Japan and the great divergence, 730–1874

https://doi.org/10.1016/j.eeh.2018.11.005Get rights and content

Abstract

Despite being the first Asian economy to achieve modern economic growth, Japan has received relatively little attention in the Great Divergence debate. New estimates suggest that although the level of GDP per capita remained below the level of northwest Europe throughout the period 730–1874, Japan experienced positive trend growth before 1868, in contrast to the negative trend growth experienced in China and India, leading to a Little Divergence within Asia. However, growth in Japan remained slower than in northwest Europe so that Japan continued to fall behind until after the institutional reforms of the early Meiji period. The Great Divergence thus occurred as the most dynamic part of Asia fell behind the most dynamic part of Europe.

Introduction

Before the publication of Maddison's (2001), The World Economy: A Millennial Perspective, most accounts of economic growth before the nineteenth century were largely qualitative. Although Maddison's estimates have been influential, they were based largely on conjecture rather than built up from contemporary data. Maddison himself was well aware that many parts of Europe and Asia during the medieval and early modern periods were quite literate and numerate, and left behind an abundance of documents containing information that could be turned into estimates of population and economic activity. Indeed, in an edited collection from the World Economic History Congress in Milan, Maddison and van der Wee (1994) provided some preliminary studies in historical national accounting that went back to 1086 in England, to 1500 in Flanders and Brabant and to 1580 in Spain (Snooks, 1994, Blomme and van der Wee, 1994, Yun, 1994). However, these studies were only very loosely incorporated into Maddison (2001: 244–265), where many of the GDP per capita estimates before 1500 are around $400 in 1990 international prices, connected up to the 1820 estimates using assumed growth rates based on a reading of the qualitative literature. The figure of $400 is seen as subsistence income, based on the assumption that most people existed at the World Bank poverty level of $1 per day (in 1990 prices), with a small rich elite pulling up the average from $365. It is important to understand that Maddison intended his estimates for the pre-modern period to be a starting point and indeed encouraged other scholars to engage in historical national accounting using contemporary data, with a view to improving his estimates, just as he had already done for the modern period in a remarkable stream of publications from Maddison (1982) to Maddison (1995).

Maddison's pathbreaking study has stimulated much progress in historical national accounting for medieval and early modern Europe, with annual estimates now available back as far as 1300 for Britain, the Netherlands, Italy, Spain, Sweden and France and back to 1500 for Portugal and Germany (Broadberry et al., 2015a, van Zanden and van Leeuwen, 2012, Malanima, 2011, Álvarez-Nogal and Prados de la. Escosura, 2013, Schön and Krantz, 2012, Krantz, 2017; Ridolfi, 2016; Pfister, 2011; Palma and Reis, 2017). For Asia, progress has been rather slower, as there is less of a tradition of collecting data and processing it within an economic framework on which to build. Nevertheless studies have now appeared for China and India based on contemporary data, showing substantial differences from Maddison's (2001) conjectures (Broadberry et al., 2018, Xu et al., 2017, Broadberry et al., 2015b). Japan, Asia's most successful economy, has been missing from this new database.

In this paper, we provide the first estimates based on contemporary data of long term growth for Japan over the period 730–1874. These estimates have important implications for the debate over the Great Divergence of productivity and living standards between Europe and Asia. Following the claim of Pomeranz (2000) that China did not fall behind the West until after 1800, most attention on the Asian side has been focused on China, with a smaller amount of literature devoted to the case of India, following the work of Parthasarathi (1998, 2011). However, a similar debate had in fact taken place over the position of Japan nearly two decades earlier, following the work of Hanley (1983, 1986). Hanley (1983) argued that Japanese living standards were as high as in the West until the nineteenth century. Furthermore, the case for Japan as the most likely Asian economy to be on a par with Europe in the early modern period is given credence by the fact that Japan was the first Asian economy to make the transition to modern economic growth in Asia after the Meiji Restoration of 1868, and went on to catch up with Western levels of per capita GDP in the twentieth century while China and India fell further behind. Although Hanley's (1983) claim of Japanese parity with the West until the nineteenth century was quickly criticised by Yasuba (1986) before fading from view, the idea of Japan following an exceptional path within Asia, sharing many characteristics with the European path, was already firmly established within the economic history literature and has remained strong (Smith, 1959, Dore, 1965, Nakamura, 1966, Hayami and Kito, 2004). Indeed, the later volumes of the Cambridge Economic History of Europe even went so far as to include separate chapters on Japan in providing a comprehensive overview of “European” development (Mathias and Postan, 1978, Mathias and Pollard, 1989). Adding Japan to the database means that for the first time, we can examine quantitatively the underlying differences between China, India and Japan which may help to shed light on why Japan caught up with Europe in the twentieth century, while India and China lagged behind.

