A Global History of Money [1 ed.] 9781003016205

Looking from the 11th century to the 20th century, Kuroda explores how money was used and how currencies evolved in tran

443 33 2MB

English Pages 228 [229] Year 2020

Report DMCA / Copyright


Polecaj historie

A Global History of Money [1 ed.]

Table of contents :
Part 1 Exchanges generate money locally
1 Peasants, marketplace and money
2 Stagnant currencies and stratified markets
Part 2 A global history of monetary delocalization
3 The ignition of monetary delocalization: an unexpected remnant of the Mongolian regime, c.1300
4 The world diversified and stratified: three paths after the global silver march, c.1600
5 Nationalized money: backstage of the international gold standard regime, c.1900
Conclusion: money as social circuit

Citation preview

A Global History of Money

Looking from the 11th century to the 20th century, Kuroda explores how money was used and how currencies evolved in transactions within local communities and in broader trade networks.The discussion covers Asia, Europe and Africa and highlights an impressive global interconnectedness in the pre-modern era as well as the modern age. Drawing on a remarkable range of primary and secondary sources, Kuroda reveals that cash transactions were not confined to dealings between people occupying different roles in the division of labour (for example shopkeepers and farmers), rather that peasants were in fact great users of cash, even in transactions between themselves. The book presents a new categorization framework for aligning exchange transactions with money usage choices. This fascinating monograph will be of great interest to advanced students and researchers of economic history, financial history, global history and monetary studies. Akinobu Kuroda is Professor of East Asian History at the University of Tokyo, Japan. His research covers comparative studies of monetary history in East Asia, India, Africa and Europe, as well as specific studies of China’s monetary history.

Routledge Explorations in Economic History Edited by Lars Magnusson Uppsala University, Sweden

The Western Allies and Soviet Potential in World War II Economy, Society and Military Power Martin Kahn Money, Currency and Crisis In search of Trust, 2000 BC to AD 2000 Edited by R.J. van der Spek and Bas van Leeuwen The Development of Modern Industries in Bengal Re-Industrialisation, 1858–1914 Indrajit Ray The Economic Development of Europe’s Regions A Quantitative History Since 1900 Edited by Nikolaus Wolf and Joan Ramón Rosés Family Firms and Merchant Capitalism in Early Modern Europe The Business, Bankruptcy and Resilience of the Höchstetters of Augsburg Thomas Max Safley An Economic History of Famine Resilience Edited by Jessica Dijkman and Bas van Leeuwen Modern Advertising and the Market The US Advertising Industry from the 19th Century to the Present Zoe Sherman A Global History of Money Akinobu Kuroda For more information about this series, please visit www.routledge.com/series/ SE0347

A Global History of Money Akinobu Kuroda

First published 2020 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2020 Akinobu Kuroda The right of Akinobu Kuroda to be identified as author of this work has been asserted by him in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested ISBN: 978-0-367-85923-7 (hbk) ISBN: 978-1-003-01620-5 (ebk) Typeset in Bembo by Apex CoVantage, LLC


List of figuresviii List of tablesix Acknowledgementsxi List of abbreviationsxiii Introduction 1  Peasants use money  1 2  Four quadrants of exchange  5 3  Exchange and institutional setting  7 4  Global history of money viewed from the ground  9 5  Bad money does not drive out good money: literature  11



Exchanges generate money locally17 1 Peasants, marketplace and money 1  Peasants exchange with peasants anonymously  19 2  Marketplace for one-time transactions  23 3  Natives preferred local currencies  27 4  Ubiquitous fractional currencies and petty exchangers  30 5 Subsequent transactions not through marketplace: another configuration between peasants and merchants  32 6  Honour substitutes for currency  35 7  Freedom or certainty: two paths (not stages) of local commercialization 39


2 Stagnant currencies and stratified markets 1  Exogenous currencies naturally stagnate  45 2  Unidirectional streams of small currencies  47 3  Stratified markets in agricultural societies  49


vi  Contents

  4  Informal agreements generate endogenous currencies  52   5  How merchants conducted business in stratified markets  54   6  Locally differentiated currencies and dematerialized money  58 PART 2

A global history of monetary delocalization65 3 The ignition of monetary delocalization: an unexpected remnant of the Mongolian regime, c.1300   1 Independent monetary systems in pre-13th century civilizations 67   2 The Eurasian silver century: emergence and collapse  73   3 The Mongolian taxation system depended on commerce  76   4  Paper monies drove out Chinese silver ingots to the west  78   5  Commensurability prevailed across Eurasia  84   6  Copper coins driven East and South  85   7 The Eurasian age of commerce: synchrony built on stratified markets 88   8  Sprouting of credit-oriented systems in Europe: an aftermath  95   9 Currency circuits with ‘Old’ coins flourished across the China Sea: another aftermath  100 10  Linkage to the second silver century  107 4 The world diversified and stratified: three paths after the global silver march, c.1600   1  A breakthrough with large silver coins  116   2 A silver century followed by a copper century: prosperity in a currency-oriented economy  119   3 States organize debts nationwide: formation of a credit-oriented economy 128   4 Local paper monies reveal differences among institutions: a comparison between England, China and Japan  132   5 Currencies, marketplaces and early industrialization: what happened to cement currency circuits?  138   6 The third path: commercial oligarchies  143 5 Nationalized money: backstage of the international gold standard regime, c.1900   1 The age of international silver dollars: an alternative to the territorial currency system  151   2 The riddle of the Maria Theresa dollar  154




Contents  vii

3 The reality of the Maria Theresa dollar’s circulation  159 4 Currencies working as complementary buffers: currency circuits survived 162 5  Paper monies nationalized peasant economies  171 6 How did banknotes reach down to the ground? The case of inter-war China  177 7 Seasonality and temporality in monetary demand still mattered: towards 1929  180 8 The paper money standard in China in 1935: unification at the top and variety on the ground  183

Conclusion: money as social circuit 1  A global history of monetary delocalization  195 2  Modern ‘common sense’ uncommon in history  196 3 The misuse of the concept of arbitrage: no equilibrium amidst streams 198 4  Escaping the teleology of monetary history  199 5  Alternative ideas about money by contemporaries  200 6  Institutions for flexibility as well as for certainty  202


Appendix206 Index208


I.1 Four quadrants of exchange 6 1.1 Frequency of daily transactions, Tongtaihao, 5th and 6th months, Daoguang 10th year (1830 CE) 25 1.2 Annual movement of rice price at rural area, Tunxi County, Anhui, China (wen per 5 sheng)29 1.3 Society with a high clustering coefficient: mutual credit 40 1.4 Society with a high preference for anonymity: local currency 41 3.1 Minting of silver coinage in London (kg) 74 3.2 Issuance of paper money by the Mongol Empire (in silver liang)81 3.3 Multiple strata of monies 86 3.4 Wheat price in grain of silver, England 92 5.1 Annual issuance of MTD from Vienna 157 5.2 Circuits of the Maria Theresa dollar in East Africa 167 5.3 Triple layer of currencies interfaced by the Maria Theresa dollar 168 5.4 Circuit of Japanese silver dollar in Southern China 169 A.1 Unidirectional streams of currency with frictions 206


1.1 Monthly sales of peasant products, Huangling village, Taiyuan, Shanxi, 1940 (yuan) 1.2 Annual sales and purchases per households in Qingyuan County, Hebei, China 1930 (yuan) 1.3 Commodities by distance in 11 marketplaces, Zouping County, Shandong, 1932 (yuan) 1.4 Exchange rates between Manillas and Sterling and the price of palm oil 1.5 Profits with rice and cotton cultivation per 1 tan mid-1880s southern Osaka, Japan (yen) 1.6 Advance and purchase between the Yamawaki family and a peasant weaver in 1850 2.1 The distribution of era names of the copper coins from the Aba hoard, Sichuan, 1870s 2.2 Currency outstanding, redeemed, in the US, June 1975 (billion US$) 2.3 The distribution of minting periods with tornesello from a Greek hoard and an excavation 2.4 Money balance of a rice area in Java around 1875 (0/00) 2.5 The entry and exit in the silver book on the 1st day of the 10th month of the 21st year of Daoguang (1841 CE), the Tongtaihao 2.6 Revenue and expenditure in wen on 1st day, 10th month, Daoguan 21st year (1841 CE), the Tongtaihao 2.7 Monies and their usages in Yingkou, China, 1930 3.1 The distribution of era names with copper coins from Chinese excavation 3.2 The distribution of minting periods of the English penny from English hoards 3.3 Casting of bronze bell in Japan, Korea and China 3.4 New markets in Holland and Japan 3.5 The funded debt of Cologne (marks)

20 24 24 30 33 34 47 48 50 51 54 55 61 69 70 87 90 97

x  Tables


Rate of interest on sales of rent charge against agricultural land, Germany (percentage of distribution for each period) 4.1 Prices in Valencia (dinar) 4.2 Total decennial imports of fine gold and silver to Seville (grams) 4.3 Imports of cotton cloths and yarns after opening ports, China and Japan 5.1 Estimates of annual production of gold and silver in the world accessible from Europe 5.2 Monies in circulation in 1879 and 1905, US (dollar) 5.3 Revenue and expenditure of the Ethiopian emperor 5.4 Monthly movements of the Maria Theresa dollar rate in Addis Ababa and silver price in London 5.5 Prices of agricultural products, 1889–1923 5.6 Paper monies in Thailand and Indochina 5.7 Currency of British East and Central Africa and the Uganda Protectorates, 1896 5.8 The proportion of the unified paper money (  fabi) among the entire currency in use at the end of 1935 (percent) 5.9 Amounts of banknotes and silver dollars possessed by native exchangers in Wuxi 1935 November (yuan) 5.10 Amounts of banknotes and silver dollars possessed by native exchangers and banks in Shashi 1935 November (yuan) 5.11 Exchange rates of 10 wen copper coin per 1 yuan at the end of 1935 5.12 Quantity of copper coins possessed by bank branches 1936 August (in pieces)

98 121 124 140 152 152 160 163 173 174 176 185 186 188 188 189


I was only able to complete this volume through collaborations which challenged conventional frameworks on society and history, with colleagues including Patrice Baubeau, Jérôme Blanc, Robert Chard, Georges Depeyrot, Georgina Gomez, Farley Grubb, Jane Guyer, Craig Muldrew, Michael North, Anders Ögren, Karin Pallaver, Masato Shizume, Ekaterina Svirina and Bruno Théret. Especially, it would have been impossible for me to make my ideas readable without support from Robert. I also owe thanks for their participation in workshops I organized, and/or advice and support for my ongoing research on global history and money, to colleagues across the world, including Gareth Austin, Ana Boldureanu, Angela Redish, Willem Wolters, Om Prakash, Doo-hwan Oh, Hans Ulrich Vogel, Richard von Glahn, Maria Alejandra Irigoin, Luca Fanttaci, Shiro Yoji, Massimo Amato, Dennis Flynn, Torbjörn Engdahl, Hiroshi Kato, Sevket Pamuk, Madeleine Zelin, David Weiman, Eric Helleiner, Hon-wai Ho, Mio Kishimoto, Najaf Haider, Wolbert Smidt, Yoshiko Nagano, George Bryan Souza, Sushil Chaudhury, John James, Katsuhiro Nishinari, Jan Lucassen, Arturo Giraldez, Ryuto Shimada, Yohei Kakinuma, Alan Stahl, George Selgin, Geoffrey Ingham, Jean Cartelier, Christian Lamouroux, Guillaume Carré, Alessandro Stanziani, André Orléan, Gilles Postel-Vinay,Vincent Bignon, Keith Hart, Nicholas Mayhew, Catherine Eagleton, Josette Rivallain, Claudia Jefferies, Joost Welten, Keiichiro Kato, Maxim Bolt, Ludovic Desmedt, Leigh Gardner, Laurence Fontaine, Michiya Nishimura, Christopher Hanes, Myung-soo Cha, Michael Schiltz, Masayuki Tanimoto, Jun Suzuki, Fabian Drixler, David Howell, Elizabeth Kaske, Elizabeth Köll, Daniel Botsman, Peter Perdue, Naomi Lamoreaux, William Goetzmann, Viviana Zelizer, Frederick Wherry, Jürgen Nautz, Catherine Brégianni, Marc Flandreau, Michael Puett, Sun Joo Kim, Elizabeth Perry, Young-hoon Rhee, Yasuo Takatsuki, Maura Dykstra, Li-chuan Tai, Ping He, Yue-jian Feng, Cong-ming Gao, Wen-cheng Wang, Sui-wai Cheung, Wen-kai He, Toshihiko Miyake, Eiji Sakurai, Makoto Nishibe, Markus Denzel, Oscar Gelderblom, Joost Jonker, Marisa Candotti, Akio Tanabe, Nobuhiro Kiyotaki, Nicola Di Cosmo, Michal Biran, Dai Matsui, Kaori Sekiguchi, Helen Wang, Mark Blackburn, Peter Spufford, Linda Twerdk, Elijah Greenstein and Brett Savage.

xii  Acknowledgements

I also thank the Institute for Advanced Studies on Asia (University of Tokyo), the Japan Society for the Promotion of Science, the Toyota Foundation, Université Paris Ouest Nanterre, the University of Tübingen, Yale University, the Harvard Yenching Institute, Labex TransferS (École Normale Supérieure, Paris), the Institute for Advanced Study (Princeton) and the Center for Chinese Studies (Taipei) for their support for my international and interdisciplinary research on global monetary history over two decades. This book is based in part on research results funded by the Japan Society for the Promotion of Science, project no. 19H01510, ‘International Collaborative Research on Variety of Exchange and Multiplicity of Money in Global History.’



British Library, Indian Office Chugoku noson kanko chosa kankokai Kahoku kotsu kabushiki kaisha shigyokyoku Kokuritsu Pekin daigaku fusetsu noson keizai kenkyusho Mantetsu Hokushi keizai chosasho Mantetsu Shanghai jimusho chosashitsu Public Record Office, Kew


1  Peasants use money As Aristotle asserted that exchange could not happen between one doctor and another, but could happen between a doctor and a peasant, trade is thought to occur only between those whose vocations are different, or those whose possessions are dissimilar.1 However, this apparently logical presumption does not accord with reality. Simply being a peasant does not guarantee that one will be sufficiently provided with products from this occupation. Modern investigations reveal that a peasant who cultivates grain purchases the grain which other peasants harvest, as we argue in detail later. Exchanges can happen between one peasant with a good harvest and another with a poor harvest even if the two grow the same products. And, peasants with no other item to give can offer their labour to other peasants in exchange for grain. A difference of vocation or a difference in items possessed is clearly not a necessary precondition for a system of exchange. Aristotle’s framework ignores those cases when one party has a surplus of an item and another a shortage. In other words, exchange can occur among people with or without specialization of vocation or possession. Importantly, it is unworkable to assume the exchanges between one peasant and another who lives at a distance and cultivates similar products. In reality, thus, the exchange without specialization can occur especially between those who work in close proximity. In other words, the trades without specialization tend to happen locally. Meanwhile, the exchanges between those living at a distance are likely to occur according to the difference in resources of each location, which enhances the division of labour between them. No thinker denies that money is a means of exchange. However, very few considered whether the difference of exchanges with specialization or without specialization mentioned earlier affects the nature of money. Assuming that there can be no significant trade exchanges between people of the same vocation, many overlooked the existence of monetary demand among those whose vocations are similar.2 Along this line, the exchanges among peasants who occupied the overwhelming majority of humanity throughout most of history often have been out of scholarly attention, or otherwise seriously underestimated.

2  Introduction

Actually, as we will confirm later, for peasant households, the proportion of exchange among local peasants whose products were similar is not low as compared to the proportion of exchange at a distance.Thus, we should not take the exchanges without specialization as so negligible as to be overlooked. A peasant exchanges with another peasant and, importantly, they make transactions in terms of money. Is there any significant difference between monetary usages among peasants and usages by others, such as merchants? The size of each individual transaction between peasants inevitably is small. Importantly, their exchanges could not have occurred at a consistent level of frequency but exhibited a significant temporal bias due to the seasonality of agricultural activity.The size of each individual transaction between urban residents, whose vocations are quite varied, is not large, but the frequency of the transactions among them has less temporal bias than those of rural peasants. In this sense, the difference between economic activities between rural and urban populations lies not in whether monetary transactions are present or absent but in whether monetary transactions are conducted seasonally or constantly. In order to clarify, we leave the monetary demand in the daily life of urban dwellers out of focus and simply assume that the monetary demand among peasants is generated locally and those among merchants emerge with no confinement to a region. The volume of exchange among merchants inevitably would have been much larger than among peasants. From a quantitative viewpoint, transactions by merchants have covered most of all transactions. From the viewpoint of frequency, however, transactions by peasants have occupied the majority of all transactions everywhere. Can any single medium sufficiently mediate both exchanges among peasants and exchanges among merchants? In practice, it is quite difficult to establish a single medium of exchange that caters to both. The former requires a large quantity of currency whose value is divisible enough to accommodate the transaction among peasants, whose demand is significantly seasonal. On the other hand, the latter necessitates a smaller quantity of currency whose value is large enough to save costs for transportation, counting and hoarding, and its demand has less temporal bias. Significantly, the seasonality of agricultural productions causes the strength of monetary demand among peasants to fluctuate in relation to that among merchants. Harvest attracts currencies from the merchants’ side to the peasants’ side and currencies become idle among local peasants in the slack season. In addition, by nature, the next harvest is always unpredictable. Thus, nobody knows how many currencies are appropriately necessary to wait for the next busy season at the level of end users, i.e., local peasants. Thus, every busy season and slack season, currencies are required to move between the peasants’ side and merchants’ side. Here is the main cause of the difficulties in substitutions between largedenomination currencies and small-denomination currencies, known in the study of economic history as the ‘denomination problem.’ Small currencies

Introduction  3

suitable for transactions among local peasants have fluctuating demand in relation to large currencies more fitting for interregional settlements by merchants. If the exchange rates of small currencies to large ones is rigidly fixed in a proportion, harvest creates a tension between the demand and supply of small currencies unless a system for flexibility is established which would increase the supply of small currencies. Unfixed rates between two currencies for small/ local and large/interregional have an advantage in that people can avoid an interruption of exchanges due to a temporal shortage of currency, since a temporal appreciation to small currencies increases liquidity to the local marketplace where peasants bring their products. That is why, independent from large currency available for distant trade, a various form of small currency functioned locally across the world and throughout history. If tensions between demand and supply for currency strikes local transactions seasonally, why do people not arrange a temporal remedy as a way of avoiding the use of cash, such as supplying credit for deferred payment? Actually, some societies chose this way. Living in the same village or dwelling along the same lane in a town, local credits enabled acquainted persons to save the use of currencies in various ways; again, across the world and throughout history. Importantly, the credit transactions and the transactions by currency are not perfectly substitutive. The former enables a buyer to defer payment to a seller, while the latter provides a seller and a buyer with instant settlement. The deferred payment is based on the expectation of subsequent transaction between a seller and a buyer, while the cash settlement enables them to freely choose with whom to enter into a relationship in the next transaction. In other words, credit entails a named relationship between a creditor and a debtor, while currency makes people freer by anonymity in making transactions. Anonymous currency or named credit is a binary option when making transactions, since they represent two basic desires by humans in different directions, freedom or certainty, as we will argue. Another difference between currency and credit is the minimum space required to generate the device for mediating transactions. It is globally common that a rural marketplace attracted peasants within a half-day’s walking distance of a given space, as we shall see later. It is widely known that, in a local marketplace, sales by credit are rare and sales by cash are dominant. People coming to a marketplace have a clear preference for one-time transactions. An anonymous environment in which currency functions needs a space attracting a certain size of potential economic participants. In reality, the local marketplace often works as a minimum unit for a native currency to mediate transactions. On the other hand, cohesive human relationships that locally create credit emerge in narrower spaces. Local credit preconditions intense interactions among the inhabitants that builds a bond that constrains subsequent transactions. From the side of the debtor, the personal link is so strong that he has to avoid any conduct that might disconnect him. From the side of the creditor, the cohesiveness leads him to respect the honour of the debtor. Under the circumstances, lenders

4  Introduction

often charged no interest. The difference in optimal space between anonymous currency and cohesive credit relates to the difference of institutional settings according to society, as we will consider later. In either an anonymous or cohesive way, with less specialization of productions, local exchanges generate money. In contrast to proximate trade, depending on the specialization of products between different regions, distant exchanges occur. In some cases, a good directly exchanges for another with no monetary unit, such as a case where x pounds of wheat barters for y bushels of raw cotton. In some cases, a commodity also trades for another directly but is measured in terms of a monetary unit, such as a case where x pounds of wheat priced at one dollar trades for y bushels of raw cotton priced at one dollar. We may call the former pure barter, while calling the latter counter trade (Tschoegl 1985). In either case, the condition of direct exchange of two commodities limits the opportunity for both possessors to reach an agreement. Apparently, the introduction of a currency that dealers from different regions can accept eliminates this limitation. Thus, merchants try to establish a currency available beyond a region by themselves. In another way, merchants create a system for deferred payment between distant places, such as bills of exchange. However, in history, merchants more often used currencies exogenously supplied. States issue the currencies with which they collect taxes and pay their expenses. As long as the exogenous currencies maintain acceptability, merchants have no incentive to make endogenous currencies. States also issue certificates proving tax payments or guaranteeing trades of goods that they monopolize. Merchants used the papers or tablets officially given to settle transactions made between distant regions under the administration of the same state. Not only merchants but also peasants used the currencies supplied by states. As long as states provide currency stability, peasants relied on it in making exchanges as well as in paying taxes. However, states were not always positive to supply currencies sufficiently to circulate at the ground level. Some states relied on forced labour and/or taxation in kind. State administrations may affect local societies to proliferate marketplaces in rural areas by increasing currency supply or may induce villagers to form a cohesive intercourse among them by discouraging monetary usages. In the case of a state relying on a currency to collect taxes and pay expenses at the ground level, currencies had to circulate among local end users (peasants) and simultaneously had to facilitate the transmission of large value at a distance, as well. States try to regulate the relationship of two currencies, but always face the difficulty of fixing them. Again, primarily, money is a means of exchange. The difference among exchanges means that the way of generating money must not be of unicity. The framework of exchange devised by Aristotle ignores exchanges among local peasants and subsequently causes scholarly attention to avoid varieties of exchange and multiplicities of money. As the manner of making money endogenously is diverse, the method of supplying money exogenously might not be uniform. When considering the nature of money, we should consider

Introduction  5

the institutional setting that adjusts bottom-up need, which generates endogenous means of exchange, and top-down enforcement, which creates a means of payment.

