Kato, Toshihiko, Hompo Ginko Shi Ron (“A History of Japanese Banks” [Tokyo: Tokyo University, 1957])Google Scholar, and Asakura, Kokichi, Meiji Zenki Nihon Kinyu Kozoshi (“A History of the Japanese Financial Structure in the Meiji Era” [Tokyo: Keiso Shobo, 1963])Google Scholar.

⁶ The point is to substitute money of token value (such as paper currency) which costs relatively little to produce, for money (such as gold or silver coins) of intrinsic value—that is, where the good being used as money has a demand in alternative uses such that, given costs of production, the price of the good in alternative uses equals its value as money. This substitution not only greatly reduces the cost of providing additions to the money supply but also frees the existing stock of specie used as money for uses which can be more productive (via the foreign-trade route). The importance of this process, not only for Japan but for the more advanced countries as well in the nineteenth century, is emphasized in Triffin, Robert, The Evolution of the International Monetary System: Historical Reappraisal and Future Perspectives, (Princeton, N. J.: Princeton University, Department of Economics, international Finance Section, 1964)Google Scholar.

⁷ Ideally, a further adjustment in estimating money supply should be made to reflect the fact that different monies did not always circulate with each other at par but at varying discounts; lack of data has prevented this adjustment.

⁸ The most comprehensive estimates of money supply since 1878 have been compiled by Shozaburo Fujino of Hitotsubashi University, in two series which have yet to be published. Some of his data for Series One, considerable discussion of them, and revision to exclude cash held by banks and bills in process of collection are in Hoekendorf, W. C., “The Secular Trend of Income Velocity in Japan, 1879–1940” (unpublished Ph.D. dissertation, University of Washington, 1961)Google Scholar. Professor Fujino is currently preparing Series Two, with substantial changes from his earlier estimate, notably the exclusion of cash held by banks and of interbank deposits. He suggests these were particularly large in early Meiji.

⁹ The government received 3.5 million yen held by daimyo as specie reserves for these notes, so the bulk of the exchange represented governmental assumption of daimyo liabilities. See Sakudo, Y., “The Reform of the Monetary System in the Early Years of Meiji,” Osaka Economic Papers, VI, No. 2 (Sept. 1957), 38–42Google Scholar.

¹⁰ Ibid., p. 33. This value was based on the outstanding issue; the Restoration government analyzed the contents of old coins and fixed their legal value.

¹¹ Included in government coins from 1871 are 5.6 million yen.

¹² Of the increase in national bank notes (see Table 1), 15 million yen was lent by one bank (Jugo Ginko, founded by 480 daimyo and representing almost half of all national-bank capital) to the government to finance its special expenditures.

¹³ Between 1868 and 1871, the standard shifted between gold and silver before officially settling on the former.

¹⁴ By 1880, 53.1 million yen in gold coins had been minted, while 50.0 million yen in gold had gone abroad as net exports. Part of this was in the form of Tokugawa coins and bullion. Nonetheless, by 1887, more than three fourths of new gold coins minted had been exported.

¹⁵ Actually this is an understatement, since for the period prior to 1888, values of imports from gold-standard countries were stated in gold-yen terms rather than in silver; compare Horie (cited in Table 1), pp. 20–21.

¹⁶ Emi's recent study (cited in n. 2) indicates that both consolidated total and central-government expenditures did not reach their peak in monetary terms until 1883 (1884 in real terms); thus most of the adjustment was initiated on the revenue side.

¹⁷ Kato, p. 52.

¹⁸ The procedure for determining the price deflator in Table 3 is the same as that used in Shinohara, M., Growth and Cycles in the Japanese Economy (Tokyo: Kinokuniya, 1962), p. 78Google Scholar. Using the Ohkawa deflator alone for this period gives virtually identical results—an 11 per cent decline—for the 1880–1882 period but 1882, rather than 1884, is the Ohkawa estimate of the trough. See Ohkawa, p. 249.

¹⁹ Although Ohkawa's long-run national income estimates are the best currently available, they have certain shortcomings for analysis both of short-run cycles and of Meiji growth. For a discussion of upward biases in Ohkawa's estimates of Meiji growth rates, see Nakamura (cited in n. 3); also, Harry T. Oshima, “Government Revenues and Expenditures in Meiji Economic Growth” in the same volume.

²⁰ At Ohkawa's suggestion, in order better to estimate sectoral influences I have used the separate agriculture and nonagriculture price deflators in Ohkawa, p. 130, rather than those of his general national income (used in his sectoral real-income estimates, p. 248).

²¹ Such fiduciary note issue was subject to a government tax; this apparently did not affect the amounts of issue when they were needed, while it served to transfer some of the central bank's profits to the government.

²² Note (Table 1 and Chart 1) that Bank of Japan notes increased 30 per cent in 1890 over 1889, offsetting completely the other pressures reducing the money supply.

²³ Bank shareholders and others closely connected frequently obtained bank loans in order to subscribe to new stock issues, with the stock as collateral. For detailed examples see Kato, Toshihiko and Ouchi, Tsutomu, Kokuritsu Ginko no Kenkyu (“Studies on National Banks” [Tokyo: Keiso Shobo, 1963])Google Scholar.