The results presented here suggest that Japan experienced substantial GDP per capita growth before 1868, thus providing a firm base for the first Asian transition to modern economic growth following the Meiji restoration of that year. Inevitably, with the current state of knowledge, the estimates lack precision. There are undoubtedly fewer sources for Asian economies than for European economies such as Britain and the Netherlands. Furthermore, those sources that are available have been subjected to less critical scrutiny by economic historians. To deal with this uncertainty, we provide data reliability assessments and sensitivity analysis. Here, we build upon the subjective error margins approach used by Farris (2006, 2009a) for Japanese population and agricultural output. This approach has also been used in historical national accounting studies for other countries (Feinstein and Thomas, 2002, van Zanden and van Leeuwen, 2012, Broadberry et al., 2018). We also consider the robustness of our main findings to alternative scenarios.

The central estimates suggest Malthusian fluctuations in Japan during the ancient period, with positive per capita income growth during periods of falling population and negative per capita income growth during periods of rising population. This was followed by positive trend growth of per capita income in Japan between 1280 and 1874, with most of the growth occurring in two phases, between 1450 and 1600 during the medieval period, and after 1721 during the Tokugawa period. Although there is a qualitative literature to support the idea of late medieval growth, the width of the error bands during the medieval period suggests caution should be exercised until further quantitative evidence becomes available to confirm this finding (Nagahara, 1981, Farris, 2006). For the Tokugawa period, however, the case for trend growth is stronger since it is based upon a much larger amount of contemporary quantitative information as well as an extensive qualitative literature that supports this interpretation (Nakamura, 1968, Hayami et al., 2004). Although our estimates suggest that Japan was improving its relative position within Asia during the Tokugawa period, growth in Japan remained slower than in northwest Europe. As a result, Japan continued to fall behind the leading European economies until after the Meiji Restoration. The Great Divergence thus occurred as the most dynamic part of Asia fell behind the most dynamic part of Europe.

Section snippets

Population

Historical demographic data allow the estimation of total population for Japan back to around 730. The data in Table 1 are taken from a number of sources that have been cross-checked and made consistent, covering the ancient, medieval and Tokugawa periods (Saito and Takashima, 2017a, 2017b). Due to the limited availability of primary sources for premodern Japan, it is not possible to construct an annual series, so data are provided for a number of benchmark years. For the ancient period

Agricultural output from the supply side

Agricultural output can be estimated directly from the supply side, using data on crops harvested or the amount of land used for crop production multiplied by crop yields. This can then be cross-checked against estimates of the demand for food derived indirectly from data on population, wages and prices. Starting with the supply-side estimates, the precise method of estimation varies by period. For the ancient and medieval periods, agricultural output is derived from data on the amount of

Agricultural output from the demand side

An alternative way of estimating agricultural output is to infer it from a demand function for food, using known trends in wages and prices. This approach can be traced back at least as far as the work of Crafts (1985), who calculated the path of agricultural output in Britain during the Industrial Revolution with income and price elasticities derived from the experience of later developing countries. The approach was developed further by Allen (2000) using consumer theory. Allen (2000: 13–14)

Urbanisation and non-agricultural production

A number of authors have used the share of the population living in towns as a measure of the growth of the non-agricultural sector. This approach began with Wrigley (1985), and has recently been combined with the demand approach to agriculture to provide indirect estimates of GDP in a number of European countries during the early modern period (Malanima, 2011, Álvarez-Nogal and Prados de la. Escosura, 2013, Schön and Krantz, 2012). With the path of agricultural output (QA) derived using

Benchmark estimates

Our benchmark estimates of Japanese GDP per capita are shown in level form in Table 9A, and in annual growth rate form in Table 9B. The levels data are shown in terms of koku, combining the benchmark estimates for GDP from Table 8 with the benchmark estimates for population from Table 1. Using this series, Japanese GDP per capita grew at an annual rate of 0.09% between 730 and 1874, cumulating to an increase of living standards by a factor of 2.7.

During the ancient period, the patterns of GDP

An Anglo-Japanese comparison

To pin down the timing and extent of the Great Divergence, we need to compare GDP per capita in Japan with Britain, where the transition to modern economic growth first occurred, and place the Anglo-Japanese comparison within its wider Europe-Asia context. Here, we project back from an estimate of GDP per capita in the late nineteenth century, expressed in 1990 international dollars, but with some important adjustments that improve on the 1874 value used by Maddison (2010). Maddison's (2010)

Conclusions

This paper provides estimates of Japanese GDP per capita for the period 730–1874, constructed from the output side, using methods developed for the estimation of GDP per capita in medieval and early modern Europe, but amended to suit Japanese circumstances and data. Our estimates for the agricultural sector are built up from direct estimates of arable land use and land productivity, and checked against trends in agricultural demand derived from the grain wages of unskilled labourers. Activity

Acknowledgements

We gratefully acknowledge the financial support of the Great Britain Sasakawa Foundation (reference number 3652), the Daiwa Anglo-Japanese Foundation (reference 09/10-95), the US National Science Foundation (NSF), and the Global COE Hi-Stat program Japanese Ministry of Education (Global COE Hi-stat program). This paper also forms part of the Collaborative Project HI-POD supported by the European Commission's 7th Framework Programme for Research (Contract Number SSH7-CT-2008-225342), and the

References (113)

  • ArthurL. Bowley

    The Measurement of the accuracy of an average

    J. R. Stat. Soc.

    (1911)
  • Mitsui Bunko

    Kinsei kōki ni okeru shuyō bukka no dōtai [Trends of Major Prices in Early Modern Japan]

    (1981)
  • S. Broadberry et al.