2  Four quadrants of exchange Rephrasing our arguments in the previous section, all exchanges are subject to two sets of binary options. The first is the degree of familiarity between participants: an exchange can occur either anonymously or within a named relationship. As we use this term, the degree of familiarity does not refer to whether a seller and a buyer are acquaintances or strangers but rather to whether or not the act of making a transaction creates a bond that constrains subsequent transactions.We use ‘anonymity’ to denote a high degree of freedom in making subsequent transactions. Through this book we will demonstrate that a currency represents anonymity. Or, to put it differently, a currency makes people freer in making transactions. In contrast, we use ‘named relationship’ to indicate a seller and buyer’s certainty that they will engage in future transactions. This expectation of subsequent transactions within the same relationship enables them to defer the settlement of each transaction. Such certainty, however, comes at a cost: namely, limits on their freedom to choose with whom to enter into a relationship within subsequent transactions. Credit entails a named relationship between a creditor and a debtor. From this volume’s viewpoint, the most important difference between currency and credit is that the former fulfils a desire for freedom, while the latter entails a different desire for certainty. The difference does not lie in whether or not there is payment of interest as we would now commonly understand it; deferred payments in many historical cases did not involve any payment of interest at all, as mentioned in the previous section. The second binary option conditioning an exchange is its distance: a trade can be made either proximately or distantly. By distance we do not refer simply to physical distance but rather to whether people can negotiate directly and multilaterally with other parties or can negotiate only indirectly and bilaterally through a third party. In other words, whether the parties concerned can negotiate all at once (proximate exchange) or they can negotiate only through a broker (distant exchange) results in different characteristics of exchanges.This criterion relates to the relative value of the exchange and its frequency. Proximate exchanges tend to be of lesser value and are made more frequently than distant ones. Distant trades may occur sporadically and be settled immediately in full, or they may occur regularly through deferred payments. Exchanges in these two ways can happen in proximate exchanges as well. However, regardless of whether they are settled immediately or through deferred payment, proximate exchanges take place constantly. Especially when a local cluster of constant exchange makes its own means independent from the means for another cluster, a kind of moneta franca becomes necessary to bridge the two local trade circuits interregionally (Guyer 2004, 37).

6  Introduction

Importantly, settlement of proximate exchanges can depend either on local currency (instant settlement) or on local credit/debt (deferred payment). The former is an anonymous means of settlement, while the latter is a named one. The settlement for distant exchanges can be the same. Thus, the two characteristics of exchanges are related to each other. There are, therefore, four different kinds of exchanges, which we can represent in a diagram with four quadrants.3 People in societies have tried to devise different devices to mediate all four kinds of exchanges. A typical case appears in Figure I.1, in which a large denomination currency is used for anonymous/distant exchanges (Quadrant I), bills of exchange for named/distant exchanges (Quadrant II), bookkeeping for named/proximate exchanges (Quadrant III) and a small denomination currency for anonymous/proximate exchanges (Quadrant IV). In reality, the four quadrants seamlessly cover all types of exchanges, and a given device thus may mediate exchanges characterized by one quadrant in one context only to function later in a different quadrant. For example, bills of exchange that were initially characterized by Quadrant II could sometimes function as currencies in Quadrant I or even in Quadrant IV in special circumstances (Ashton 1945). The unit of account works in all four quadrants. Significantly, however, the four quadrants do not always share the same unit of account even under the same sovereignty. In other words, it is not easy for a state to establish a universally accepted unit of account covering all four kinds of exchange. The diversification of monetary unit outcomes from two natures of money proceed as follows: First, the separation of a monetary unit often appeared due to different patterns in the physical circulation of currencies. This diversification results from considerable differences between the sporadic and expensive exchanges that characterize Quadrant I and the frequent and fragmental exchanges of Quadrant IV. The coexistence of different monetary units between local currency (Quadrant IV) and interregional currency (Quadrant I) were popular until the end of 19th century, such as copper coins in terms of wen and silver by weight

II. Named/Dista nt Bills of Exchange

I. Anonym/Dist ant Large Currency

III. Named/Proxi mate Book Keeping

IV. Anonym/Pro ximate Small Currency

Figure I.1  Four quadrants of exchange

Introduction  7

in terms of liang in China (Kuroda 2018). The difference of units mostly represented the unfixed proportion in the conversion of two currencies. Considering the difficulty in catering to both proximate exchanges and distant exchanges mentioned earlier, it is not so strange that the exchange among peasants and that of merchants require different monetary units. Second, the divergence results from the presence or absence of physical substance. While almost no item representing the unit circulates in Quadrants II and III, currencies with significant physical substance denominated in the unit mediate transactions in Quadrants I and IV. Owing to the stagnant nature of currency, for which we will argue later, currencies chronologically become scarcer and actual physical contents of currencies tend to deviate from the unit of account in which people make transactions.4 In extreme instances, which often happened, the unit continued to work after the item that had originally represented the unit had disappeared from the marketplaces. ‘Imaginary money’ frequently appeared across the world and throughout history. As a result, it was not uncommon that merchants kept their book with disappearing currency factored in (Quadrant III), meanwhile locals already traded with new currencies (Quadrant IV). The same relationship sometimes happened in the case of distant exchange (Quadrant II and I) as well, such as that particular old silver dollar remained as a monetary unit in a distant trade of a particular commodity while actual payments were made by new currencies. Interestingly, unlike the relationship between distant/anonymous exchange (Quadrant I) and proximate/anonymous exchange (Quadrant IV), the case of different monetary units between distant/named exchange (Quadrant II) and proximate/named (Quadrant III) did not appear so prominently. The distinction between Quadrants II and III, both of which feature no significant physical substance, is especially fuzzy.

3  Exchange and institutional setting So far, given the condition that peasants can exchange their products by themselves, we have argued for a variety of exchange and multiplicity of money. However, peasants of all societies do not have the freedom to participate in the marketplace to the same degree. In other words, to what extent peasants can exchange their products differs by society and according to period. An exchange means that items equivalently valued by a unit are taken and given, usually simultaneously, between two parties. If the items are taken and given with no measurement of value exchange often and at different times, we used to define them as mutual gifts. In some cases, as Mauss (2012) insisted, anything more than the equivalent should have been expected to return. In some cases, however, presenters expected an equivalent gift in return. In this sense the boundary between ‘exchange’ and ‘gift’ is blurred. On the other hand, a unit of account can be used to measure values of items and services in a redistributive system as well as pricing them in a market system. Unlike what happens in exchanges between independent persons, an organization such as a village community or an authoritarian state decides the values when redistributing

8  Introduction

goods and services. As long as the same unit was used to measure values in the marketplace, a redistributive system and a market-based economy could interact with each other. We may call a social system depending on mutual gifts a ‘custom economy,’ a system of exchange a ‘market economy’ and a system of redistribution an ‘order economy’ (Hicks 1969). Noting that exchange is just one form among the three ways of connecting production with consumption, through this volume, we keep the focus on the variety of exchange that appeared throughout global history. The boundaries among the three economies are fuzzy, and the three activities are interactive. An institutional setting including a state system not only determines the relationship among the three ‘economies’ mentioned earlier but also configures the combination of the four types of exchange differently according to society. A focal point in the institutional setting is what works as the standard unit for proximate exchanges at the ground level. As already argued in the previous section, the population and the space for local credits tends to be less than those of local currencies are. Different ways of proximate exchanges relate to the diversity of human communication. As late as the early modern period, oral contracts had been crucial for named/ proximate exchanges (Quadrant III). Payment by transfer from one account to another through a financier became popular in late Middle Ages Europe. However, in this procedure, an oral statement in front of the banker continued to be common. In the case of Venice, the oral statement remained compulsory (van der Wee 1977, 311). Litigation records in local market towns and manors in 17th century England reveal that most transaction agreements were oral. This direct and instant exchange through face-to-face contact should be distinguished from indirect and delayed exchange through some sort of a broker. Historical facts across the world tell us that the space and population for anonymous/proximate exchange (Quadrant IV) is wider and larger than the space and population for which people can communicate only orally. ­Fourteenth-century English counties had marketplaces that served an area of 20 or 30 square miles (Hilton 1983, 168). A warlord in late 16th-century Japan prohibited the establishment of a new rural market within 12 kilometres of an existing market (Toyoda 1952, 318). Tenants in 18th century France carried grain to the nearest market town or to any other place within a radius of three leagues (Le Goff 1973, 98). Rural markets in early 20th century China attracted people who lived within half-day’s journey (Kuroda 2018). All suggest that most villagers could attend the venue for exchange within an easy day’s travelling distance. Importantly, a typical space for the circulation of endogenous currency appeared to be around the same size. For example, native notes issued by shops in 1930’s Tongshan, China, circulated as widely as 40 or 50 square kilometres in rural areas, as we will see in detail in Chapter Five (Jin 1931, 24). Not only the exchange of goods but also the exchange of services has variety. In agricultural local societies, one of the most essential issues is how to manage the labour shortage during the busy season, such as the harvest.Village life that

Introduction  9

depends on face-to-face contact with neighbours obviates hiring a stranger as a day laborer even during the busy season. Meanwhile, the rural society depending on an accessible marketplace had no reservations about hiring day laborers. The latter needs cash payment on the day work is performed, while the former inclines a person to offer employment for a longer period of time, weekly and monthly, unless they conduct mutual aids with other villagers when additional labor is necessary.The difference in the duration of employment determines the appropriate denominations of coin differently between the two parties.The latter requires currencies whose value is divisible enough to mediate the demand/ supply of labour which fluctuates day by day. Thus, actually, the denomination of currencies popularly circulating is dependent on social relationships among the end users of the currencies.The diversity clearly indicates that money is not neutral to human associations at all. The different extent of space for local credit and local currency reflects also the different size of the population conducting trades. Some contemporary social experiments reveal that increasing the population involved in exchanges changes the measure of human communication. A late report of local money movement in contemporary Buenos Aires interestingly showed that there was clearly a boundary between a stage of keeping records among a small number of the well-acquainted and another stage of issuing tangible devices for circulation among a larger number of participants (Gomez 2009). The initiating condition of this experience implies that the boundary between personal exchanges and mutual gifts is fuzzy. The following process implies that, as the frequency of transaction increased, exchanging goods between the acquainted through keeping records transformed binary personal exchanges towards multilateral exchanges. It shows that a society changes its basic unit for proximate exchanges in a short time according to circumstance.

4  Global history of money viewed from the ground The concept of four quadrants of exchanges gives us a new viewpoint for globally revisiting the history of money. The standpoint enables us to understand that money reflects an institutional setting that combines bottom-up needs of means of exchange with top-down enforcement of means of payment.5 The binary option by familiarity signals the different ways of organizing society. Another binary option by distance signifies to what extent a state interferes with a society at the ground level. This multilateral scope makes clear that a global history must have both phases of convergence and divergence and it should go beyond a history of ‘globalization’ which focusses on convergence of any index such as commodity prices (O’Rourke and Williamson 2000). Worldwide transformation of the configuration of the four quadrants reveals that global monetary history is marked by three turning points: c.1300 under the Eurasian Mongol regime, c.1600 with the global silver march and c.1900 under the international gold standard system. All three periods experienced significant growth of large currencies mediating anonymous/distant exchanges

10  Introduction

(Quadrant I). Importantly, societies seriously re-configured the four types of exchanges in the period of the contraction of Quadrant I following the period of expansion.The difference in the ways of softening the destruction by shrinking large currencies, eventually, formed different types of economies. Interaction between different economies molded a world economy, and a reorganization of them shaped another type of world economy. Each civilization followed its own pattern in configuring the four quadrants of exchange until the 13th century.Then the Mongol regime established a common unit of account denominated in silver for long-distance exchanges across the Eurasian landmass. The common unit of account in terms of silver did not accompany a popular circulation of silver items and did not seriously affect the method of exchange at the ground level. The significant contraction of distant trade after the collapse of the Mongol regime, however, precipitated the divergence between currency-dependent exchange and credit-dependent exchange at the ground level and the discrepancy paved the way for the global silver march starting from South America and terminating in China in the late 16th century. Global silver flows of unprecedented size from the 16th century through the 17th century significantly inflated the sphere of anonymous/distant exchange. The excessive liquidity in Quadrant I could not but affect the way of proximate exchanges. Importantly, the contraction of the silver flow after an extraordinarily affluent currency supply in Quadrant I divided local economies across the world in three directions: (1) a currency-oriented path (increasing local currency supply in anonymous/proximate exchanges), (2) a credit-oriented path (intensifying the role of local credit in named/proximate exchanges) and (3) dependence on commercial oligarchies that monopolized access to anonymous/distant exchange.Through interactions among these three paths, state systems of organizing debts nationwide (the named/distant exchanges linked to the named/ proximate exchanges) emerged and the banking system originating from named exchanges began to supply anonymous currencies in the form of banknotes. The international gold standard system that dominated before and after 1900, when national economies had central banking systems, which blurred the boundaries between the four quadrants within each national territory, extended their organizations out to assimilate the rest of the world in which the functional division of the four quadrants apparently still worked. Covering a multiplicity of exchanges within a territory, in tandem with a managed currency system, state monopoly of monetary supplies began to prevail across the world. Through this process, the notion of a single currency under a sovereign authority gained common understanding. Organizing the means of exchange in endogenous ways from the ground level became illegal, transitional and negligible, though it did not completely disappear. Passing through the three epochs mentioned earlier, proximate exchanges eventually lost their opportunity to generate their own money. From the opposite viewpoint, distant exchanges organized by states finally absorbed autonomous liquidity from the ground level. Through the transformation, in contrast with the fact that empires allowed local currency circuits to flourish, nation state systems strengthened the exchanges through credit in which local and

Introduction  11

distant exchanges (Quadrant II and III) had less tension between them than exchanges through currency (Quadrant I and IV). As a result, now, we take it for granted that states maintained a monopoly on money. History tells us, however, that until recently no sovereign state succeeded in excluding the money endogenously made at the ground level. On the other hand, history also reveals that endogenous money generated by exchanges among end users could not by itself build up an integral system beyond local circuits. It is difficult for a single medium, provided exogenously or supplied endogenously, to satisfy the monetary demand for various exchanges. Thus, by nature, one form of money cannot completely substitute another. That is why monies often cooperated in a complementary relationship throughout history (Kuroda 2008a, 2008b).

5  Bad money does not drive out good money: literature The concept that bad money drives out good money is so popular that people often mention it when considering monetary transformations. Very few, however, pay attention to the idea that preconditions a substitutive relationship between the two monies. In theory, unless bad money functions perfectly in the same way as a good money, the introduction of bad money would not necessarily drive out good money. In history, a new currency of inferior content often circulated together with an old currency of superior content, as we will see throughout this volume. As suggested in previous sections, it is not easy for a single device to mediate all kinds of exchange. It means that one money can do what another cannot, and vice versa. Under such conditions, the multiplicity of money is not accidental but functional, since those monies comprise a complementary relationship, not the substitutive one which Gresham’s Law presumes. Even when bad money supplied the demand for mediating transactions among end users, it often failed to adapt sufficiently to other demands. That is why bad money often worked in concert with good money.6 In history, peasants comprised the majority of the end users of currencies. It is true that some peasants were either self-sufficient or dependent enough on the communal redistribution of products to have few occasions to use currencies. Unlike the image of peasants projected by the arguments along ‘the moral economy’ or ‘rational peasant,’ however, significant numbers of peasants engaged in monetary transactions by their own decisions. The point is that their business has great seasonality. Relating to this point of monetary demands by peasants, this book revives an argument that attracted attention in the past but has now become almost overlooked. Before World War I, some scholars paid attention to monetary tensions caused by seasonality (Kemmerer 1910). However, by decreasing the effects of seasonality, industrialization made proximate exchanges less different from distant exchanges in this regard. In such a situation as the 1920s USA, seasonality in economic activity was still taken seriously as a factor in causing unemployment, though it was not considered a cause of monetary problems (Kuznets 1933).

12  Introduction

In agricultural societies especially, proximate exchange trade is more affected by seasonality than distant exchange. As we will see throughout this volume, agricultural cycles seriously affect monetary demands.7 Owing to the physical movement of currencies, seasonal fluctuations often brought tensions between supply and demand of currency to the local market. In contrast, credit systems usually worked to absorb temporary oscillations, though it was always accompanied by the risk of non-performance. As we will see, even in the early 20th century US the harvest in autumn caused interest rates in New York to spike due to a large flow of silver coins to the West (Miron 1986). When the inelasticity of supply with a local currency is significantly lower than with interregional currency, the opportunity cost can be lower with a local currency than without a local currency. Seasonality and temporality of monetary demands by end users of currencies is a key factor in understanding the multiplicity of money in history.This point gives us a clue to the latest topic argued on money in history. Recent studies on the ‘denomination problem’ in economic history have revealed difficulties in substitutions between large denomination currencies and small denomination currencies. In the case of early modern Western Europe, Redish (2000), building on Cipolla (1967), the focus has been on minting costs and the intrinsic metallic content of coins when arguing that there is an imperfect substitution of denominations; in other words, one denomination cannot do what another does. Sargent and Velde (2002) have also created a model based on the expectations of coin holders about the exchange rate between large and small coins and have found that shortages and the depreciation of small change occur simultaneously. Their emphasis on the importance of steam-press minting in the creation of a unified currency system, however, is inconsistent with the fact that private local tokens were in popular circulation in England from the late 18th century to the early 19th century (Whiting 1971; Selgin 2008). It is important to note that the imperfect substitution of denominations is not a phenomenon unique to metallic coinage, and that it has varied according to local circumstances around the world. As we will see in Chapter Four, local preference for small denomination currencies appeared even in the case of paper monies. This book examines the global prevalence of complementary relationships between local, small denomination currencies and interregional, large denomination currencies. I show that the combination of a strong seasonality in monetary demand and the disinclination of small currencies to assemble on demand in marketplaces has often resulted in shortages of the means of payment and thus led local traders to create autonomous currencies to ease those shortages. For example, in late Imperial China, copper coins seriously appreciated against silver during the harvest, but the chronological shortage of copper coins caused native notes issued by local dealers to circulate locally. These arguments are supported by the high wastage rate of currencies (a phenomenon discussed, though not emphasized, by Redish) and by variations in the degree of wastage of different denominations. The co-existence of different types of currencies and the relationship between currency and credit have also been studied theoretically. Developing

Introduction  13

search theory, Kiyotaki and Wright (1993) have demonstrated the theoretical possibility of dual currency circulation under conditions in which a currency of universal acceptability and a lower return rate is balanced against another of partial acceptability and a higher return rate. Shi (1996) has extended the model to incorporate credit and generated multiple equilibria between currency and credit depending on the means of repayment. Meanwhile, historical cases of local tension between demand and supply for currency, such as the examples mentioned earlier, suggest significant spatial bias in the probability that demand for currency meet its supply. This book will also examine two cases in which, regardless of the return rate, currencies with a high acceptability in one place did not have the same degree of acceptability over a broader area. Under certain conditions, local credit could act as a substitute for these currencies. Although theoretical studies of the relationship between credit and currency often assume that charging interest provides an incentive to the credit provider, in most of the cases studied in this book credit did not bear interest, suggesting that it served a different purpose.This book also examines the social foundations of local credit networks in which lenders did not charge interest. The multiplicity of money inevitably leads us to doubt the idea that only states can provide money, since a sovereign power only accepts its own money. Theoretically, Hayek (1976) argued that money could be stably provided without state intervention. However, he did not know that monies do not necessarily compete, as they can also work together in complementary relationships. In an historical study, Helleiner (2003) revealed that a single currency system within a state is a product of modern history. The current book shows why premodern monies were in a state of heterogeneity. In situations in which the state already monopolized money, people could differentiate fungible coins according to the purpose of expenditure, as demonstrated by Zelizer (1994) who showed that ordinary people in the early 20th century US put earmarks on their money to designate it for different uses.The historical examples in this book show that such earmarking was the result of monies losing their variety due to the disappearance of complementary currencies. The multiplicity of money also means that the market does not unify monies by itself. As demonstrated by Desan (2014) in her reconsideration of ­English monetary history, transformations of monetary systems have historically resulted from political and social conditions rather than the evolution of market activities. Exploring the idea that money is a chain of credit/debt relationships. Ingham (2004) understood the emergence of Bank of England notes as a transformation of the sovereign’s personal debt into public debt. The system in which central banks monopolized banknote issuance spread across the world, but such methods of organizing private debts under public debt did not proliferate in the same manner. In reality, neither top-down systems designed by states nor bottom-up procedures of collective individual choices can create a stable means of exchange. Investigating cases of monetary issues in West Africa, Guyer (2004) gave us the important insight that certainty in transactions does not always bring stability. Rather than considering either a macro or micro viewpoint to be decisive, we

14  Introduction

should instead pay closer attention to the mezzo-level conditions that create patterns of exchange. In sum, this book provides the following new understandings of money. (1) The denomination problem (the imperfect substitution of large and small denomination currencies) is due to the multiplicity of money. (2) The plurality of money results from the variety of exchanges. (3) The different features of proximate and distant exchanges give currencies different degrees of acceptability in local and interregional trade. (4) The extent to which inhabitants make transactions cohesively or anonymously determines whether they will tend to make exchanges through either instant settlement or deferred payments. (5) Anonymous exchanges result in piles of inactive currencies, while cohesive ones conceal non-performance. Beyond pure monetary issues, also, a revision of global monetary history from the ground level will reveal the shortcomings in the conventional framework for understanding human activities. Unification of money seems the obvious way to reduce transaction costs significantly, since we can save the costs of conversion between monies. The geographical expansion of a single currency seems to promise more effective specialization according to comparable advantages as the idea of the optimal currency zone by Mundell proposed. A fixed monetary system appears to increase the degree of certainty enough to attract traders to invest with greater confidence (Mundell 1961). Certainty, however, does not always bring stability. Unfixed monetary institutions might have provided flexibility in a situation of unexpected fluctuations. A social institution is established not only for increasing certainty, as is now popularly thought (North 1990), but is also generated for maintaining flexibility against an unknown future. Humans have always been wise to balance certainty and flexibility.