²⁴ This was particularly true between 1890 and 1900, when loans from the Bank of Japan ranged between 10 and 18 per cent of commercial bank deposits plus net worth; thereafter, they declined to between 2 and 6 per cent. See Yoshino, T., “Waga Kuni Shichu Ginko no Obaaron ni Tsuite” (“Overloan of Commercial Banks in Japan” [Bank of Japan, Chosa Geppo, Feb. 1954]), pp. 1–39Google Scholar.

²⁵ Compare Emi, especially ch. iii and pp. 140–41.

²⁶ The only available indexes of export and import unit prices and volume, while not very good, are found in Oriental Economist, Foreign Trade of Japan, a Statistical Survey (Tokyo: Toyo Keizai, 1935), p. 698Google Scholar.

²⁷ Money wages from Ohkawa, p. 244.

²⁸ Net value added in textiles rose from 10.7 million yen (26 per cent of value added in factory manufacturing) in 1890 to 28.9 million yen (36 per cent) in 1895; compare Ohkawa, p. 88. During this period the number outstanding of installed spindles more than doubled, from 277,895 to 580,945; this continued die investment trend begun in 1886 (71,604 spindles). See Clark, W. A. Graham, Cotton Goods in Japan (Washington: U.S. Department of Commerce Special Agent Series No. 86, 1914), p. 40Google Scholar. See also this work and Shinohara, ch. iii, for a discussion of the shifting pattern of trade.

²⁹ An important point was that the fluctuation of the world price of silver as compared to gold was relatively independent of any action Japan might take. While depreciation had been beneficial, there was some fear that silver might not depreciate further but, instead, perhaps even appreciate.

³⁰ See, for example, Soyeda (cited in Table 1, Sources), IV, 536–37. This author, a senior member of the Ministry of Finance, was a member of the commission discussed in the following paragraph. See also Droppers, Garret, “Monetary Changes in Japan,” Quarterly Journal of Economics, XII, No. 2 (Jan. 1898), especially 178–79.Google Scholar

³¹ Some 85 per cent of the indemnity was allocated for military purposes; see Matsukata, p. 220. The government also expanded its railroads, telegraph, and ironfoundry and munitions facilities, financed also by increased consumption taxes and domestic bond issue.

³² Just after the Sino-Japanese War, the proportion had been above 80 per cent. See Fuji Bank Bulletin, p. 44. Data on the amount and proportion of official specie reserves held abroad from 1903 are available in Shimpo Sha, Toyo Keizai, ed., Meiji Taisho Kokusei Soran (“Survey of the State of the Nation in the Meiji-Taisho Period” [Tokyo: Toyo Keizai, 1927]), Table 163, p. 139Google Scholar.

³³ However, Japan's expanding trade with China (still on silver) now became influenced by fluctuations in the gold-silver price ratio. This was particularly noticeable in 1907, when silver appreciated. Compare Shiraishi, Takashi, “Nihon Boeki Kozo no Bunseki to Koeki Joken” (“Structural Analysis of Japanese Foreign Trade and Terms of Trade”), in Kojima, Kiyoshi, ed., Ronso: Keizai Seicho to Nihon Boeki (“Debate: Economic Growth and Japanese Foreign Trade” [Tokyo: Kobundo, 1960])Google Scholar.

³⁴ While the real national income data in Table 3 suggest there was not an absolute decline until 1899, contemporary descriptions and other indicators suggest an earlier trough. The Ohkawa money national income for 1898 is clearly overly high, implying as it does a real national income increase of some 32 per cent over the previous year.

³⁵ Rules which neither Japan nor other countries invariably followed. See Triffin; also, Arthur Bloomfield, I., Monetary Policy Under the International Gold Standard: 1880–1914 (New York: Federal Reserve Bank of New York, 1959)Google Scholar.

³⁶ A decline of real national income in 1902 of 13 per cent, according to Table 3, is probably an exaggeration. The timing may also have been earlier.

³⁷ The government attempted a £10 million sterling bond issue in London in June 1898 but with disastrous results; the underwriting syndicate had to take over 90 per cent of the issue. See Fuji Bank Bulletin, p. 42. It is not coincidental that Japan's greatly enhanced ability to borrow in the London capital market followed the signing of the Anglo-Japanese Alliance in 1902. The question of foreign borrowing was also raised in 1880 by Finance Minister Okuma, who suggested a 50 million yen foreign loan to provide additional specie. This suggestion was firmly rejected.

³⁸ Estimated from investment figures in Henry Rosovsky, Capital Formation in Japan (New York: Free Press, 1961), pp. 9, 15; and from the current-account deficit in the balance of payments (Bank of Japan, Historical Statistics [cited in Table 3, Sources], p. 101). Including military investment the ratio would be about one third. The deficit was between 25 and 30 per cent of gross domestic capital formation (excluding military); actually, the increase in foreign loans was greater than the current-account deficit.

³⁹ Aside from the issue of allowing long-run depreciation to alleviate the balance-of-payments constraint on growth, a further issue is whether depreciation should be brought about by flexible exchange rates or by a system of fixed but adjustable rates. I do not choose to consider this issue here though, given the administrative problems and time lags in policy formulation and implementation in underdeveloped countries, flexible exchange rates may be more efficient.

⁴⁰ Without at the same time giving up its tariff and customs autonomy, as China was forced to do. Inasmuch as a major objective of Japanese foreign policy until successful attainment in the 1890's was to regain tariff autonomy and to end extraterritoriality, it seems unlikely that Japan would have been willing to assume anew such obligations.