    British Economic Growth, 1270–1870

    (2015)
  • S. Broadberry et al.

    China, Europe and the great divergence: a study in historical national accounting

    J. Econ. History

    (2018)
  • S. Broadberry et al.

    Growing, Shrinking and Long Run Economic Performance: Historical Perspectives on Economic Development

    (2017)
  • AgathaL. Chapman

    Wages and Salaries in the United Kingdom 1920–1938

    (1953)
  • N.F.R. Crafts

    British Economic Growth During the Industrial Revolution

    (1985)
  • A. Deaton et al.

    Economics and Consumer Behaviour

    (1980)
  • J. de Vries

    European Urbanization, 1500–1800

    (1984)
  • R. Dore

    Education in Tokugawa Japan

    (1965)
  • M. Elvin

    The Pattern of the Chinese Past

    (1973)
  • M. Endo

    Shokunin no rekishi [History of Craftsmanship]

    (1956)
  • W.W. Farris

    Japan's Medieval Population: Famine, Fertility, and Warfare in a Transformative Age

    (2006)
  • W.W. Farris

    Daily Life and Demographics in Ancient Japan

    (2009)
  • W.W. Farris

    Japan to 1600: A Social and Economic History

    (2009)
  • C.H. Feinstein

    National income, Expenditure and Output of the United Kingdom, 1855–1965

    (1972)
  • C.H. Feinstein et al.

    A plea for errors

    Hist. Methods

    (2002)
  • K. Fukao et al.

    Regional Inequality and Industrial Structure in Japan: 1874–2008

    (2015)
  • K. Fukao et al.

    Kanmatsu furoku: Seisan, bukka, shotoku no Suikei [Appendix: estimation of production, prices, and income]

  • S.B. Hanley

    A high standard of living in nineteenth-century Japan: fact or fantasy?

    J. Econ. History

    (1983)
  • S.B. Hanley

    Standard of living in nineteenth-century Japan: reply to Yasuba

    J. Econ. History

    (1986)
  • T. Harada

    Chūsei ni okeru toshi no kenkyū [Study of Urban History in the Medieval Period]

    (1942)
  • A. Hayami

    Keizai shakai no seiritsu to sono tokushitsu [The emergence of economic society and its characteristics]

  • A. Hayami

    Japan's Industrious Revolution: Economic and Social Transformations in the Early Modern Period

    (2015)
  • A. Hayami et al.

    Demography and living standards

  • A. Hayami et al.
  • HuangP.

    The Peasant Economy and Social Change in North China

    (1985)
  • M. Iwahashi

    Kinsei nihon bukkashi no kenkyū [Research on the History of Prices in Early Modern Japan]

    (1981)
  • M. Iwahashi

    The institutional framework of the Tokugawa economy

  • T. Iyanaga

    Nihon kodai shakai keizaishi kenkyū [Study of Japanese ancient Socio-Economic History]

    (1980)
  • M. Kamata

    Nihon kodai no jinkō ni tsuite [population in ancient Japan]

    Mokkan Kenkyū

    (1984)
  • M. Kimura

    La Revolución de los precios en la cuenca del pacífico, 1600–1650 [The Price Revolution in the Pacific Basin, 1600–1650]

    (1987)
  • H. Kito

    Nihon nisen nen no jinkōshi [A Two Thousand Year History of the Japanese Population]

    (1983)
  • H. Kito

    Meiji izen Nihon no chiiki jinkō [Japan's population before the Meiji period]

    Jōchi Keizai Ronshū [Jochi Econ. Rev.]

    (1996)
  • H. Kito

    Jinkō kara mita Nihon no rekishi [Japanese History from a Demographic Perspective]

    (2000)
  • H. Kondo
  • O. Krantz

    Swedish GDP 1300-1560: A Tentative Estimate

  • Kunaichō Shōsōin Jimusho (ed.), (1988) Shōsōin komonjo eiin shūsei 1 [Photographic Reproductions Collection of Ancient...
  • K. Kuroita

    Ruijūu sandaikyaku and Kōnin kyakushō (Shintei zōho Kokushi taikei 25, Shinsōban) [Assorted Regulations from Three Reigns and Extracts of the Kōnin Regulations (The 25th Volume of the Supplementary compendium of Japanese history, Newly Revised Edition)]

    (2011)
  • Cited by (23)

    • Political centralization, career incentives, and local economic growth in Edo Japan

      2022, Explorations in Economic History
      Citation Excerpt :

      According to the time-series estimation of Saito and Takashima (2016) using backward projection, the shares of primary, secondary, and tertiary sectors in GDP were all highly stable between 1721 and 1874.17 In another estimation, the share of primary output in Japan’s GDP had always been over 60% until the 1850s while the share of urban population was barely above 10% throughout the entire Edo Period (Bassino et al., 2019). While these estimates represent national level averages, subnational evidence also suggests that a focus on agricultural output is appropriate to measure economic growth during the Edo Period.

    View all citing articles on Scopus
    View full text