Notes 1 ‘An association for exchange is formed not from two doctors but from a doctor and a peasant,’ thought Aristotle in his treatise on money (Aristotle 2004, 125). Numerous thinkers on money throughout history, including great names such as Thomas Aquinas, Karl Marx, Knut Wicksell and Karl Menger, constructed their theories almost entirely along this line. 2 A typical example is the framework of Smithian Growth, which suggests that as ‘the extent of a market’ increases, a division of labour spreads horizontally beyond villages, municipalities and nations. The concept puts emphasis on the proposition that specialization enhances the economy of size. Specialization means the division of labour. The division of labour is limited by ‘the extent of the market.’The degree of disintegration in production is reflected in the extent of the market. The extent of the market increased according to the reduction of transportation cost. Sticking to a concept of specialization according to comparative advantages, this idea takes no account of the possibility that a human might organize his or her labour for different types of work, for example cultivating rice in spring and summer while spinning cotton in winter. Stigler 1951; Kelly 1997. 3 I thank Angela Redish for suggesting that I formulate two binary options into a matrix. 4 In early modern Europe, which was suffering from a chronic shortage of mono-unit currency and an insufficient metal supply, the debasement of mono-unit currency was inevitable in order to avoid a possible deflation. Fantacci 2008.

Introduction  15 5 As Théret argued through Argentina’s cases of federalism, since there can be an overlapping hierarchy of power, the enforcement of the means of payment can be multiple and complementary (Théret 2018, 87). The multiplicity of power in monetary issuance was also found in the case of the relationship between the Imperial Estates and the Imperial Diet under the Holy Roman Empire. North 2016. 6 For example, in the 17th century Netherlands East Indies, ‘heavy money’ of 60 stivers and ‘light money’ of 75 stivers co-existed. Wolters 2008. 7 Detailed analysis of remittance by post reveals what monies were actually sent in 19thcentury Sweden. An interesting result of the study is that the issuance of local banknotes corresponded with a seasonal demand for money. Engdahl and Ögren 2008.

References Aristotle. 2004. The Nicomachean Ethics. London: Penguin. Ashton,TS. 1945.‘The bill of exchange and private banks in Lancashire, 1790–1830,’ Economic History Review 15(1–2): 25–35. https://doi.org/10.1111/j.1468-0289.1945.tb00705.x Cipolla, C. 1967. Money, Prices and Civilization in the Mediterranean World: Fifth to Seventeenth Century. New York: Gordian. Desan, C. 2014. Making Money: Coin, Currency, and the Coming of Capitalism. Oxford: Oxford University Press. Engdahl, T. and Ögren, A. 2008. ‘Multiple paper monies in Sweden 1789–1903: Substitutive or complementary?’ Financial History Review 15(1): 73–91. https://doi.org/10.1017/ S0968565008000061 Fantacci, L. 2008. ‘The dual currency system of Renaissance Europe,’ Financial History Review 15(1): 55–72. https://doi.org/10.1017/S096856500800005X Gomez, GM. 2009. Argentina’s Parallel Currency:The Economy of the Poor. London: Pickering & Chatto. Guyer, JI. 2004. Marginal Gains: Monetary Transactions in Atlantic Africa. Chicago: University of Chicago Press. Hayek, FA. 1976. Denationalisation of Money. London: Institute of Economic Affairs. Helleiner, E. 2003. The Making of National Money: Territorial Currencies in Historical Perspective. Ithaca: Cornell University Press. Helleiner, GK. 1966. Peasant, Agriculture, Government, and Economic Growth in Nigeria. Homewood, IL.: R.D. Irwin. Hicks, JR. 1969. A Theory of Economic History. Oxford: Clarendon Press. Hilton, R. 1983. A Medieval Society: The West Midlands at the End of the Thirteenth Century. Cambridge: Cambridge University Press. Ingham, G. 2004. The Nature of Money. Cambridge: Polity Press. Jin, WJ. 1931. Tongshan nongcun jingji diaocha. Zhenjiang: Peasant Bank of Jiangsu Province. Kelly, M. 1997. ‘The dynamics of Smithian growth,’ Quarterly Journal of Economics 112(3): 939–64. Kemmerer, EW. 1910. Seasonal Variations in the Relative Demand for Money and Capital in the United States. New York: Arno Press. Kiyotaki, N. and Wright, R. 1993. ‘A search-theoretic approach to monetary economics,’ American Economic Review 83(1): 63–77. Kuroda, A. 2008a. ‘What is the complementarity among monies? An introductory note,’ Financial History Review 15(1): 7–15. https://doi.org/10.1017/S0968565008000024 Kuroda, A. 2008b. ‘Concurrent but non-integrable currency circuits: Complementary relationship among monies in modern China and other regions,’ Financial History Review 15(1): 17–36. https://doi.org/10.1017/S0968565008000036

16  Introduction Kuroda, A. 2018. ‘Strategic peasant and autonomous local market: Revisiting the rural economy in modern China,’ International Journal of Asian Studies 15(2): 195–227. https://doi. org/10.1017/S1479591418000049 Kuznets, S. 1933. Seasonal Variations in Industry and Trade. New York: National Bureau of Economic Research. Le Goff, TJA. 1973. ‘An eighteenth-century grain merchant: Ignace Advisse Desruisseaux,’ in Bosher, JF. (ed.) French Government and Society 1500–1850. Essays in Memory of Alfred Cobban, 92–122. London: Athlone Press of the University of London. Mauss, M. 2012. Essai sur le don: forme et raison de l’échange dans les sociétés archaïques. Paris: Presses universitaires de France. Miron, JA. 1986. ‘Financial panics, the seasonality of nominal interest rate and the founding of the Fed,’ American Economic Review 76(1): 125–40. Mundell, R. 1961. ‘A theory of optimum currency areas,’ American Economic Review 51(4): 657–65. North, DC. 1990. Institutions and Institutional Change and Economic Performance. Cambridge: Cambridge University Press. North, M. 2016. ‘Monetary reforms in the Holy Roman Empire in the fifteenth and sixteenth centuries,’ in Fox, D. and Ernst, W. (eds.) Money in the Western Legal Tradition: Middle Ages to Bretton Woods, 191–9. Oxford: Oxford University Press. doi:10.1093/acprof: oso/9780198704744.003.0010 O’Rourke, KH. and Williamson, JG. 2000. ‘When did globalization begin?’ NBER Working Paper, No. 7632. Redish, A. 2000. Bimetallism: An Economic and Historical Analysis. Cambridge: Cambridge University Press. Sargent, TJ. and Velde, FR. 2002. The Big Problem of Small Change. Princeton: Princeton University Press. Selgin, GH. 2008. Good Money: Birmingham Button Makers, the Royal Mint, and the Beginnings of Modern Coinage, 1775–1821. Ann Arbor: University of Michigan Press. Shi, SY. 1996. ‘Credit and money in a search model with divisible commodities,’ Review of Economic Studies 63(4): 627–52. https://doi.org/10.2307/2297797 Stigler, G. 1951. ‘The division of labor is limited by the extent of the market,’ Journal of Political Economy 59(3): 185–93. www.journals.uchicago.edu/doi/10.1086/257075 Théret, B. 2018. ‘Monetary federalism as a concept and its empirical underpinnings in Argentina’s monetary history,’ in Gomez, GM. (ed.) Monetary Plurality in Local, Regional and Global Economies, 84–113. London: Routledge. Toyoda, T. 1952. Chusei Nihon shogyo shi no kenkyu. Tokyo: Iwanami. Tschoegl, AE. 1985. ‘Modern barter,’ Lloyds Bank Review October: 32–40. van der Wee, H. 1977. ‘Monetary, credit and banking systems,’ in Rich, EE. and Wilson, CH. (eds.) Cambridge Economic History of Europe, vol. 5, 290–392. Cambridge: Cambridge University Press. https://doi.org/10.1017/CHOL9780521087100.006 Whiting, JRS. 1971. Trade Tokens: A Social and Economic History. Newton Abbot: David & Charles. Wolters, W. 2008. ‘Heavy and light money in the Netherlands Indies and the Dutch Republic: Dilemmas of monetary management with unit of account systems,’ Financial History Review 15(1): 37–53. https://doi.org/10.1017/S0968565008000048 Zelizer,V. 1994. The Social Meaning of Money. New York: Basic Books.

Part 1

Exchanges generate money locally

1 Peasants, marketplace and money

1  Peasants exchange with peasants anonymously This volume pays attention most importantly to exchanges among peasants, who represented most humanity throughout history, since substantial exchanges among peasants generated money at the ground level. Before considering a variety of monetary usage in the bottom portion of societies, we should confirm how seriously ordinary people in the period before industrialization, even in rural localities, engaged in exchanges. This chapter demonstrates the importance of proximate exchanges made by ordinary peasants with other nearby peasants and how money was accordingly generated endogenously for these local exchanges. Unlike the image of anti-market peasants in the moral economy theory proposed by J. Scott, or the different image of corporative peasants in the rational peasant argument advanced by S. Popkin, typically, the case of Chinese peasant households showed market-dependent and individual strategies (Scott 1976; Popkin 1979; Kuroda 2018). Mostly, indeed, peasants consume their products themselves. They, generally speaking, eat grains in the home rather than sell them. However, the tendency towards self-consumption by cultivators does not exclude the possibility that they need to purchase grains in addition to their own products. In reality, a peasant cultivating a grain sometimes purchases the grain from another peasant. For example, in the case of the Huangling village, Taiyuan County, Shanxi province, early 20th-century China, millet was the main grain cultivated by villagers. Among 71 peasant households who produced millet, 15 peasant households also purchased millet (KAKO 1944, appendix 12–15, 18–21). Not rarely, across the world, peasants purchased grains that they planted most commonly in their district. Even if we assume an extreme case that all peasants in a region cultivate a grain, some may have a good enough harvest to enjoy a surplus, while some may suffer shortages due to a bad harvest. As considered already in the introduction, potentially, exchanges between peasants can occur proximately without specialization of production. Peasants consumed, sold and reserved their products in different ways according to society and period. Although we shall make a comparison between different societies later, here, we focus on the pattern that peasants had a certain

20  Exchanges generate money locally

degree of freedom to sell crops by their own will. Where and how did peasants sell the grain? Take a closer look at the case of the Huangling village mentioned earlier. There, local peasants brought grains to the marketplace which was held every other day in their village. Note that all villages did not always hold a marketplace. Usually, a marketplace in China attracted peasants from ten to 20 villages located within half-a-day commutation, as mentioned in the introduction. The Huangling village happened to have a marketplace in it. As far as the main crop was concerned, peasants usually sought to bring them to the marketplace and directly negotiate prices with buyers. The majority of modern investigations of peasant households in early 20th-century China show that peasants preferred selling products in a marketplace to selling to merchants in their own homes because they believed that they could realize appropriate prices in a marketplace while they might be cheated by the merchants at home.1 Usually trades in the marketplace were settled by cash on the spot. Meanwhile, the peasants in the Huangling village also produced vegetables for Taiyuan (the provincial capital,Yangqu County). Easy access to the provincial capital was a specific advantage for the peasants in this village.Villagers sold vegetables to merchants coming from the town to their homes and sometimes sold vegetables in the town by themselves. As we will discuss later, depending on situation, Chinese peasants strategically chose whether to bring their products to the marketplace or to sell directly to particular merchants. As Table 1.1 shows, importantly, either type of sale, grains or vegetables, had strong seasonality, though the bias towards grains was slightly bigger than that in vegetables. Agriculture all over the world commonly experiences a considerable degree of seasonality. There is no agriculture without a slack season when little labour supply is needed. Meanwhile, in the busy season such as the cultivation and harvest, as much labour as is required must supply the need. How to organize Table 1.1  Monthly sales of peasant products, Huangling village,Taiyuan, Shanxi, 1940 (yuan)

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Total

Grains (millet)



Sold in village

Sold for Taiyuan

Sold in yard

0 0 24 57 0 17 115 214.5 90.7 377.52 15 297.4 1208.12

20 0 7.5 0 0 100.8 103.3 99.67 369.97 401 70 149.67 1321.91

20 0 31.5 57 0 117.8 218.3 314.17 460.67 778.52 85 447.07 2530.03

0 0 16.5 42 0 3.75 126.25 192 137.2 409.02 15 284.9 1226.62

20 0 0 0 0 70.75 48.75 122.17 275.17 269.5 70 132.17 1008.51

0 0 15 15 0 43.3 43.3 0 48.3 100 0 30 294.9

Source: KAKO 1944, appendix 16–17.

Peasants, marketplace and money  21

both idle and busy times determines the characteristic of the peasantry differently society by society. During busy seasons, when facing a labour shortage, Chinese peasant households typically hired day labourers. Those who wanted to be employed and who wanted to employ both gathered at a particular place, such as the end of a bridge in the early morning and negotiated conditions: the wage of the day and whether lunch would be provided. Liu Dapeng, a literatus in Chiqiao village, which also belonged to Taiyuan County, as the Huangling village did, wrote of a typical case in his diary in which hundreds of men gathered at a bridge to look for employers during the harvest season. Liu himself was not a largescale landowner but a farmer with a middle-level-size holding in the region. However, Liu himself often hired several labourers to cultivate land and harvest crops. He sometimes recorded the amounts of wage payments in his diary, but not the names of the day labourers. His descriptions suggest that, unlike the long-term labourers called changgong, day labourers and their employers made a one-time transaction that entailed no subsequent relationship. Liu’s descriptions reflect a common situation in modern China (Kuroda 1996). Take another example from an investigation of a rural market, Wulibao, Yidu County, Shandong province. The report reveals that at every harvest for wheat in May and for sorghum in July, 30 to 500 job seekers would gather in a temple for day labour. Wages fluctuated day-by-day according to demand and supply Nishiyama 1942, 14–15). In addition to day labour markets, a custom of mutual labour exchange between peasants, called huangong, also existed in order to ease the tension between labour supply and demand during agricultural busy seasons, though it was not at all common across China overall.2 Thus, the provision of labour through ready cash payments dominated in making up labour shortages due to agricultural seasonality. In most cases day labour wages were paid in cash on the day without deferment. In the case of day labour markets in rural areas, most employers and employees mostly lived in the same village, or in neighbouring villages. For example, the diary by Liu Dapeng on 15 July 1918 records that by the bridge, called Yurangqiao, 200 workers gathered and were employed in Xizhen, Huada and Yingdi, in addition to Chiqiao (Liu 1990, 263.). Mid-July is a busy season for harvesting spring wheat. Two hundred day labourers in the region of four villages in which probably around 300 households lived was a considerable number.3 Considering the villages are located close to each other, we can assume that employers and employees were not complete strangers. However, importantly, there was a clear tendency for those who wanted to be employed to go to the mustering place outside the village to seek employers. Even if eventually a job seeker was employed by an employer within the same village, they rarely looked for jobs without going to the day-labour market (CNKC vol. 3, 194). This tendency for Chinese peasants to avoid a subsequent relationship in seeking day labour marks a clear contrast with Japanese villagers, as we will see in a later section.

22  Exchanges generate money locally

Interestingly, the investigations of incomes and expenditures of peasant households in the 1930s revealed that some households receiving income from being employed had expenditures for employing others as well.4 The relationship between an employer and an employee did not always reflect inequality between the affluent and the indigent but rather showed an adjustment between households which had surplus labour and others which had labour shortages at a particular point in time. The relationship between them might be reversed at a different time. However, direct exchanges of labour without wage payments between neighbouring peasants did not often occur. We can find a strong tendency in China for peasant households to choose one-time transactions rather than establishing subsequent cooperation with a particular household. The wages for day labour depended on demand and supply on the day, as already mentioned, but there was clear seasonality in the movement of wage levels. For example, in the case of one village in Shandong, the wage for day labour in August was 80 cents, while in January it was 20 cents (CNKC vol. 3, 194). It is often assumed that conditions of low opportunity cost forced peasants to add to their working hours in farming at an unfavourable return per working hour, even if below subsistence level (Huang 1985, 195, 201). If one considers only the wages during the season of lowest demand, the argument of farming work at a sub-subsistence level might seem persuasive. However, in the slack seasons with unfavourable conditions for agriculture, peasants quite naturally would try to find any available opportunities for income outside agriculture. Nobody can be categorized as a peasant unless they practice agriculture on land. However, it does not mean that the income of the household comes only from agricultural work. One survey showed that, among 99 day-labourers who lost agricultural employment, 91 engaged in non-agricultural work, while only eight, including gamblers and beggars, were categorized as unemployed. Interestingly, the most popular vocation was small peddler, which amounted to 37 labourers among 99 (Amano 1935, 210). An investigation in the Huangling village revealed that the peasants had no shortage of opportunities to earn cash income from non-agricultural engagements. Among 81 peasant households, in addition to 23 engaging in agricultural labour, 74 worked as coolies, 37 made shoe soles, ten worked as peddlers, four engaged in making noodles, two worked as village police and so on. In particular, shoe sole-making absorbed a considerable amount of female labour from nearly half of peasant households.5 In other villages in the same region, some produced coarse paper, some worked in local coal mines, and the like (Kuroda 1996). It is true that farming land, especially cultivating grains, was the principal occupation for most of villagers. However, their work on arable land was subject to a high degree of seasonality, and peasants organized their labour in slack times to fulfil a variety of side businesses as often as possible. Thus, we confirmed a case that peasants exchanged goods and services proximately by cash rather than mutual gifts and cooperative services. Importantly, although the exchanges were made within a walkable distance, peasants

Peasants, marketplace and money  23

conducted anonymously enough to negotiate prices day by day. In order to seek for an agreeable rate both for a seller and a buyer, they needed currencies whose value were fragmented enough to settle the negotiation in a minimum unit. In addition, the demands for cash surged seasonally. A large amount of small currencies was required on hand at marketplaces in busy seasons while most of them became inactive in slack seasons. We shall confirm the ubiquity of fragmental currencies and the strong seasonality of the demands for them across the world later. We also confirmed the multiplicity of zones for exchanges. In the agricultural busy season Chinese peasants need an exchange of labour on a scale of several villages; in selling their main harvest they need an exchange of crops on a scale of around 20 villages and in selling by-products for specific uses they needed an exchange of commodities on a scale of county or larger. The give-and-take of surplus goods and labour across proximate households with no specialization operated in the first two sizes of zone, while, in the largestsize zone, an inter-household division of labour with specialization dominated. Thus, according to these multiple exchange zones, Chinese peasants combined two kinds of trade: with or without specialization. In the following section we will examine the marketplace for peasants.

2  Marketplace for one-time transactions Working on the assumption that a transaction between different occupations and/or dissimilar products can increase productivity and enhance economic development, one would conclude that exchange among local peasants whose products are similar must be negligible. However, such an assumption seriously underestimates the actual size of exchanges among peasants. Table 1.2 shows the amount of trade both for inside-zone and for outside-zone exchange in Qingyuan County, Hebei province, China. Here the zone represents the area in which the local town attracted sellers and buyers from 4,000 peasant households. The grains sold were mainly for outside the zone, 211,600 yuan (52.9 yuan per household), but some were for inside the zone, 44,880 yuan (11.22 yuan per household). However, non-grain agricultural products were sold and bought completely inside the zone. The total amount of sales inside the zone, 320,888 yuan, was larger than those outside the zone, 230,800 yuan. In the previous section we know that Chinese peasants combined two kinds of trades: proximate exchanges with no specialization and distant exchanges with specialization. The breakdown of the amount in Table 1.2 suggests that proximate exchanges might have outweigh distant ones. Let us take a look at a lower level than a local town, that is, marketplaces in rural areas which are located closer to peasant households. Table 1.3 shows the value of commodities sold in 11 rural marketplaces in the same county, Zouping, Shandong province, by the distance of transportation from shipping places. Nearly half the value was derived from agricultural products, namely

24  Exchanges generate money locally Table 1.2 Annual sales and purchases per households in Qingyuan County, Hebei, China 1930 (yuan) Sales

Grains Agricultural byproducts Others Total


Within district

Outside district

Within district

11.22 43.28

52.9 4.8

11.22 43.28

28.22 87.72

0 57.7

28.22 87.72

Outside district 0 0 35.55 35.55

Source: Cui 1990, 57.

Table 1.3 Commodities by distance in 11 marketplaces, Zouping County, Shandong, 1932 (yuan) distance

Sunjiazhen Wangwuzhuang Huilizhuang Huagou Tianjiaguanzhuang Yaozhuang

< 25 km 25–50 km 50–150 km 150 km < Total Surplus

15,690 720 6,690 6,850 29,950 1,430

3,710 120 280 460 4,570 2,850

2,485 210 375 830 3,900 1,070

4,870 170 500 2,480 8,020 1,720


720 70 0 410 1,200 240

70 225 750 160







465 30 40 275 810 120

4,135 300 1,420 3,195 9,050 -780

5,885 180 730 2,210 9,005 2,765

6,090 280 380 1,960 8,710 3,470

1,389 20 50 620 2,079 699

44,505 2,080 10,485 18,895 75,965 13,045

Source:Yang 1934.

grains, produced from an area within 25 km from each marketplace. Adding the handicraft, native cloth, from within 25 km the major local peasant products represented nearly 60 percent of the entire commodities trade. The balance between local products and interregional products was favourable for locals, except in one marketplace, Xiaodian. In the largest case, Wangwuzhuang, twothirds of the entire value was a ‘surplus’ of local products. Generally speaking, petite dealers connecting rural markets with towns sold local products to towns and used the cash they earned to purchase products that could be sold in rural markets, rather than keep the cash.6 It might be safe to say that the ‘surplus’ in the table reflected the amount of intra-exchange between locals gathering in the marketplace, which exceeded the amount of local commodities sold to outside buyers in exchange for commodities from outside. In short, the data

Peasants, marketplace and money  25

suggests that peasants exchanged with proximate peasants more than with merchants engaging in distant trade. What were the transactions made by peasants like? A category of account book from Tongtaihao, a grocery merchant in Daliu zhen town, Ningjin County, Hebei province, gives us an interesting glimpse. The daily frequency of transactions periodically surged on the second day and the seventh day in every ten-day period (Figure 1.1). The account book called churu qian liushuizhang (hereafter ‘cash book’) appears to have been used to record all the Tongtaihao transactions, except for miscellaneous purchases for daily use and the like. Most of the names in the entries are individual and rarely appeared twice in the same year. On days other than the second and seventh the merchants’ names in the same town made up most of those appearing in the account books.7 We will consider the settlements among the merchants in the same town in Chapter Two. The difference by day strongly suggests that the town had a five-day market held on the second and seventh days and that the market attracted a number of individual petite dealers. We should remember that petite dealer was an important side-business for peasants, as demonstrated in the last section. The individual persons appearing in the account book from the Tongtaihao must be the same as dealers called xiaofan in the reports of later period (KPDF 1942, 39). Sometimes Tongtaihao sold on credit and purchased on debt with the petite dealers. Even in those cases, there was a clear tendency for both sides to settle 120 100 80 60 40 20 0

Figure 1.1 Frequency of daily transactions, Tongtaihao, 5th and 6th months, Daoguang 10th year (1830 CE) Sources: Daoguang shinian siyue churu liushui zhang and Daoguang shinian liuyue churu qian liushui zhang. Chinese National Library, Wenjinge, no. 49120–144 and no. 49120–145.

26  Exchanges generate money locally

transactions as late as the next market day, or five days later. In the case of such deferred settlements they appeared to have charged no interest.8 Even if sales credit existed between the grocery shop and peasant-dealers, deferred payments consisted of a very small part of the overall transactions. During the month between the 18th day, 5th month, Daoguang 10th year (1830 CE) and 17th day, 6th month, Daoguang 10th year, only 918,554 wen was recorded as sums sold on credit and purchased on debt with the petite dealers, while all transactions of the Tongtaihao amounted to 23,616, 946 wen. The deferred payments occupied as much as five percent of the entire deals. In addition, among 57 persons who made deferred payments to the Tongtaihao during the period, only three appeared in the account book for the deferred payment in the same period of the next year, Daoguang 11th year. During the period from the 12th day, the 1st month, Daoguang 10th year and 12th day, 7th month, Daoguang 10th year, a total of seven months including an intercalary month, 420 persons appeared only once among 483 deferred payments recorded in the account book.9 We can see that cash transactions were so dominant that sales credit rarely repeated between the same buyer and seller. Thus, the preference for one-time transactions was particularly strong in the case of transactions through the marketplace. This finding coincides with the observations through modern rural investigations that peasants (and petite traders) made no subsequent transactions with particular merchants (CNKC vol. 4, 228–9). What was the merit in persisting with one-time transactions for peasants? Rural investigations such as Chugoku Noson Kanko Chosa always show that Chinese villagers were very keen observers of price movements in the marketplaces they accessed. This observation is supported by Liu Dapeng’s diary, in which he frequently recorded the price of grains in rural market, and described the movement of prices in a local town, the Jinci zhen, which was located within walking distance from his village, Chiqiao.10 Importantly, when they judged that the price in the marketplace was not favourable, they often brought their products back home and waited for prices to spike. Some cottongrowing peasants did not harvest raw cotton all at once, but batch-by-batch every five days in order to wait for favourable market conditions (CNKC vol. 3 324). Investigations of peasant households of three villages in Northern China, 1939–1940, show that peasants retained a quarter of the value of all products neither consumed nor sold. In Michang village in Hebei, 47.4 percent of total products were sold, 26.3 percent consumed and 25.8 percent reserved (MHKC 1939, 26, 38). In Ma village in Hebei, 43.5 percent were sold, 34.1 percent consumed and 22.4 percent reserved (MHKC 1940, 20, 26). In Wuguan village in Henan, 53.8 percent were sold, 28.5 percent consumed and 17.7 percent reserved (MHKC 1942, 16, 23). Those villages mainly cultivated grains and raw cotton. Naturally, grains occupied the majority of the reserved products. In the case of Michang village 50 percent of grains were neither sold nor consumed (MHKC 1939, 38–9). However, of the raw cotton that had been thought to be completely sold soon after the harvest by

Peasants, marketplace and money  27

peasants desperately seeking cash, in reality one tenth was also left unsold or unconsumed. Interviews with villagers through rural investigations in Hebei, Shandong, Shanxi and Jiangnan revealed four points: first, villagers seldom exchanged one commodity for another commodity, in other words, they rarely engaged in direct barter, but usually exchanged through transactions in cash; second, villagers avoided selling their products within their village as long as they could go to a marketplace; third, villagers showed no tendency to borrow money in the same village; fourth, villagers did not purchase necessities from the same merchants to whom they sold their products (CNKC vol. 4, 228). All four features support the conclusion that Chinese peasant households heavily depended on one-time transactions through multiple and direct negotiations in anonymous settings. A marketplace accessible by half-a-day’s travel provided an indispensable matching opportunity among local peasants. A type of peasant tended to increase the sale of products to the outside as they increased cash income, while another type kept selling grains within the local market regardless of the amount of cash income. Apparently, the former needed the merchants’ connection with the outside market.We already saw the case that the peasants in the Huangling village sold vegetables to merchants coming from town. In the case of Jiading district, Shanghai, cotton growers usually sold raw cotton in their home on terms offered by dealers (MSJC 1940b, 66). Peasant households in Jiading had other income sources from Shanghai besides cotton crops. We may call the former a broker-dependent (or distant-trade-oriented) peasant, the latter a broker-independent (or local-trade-oriented) peasant. By keeping open the option to depend on a broker from outside, however, Chinese peasants, initially, maintained flexibility against unknown future situations by avoiding subsequent transactions. These decentralized features in the transactions at the village level were also mostly shared by the transactions at the town level between merchants residing in towns and petite dealers coming from villages. One-time transactions were dominant not only at the level of proximate exchanges among villagers but also at the level of less-proximate exchanges connecting villages and towns. The autonomy of dealers in lower-level markets, as opposed to those of upper-level markets, appears more clearly in the means of exchange, that is, money.

3  Natives preferred local currencies According to the Decennial Report published by the Imperial Maritime Customs, China, in 1912 the peasants in the mid-Yangzi River region preferred to receive 1,000 copper coins in payment, rather than one silver dollar valued at 1,300 copper coins (Imperial Maritime Customs 1912, 283). At the time, silver dollars were available to settle interregional transactions, but they were rarely used in local transactions, especially among peasants in the region. This striking

28  Exchanges generate money locally

preference for copper coins by peasants vividly shows how autonomous local trades were compared to distant trades. Some may suggest that in this situation merchants could have profited from the value differential due to an information asymmetry between peasants and merchants. Logically we might assume that a merchant would exchange ten silver dollars for 13,000 copper coins in a treaty port such as Hankow, bring the copper coins to rural markets and exchange them for 13 silver dollars. However, it was impractical for any merchant to do this, since the merchant would have little chance of finding sufficient silver dollars in rural marketplaces, and they would therefore suffer the transportation cost to no purpose. In fact, merchants had to do the opposite: exchange interregional currencies for local currencies in order to acquire peasant products in rural marketplaces, and increased demand for such products, especially after harvests, often caused local currencies such as copper coins to appreciate. Consequently, merchants had to pay more in order to secure the means to pay the peasants. While it may seem that one solution would have been for an urban merchant to leave some local currencies in the rural marketplaces to prepare for subsequent business opportunities, this would not have been good business. In rural regions, monetary demand had a strong seasonality. As we have seen earlier, Table 1.1 shows that the majority of sales of peasant products occurred within three months of the harvest. Currencies that mediated transactions with peasants would remain stagnant for the next nine months. Thus, it was more reasonable for merchants to exchange their interregional currency for local currency only in the season when they were acquiring peasant products. Why were silver dollars (an interregional currency) so unpopular among peasants? In cases where peasants sold their products to a larger market mainly through merchants, in the way of broker-dependent peasants argued in previous section, currencies used in interregional settlements might have had more circulation among peasants. However, as we have already seen in Table 1.2, the majority of sales and purchases by Chinese peasants occurred within rural market zones. In the case of Qingyuan County, the peasants sold their grains for commodities from outside the zone, but they exchanged the majority of their products for the products of other peasants in the same zone. ‘Others’ in Table 1.2 indicates handicraft products made by peasants during their slack season. Thus, peasants required the means to make exchanges with other peasants in the same zone more often than they did with merchants from outside the zone.The means of payment among peasants had to be of small value, appropriate to the size of these transactions. We can see one case showing the discrepancy between the two levels of markets associated with silver and copper coins in mid-19th-century China. From 1830 to 1850 the circulation of silver decreased, likely resulting from China’s international trade deficit due to increasing opium imports.The account books of the Tongtaihao recording the entry and exit of silvers reveal the number of silver ingots, baoyin (50 liang), that entered the Tongtaihao store in nine different years: 255 ingots in 1821, 233 in 1823, 509, in 1828, 361 in 1830, 528 in 1831, 968 in 1833, 504 in 1841, 168 in 1846 and 176 in 1848. After a peak in 1833

Peasants, marketplace and money  29

the number dropped to less than one-fifth of the peak in 1846.11 The merchants did not set a fixed exchange rate between silver and copper coins; the rate changed daily, and even more than once within the same day. As a result, silver appreciated substantially against copper coins. According to the account books of the Tongtaihao, one liang of silver was exchanged for 2,590 wen in the local unit of copper coins (jingqian) in 1821, 2,820 wen in 1,830, 2,970 wen in 1841, 3,580 wen in 1844 and 4,800 wen in 1848.12 The value of copper coins in terms of silver depreciated by nearly half. At the time, silver was flowing out of China in significant quantities. It is safe to assume that under the contraction of silver flows, merchants depended to a greater degree on multilateral counter-trade through book transfers to keep their transactions in the same town, as we will see later. What happened to proximate exchanges among local peasants then? Did they suffer from the depreciation of copper coins? In spite of the serious depreciation of copper money against silver, however, prices in terms of copper coins remained surprisingly stable for almost half a century. Figure 1.2 shows the price movement of rice in one rural area, Tunxi County, Anhui province, in the first half of the 19th century. Rice prices in copper coin at the end of the 1840s were almost the same as those at the beginning of the 1820s. In the district, many clans annually held a ritual for praying to their ancestors in early spring, at the festival known as qingmingjie. They would record the prices of items they bought for this ritual in their account books. Importantly, in this case we can trace the movement of prices year-by-year without having to adjust for seasonality. Grains including rice and beans were traded in terms of copper cash, while imports from outside regions were made in terms of silver. Thus, the appreciation of silver must have hampered the purchase of products through distant trade. In addition, since tax payments were made in terms of silver, the shortage of silver caused 300 250 200 150 100 50 0










Figure 1.2 Annual movement of rice price at rural area, Tunxi County, Anhui, China (wen per 5 sheng) Source: Jinshan sihui shouzhi bu. Institute of Economics, Chinese Academy of Social Sciences, no. si36 han549.

30  Exchanges generate money locally Table 1.4  Exchange rates between Manillas and Sterling and the price of palm oil

1912 1920 1927 1932

No. of manillas to £

average price of per ton of palm oil Liverpool (ex quay)

Equivalent price per ton of palm oil in manillas

160 80–100 120–160 240–260

£30.1 £68 £34.2 £18.11

4,816 5,440–6,800 4,104–5,472 4,364–4,708

Source:Vice 1983, 16.

difficulties for taxpayers. However, considering the high proportion of proximate exchanges as compared to distant exchanges for peasant households, as shown in the statistics of Qingyuan County (Table 1.2), keeping price movements stable within proximate exchanges must have been beneficial for ordinary peasants in rural China. The unfixed relationship between silver and copper coins in 19th-century China must not be an isolated case but one of very many similar phenomena across the world, as we will see throughout this volume.Was it common that an autonomous currency kept stabilizing local prices? Let us see another example. In spite of the British policy demonetizing manillas (an indigenous rod-shaped currency made of brass, produced in Birmingham and Nantes after the 19th century), manillas continued circulating among locals in early 20th-century British West Africa. As Table 1.4 shows, between 1912 and 1920, when the palm oil price in terms of sterling in Liverpool rose, the exchange rate of manillas against sterling also rose, while between 1920 and 1930 when the palm oil price fell, the exchange price of manillas fell (Vice 1983, 16). Thus, the price of palm oil (the most important product for peasants in British West Africa) were more stable when valued in manillas than in sterling in this period. Monthly movement of the exchange rates between manillas and sterling clearly shows the demand for manillas rose during the harvest season of palm oil (Vice 1983, 15), in the same way as did copper coins in China and cowries in India. The presence of an independent local currency mediating exchanges of local products absorbed the extreme instability of price movements in distant trade and eventually offered a haven to local exchanges (Helleiner 2003, 183).

4  Ubiquitous fractional currencies and petty exchangers It was globally common that the exchanges in rural marketplaces required currencies whose values were small enough to negotiate prices at a minimum unit. Depending on the situation, even the value of copper coins was not fractional enough for end users to graduate prices. The observations on zinc zapeks in early-20th-century Indochina show a typical feature of a super-fractional currency.13 They were convenient for ordinary people since the value was minute

Peasants, marketplace and money  31

enough to purchase a slice of papaya and drink a cup of tea (Robequain 1944, 137). Responding to the demand at the bottom of society ‘private individuals could coin zinc sapeks, thus meeting a scarcity in the means of trade as they would meet a scarcity of nails or fishing nets – by manufacturing them’ (Robequain 1944, 138). Locals made currencies by themselves if they needed. Zinc coins were not the most fractional currencies in the world. A British observer noted that in early 19th-century India shell monies (cowries) were dominant in the markets. The value of one cowry was so minute that 80 cowries were equivalent to one small copper coin and no less than 5,120 cowries to a silver rupee. A farthing, equal in value to more than 50 cowries, was hardly of any use as money. The usage of so minute a denomination of money meant to the British observer that the situation showed the degree of ‘the poverty of the people, and of the low prices of wages, and necessities of life among them’ (Anon 1837, 44). Leaving aside the question of whether this situation was an accurate index of the degree of poverty, the observer failed to link this to the peculiar situation of market activities by ordinary people, which were as follows. The author also described how the money-changer went to a marketplace ‘with a bag of cowries on his head; or, if a rich man, with a loaded ox, which, if strong, may carry to the value of a hundred and fifty rupee.’ In the early hours of the market, he sells cowries for silver, to the people who wish to purchase goods, and in the evening the various hucksters bring back their cowries, and change them for silver. In the morning, the money-changer usually gives 5,760 cowries for a rupee, and in the evening he gives a rupee for 5,920 cowries. (Anon 1837, 47) This situation in rural markets suggests that there was such strong demand for cowries in purchasing local commodities that the exchangers could make a profit every day.The demand for a super-fractional currency showed the necessity for negotiating prices between the seller and buyer on a very tiny scale. Cowries were ideally fragmental enough to create a minute gradation of prices in fractional units of commodities, such as a slice of papaya and a cup of water. Even copper coins were not fragmental enough to make a gradation for purchasing items in minimum units. Thus, the popularity of small denomination currencies in the marketplaces reflected the strong necessity for subdivisions in negotiating prices rather than the lowness of prices. The presence of numerous petty exchangers showed there was no monopoly in this business. There is no big difference between cowrie-exchangers and papaya-dealers with respect to the point that their profits depended on the tiny scale of retail transactions, though, of course, the cowries remained, and the papaya was consumed after transactions. A coupling between small denomination currencies and independent end users gathering in local markets was very often found across the world. For example, in 1896, at Obegu, East Nigeria, ‘an extremely large and important

32  Exchanges generate money locally

market is held every four days, people coming from immense distances to it.’ At Bende every fair attracted 20,000 to 30,000 people from the hinterlands (Guyer 2004, 73). Cowries dominated local markets in West Africa and East Africa in the same way as in Bengal and Orissa, India, as mentioned earlier. We return to Chinese villages again in order to confirm concrete information on the relationship between marketplace, small currency and petty exchangers. In the village of Lujiasai in Zunhua County, Hebei province, the currency most frequently used was copper coins as was the case across Northern China. People in both Lujiasai and in the town of Zuojiawu, where the villagers sold their products, conducted trade in cash, although within the village copper coins were rarely circulated except for purchasing salt when it was time to make bean paste. This is not surprising if currency was mostly used when villagers went to market (Ito 1936, 336–9). More information about currency use can be found by looking at a rural market, the Madianji fair in Shandong. According to a survey made in 1933, peasants brought a small quantity of products to the market, such as between 10 and 20 eggs, from places typically 90 minutes away on foot. Even in April, the off-season, the marketplace was held six times a month and attracted more than 2,000 people each time. The size was sufficiently large that most transactions would have been between strangers. A number of merchants came from the treaty port of Qingdao, and they had to convert the silver coins they brought with them into the copper coins popular among the peasants attending the market. Conversion was unavoidable, since the silver dollar was of too high a value for small transactions in a marketplace where 100 eggs cost around one silver dollar. The gap between the two currencies made the presence of smallscale money exchangers indispensable, and the Madianji fair had 30 persons so engaged. Furthermore, the exchange rate between silver and copper fluctuated according to supply and demand, day-by-day, even as late as 1933, which was just two years before the national government introduced a standard system using paper money and announced a unified monetary system across China (Mizuno 1935, 65–8, 98). We shall argue the situation again in Chapter Five. In contrast, a society favouring lump transactions would have no demand for fragmental currencies. A combination of the dominance of big coins and long intervals between wage payments, monthly for example, meant that most ordinary people were unfamiliar with the use of cash. Nor did an economy depending on proximate/named exchange have any need to generate local currencies. Those systems could avoid a situation in which a huge amount of shell monies was piled up and traders or officers had difficulties transporting them over a long distance. However, the super-fragmental currencies were an instrument of ordinary locals to keep their freedom to choose and negotiate in exchange.

5 Subsequent transactions not through marketplace: another configuration between peasants and merchants A marketplace provides local peasants with a place in which they exchange monies anonymously. Depending on a one-time transaction base, peasants

Peasants, marketplace and money  33

maintain the freedom to negotiate prices. Some societies, however, put less of a priority on such freedom. Before industrialization, villages in China and Japan offer contrasting features. Both were agrarian societies which, in general, should have organized labour supplies according to season, such as responding to increasing labour demand during the harvest. However, the different methods of labour organization reflect specific characteristics of each society. Table 1.5 reveals that peasants in the region of present-day south Osaka used hired labour.This must have included day labour for rice planting in late spring, spreading water in cotton fields in summer and harvesting in autumn. However, unlike rural China, as far as agricultural day labourers were concerned, the pattern of one-time employers occasionally engaging one-time employees rarely appeared in rural Japan. For example, the diary of a landowner in Yamashiro (present-day southern Kyoto) recorded the names of day labourers as well as the amounts of payments. In four different years the same people appeared, and all seemed to live nearby. No such document is known in China, but it is not improbable that the same person might have been employed for day labour many times, by the same employer within a Chinese village. There was, however, a conspicuous difference between Japan and China. The day labourers in Yamashiro were paid later, in one extreme case, two months later (Tanimoto 2018). The situation mentioned earlier shows that hiring day labourers in Japanese villages was not one-time transaction but continual transaction. Deferred payment for day labour would have been beyond imagination in contemporary China. Thus, generally speaking, in Japanese villages a villager could not easily make a one-time transaction with another. A set of institutions underpinned this feature. A village in Tokugawa, Japan, was a unit of taxation. Importantly, unlike as in Qing, China, the assessment for land taxation was readjusted so frequently that the relationship between tax collectors, domains and villages was always in tension. It was common for leading households in a village to cover unpaid

Table 1.5 Profits with rice and cotton cultivation per 1 tan mid-1880s southern Osaka, Japan (yen) Rice










1.9 3.735

2 1.96

6.2 4.88

8.55 7.825











Sale Expenditures fertilizer hired labour

Source: Shibahara 1981, 170.

34  Exchanges generate money locally

taxes for destitute households. A village worked as an entity, something much more than a collection of peasant households. It possessed common land for uses such as collecting firewood. It also had its own laws which applied only among villagers and which even the warrior rulers could not ignore. Under these circumstances, various forms of labour exchange in proximate locations such as yui (labour exchange for agricultural work) dominated Japanese village life. Under such a cohesive atmosphere, it is no exaggeration to say that day labour for a cash payment settled on the day would have seemed unsocial. The dominance of subsequent transactions affected the relationship between peasants and merchants. In Japan, basically, merchants purchased products such as raw cotton in cash, as in China. Unlike commission-oriented merchants in China at the same time, however, in Japan merchants from towns sometimes made advances to peasants to secure commodities.Typically, this manifested itself as an outlay system.Table 1.6 shows the account book of a cotton merchant, the Yamawaki family, showing transactions with a peasant household which wove cotton cloth in Kawachi, in present-day south Osaka, then the most commercialized region in Japan. Probably this part of the account book did not cover the entire business between the merchant and the weaving peasant, but we can

Table 1.6  Advance and purchase between the Yamawaki family and a peasant weaver in 1850 Date

Advance raw cotton (kin)

3.2–3 3.13 3.16 3.28 4.12 4.12 4.12 4.13 5.14 5.14 6.9 6.9 6.28 6.28 6.29 7.24 8.8 8.8 Total

Purchase value (monme)













cash (monme)

Source: Nakamura 1991, 152–3.

value (monme)



1 1

16.5 16.5







1 1

16.6 16.8



2.76 140

cotton cloth (hiki)



cash (monme)



Peasants, marketplace and money  35

find a clear tendency. Repeating the advance of 10 kin of raw cotton and the purchase of 1 hiki of cotton cloth from the third month to the eighth month, the value both of raw cotton and of cotton cloth was kept almost at the same level. The merchant is likely to have continued being the primary customer of the cloth the peasant wove. It is probable that the peasant also cultivated raw cotton on their land, but their weaving capability exceeded the amount of their own crop, thus the merchants provided raw cotton collected from other districts.The merchants might have made advances to peasants from whom raw cotton was purchased after the harvest. As Table 1.5 shows, cotton farmers needed more cash for fertilizer than did rice farmers. It was a typical pattern for merchants to advance fertilizer, mostly from sardines, to peasants before cultivation and receive raw cotton after the harvest. To complete the transactions little cash was used between merchants and peasants. A comparison between the account books of the Yamawaki family in 1850 and 1858 suggests that the proportion of outlay and advances increased while that of purchasing cloth in cash from petite traders or peasants decreased (Nakamura 1991, 160–1). The prospect of subsequent contracts between sellers and buyers lessened the necessity of cash and enabled drafts to circulate, as we will see later. Here we see a substantial difference between China and Japan, both of which had developed rural commercialization dependent on a handicraft cotton industry. We find that it is quite similar in both countries that merchants brought materials to rural regions and peasants manually produced goods in agricultural slack seasons. However, as the case of Gaoyang (a district representative of the handicraft cotton industry in China) vividly shows, besides directly selling cloth to merchants, peasant weavers always sold their products in local marketplaces, as well. Materials were also mainly purchased by peasants through local marketplaces where merchants brought them.14 Thus, the businesses were far more negotiable in Chinese villages than in Japanese ones.

6  Honour substitutes for currency Institutional factors, sometimes, happened to support the named exchanges and consequently enhanced subsequent transactions. The case of early modern England reveals a clear pattern: large numbers of litigation records in local courts of law offer snapshots of the kinds of social devices that operated in transactions. Litigation records in local market towns and manors show how deeply ordinary people depended on credit transactions in daily life. Most credit took the form of unsecured small debts and consisted mostly of sales credit, wages, rent, taxes and other payments due.These might have been listed in an account book, in a diary or on a loose slip of paper, but most transactions were undoubtedly only oral. Ordinary people often took on debts of less than £1. In other words, credit transactions were used frequently enough to render cash transactions rare. Indeed, the most important use of currency was for very small transactions between strangers (Muldrew 1998, 101).

36  Exchanges generate money locally

It cannot have been mere coincidence that most people brought lawsuits against members of their own community, whether urban or rural. When a villager made an agreement to sell his produce at a local marketplace, he tended to call upon a member of his village community to witness the agreement (Moore 1981, 288) – a phenomenon that indicates the importance of community in credit transactions. The number of lawsuits brought to local courts appears to have peaked in the latter half of the 17th century. In the case of the town of Kings Lynn, on average, each household launched lawsuits nearly three times a year. However, it seems to have been the relation between credit and community, rather than fear of punishment, that effectively led local people to honour their debts. A person had to avoid any conduct that might disconnect him from the highly personal links of the community (Everitt 1967, 567). We have no statistics about the proportion of credit to cash transactions among ordinary people, but anecdotal evidence gives the impression that the use of currency was very rare. In this sense, the inheritance system of England, which required people to leave an inventory of their goods, provides us with vivid information about the form of their assets. Notably, credit in various forms, such as bonds, bills obligatory and book credit, was predominant, and, in contrast, only modest amounts of cash appear in these inventories. According to Kerridge, who studied inventories mainly from Liverpool, ‘in the period 1538–1660 the ratio of coins to debts by bond, bill of obligatory and book was on average 1:9’ (Kerridge 1988, 98). A sample of inventories from London merchants and the trading elite shows that the ratio of trade credit, loans and other investments to cash was 15:1 (Muldrew 1998, 100). Of course, the ratio between credit and cash varied by class, occupation and geographical location. The cases mentioned earlier mainly reflect the patterns of urban merchants. In addition, forms of assets do not always correspond to forms of settlement. Nevertheless, it is safe to say that in early modern England transactions depended heavily on credit supported by communal trusts. Named connections rather than anonymous relationships propelled exchanges in early modern England. The decrease of local markets in early modern England precipitated a nationwide shift from anonymous to named exchanges. In regions of 14th-century England, such as Essex, the density of rural markets appears to have been almost proportionate to population density. There, dispersed settlements – with nonagricultural activities carried on in villages – blurred the distinction between the ‘rural’ and ‘urban’ economic spheres (Poos 1991, 56). At the end of the 13th-century market centres serving an area of 20 or 30 square miles existed in every English county. That means that most villagers could have access to a marketplace located within one day’s travel on foot (Hilton 1983, 168). However, in the 16th century, officially licensed local markets may have dwindled to a third the amount of those in the 13th century (Everitt 1967, 467). In the case of Winchester, the fading away of some markets in the surrounding area brought great advantages to Winchester.There was a tendency for internal trade

Peasants, marketplace and money  37

to become concentrated in one or two towns in each county through the 17th century (Clark and Slack 1976, 184). Roughly speaking, markets in villages disappeared, while the boroughs, which were populated by privileged residents, were able to survive. The borough as an institution successfully differentiated the market town from the market village (Hilton 1985, 5). On the basis of the different degrees of seasonality in marriages among parishes across England, it is also argued that through the 17th century a division of labour between grain-producing regions and grain-consuming regions developed. The distance between parishes harvesting grains and parishes consuming them became greater in the late 17th century and early 18th century than in the early 17th century (Kussmaul 1990, 99–100). Though this is not perfectly certain, regional specialization might have advanced in tandem with enclosure movements. We may interpret the changes in England as follows: the spatial concentration of markets led local people to rely on named transactions within their communities and only occasionally engage in transactions with outsiders. Since currency was only used in these transactions with outsiders, the general usage of currency therefore also diminished. This causal relation between the reduction in the number of local markets and increased dependence on named transactions among locals is also evident in societies other than England. Again, let us visit early modern Japan, which provides us with a parallel case. Depending on the situation in 18th and 19th centuries, in the previous section, we highlighted the importance of subsequent transactions in Japan. However, marketplaces had played important roles in an earlier period in Japan, as well. Place names referring to markets held on a particular day of the ten-day week, such as Yokka-ichi (‘4th day market’), can be found all over Japan. This suggests the ubiquity of periodic rural markets throughout the country, with peddlers sequentially trading along chains of periodic markets in which peasants sold their products. However, this began to change in the 17th century, after the Tokugawa house established a system of rule (the so-called bakuhan system) in which the shogunate ruled at the national level and the daimyo (domain lords) ruled at the local level. As the daimyo consolidated local power, they raised barriers along their domain borders and strictly inspected people and goods crossing them, thereby limiting the autonomy of local markets in rural areas. The castle town of each daimyo became the administrative centre of the domain, and merchants with a privileged relationship with a specific domain became dominant in local business there. The development of privileged, urban-based merchant activity subsequently inhibited the activities of village-based peddlers. In Ueda domain (in the present-day Nagano prefecture), there were five periodic rural markets at the beginning of the 17th century, but three out of the five had disappeared by the beginning of the next century (Oishi 1975, 29–65). In mid-17th-century Musashi and Sagami, located in the western part of the Kanto region, there had been more than 60 rokusai-ichi, a marketplace that was held every five days. A century and a half later, many of these had disappeared or had been reduced to a single fair held just once or twice a year (Toyoda 1952,

38  Exchanges generate money locally

312–23). This disappearance of marketplaces was a tendency common across Japan. By the late 19th century, local periodic markets such as rokusai-ichi had almost entirely vanished from Japan, though some remained in Niigata prefecture (Ishihara 1987, 344–62). How did the disappearance of local marketplaces influence transactions? A study of the grain trade in Iiyama domain makes it clear that, in the 17th century, it was common for sellers and buyers of grains to gather at local markets where middlemen brokered deals for a commission. Spot trade still dominated the local market, and traders possibly had to face anonymous strangers. However, by the late 18th century, such unsolicited business had become unpopular. Grain merchants in the domain began to make future contracts at a fixed price with particular buyers, especially those from outside the domain. These contracts for future trade were often sold to third parties as drafts. As such business grew in popularity beyond the grain trade in the early 19th century, the use of drafts in local commerce also increased (Tawada 2007, 95–104, 182–209). The proliferation of drafts was followed by the emergence of highdenomination copper coins with a face value of 100 wen, the tempo tsuho. The tempo tsuho possibly worked as a standard of value in the grain trade more often than it was actually circulated. As seen in the account book of a grain merchant in Iiyama domain reveals that prices were recorded not in terms of how many coins (mon) were equivalent to a unit of grain (sho), but in terms of how much grain was equivalent to a certain amount of coin. The account book of a day in 1843 records the price of refined rice equivalent to one tempo tsuho coin (100 mon) 1.2 sho (1.8 litre), rice cake equivalent to one coin 1.05 sho and refined wheat equivalent to one coin 1.7 sho (Tawada 2007, 97). Given the situation, we may suppose that the account book suggests that small denomination coins lower than 100 mon were seldom used. That is why pricing was not made in terms of copper coins to grain volume, sho, but conversely in terms of units of grain to one coin (the tempo tsuho). In tandem with this reduction of dependency on currency, mutual credit among villagers increased in importance. Under the bakuhan system it became common for wealthier households in villages to shoulder more of the taxation burden in order to provide relief to peasants unable to pay their share. By assuming the role of lender in tax payments, village leaders began to work as credit suppliers in transactions within villages in the late 17th century (Makihara 2001). Interestingly, a village moneylender in a village near Osaka admitted overdrafts in the case of lending to peasants in the same village, while he always made drafts in lending to outside villages (Fukuyama 1975, 68). The leaders were also responsible for ensuring that villagers observed the village laws, and they often served as mediators in disputes (Mizumoto 1993), though disputes over nonpayment were not as common as in England. The forms of assets held by merchants also reveal the superiority of credit. For example, the list of assets of the Takahashi family, a lacquer trader residing in Inaniwa, Akita domain, reveals that currency occupied only a small portion of their total assets. The family’s accumulated currency, including tempo tsuho, only

Peasants, marketplace and money  39

amounted to 5,349 kan (1000 wen) out of its total assets of 44,810 kan. In the list of assets, pledges for pawning accounted for 6,032 kan, grain for 1,258 kan, lacquer for 1,513 kan and credit for 27,992 kan (Akita Prefecture 1963, 166) England and Japan thus shared several parallel features: the disappearance of local marketplaces; rising dependence on named credit – or independence from currency – in the making of transactions and the accumulation of assets; and a general preference for repeating transactions with one’s best customers as opposed to one-time transactions with strangers. These features stand in stark contrast to the case of China. For example, the numbers of rural marketplaces per county in Shandong and Hebei increased from 15 in Shandong and 10 in Hebei before 1735 to 20 and 12 before 1795, to 23 and 13 before 1861, 29 and 14 before 1911 and to 42 and 21 before 1949 (Qiao 1998, 351). The density of marketplaces increased continuously, rising particularly in the early 20th century when the introduction of peasant products to international trade stimulated commercial activity in rural China. Coastal areas, such as Shandong, experienced greater involvement in such trade, which accelerated the emergence of new rural marketplaces in those regions. In China, the intensified commercialization of the peasant economy was accompanied by the proliferation of new marketplaces.

7 Freedom or certainty: two paths (not stages) of local commercialization Currency shortages have occurred across the world and throughout history, as we will see throughout this volume. When suffering from liquidity shortages, local end users have generally responded in one of three ways: increasing the supply of the currency available for proximate exchanges, whether new or existing; enhancing systems that provide local credit; and making ­counter-trades. A counter-trade means that goods could be bilaterally exchanged between owners in terms of a unit of account, without any usage of currency. The latter two methods avoid the use of currency, and money functions only as a unit of account. Meanwhile the first and third methods support one-time transactions, while the second tends to generate subsequent transactions through named relationships. It is safe to assume that the first two ways can more flexibly increase monetary supplies according to demand than the third method, which places a priority on exchanges of equal value in commodities rather than on developing the scale of transactions. We may depict these two paths in general terms as follows. Assume a case where person A can make a deferred payment to person B, and also to person C, and that the same is possible between B and C. Call the triangle of debt connections a cluster, to use the terminology of network theory. A village in which debt transactions concluded through oral agreements among villagers are dominant may be thought to have a high clustering coefficient (Figure 1.3, containing nine triangles). The hexagon in the figure indicates a village of six households, while the circle shows a marketplace that villagers can directly

40  Exchanges generate money locally

access. A solid line signifies a deferred payment through personal trust, while a broken line shows an anonymous exchange. In contrast, a village in which peasants tend to make anonymous sales with cash in the rural market would have a low clustering coefficient (Figure 1.4, two triangles). Both society X, with its high clustering coefficient, and society Y, with its high dependence on anonymity, can increase the frequency of transactions in terms of the unit of account per household. As X and Y have the same number of lines in total (14), credit and currency can together make up the total amount of money necessary to mediate transactions in a given sphere. In other words, credit can supplement a shortfall in the currency supply. However, the social consequences of the named connection sustaining credit and of the anonymous contacts that made currency indispensable were in tension with each other. Strengthening personal ties and increasing the chances of anonymous transactions are contradictory forces operating within the same circuit. The contrast of images between a dense cluster in Figure 1.3 and a loose one in Figure 1.4 shows this visually. Putting aside counter-trade for now, we can evaluate most rural groups engaging in exchange according to the density of such clusters (Kuroda 2013). In other words, taking the four quadrants in Figure I.1 in the introduction, transactions in society X amplify Quadrant III, while those in society Y magnify Quadrant IV. Both societies can have the same frequency of monetary transactions, even though, depending on their social relations, their paths of commercialization will differ. Using Figures 1.3 and 1.4, we already theorized the contrast between the first method and the second method. As represented by the four quadrants of Figure I.1, a cohesive society will see an amplification of Quadrant III (named/proximate), while the latter will see a magnification of Quadrant IV

Figure 1.3  Society with a high clustering coefficient: mutual credit

Peasants, marketplace and money  41

Figure 1.4  Society with a high preference for anonymity: local currency

(anonymous/proximate). Though the same frequency of transactions will take place in both societies, these variations in social relations will diversify the paths of commercialization. Preference for one-time transactions or preference for subsequent transactions indicates whether people put emphasis on freedom or on certainty. Institutions determine which alternative a society orients towards and how the society associates the two elements. The institutions which precondition the means of transaction may vary between proximate exchanges and distant exchanges. Thus, money, a means of exchange, works variously society by society according to a particular set of institutions.

Notes 1 CNKC vol. 3, 371, vol. 4, 229. MSJC 1940a, 105–6. 2 Fukutake 1976, 165. In the investigation of Jiading district, Shanghai, we find two cases of labour exchanges between two peasant households in rice cultivation, while 16 households in the same village hired day labour. MSJC 1940b Table 2. 3 The number of households of the villages was estimated according to Chang 1942. 4 In the case of Jiading, Shanghai, 6 peasant households out of 25 households from which labour was hired employed labour themselves. MSJC 1940b Table 2. The case of Huangling village, Shanxi, shows that among 25 peasants households which had members being hired for labour, eight households also employed labour, most of which was probably for ploughing with livestock. KAKO 1944, 22–5. 5 Merchants in Taiyuan town provided domestic female with materials for shoe soles. It took almost one day for one worker to complete one pair of soles. KAKO 1944, 92, appendix 4–5.

42  Exchanges generate money locally 6 Petite dealers who sold bean cake in Jinjing County town mostly used the proceeds to purchase kerosene oil or tobacco, and rarely brought cash back home. KPDF 1942, 67. 7 The gazetteer of Ningjin County published in the late 19th century wrote that the Daliu zhen town held open market on the day of second and seventh of every ten days. Zhu 2004, vol. 2, 25a. 8 This analysis is based on the account books of short sales credit by the Tongtaihao in 1831, Daoguang shinian zhengyue zanfuji zongzhang, Chinese National Library, Wenjinge, no. 49120–166. Further investigations are required, but the same tendency appears to be shared by the account books for other years. 9 Daoguang shinian zhengyue zanfuji zongzhang, no. 49120–166 and Daoguang shiyinian zhengyue fuji zongzhang, no. 49120–168. 10 For example, see Liu 1990, 300. 11 Tongtaihao zhangbu. Chinese National Library, Wenjinge, no.131001–1,2,4,5 and Daoguang shiyinian lixi zhang Tongtaihao, no. 49120–142,148,153; Tongtaihao, Daoguang ershiyi nian yin churu, Institute of Economics, Chinese Academy of Social Sciences, no. qitadiqu han2 c024; Tongtaihao, Daoguang ershiba nian yin churu, no. qitadiqu han2 c029. 12 All are taken from a rate early in tenth month of each year. According to jingqian custom, one copper coin was counted as two wen. Tongtaihao zhangbu, no. 131001–2,5; Tongtaihao, Daoguang ershiyi nian yin churu, no. qitadiqu han2 c024; Tongtaihao, Daoguang ershisi nian yin churu and Tongtaihao, Daoguang ershiba nian yin churu, no. qitadiqu han2 c029. 13 We may trace the history of super fractional currencies in South East Asia as far back as the late 16th century, when the Dutch encountered lead coins in Java. They were so cheap that 12,000 to 13,000 pieces were equivalent to one Spanish silver coin. Mollema 1935, 211. 14 Wu 1936, 244–5. Cotton cloths made in Gaoyang extended its market across North China in the early 20th century. See Grove 2006, 99–120. The expansion of interregional trade did not accompany a monopoly of collecting cloths from local petti producers by brokers.

References Akita Prefecture. 1963. Akita-ken shi, shiryo, kinsei, vol. 2. Akita: Akita Prefectural Office. Amano, M. 1935. Santo nogyo keizai ron. Dalian: Minami manshu tetsudo kabusiki kaisha chosakai. Anon. 1837. Sketch of the Commercial Resources and Monetary and Mercantile System of British India, with Suggestions for Their Improvement, by Means of Banking Establishments. No place: No publisher. Chang,YF. 1942. Shanxi Jinquanxian zuixin zhuyao chupin ji hushu juli jianyao ditu. No place: No publisher. Clark, P. and Slack, P. 1976. English Towns in Transition 1500–1700. Oxford: Oxford University Press. Cui, XL. 1990. ‘Jiating, shichang, shequ: Wuxi Qingyuan nongcun shehui jingji bianqian de bijiao yanjiu (1929–1949),’ Zhonguo jingji shi yanjiu (1): 42–65. Everitt, AM. 1967. ‘The marketing of agricultural produce,’ in Thirsk, J. (ed.) The Agrarian History of England and Wales, vol. 4, 466–592. Cambridge: Cambridge University Press. Fukutake, T. 1976. Chugoku noson shakai no kozo. Tokyo: Tokyo daigaku shuppankai. Fukuyama, A. 1975. Kinsei noson kin’yu no kozo. Tokyo:Yuzankaku. Grove, L. 2006. A Chinese Economic Revolution: Rural Entrepreneurship in the Twentieth Century. Lanham, MD: Rowman & Littlefield. Guyer, JI. 2004. Marginal Gains: Monetary Transactions in Atlantic Africa. Chicago: University of Chicago Press.

Peasants, marketplace and money  43 Helleiner, E. 2003. The Making of National Money: Territorial Currencies in Historical Perspective. Ithaca: Cornell University Press. Hilton, R. 1983. A Medieval Society: The West Midlands at the End of the Thirteenth Century. Cambridge: Cambridge University Press. Hilton, R. 1985. ‘Medieval market towns and simple commodity production,’ Past and Present 109: 3–23. https://doi.org/10.1093/past/109.1.3 Huang, PCC. 1985. The Peasant Economy and Social Change in North China. Stanford, CA: Stanford University Press. Imperial Maritime Customs, China. 1912. Decennial Reports: 1902–11, vol. 1. Shanghai: Imperial Maritime Customs. Ishihara, H. 1987. Teikiichi no kenkyu: kino to kozo. Nagoya: Nagoya daigaku shuppankai. Ito,T. 1936. Kito chikunai nijugo kason noson jittai chosa hokoku.Tianjin: Kito chiku noson jittai chosahan. Kahoku kotsu kabushiki kaisha sigyokyoku. 1944. Hokushi noson no jittai: Sanseisho Shinsenken Koryoson jittai chosa hokokusho. Tokyo: Ryubun syokyoku. Kerridge, E. 1988. Trade and Banking in Early Modern England. Manchester: Manchester University Press. Kokuritsu Pekin daigaku fusetsu noson keizai kenkyusho. 1942. Santosho Sainei kenjo wo chushin to seru nosanbutsu ryutsu ni kansuru ichi kosatsu. Beijing: Kokuritsu Pekin daigaku fusetsu noson keizai kenkyusho. Kuroda, A. 1996. ‘20 seiki shoki Taigen-ken ni miru chiiki keizai no genki,’ Toyoshi kenkyu 54(4): 685–718. Kuroda, A. 2013. ‘Anonymous currencies or named debts? Comparison of currencies, local credits and units of account between China, Japan and England in the pre-industrial era,’ Socio Economic Review 11(1): 57–80. https://doi.org/10.1093/ser/mws013 Kuroda, A. 2018. ‘Strategic peasant and autonomous local market: Revisiting the rural economy in modern China,’ International Journal of Asian Studies 15(2): 195–227. https://doi. org/10.1017/S1479591418000049 Kussmaul, A. 1990. A General View of the Rural Economy of England, 1538–1840. Cambridge: Cambridge University Press. Liu, DP. 1990. Tuixiangzhai riji. Qiao, ZQ. (ed.). Taiyuan: Shanxi renmin chubanshe. Makihara, S. 2001. ‘Kan’eiki no kin’yu to chiiki shakai,’ Rekishigaku kenkyu 747: 18–33, 63–4. Mantetsu hokushi keizai chosasho. 1939. Showa 14 nendo noka keizai chosa hokoku: Hojunken Gishochin Beishoson. Dalian: Minami manshu tetsudo chosabu. Mantetsu hokushi keizai chosasho. 1940. Showa 14 nendo Noka keizai chosa hokoku: Hojunken Gishochin Beishoson. Dalian: Minami manshū tetsudō chosabu. Mantetsu hokushi keizai chosasho. 1942. Showa 15 nendo Noka keizai chosa hokoku: Hojunken Gishochin Beishoson. Dalian: Minami manshu tetsudo chosabu. Mantetsu Shanghai jimusho chosashitsu. 1940a. Kososho Jojukuken noson jittai chosa hokokusho. Shanghai: Minami manshu tetsudo Shanghai jimusho. Mantetsu Shanghai jimusho chosashitsu. 1940b. Shanghai tokubetsushi Kateiku noson jittai chosa hokokusho. Shanghai: Minami manshu tetsudo Shanghai jimusho. Mizumoto, K. 1993. Kinsei no kyoson jichi to gyosei. Tokyo: Tokyo daigaku shuppankai. Mizuno, K. 1935. Santo no ichi noson ni okeru shakai keizai jijo. Dalian: Minami manshu tetsudo. Mollema, JC. 1935. De eerste schipavaart der Hollanders naar Oost-Indië 1595–97. The Hague: M. Nijhoff. Moore, EW. 1981. ‘Medieval English fairs: Evidence from Winchester and St. Ives,’ in Raftis, JA. (ed.) Pathways to Medieval Peasants, 283–99. Toronto: Pontifical Institute of Medieval Studies.

44  Exchanges generate money locally Muldrew, C. 1998. The Economy of Obligation. London: Macmillan. Nakamura, S. 1991. Nihon shoki shihon shugi shiron. Kyoto: Mineruva. Nishiyama, T. 1942. Santo no ichi shushichin no shakai teki kozo: Ekito ken Goriho no chosaki. Beijing: Institute of Rural Economy, Beijing University. Oishi, S. 1975. Nihon kinsei shakai no shijo kozo. Tokyo: Iwanami. Poos, LR. 1991. A Rural Society After the Black Death: Essex, 1350–1525. Cambridge: Cambridge University Press. Popkin, S. 1979. Rational Peasant: The Political Economy of Rural Society in Vietnam. Berkeley: University of California Press. Qiao, ZQ. (ed.). 1998. Jindai Huabei nongcun shehui bianqian. Beijing: Renmin chubanshe. Robequain, C. 1944. Economic Development of French Indo-China. Ward, IA. (trans.). London: Institute of Pacific Relations. International Secretariat. Scott, JC. 1976. The Moral Economy of the Peasant: Rebellion and subsistence in Southeast Asia. New Haven:Yale University Press. Shibahara,T. 1981. Nihon kindaika no sekaishiteki ichi: Sono hohoronteki kenkyo.Tokyo: Iwanami. Tanimoto, M. 2018. ‘Peasant society in Japan’s economic development: A special focus on rural labour and finance markets,’ International Journal of Asian Studies 15(2): 229–53. https://doi.org/10.1017/S1479591418000050 Tawada, M. 2007. Kinsei Shinshu no kokumotsu ryutsu to chiiki kozo. Tokyo:Yamakawa. Toyoda, T. 1952. Chusei Nihon shogyo shi no kenkyu. Tokyo: Iwanami. Vice, D. 1983. The Coinage of British West Africa and St. Helena 1684–1958. Birmingham: Peter Ireland Format. Wu, Z. 1936. Xiangcun zhibu gongye de yige yanjiu. Shanghai: Shangwu yinshuguan. Yang, QK. 1934. ‘Shiji xianxiang suo biaoxian de nongcun zijizizu wenti,’ Dagongbao, 30 August: 11. Zhu, JY. 2004. Guangxu Ningjin xianzhi. Nanjing: Fenghuang.

2 Stagnant currencies and stratified markets

1  Exogenous currencies naturally stagnate In the previous chapter, we confirmed that peasants exchanged with peasants proximately and that in the case of one-time transactions peasants needed currency on hand at marketplaces. We already showed that in some cases locals created currencies by themselves, such as zinc coins in Indochina. History, however, tells us that more commonly currencies were supplied exogenously rather than endogenously. The most common suppliers of currencies have been rulers who needed to collect taxes from people. Some may assume that tax-payability with currencies should entice people to more easily accept the currency as legitimate. Did currencies issued by governments, however, reach end users and return to the issuers? In theory, currencies should circulate, but modern investigations reveal that they often stagnate. A 1995 study estimates that, out of 277 billion dollars in total held in the US in 1992, households held 76 billion dollars, business 8 billion dollars, underground economy 36 billion dollars and destroyed or forgotten 10 million dollars, but remaining 147 billion US dollars of cash was unaccounted for. (Haughton 1995, 599). Some amount of cash may flow abroad clandestinely, but the missing quantity is too large for this explanation to be persuasive. Now firms make transactions mainly through bank transfers, drafts and the like. Even individual households do not need to use so much cash for their activities. From the viewpoint of estimated transaction demands, a huge amount of cash exists unnecessarily outside banks. Though the statistics were not then known, during the Great Depression in the 1930s John Hicks had already questioned why people would keep a currency that provided no return in value when they could instead purchase bonds that earned interest (Hicks 1935). He apparently considered only the intentional saving of currency and overlooked the possibility of the unintentional stagnation of currency. An investigation of coins collected from bank branches across the UK in 1968 provides us with important information on actual circulation of currencies.1 Before the introduction of decimal denominations in 1971, in cooperation with major banks such as Barclays, the Royal Mint surveyed how many

46  Exchanges generate money locally

coins of each denomination crossed the counters of banks in one day and the distribution of the dates of issuance among coins. This was necessary for the Mint to measure how many coins should be supplied under the new denomination system. Since the Mint had statistics on the quantities of minted coins of each denomination for each year, they could estimate how long a coin had remained outside the banks. The assumption was that the older the date of a coin, the less frequently that coin returned from holders to banks, but the question remained: to what extent was this true? The survey revealed that, besides a portion disposed of due to physical damage, which accounted for just 0.1 percent,2 two out of every 100 coins failed to reach the counters of the banks annually, or, in other words, those coins became economically inactive. This meant that half of all coins present in a given period would, within 33 years, become inactive.The wastage rate of coins varied by denomination. In the case of half pence coin, the smallest denomination, 37 coins out of 1,000 went missing annually. In the case of one pence, three pence and six pence, 22 coins, 17 coins and 21 coins out of 1,000 disappeared. Meanwhile, in the case of one shilling, two shilling and two shilling and half, only four coins, 12 coins and nine coins out of 1,000 went away.3 Apparently, the wastage rates of smaller denominations were larger than those of higher denominations (De Glanville 1970). An investigation of coins sampled from bank branches across the US in 1962 showed similar results. The average wastage rate was, incidentally or not, 2 percent.The wastage rate of the 10-cent coin, the smallest denomination in the sample investigation, was the highest (Patterson 1972). People do not save cheap, bulky coins as assets; we can therefore assume that nobody held such coins for any particular purpose. In other words, though some currency becomes intentionally hoarded by holders, a large amount of currency also becomes unintentionally inactive. The social science mainstream has remained blind to the stagnant nature of currency. Currency is easy to distribute but difficult to recall on demand; this is the flip side of currency’s ‘acceptability’ – its capacity to circulate anonymously. The wastage rates mentioned earlier were determined by the investigation of coins passing over the counters of banks. Banking systems work to collect currencies once distributed among numerous end users. Even in the UK and the US, which had the world’s most dense banking systems in the 1960s, two out of every 100 coins were losing their function every year. It is quite easy to imagine what happened to the circulation of currencies in the situation where banking systems do not organize the majority of households. Large portions of currency must have become stagnant every year and remained in the hands of end users. In places such as Nigeria, 60 percent of currencies, once issued, would not go back through the banking system again (Guyer 2004, 3). We may apply this same assumption to economies in the past, as well as to the situation of contemporary economies dominated by informal sectors. Archaeological findings prove it. Excavations of coin hoards have proved that a proportion of copper coins in China did stay among end users for a long time.4 Table 2.1 shows the contents

Stagnant currencies and stratified markets  47 Table 2.1 The distribution of era names of the copper coins from the Aba hoard, Sichuan, 1870s All Periods Tang 618–907 Song 960–1279 Ming 1368–1644 Qing 1644–1911 Foreign Qing Period Shunzhi 1644–1661 Kangxi 1662–1722 Qianlong 1736–1795 Jiaqing 1796–1820 Daoguang 1821–1850 Xianfeng 1851–1861 Tongzhi 1862–1874

7 30 11 16,491 13 5 176 5,400 2,776 4,787 1,945 1,402

Source: Li and Chen 2015, 33–9.

of a hoard from the Aba district, Sichuan province, which appears to have been abandoned in the 1870s. Qianlong coins issued almost 100 years earlier constituted one-third of the hoard. We will see similar features in hoards from 14thcentury China in Chapter Three. Over more than five centuries, copper coins worked in the same fashion. The physical presence of coins does not always mean that they are fully circulating. The statistics showing the high wastage rate of currencies are clear.The stagnancy conventionally would suggest that these currencies eventually became inactive. And yet we know that people took great care with their currencies. As Arthur Smith described vividly, merchants in Chinese villages would count all copper coins in their possession for several hours after closing their shops every evening (Smith 1899, 51). Nor is counting currencies for many hours entirely an activity of the past. For example, in present-day Nigeria, during an instance when two million naira ($25,000) was transferred in cash, tellers spent two days counting it (Guyer 2004, 3).Thus, from the past to the present, each user of currencies is keen to grasp how much currency is in their possession. High wastage rates of currencies from the viewpoint of the state as an issuer means that a large part of the currency flow became disarticulated from the main return streams to the issuer, and not that a significant proportion of currencies became completely inactive. The proliferation of copper coins in China means, rather, that they were staying with a number of end users and became disarticulated from the return stream to the interregional (or upper-level) market.

2  Unidirectional streams of small currencies As we saw, even with the development of banking systems which work well to collect idle currencies, a large quantity of coins is annually disengaged from major currency circuits, and this has been traced by surveys. The statistics from the investigations in the 1960s described earlier reveal another tendency that

48  Exchanges generate money locally

has not been considered.The survey happened to find that small-denomination coins in particular disappeared faster than large-denomination ones. These statistics indicate that the propensity of a currency to assemble can vary. The outcome of this modern experiment offers a new perspective on the monetary history of periods when there were no banking systems and when there was a greater dependency on coins. Two other contemporary statistics confirm how differently small currencies circulated from other currencies. Statistics from the US in 1975 make clear that the proportion of redemption of outstanding one-dollar notes was far higher than that of 100-dollar notes (Table 2.2). Another survey revealed that the demand for small notes has a conspicuous seasonality, such as the surges of their circulation in November and December because of holidays and related travel in the US (Kimball 1981, 44). These data apparently demonstrate that small denomination notes are used more frequently and circulate with seasonal bias. Can we assume that small currencies, once distributed in a local market during a busy season, would be assembled during a slack season and be transported to other places with high demand? Besides the cost of collection from numerous end users, obviously, the bulkiness of small denomination currencies makes their transportation costs so high that their owners hesitate to use them for distant payment. For example, under the Tang Dynasty (618–907) a Chinese bureaucrat lamented that the transportation of an amount of copper cash from a nearby region to the capital cost more than the value of the copper cash itself (Miyazawa 2007, 153). This lament has been repeated up to recent times. A British man in colonial West Africa failed to sell his unserviceable horse since its value in cowries would require 15 porters to carry it, and to whom he would have had to pay all the money they were hired to carry (Vice 1983, 13). The excavation of accumulated coinages weighing more than one ton has not been rare in the case of Chinese copper cash hoards. For example, an excavation from Xisaishan, a site dated to the Southern Song period, contained 110

Table 2.2  Currency outstanding, redeemed, in the US, June 1975 (billion US$) Denomination



Estimated to be in active use

$1 2 5 10 20 50 100 500 and over Total (billions)

1.7 0.0 1.9 3.7 5.9 0.8 1.0 0.0 15.0

2.6 0.1 3.6 10.2 26.8 7.7 21.3 0.4 72.7

0.0 0.1 0.7 4.5 17.8 6.4 19.8 0.4 49.7

Source: Anderson 1977, 26.

Stagnant currencies and stratified markets  49

tons of copper cash (Miyake 2005, 38), although the site is thought to have been a military base which might have stored currencies in larger quantities than civilian hoards. Huge quantities of copper cash must have been moulded at the Imperial mints and distributed across the empire. It was easy to distribute this cash, but difficult to recall it. It is hard to imagine that those bulky coinages of only fragmental value could make the return trip to the central place from which they were issued. Even if the exchange between a central marketplace and an end-marketplace is balanced in value, some smaller denomination currencies would still tend to stagnate at the bottom of a market hierarchy while larger denominations would tend to stay near the central place of issue. In fact, throughout the two millennia of the Chinese Empire, copper cash operated locally while silk and silver circulated interregionally, working in a complementary way. This unilateral movement of small currencies appeared in various ways throughout history and across the world. For example, in early 20th-century Kenya under British rule, the majority of currencies introduced in inland districts were small denomination coinages, particularly one-cent coins (Pallaver 2018). We cannot imagine that most one-cent coins distributed across Kenya could eventually return to Nairobi or any other centres. Excavations of one coin show us a case of the unilateral movement of small currencies in the medieval era. The Venetian Republic issued torneselli only for use in its Eastern Mediterranean colonies. As far as the latter half of the 14th century was concerned, torneselli are thought to have almost solely mediated transactions inside these colonies. The coins were 11 percent silver on average. They replaced soldini, which had been the basic low denomination coin previously and which had contained more silver (Stahl 1985, 8–9). It could not be mere coincidence that most torneselli were issued in the latter half of the 14th century when the silver supply fell considerably.This was part of a Eurasia-wide transformation, as we will see in greater detail later (Chapter Three). The hoard from Greece contains coins from a period longer than 100 years, though most had been issued within a 70-year period as shown in Table 2.3. Two torneselli seem to have been used to purchase a loaf of bread in the second half of the 14th century in Crete (Stahl 1985, 65). Torneselli were not designed to be recalled to the minting place. They were brought from Venice to the colonies and made no return trip. This official, unidirectional supply, however, was not always sufficient for local demands. The coexistence of a lot of torneselli from mints other than Venice reveals that local demand for the small currency stimulated unofficial supplies from places closer than Venice. Locals endogenously supplemented what exogenous currency failed to supply.

3  Stratified markets in agricultural societies Assume a market system in an agricultural society stratified between an interregional market and local markets. The former functions to collect agricultural products from the latter with strong seasonality, while the former distributes

50  Exchanges generate money locally Table 2.3 The distribution of minting periods with tornesello from a Greek hoard and an excavation Chalkis 1343–55 1355–68 1368–82 1382–1400 1400–14 1414–23 1423–57 1462–71 Total

7 169 846 2,729 725 294 4 18 4,792

Acropolis 48 87 150 42 12 3 342

Source: Stahl 1985, 24–5.

commodities to the latter at a relatively consistent rate. An investigation of 19th-century Java peasant households has revealed an asymmetric feature between seasonally fluctuating movements of selling products by peasants and flat movements of purchasing goods (Table 2.4). As already mentioned, the investigation of peasant households in the Huangling village in early 20th-century Northern China explicitly showed that most monetary income was obtained in the few months during and after the harvest. The same tendency can be confirmed in other peasant economies such as Java, as we saw in Table 2.4, though the peasants in Java were more dependent on credit/debt. The only difference between Northern China and Java is that, in the former, the busy season came with the millet harvest in autumn, while, in the latter, it came with the rice harvest in spring. Considering the high seasonal demand for money in rural areas, it is not surprising that the amount of coins issued by the Reserve Bank of India would surge every winter (Reserve Bank of India 1938, 49). The interest rates of cities in China rose after harvest seasons as well. For example, according to a Chinese magazine in the 1920s, the monthly interest rate in Ji-nan, Shandong, was 1 percent in spring, 0.8 percent in summer, 1.8 percent in autumn and 1.5 percent in winter.5 The distribution of currencies from interregional markets to local ones, or, looking at it in a different way, from merchants at the wholesale level to peasants at the retail level, was concentrated within a few months. In other words, currencies poured into small households over a short temporal period. At the same time, monthly movements of their expenditures were flat rather than dynamic as we saw in the case of 19th-century Java in Table 2.4. It means that the velocity in the return flow of currency from local markets to interregional ones, or from peasants to merchants, was more constant than the downward flow from city to village. In addition, we may assume that, while large-denomination currencies were needed in interregional markets, small ones were in demand in

Stagnant currencies and stratified markets  51 Table 2.4  Money balance of a rice area in Java around 1875 (0/00) Month

Money from crops

Credit obtained

Debt repaid




March April May June July August Sep Oct Nov Dec January February Year

50 200 400 150 20 20 90 30 10 10 10 10 1000

0 0 0 0 0 0 0 50 80 60 10 0 200

0 100 150 50 0 0 0 0 0 0 0 0 300

0 0 0 100 80 50 40 10 10 10 0 0 300

40 50 70 50 40 40 40 40 40 30 30 30 500

0 0 0 0 0 0 0 30 40 30 0 0 100

Source: Wolters 2005, 181.

local markets. The downward flow of money needs to accommodate exchange in fractional currencies, while the return up-flow must allow for the reverse exchange into larger currencies. Considering the transport, exchange and other expense costs per unit, distributing currencies could be easier than collecting the same currencies in return. Thus, given the need to find suitable denominations, and the temporality of demand for different currencies, the relationship between the distributed flow of currencies to the lower market and the return flow to the upper market can be thought of as asymmetric. Even if the annual amount of purchases by peasants from merchants in cities might have been equivalent to that of sales in the opposite direction, the components of the currency flows would be different. Thus, though exchangeable, the streams of large-denomination currencies and those of small-denomination currencies were often separate. The heterogeneity between interregional and local markets often appeared in the relationship between currency supplies and price movements. Through the 16th and 17th centuries, China absorbed a significant quantity of silver in exchange for silk and porcelain, as did India in exchange for cotton. During this period, however, China and India recorded no clear continuous spikes in prices. Some have argued that precious metals were not used as money but instead hoarded for non-monetary purposes.6 The same argument was made for the Atlantic slave trade. The trade must have brought a huge quantity of precious metals to Africa, but there were no significant price spikes there (Hogendorn and Johnson 1986, 146). Some argued that precious metals did not work as currency but functioned merely as hoard in Asian and Africa (Kindleberger 1989, 62–4, 70–2). However, once we recognize the asymmetry between interregional and local markets, it would not seem strange that a large quantity of precious metal imports was not accompanied by ‘hyperinflation’ in Asia and Africa in the early

52  Exchanges generate money locally

modern period. Importantly, China and India experienced mild and long-term inflation through the 18th century when copper coins, cowries and other small currencies for local usages increased in supply, which we will see in more detail in Chapter Four. Following liquid dynamics, we can theorize the observations mentioned earlier as follows. As far as the flow between upper-level (interregional) markets and lower-level (local) markets is concerned, the monetary current down to lower-level markets did not match the current to the upper-level markets. Like a vortex appearing on the edge of a stream, a currency flow entering the seasonal cycle of a lower-level market would separate from the currency currents among merchants and between cities and then could not easily return to the upper-level market (Appendix, Model I). These currency flows separated from the mainstream might appear to be ‘hoarded’ because of their diminished connection with business activities in upper-level markets. We must, however, distinguish such involuntary ‘hoarding’ from voluntary hoarding based on the holder’s choice, such as Keynes assumed. The large amount of coins lost for unaccountable reasons (as mentioned earlier) suggests that it was not easy for the former to rejoin the main current. The argument here is that the frictions produced by varying streams of currencies will inevitably result in an ‘unaccountable loss’ of currencies. From this perspective, such lost currencies are now accountable.

4  Informal agreements generate endogenous currencies If the supply of exogenous currencies fails to satisfy monetary demand for any reason, locals will not hesitate to supplement the shortage by themselves, as we already suggested in mentioning the case of imitation tornesello. History is full of incidents in which local currencies were generated by fronting an emergent ‘famine’ of cash. Let’s take a look at some cases. In 1606, Quanzhou, the largest port city in Southern China in the period, faced a famine due to poor harvest of grains and the increased circulation of illegal coinage. In order to stabilize the local market, the city’s magistrate tried to set a maximum price for grains to make it easier for locals to buy them. He also prohibited the use of unofficial coins. Though his actions may seem appropriate from a modern perspective, to his contemporaries he appeared to be blind to reality. The traders at the port protested the price controls and the prohibition of private coinage and went on strike, causing the market to close. Those in opposition to the magistrate’s rules insisted that price controls discouraged maritime traders from importing grains and that the private coins would disappear as soon as they became unnecessary. The magistrate lifted the regulations, which encouraged traders to import larger quantities of grain, causing market prices to decrease to a normal level, whereupon private coins disappeared (Chen MR. 1995–97.). This process indicates that increasing the amount of privately made coins allowed prices to rise high enough to absorb grains from other regions.When facing famine, in addition to securing actual food supplies,

Stagnant currencies and stratified markets  53

it is also important to avoid disturbing food distribution patterns. In the case of Quanzhou, the local market generated sufficient liquidity to import grains from outside the city. The magistrate failed to notice that the system worked locally and also could not distinguish stability from formality. Depending on the situation, local currencies appeared in a variety of forms other than metallic coins. In the case of China after the 18th century onward, notes issued by local dealers became popular. What they typically called, qianpiao (notes in terms of copper coins). In 1929, Xianrendu, in Laohekou County, Hubei province – a town in the mid-Yangzi region – suffered a serious currency shortage. Soon after, notes issued by 36 different shops began to circulate in town, even though the area comprised only around 300 households. The stores that issued the notes included 13 grocers, nine grain dealers, two seed oil mills, two medicine dealers, three liquor shops, two tea houses, a wholesale trader, a tofu shop, a rice cake shop, a silver dealer and a butcher (Chen XR. 2012, 116). It seems that all kinds of shops in the town issued their own notes, and that the issuers included no official financiers. In other words, any type of shop functioned or could function as an issuer of currency. In Chapter Four we shall return to native notes in China in comparison with other local paper money in England and Japan. China is not the only example of such locally organized currencies: societies already dependent on modern banking systems, such as the 20th-century US, have seen similar occurrences. On 7 December 1931, the only bank in Tenino, Washington, the Tenino Citizen Bank, closed its teller services, paralyzing the town’s entire business sector. This was not an isolated case: In the last quarter of that year, 1,055 banks across the US suspended their services against the backdrop of the Great Depression. However, the emergency measure adopted by the Tenino Chamber of Commerce to counteract the bank’s closure was an uncommon one. It issued certificates in an amount equivalent to 25 percent of the bank’s total deposit value. Some of these certificates had a face value of 25 cents and were printed on wooden tablets. This wooden money went into circulation, though only within the town. Needless to say, its function as a currency ended once the formal currency system was reintroduced (Preston 1933). The Chamber of Commerce took the initiative in the case of Tenino and the acceptance of wooden currency was not backed by any regulations. There was therefore still a possibility that the bank deposits would become nonperforming debts.Thus, it was neither official support nor the intrinsic value of the tablets, but a loose agreement among locals that sustained the acceptability of the wooden currency. Such informal and loose agreements have emerged around the world in various forms throughout history. Local economies have often encountered problems with small denomination currencies in particular, rather than currency in general, due to the former’s tendency to experience shortages. Since the majority of exchanges in the local marketplaces are wage-goods-level transactions, small denomination currencies have naturally been in greater demand.

54  Exchanges generate money locally

5 How merchants conducted business in stratified markets Under the condition of stratified markets mentioned earlier, how do merchants engaging in trades connecting an interregional market and a local market manage the settlement for both markets? Here we investigate in greater detail the account books of a grocery shop in early 19th-century China, Tongtaihao, which we mentioned in Chapter One. The account books contain volumes used specifically to register the receipt and payment of silver; these were called churu yin liushui zhang (hereafter ‘silver book’). Looking through the account books, we can see how silver ingots worked in actual businesses in traditional China.Table 2.5 shows all items in the silver book recorded on the 1st day of the 10th month of the 21st year of Daoguang (1841 CE). The silver book records the entry and exit of several types of silver ingots (baoyin, yankeyin, ganbaiyin and baiyin), but baoyin ingots weighing 50 liang occupied the majority of documented currency in value. Moreover, unlike other kinds of silver, each baoyin ingot was given a serial number upon receipt, beginning with ‘1’ for the first ingot received that year. In Table 2.5 the baoyin ingot received from the shop Longxianghao was the 393rd baoyin ingot to enter Tongtaihao in that year. On the same day, that 393rd baoyin ingot was paid to another shop, Huangxueguang. A baoyin ingot commonly weighed 50 liang, but variations in the fineness and weight of the ingots led traders to value them at a premium or discount. Baoyin ingots were popular, but they were not at all fungible – in other words, one baoyin ingot could not be exchanged for another. The 393rd silver ingot thus moved from Longxianghao to Huangxueguang via Tongtaihao, and we can assume that, accompanying the ingot’s movement, Tongtaihao sold a commodity to Longxianghao and purchased another from Huangxueguang. The total value of the two silver ingots numbered 393 and

Table 2.5 The entry and exit in the silver book on the 1st day of the 10th month of the 21st year of Daoguang (1841 CE), the Tongtaihao From Longxianghao Bitongxing Tianxianghao


baoyin 393 Baoyin 394 Baoyin 395 Baoyin 396









50 50

Tianxianghao 297,560 593,610


liang baoyin 394 baoyin 393 baoyin 395 baoyin 396


50 50


50 50

302,560 605,620

Source: Daoguang ershiyi nian yin churu, Tongtaihao. Institute of Economics, the Chinese Academy of Social Sciences, no. qitadiqu han2 c024.

Stagnant currencies and stratified markets  55

394 was, at the time of receipt, 297,560 wen of copper coins. At the time of payment, however, their combined value was recorded as 303,060 wen with the additional note, ‘two months.’ Another account book, meanwhile, clarifies this gap between the value at the time of receipt and at the time of payment. A category of account book called churu qian liushuizhang (hereafter ‘cash book’) appears to have been used to record all Tongtaihao trades, except for miscellaneous purchases for daily usages, etc. Table 2.6 shows all of the transactions recorded in the cash book on the 1st day of the 10th month of 1841 (the same day as seen in Table 2.5).The total transactions that day amounted to more than two million wen of copper coins (or 2,000 chuan, a unit of coin string), which was equivalent to around 700 liang (19 kg) of silver. Though the grocery store was located in a local town, the scale of its business was quite large. The cash book included entries for both revenue (ru) and expenditures (chu). Importantly, equal amounts of wen copper coins frequently appeared as entries in both the revenue and expenditure columns. We signify the cases in italic and bold style in the table. This suggests that these transactions may have been made not through the exchange of currency, but through transfers recorded

Table 2.6 Revenue and expenditure in wen on 1st day, 10th month, Daoguang 21st year (1841 CE), the Tongtaihao Revenue copper coin from Zengtaidian from Simeidang copper coins from Yandian from Hufang from Longtaihao copper coin from Longxianghao from Bitongxing from Tianxianghao silver from Tianxianghao from Shuangchengdian from Wangfasheng from Chengli Fashenghao from Jushenghao from Huanghesheng from Huanghesheng from Jushenghao from Tianxianghao Total

Amount (wen) 1,000 2,000 1,480 450 4,000 128,000 5,500 14,400 146,262 149,788 297,560 605,620 150,000 40,000 8,000 28,000 50,000 72,000 9,000 40,000 317,872 2,070,932


Amount (wen)

to Yinzhongli to Longtaihao to Tongxi-youfan to Tangji to Liubingchun to Liu Xiang to Tangji to Yuanshuzhai To Yuanqingao silver to Huangxueguang to Tianxianghao to Yongshunhao to Jiangtongsheng

70,000 1,000 2,000 2,000 1,480 450 4,000 128,000 5,500 593,610 303,060 302,560 150,000 40,000

to Sunmouyou to Tianxianghao

8,000 150,000

to Tongdian youfang to Longxianghao travel costs to Tianxianghao to Liumoumu Total

9,000 40,000 80 317,872 1,000 2,129,612

Source: Daoguang ershiyinian bayue churu liushui zhang. Chinese National Library,Wenjinge, no. 131001-10.

56  Exchanges generate money locally

in the account books of different merchants. For example, Tongtaihao received a commodity valued 2,000 wen from the Zengtaidian and, on that same day, paid the same amount to the Tongxi oil producer. Probably, Zengtaidian likewise recorded 2,000 wen on the expenditure side of their account books, while Tongxi recorded 2,000 wen on the revenue side of their books. It was not necessary to use currency as long as the amounts of the two transactions were exactly the same. Consider the following three cases: Tongtaihao received 146,262 wen from Longxiang; it received 149,788 wen from Bitongxing; and it received 297,560 wen from Tianxianghao.According to the silver book of the same day (Table 2.5), these revenues were received in the form of silver ingots. In the transaction with Longxiang, Tongtaihao received 146,262 wen through the receipt of the 393rd baoyin ingot, most likely in exchange for commodities. On the same day, Tongtaihao used the ingot to pay Huangxueguang. However, Huangxueguang probably did not give Tongtaihao any commodities on that day, but instead promised to deliver goods two months later. Therefore, 7,010 wen was added as interest. This interest rate was almost 1 percent per month. In the case earlier, in which the three parties used a silver ingot in their exchanges, the Tongtaihao/Longxiang transaction and the Tongtaihao/ Huangxueguang transaction are disconnected. In contrast, the majority of Tongtaihao’s transactions, as recorded in the cash book, were made in pairs of equal amounts of currency.This allowed the merchants to avoid using currency and is called the multilateral clearance system. The cash book records only three pairs of revenue and expenditure for the 8th day of the 8th month of the 21st year of Daoguang (1841 CE). These entries allow us to confirm our interpretation of the account books. Tongtaihao made three transactions on that day: it purchased goods whose value was 95,000 wen from Longtaihao and sold goods valued at 95,000 wen to Yishengheng; it purchased goods whose value was 1,400 wen from Longtaihao and sold two pieces of cloth, lubaibu, to Tangji; and it purchased goods whose value was 159,276 wen from Longtaihao and paid 159,276 wen in silver to Longtaihao. In most cases we cannot identify what commodities were sold and purchased, but in the case of the sale to Tangji, the account book noted two pieces of lubaibu, a kind of cloth, probably white cloth made in Shandong.7 Most of the dealers appearing in the account books presumably carried out business in the same town as Tongtaihao. For cases in which the grocery shop engaged in transactions with traders from outside the town, the place name is recorded in front of the name of the merchant shop, such as the Fashenghao shop in Chengli. Importantly, for such business with outside merchants, silver was oftener used to settle the transaction. As seen in Table 2.6, transactions valuing a total of 1,450 chuan (almost twothirds of the day’s total transactions, which were worth 2,070 chuan) were carried out through multilateral counter-trade. Though copper coin (chuan) was used as the unit of account, only 16 chuan of copper coins – or less than 1 percent of the sum of all transactions – exchanged hands on that day. Thus, the

Stagnant currencies and stratified markets  57

merchants carried out business in terms of copper coins, but they did not rely so much on material coins. Silver ingots circulated more than copper coins, but transactions that used silver occupied, at most, 30 percent of the day’s transactions. Through the 21st year of Daoguang (1841 CE), silver ingots and copper coins were used in roughly the same proportion for Tongtaihao’s transactions. As we argued in Chapter One, China fronted silver outflow due to trade deficits from the 1830s through the 1850s. Now we can assume that silver outflow in the period asymmetrically impacted distant and proximate exchanges. Shortages of silver must have made the settlement of distant transactions more difficult and therefore forced dealers to depend on bilateral counter-trade for distant exchanges. At the same time, by substituting multilateral book transfers for the exchange of silver ingots, merchants in the same town could, to some extent, continue to carry out business in proximate exchanges despite the liquidity shortage. In traditional Chinese business history, there are other examples of similar exchanges carried out through transfers recorded in account books. The bestknown case is that of the multilateral clearance system used by Chinese money exchangers. In the mid-19th century, exchangers based in Ningbo, a major financial centre, are thought to have made deals primarily through book transfers. From the late 19th to the early 20th centuries, exchangers in Shanghai, most of whom came from Ningbo, preserved the system. Various systems of imaginary silver liang – a unit of account independent from tangible silver – functioned in cities throughout late 19th-century China. The silver accounts allowed financiers to carry out multilateral book transfers without depending on the use of material silver. Historians have long been familiar with the multilateral clearance system used by financiers in 19th-century China. Chinese merchants called the custom guozhang. The discovery of the use of multilateral book transfers among merchants who dealt with commodities in a local town, however, suggests that such transfer systems may have been far more popular than previously thought. This case confirms that currency-less transactions are not necessarily credit/ debit ones. Merchants in towns such as Tongtaihao exhibit a strong tendency to equalize the value of sales and purchases in order to avoid carrying over an unsettled balance to the next day. The strong preference for instant clearance becomes apparent when we compare the account books of Tongtaihao with those left by Taiyihao (or Taiekigo in Japanese), a dry-seafood trader in Nagasaki, Japan. This merchant mainly engaged in the export of dry seafoods collected in Japan to Taiwan and southern-coastal China.The account books made at the turn of the 20th century share a basic structure with the account books from Tongtaihao. They followed the Chinese-style accounting system. However, though sometimes it appeared, we cannot clearly find a tendency to complete settlements on the same day. The biggest difference between Tongtaihao and Taiyihao is that the latter’s business depended heavily on the banks. Keeping the traditional Chinese account style, Taiyihao made settlements basically through its current accounts in Japanese banks such as the Yokohama Specie

58  Exchanges generate money locally

Bank, which was transfer-dominated, and the Eighteenth Bank, which represented the local business circuit in Nagasaki. Comparing this with the case of Chinese merchants in Nagasaki, we can understand that the multilateral clearance system among merchants in Ningjin originated from financial conditions rather than the bookkeeping system itself.8 This system was thus clearance-oriented, rather than credit-oriented, since the merchants seemingly hesitated to make deferred settlements. Though the system appeared to work similarly to the cases in Quadrant III of Figure I.1 (named/proximate), it actually functioned more with the attributes of Quadrant IV (anonymous/proximate), as dealers operated under the principle of engaging in one-time transactions, even with nearby merchants. Although the types of sources vary from the Chinese case, the situation of Japanese merchants in the early 19th century offers a clear contrast. At the time, in accordance with domain edicts, many merchants made lists of their assets. As we already saw the case of the Takahashi family, the majority of assets took the form of credit. In short, merchants under the Tokugawa regime tended to establish the best customer relations, through which they issued and received drafts instead of settling in cash. Meanwhile, in China, a preference for onetime transactions prevailed among merchants in towns and among peasants in rural marketplaces. Between sellers and buyers, there are two ways to avoid or reduce using currency in exchanges. One way is to defer payment: credit/debt. In Chapter One we saw cases of deferred payments made locally; however, deferred payments can also operate for distant exchanges. Another way to save currency is to simultaneously exchange one commodity for another of the same value. We call this form of exchange ‘counter trade’ as we defined it in Chapter One. Countertrade, rather than ‘barter,’ is a more appropriate term for this form of exchange because it clarifies the fact that the two commodities are exchanged on the basis of values calculated in terms of money. Barter, on the other hand, can include exchanges that are not evaluated in terms of money. Bilateral countertrade, both proximate and distant, is a phenomenon that has appeared across the world and throughout history. However, the case among Tongtaihao and other merchants in the same town vividly demonstrates that proximate counter-trade has occurred multilaterally rather than bilaterally. Multilateral counter-trade, thus, happened among local merchants sharing a book transfer system, but it could not easily be established among merchants engaging in distant trade.This multilateral clearance system through account books conditioned the establishment of local imaginary money, as we shall see in the next section.

6 Locally differentiated currencies and dematerialized money As the case of Tongtaihao in the previous section shows, local merchants created a functional differentiation among currencies. By functional we mean that the differentiation occurs to ease monetary difficulties that impede transactions. Political and social factors can also differentiate currencies, but we will clarify

Stagnant currencies and stratified markets  59

different reasons. What kinds of difficulties can be solved by the differentiation of currencies? One apparent difficulty is the seasonal clash between monetary demand and supply. Consider some cases in history. Currency often appreciated in each harvest season, typically in autumn. When many independent peasants want to sell their products with spot transactions in cash, rural marketplaces require a plentiful supply of currency. Unless a sufficient stock of currency is reserved in rural marketplaces, a significant quantity of currency must move from towns to these marketplaces.The interest rate in cities thus usually rises in the harvest season, as urban areas suffered from currency shortages due to this outflow to rural areas. However, if one currency is used in the collection of peasant products while another mediates transactions in towns, the monetary tension between demand and supply in the harvest season can be eased. Such functional differentiations of currencies have often occurred in history. For example, copper coins in China and cowries (shell monies) in Bengal tended to increase in value against silver during the harvest season (Mahapatra 1969–1970). Both were fractional in value, and thus suitable for distribution among peasants, but they were unsuitable for merchants in towns. Thus, silver worked to mediate transactions between urban traders in both countries. Such precious metal monies were more appropriate for large-size transactions. Under this condition, with unfixed exchange rates between them, copper coins (or shell monies) and silver could work independently enough to facilitate exchanges in both rural and urban circuits. In this case, copper coins (or shell monies) paired with peasant products while silver paired with urban trade. However, this differentiation of currencies was more than a simple division between rural and urban trade or between small- and large-size transactions. Even two different kinds of silver coins could perform different functions. For example, in some districts of Bengal, the Arcot rupee appreciated against the Sicca rupee during the harvest season. Unlike the Sicca rupee, which was preferable for tax payments and remittances to Calcutta, the Arcot rupee in Bengal was more popular among peasants and artisans. During tax collection seasons, the Sicca rupee appreciated against the Arcot rupee (Mitra 1991, 72–3). We can find that this seasonality also affected relationships among different silver monies. Also, we know the case of late 18th-century Jessore, where the Sicca rupee was used to trade for rice while the French Arcot rupee was used in the salt trade. Similar differentiation between silver coins appeared in China as well. In Jiujiang, an inland port in the mid-Yangzi region, the Mexican silver dollar and the Japanese silver dollar were in circulation with other currencies in the early 20th century. The dealer who brought porcelain from Jingdezhen to Shanghai used the Mexican dollar, while the import of kerosene oil from Shanghai to Jiangxi depended upon the Japanese silver dollar. We will also consider these cases in detail in Chapter Five. In the cases mentioned earlier, the value of two silver coins thus fluctuated differently according to demand and supply, despite sharing the same intrinsic content: silver. These oscillations in the exchange rate between two kinds of coins of the same metal strongly suggest that the value of each coin derived

60  Exchanges generate money locally

from the demand for the transactions they mediated, rather than from their intrinsic contents. That is why, in history, a particular silver coin often appreciated far above the value of its silver content. Later, we will see a typical case in the circulation of the Maria Theresa dollar in the Middle East and Africa (Chapter Five). In history, thus, a circuit of trade dealing with particular goods often appears to pair with a particular currency. Multiple currencies circulated even in one town for different goods.We call this coupling a currency circuit (Kuroda 2008). A combination of three conditions generates a currency circuit. First, a large number of independent dealers must be engaged in the exchanges, since a small number of traders do not always need to use currencies to make transactions. Second, exchanges must be dominated by spot transactions (one-time transactions), as deferred payments or future trades do not require payment in currency. Third, traders must find it difficult to provide a currency on demand, otherwise, they would prefer to use a universal currency. It is important to note that one person exchanging in a currency circuit does not engage exclusively in that one-currency circuit.Various characteristics of different exchanges require a parallel variety of devices. For example, in the colonial Belgian Congo, the currency used for purchasing canoes was not the same currency that was used for buying fish (Guyer 2004, 37). A fisherman must have lived crossing between two currency circuits. A situation of multiple currency circuits has the precondition of an open economy, rather than an economy closed by race, religion or other factors. In the same way that endogenous currencies were generated by informal loose agreements among locals, as seen in Section Four, currency circuits did not depend on enforcement by authorities. As we will confirm throughout this volume, such a division of functions among monies was ubiquitous throughout history. However, the function of each currency was not self-evident. Fortunately, for early 20th-century Yingkou, China, as shown in Table 2.7, foreign observers left descriptions distinguishing one currency’s role from another. Daily transactions of fragmental value in the town were mediated by copper coins. Small-size transactions sometimes were dealt with in former provincial silver dollar notes, which were mainly used for the trade of local products collected by carts.The local products collected by carts were also mediated by current silver dollar notes, which were used for settling deals in terms of local silver liang. The local silver liang, called guoluyin, was used in transactions for local products collected by train. Unlike those collected by carts, transactions for products transported by train were relatively large. The guoluyin is a silver unit, but there was no actual silver representing the unit. It worked only as a unit of account among major merchants in the town. Merchants used a multilateral clearance system shared through their account books, and thus rarely made payments in cash. Current silver dollar notes were only paid when they needed to settle accounts, usually at the end of the financial season. The system appears similar to what we saw in the case of Ningjin, though the unit of account used by Ningjin merchants was recorded in terms of copper coins.

Stagnant currencies and stratified markets  61 Table 2.7  Monies and their usages in Yingkou, China, 1930 Currency and unit of account


Former fengtianpiao [provincial silver dollar notes] Current fengtianpiao [provincial silver dollar notes] Silver dollars Guoluyin (a local silver tael) ‘Chosen’ banknotes (gold yen) ‘Yokohama Specie’ banknotes (silver yen) Copper coins (wen)

Local products collected by carriage, small transactions Local products collected by carriage, settlement in terms of local silver tael Maritime Customs, remittances by foreigners Local products collected by train, cotton Japanese import goods, railway fares Bills of exchange Small-value daily transactions

Source: Hirakawa 1936, 76.

Meanwhile, remittances made by foreigners and payments of maritime customs fees required the use of silver dollars, while dealing with Japanese imports or paying South Manchuria Railway fares required Chosen banknotes denominated in terms of gold yen. However, another Japanese banknote also had an important role in the city: Yokohama Specie banknotes denominated in terms of silver (not gold) yen worked as bills of exchange. The economy of Yingkou had a close relationship with Shanghai, where a local silver liang, called guiyuan, dominated in trades among major merchants. For the sake of making settlements with other Chinese cities, Yingkou merchants used the Japanese banknotes denominated in silver as bills of exchange. After the introduction of the gold standard in 1895, Japan did not use currency denominated in silver, but the special banknotes continued to circulate in China to settle inter-city transactions. The exchange ratio among monies changed daily, reflecting the fluctuating supply and demand for each currency. Major merchants in the city did not depend on either currency. They primarily conducted business through book transfers in which a local unit of account with no substance was used. The imaginary aspect of the settlements enabled merchants to be indifferent to the circulating currencies and thereby avoid disturbances caused by sudden currency shortages. The local unit of account with no substance, guoluyin, worked as a numeral to neutrally value all currencies. Multilateral clearance systems among merchants shared through their account books were not always accompanied by imaginary money. For example, the merchants in Ningjin booked their transactions in terms of copper coins that actually circulated.The coexistence of numerous currencies behaving independently – as in Yingkou – might have caused merchants to invent a device to measure them. Or the existence of imaginary money might have encouraged the creation of new currencies, which would have accelerated the multiplicity of currency. Imaginary money thus could work only in association with coexisting currencies and was therefore part of a system of complementary monies. Imaginary monies shared by local merchants were not found only in China. We will see later that imaginary money existed across the world more frequently

62  Exchanges generate money locally

than is generally recognized. Here, let us confirm a similar situation in the regions under the Ottoman Empire. For the purposes of providing for the suspension of silver coinage and adopting the gold standard (equivalent to 4.40 pre-war gold dollars) the monetary unit of the empire, the Ottomans issued the Kararname gold pound of 1296 (1880 CE). The pound was declared equivalent to 100 piasters sagh (literally ‘good, sound’), and a silver piaster (1/100th of a pound) was put into circulation. However another piaster, ghursh shruk (‘defective piaster’), having no existence as a coin and without legal status, was used as the unit in market prices. Its value in terms of the Turkish gold pound varied from district to district. For example, the following rates for the Turkish pound in piasters (shuruk) prevailed in 1914 in five of the principal cities: Tripoli, 123; Hom, 124; Beirut, 124.625; Sidon, 125; Damuscus, 130.75 (Said 1935, 24). Moreover, in certain places, in one market it had more than one value according to the type of goods traded. Until the early 20th century there were full currency circuits and imaginary monies across the world. Foreign observers coming from countries which had already established a territorial single currency described the native monetary systems with currency circuits and imaginary money as chaotic and eccentric. For example, in 1889, J. Edkins lamented ‘prominent annoyances’ when using money in China.9 Even Edkins admitted, however, that the Chinese people were not burdened by these conditions. Behind the ‘chaotic’ appearance, they were surely orderly and endogenously mediating exchanges in their own ways.

Notes 1 The original idea of this investigation is as follows. ‘The procedure, briefly, is that we ask some 40 banks, distributed regionally, to open on a certain day a random bag of coins of each denomination received from customers in the normal course of business, and record on forms, to be provided, the numbers of coins of each date. When these figures are related to the number of coins of each date actually issued, it is possible to deduce statistically a circulating figure, and a wastage factor.’ PRO, MINT20/3965 ‘Decimal coinage’ by D.H.S. Biggs, 25 July 1967. 2 PRO, Mint 20/3965, 20 May 1969. 3 PRO, Mint 20/3965. 4 In 1912, a year of low cash reserves, the French West African government found 287 franc worth of 5- and 10-centime pieces dated 1855, struck with the image of Napoleon III among collected coins. Manning 1982, 312. 5 ‘Ji-nan zhi jinrong jiguan yu tonghuo,’ Zhongwai Jingji Zhoukan 84:9. 6 The multiple strata of monetary circulation described here reveal the defects inherent in two common viewpoints. Frank (1998), for example, pays attention only to horizontal movements of silver, or trade balances, whereas Kindleberger (1989) dichotomizes the usage of silver into coinage and hoard. 7 Daoguang ershiyinian liuyue churu liushui zhang, Chinese National Library, Wenjinge, no. 131001-9. 8 The documents including the account books from Taiekigo are held in the Nagasaki Museum of History and Culture. They are registered as Taiekigo kankei shiryo (‘materials relating to Taiekigo’). 9 North China Herald, 1889, fol. 411, cited in Kann 1926, 415.

Stagnant currencies and stratified markets  63

References Anderson, PS. 1977. ‘Currency in use and in hoards,’ New England Economic Review 2: 1–30. Chen, MR. 1995–97. ‘Quannan zazhi,’ in Siku quanshu cunmu congshu bianzuan weiyuanhui (ed.) Si ku quan shu cun mu cong shu, vol. 247, 859. Jinan: Jilu shushe. Chen, XR. 2012. Minguo xiaoquyu liutong huobi yanjiu. Beijing: Zhongguo shehui kexue chubanshe. De Glanville, RG. 1970. ‘The numbers of coins in circulation in the United Kingdom,’ Studies in Official Statistics, research series 2: 1–14. Frank, AG. 1998. ReOrient: Global Economy in the Asian Age. Berkeley: University of California Press. Guyer, JI. 2004. Marginal Gains: Monetary Transactions in Atlantic Africa. Chicago: University of Chicago Press. Haughton, J. 1995. ‘Adding mystery to the cases of the missing currency,’ Quarterly Review of Economics and Finance 35: 595–602. https://doi.org/10.1016/1062-9769(95)90056-X Hicks, JR. 1935. ‘Suggestion for simplifying the theory of money,’ Economica, new series 2: 1–19. https://doi.org/10.1016/B978-0-12-663970-4.50005-0 Hirakawa, S. 1936. Manshu ni okeru tsuka kin’yu no kako oyobi genzai. Fengtian: Manshu jijo annaisho. Hogendorn, J. and Johnson, M. 1986. The Shell Money of the Slave Trade. Cambridge: Cambridge University Press. Kann, E. 1926. The Currencies of China: An Investigation of Gold & Silver Transactions Affecting China, with a Section on Copper. Shanghai: Kelly & Walsh. Kimball, RC. 1981. ‘Trends in the use of currency,’ New England Economic Review, Federal Reserve Bank of Boston, September–October: 43–53. Kindleberger, CP. 1989. Spenders and Hoarders:The World Distribution of Spanish American Silver, 1500–1750. Singapore: Institute of Southeast Asian Studies. Kuroda, A., 2008. ‘Concurrent but non-integrable currency circuits: Complementary relationship among monies in modern China and other regions,’ Financial History Review 15(1): 17–36. https://doi.org/10.1017/S0968565008000036 Li, J. and Chen, YJ. 2015. ‘Sichuan Aba luchang Qingdai qianbi jiaocang qingku jianbao,’ Zhongguo qianbi (1): 33–9. Mahapatra, PR. 1969–1970.‘Currency system in medieval Orissa,’ Quarterly Review of Historical Studies 9(2): 72–80. Manning, P. 1982. Slavery, Colonialism and Economic Growth in Dahomey, 1640–1960. Cambridge: Cambridge University Press. Mitra, DB. 1991. Monetary System in the Bengal Residency. Calcutta: Firma K.L. Mukhopadhyay & Sons. Miyake, T. 2005. Chugoku no umerareta senka. Tokyo: Doseisha. Miyazawa, T. 2007. Chugoku dosen no sekai: dosen kara keizaishi he. Kyoto: Shibunkaku. Pallaver, K. 2018. ‘Paying in cents, paying in rupees: Colonial currencies, labour relations and the payment of wages in Kenya (1890–1920),’ in Hofmeester, K. and de Zwart, P. (eds.) Colonialism, Institutional Change and Shifts in Global Labour Relations, 295–325. Amsterdam: University of Amsterdam Press. Patterson, CC. 1972. ‘Silver stocks and losses in ancient and medieval times,’ Economic History Review, second series 25(2): 205–33. Preston, HH. 1933. ‘The wooden money of Tenino,’ Quarterly Journal of Economics 47(2): 343–8. doi:10.2307/1883693 Reserve Bank of India. 1938. Report on Currency and Finance for the Year 1937–38. Bombay: Times of India Press.

64  Exchanges generate money locally Said, H. 1935. Monetary and Banking System of Syria. Beirut, Syria: American Press. Smith, A. 1899. Village Life in China: A Study in Sociology. New York: Fleming H. Stahl, A. 1985. The Venetian Tornesello: A Medieval Colonial Coinage. New York: American Numismatic Society. Vice, D. 1983. The Coinage of British West Africa and St. Helena 1684–1958. Birmingham: Peter Ireland Format. Wolters, W. 2005. ‘Southeast Asia in the Asian setting: Shifting geographies of currencies and networks,’ in Kratoska, PH., Raben, R. and Nordholt, HS. (eds.) Locating Southeast Asia: Geographies of Knowledge and Politics of Space, 175–202. Singapore: Singapore University Press.

Part 2

A global history of monetary delocalization

3 The ignition of monetary delocalization An unexpected remnant of the Mongolian regime, c.1300

1 Independent monetary systems in pre-13th century civilizations This chapter will indicate that, contrary to the popular idea that global economic transformation started only after the European expansion to the Americas, in fact a common unit of account denominated in silver which the Mongol regime happened to establish for long-distance exchanges across the Eurasian landmass had already ignited a global monetary delocalization. Unlike studies focusing only on unification, such as the pioneering work by J. Abu-Lughod, it will put emphasis on the difference between a vertical synchronization of currencies within a state’s borders, and a horizontal integration outside its borders (Abu-Lughod 1989). Until the mid-13th century, there was no commonly accepted unit of account across Eurasia, even in long distance markets. Following the collapse of the Roman Empire, where gold, silver and bronze had circulated, Western European minting came to depend exclusively on silver. In the eastern Mediterranean, including the Byzantine world, the Roman system continued, while in West Asia and South Asia precious metals were used as currency alongside non-precious materials. Although preferences varied locally, broadly speaking in West Asia silver served as the most popular currency and unit of account. A variety of coins sharing the name dirham and composed of a silver-copper alloy or silver-coated copper, circulated widely in the mid-13th century. The continuous debasement of silver, known to numismatists as the ‘silver famine’, caused some regions of West Asia to revive gold as a monetary medium. In the eastern part of Eurasia, however, the use of silver as a currency and monetary unit was quite limited. Chinese dynasties had almost never minted silver coins, depending instead on copper coinage for local exchange and silk for interregional transfer. Silver ingots were used only as a supplement to silk. Surrounding countries followed Chinese practice, and documents from East Turkestan strongly suggest that both taxation and commerce in the region were conducted in terms of cloth or copper coins.1 Commodities as well as metallic currencies worked as mediums of exchange.All over the world certain grains or cloth were commonly used to settle transactions in local markets. For example, the 12th-century islanders of Rügen, off the coast of

68  Global history of monetary delocalization

Germany, used linen as currency (Sargent and Velde 2002, 11).Thirteenth-century peasants in Xiuzhou (Jiaxing) County, China, brought rice to a market town,Weitangzhen, to exchange for salt or oil (Fang H. 1971, 698). Unlike simple barter or the counter trade, in which two parties directly exchanged commodities with no medium, these commodities worked as mediums that facilitated exchange. Even in long distance trade, many merchants used portable commodities like spices rather than precious metals (Lopez and Raymond 1955, 145). In the case of Europe, such commodity currencies played an important role in local transactions until later than is commonly imagined. For example, as late as 1760 in the Catalan area of the Pyrenees, villagers would take sacks of grain to the market to pay for their purchases (Vilar 1976, 25). Indeed, in early modern Europe, ordinary people were not particularly dependent on coinage. The dominance of big coins whose value was as large as one day’s wages, combined with long, even monthly, intervals between wage payments, meant that most ordinary people were unfamiliar with the use of cash.2 A crucial difference between the Euro-Mediterranean world and China lay in the relationship between metallic and commodity currencies. Broadly speaking, and bearing in mind individual variation, medieval Europe lacked sufficient small-denomination currencies. England was an extreme case, for the shortage of small change caused bakers to change the size of loaves of bread according to the price of wheat (Spufford 1988, 238). In such a situation, the circulation of precious metal currencies and the mediation of exchange by commodity currencies operated separately, but in tandem. The gap between these currencies was so large that the two spheres were complementary rather than substitutive. Except for urban residents, who depended more on daily purchases of food, formal currency supplies had little bearing on the lives of ordinary people. In East Asia by contrast, as far as proximate exchanges were concerned, the circulation of copper coins and commodity currencies overlapped, making their relationship substitutive rather than complementary. The authorities had to consider the danger that a diminished supply of copper coins could adversely affect grain transactions at the bottom levels of society. Symbolically, the Chinese imperial government distributed copper coins to villages suffering from famine, even at the end of the 19th century (Li 1984, 7–10).3 Clearly, a sufficient supply of copper cash was thought to encourage the trade of grain in local marketplaces. Monetary systems dependent on copper coinages and grains, or those independent of portable materials such as precious metals, were inconvenient both for merchants engaging in long-distance trade and for central governments that collected taxes from locals. Though gold was used in China to settle tax transactions between the Western Han Empire (202 BCE–8 CE) and local governments, after the collapse of the empire, the metal was only kept as treasure. Instead, silk came to play the principal role in the transfer of tax revenues from local authorities to the central government. Until the Tang period (618–907), silk continued to serve as the most important instrument for bulk transactions (Kuroda 2017). For example, agrarian dynasties often obtained

The ignition of monetary delocalization  69

horses, which were a military necessity for Chinese states, from nomads in exchange for silk. Thus, in contrast to Western Europe, in which metallic coins (silver) worked for distant trade and textiles (linen) for local trade, in China, textiles (silk) for distant trade and metallic coins (copper) formed a complementary relationship. The supply of metallic coins issued by the authorities therefore differed in the east and in the west. Put simply, in China the supply of currency was a measure of the emperor’s virtuous rule over commoners, whereas in EuroMediterranean societies it was a means for the authorities to supplement their budgets. Bian min (‘giving convenience to the people’) was a phrase frequently used in monetary affairs in China (von Glahn 1996, 82). The common people could only benefit from a currency, such as copper coins, that could reach down to their level. That is why the dynasties often issued copper coins, even if the state suffered a loss. For example, according to an essay written in 1125, the intrinsic value of 1000 wen of copper coins was 600 wen, but, adding the cost of transport, salaries and other expenses, the total cost to supply 1000 wen of copper coins amounted to 1400–1500 wen. Sometimes the metallic value of copper coins alone exceeded the official face value (Kuroda 2000, 188–9). In contrast, because it is necessarily expensive to supply small change, EuroMediterranean authorities tended to issue small-denomination currency only when substantial seigniorage could be gained (Stahl 2000, 376). Mint staff in England thus preferred to strike 64 penny groats from a pound of bullion, rather than 960 farthings, since the supply of the former cost much less than that of the latter (Mayhew 2007, 218). The excavations of coins reveal to us a clear outlook of the differences in the institutions within which the currency circulates. Here we compare two types of coinage which were unearthed from hoards or found in lost caches. One is copper coins unearthed from China (Table 3.1). Another is the penny from Table 3.1 The distribution of era names with copper coins from Chinese excavation Year -618 -907 -960 -1004 -1054 -1101 -1126 -1201 -1260 Total latest era Source: Miyake 2005, 104–5.

Yankeshan 27 866 12 472 2,706 7,156 2,018 212 13,469 1131–63 CE

Wenqiaoyidui 57 3 32 213 277 95 45 4 726 1260–65 CE

70  Global history of monetary delocalization Table 3.2 The distribution of minting periods of the English penny from English hoards Beaworth William I (1066–87)

William II (1087–1100)


York Minster


31 34 11 6,457

6 5 30 97 164

Source: Grierson 1975, 134.

England (Table 3.2). Both types circulated and were buried in the medieval period and then unearthed in the modern period. Importantly, the two worked as a mono-unit coins in each region and were not accompanied by currencies of smaller divisions. In China, although a large amount of paper money was in circulation, and silver ingots and silk also worked as money under the Song (960–1279), it was copper coins that mediated the largest proportion of transactions, especially under the Northern Song (960–1127). Although silver and gold coins of large denominations began to be minted in the 13th century, medieval England generally depended on a monocurrency system in which the penny dominated monetary usage. An apparent difference between the two coinages was their metallic content. Copper coins under the Song was bronze, an alloy of copper, lead and tin, in which copper occupied 60–80 percent. English pennies were silver coins whose fineness remained among the highest of all coins in Europe. From the two tables we can easily see a clear contrast in the range of dates in which Chinese copper coins were issued, and that in which English pennies were issued. The issuing dates of Chinese copper coins ranged across a period greater than 300 years. The hoard even included more than a few coins minted under the previous Tang Dynasty, which would have been more than 300 years older than the newest coins in the hoard (though it is possible that some of these may have been imitations produced in later periods). Meanwhile, English hoards held pennies issued within only a 20-year period. We can infer that the preciousness of the material of a coin has an inverse relation to the length of the period in which it circulates. English excavations are full of silver coins issued only within a short range of time, the Chinese examples containing no precious metals had been issued over a long period.

The ignition of monetary delocalization  71

Each table shows both one excavation with a large quantity of coins as well as one or two cases with only a small quantity. The presence of large quantities of coins suggests that they were intentionally saved; in other words, these hoards are examples of a stock of coins, whereas caches of smaller quantities were perhaps unintentionally lost and therefore may be taken as examples of flow. Seven hundred twenty-six pieces from the Wenqiaoyidui excavation in China may seem to be a large quantity, but, considering the low value of bronze coins, this quantity is still too small to be considered an intentional hoard. Note that regardless of whether the quantity of coins in each excavation was large or small – in other words, whether the coins were accumulated intentionally or unintentionally – the range of issuing dates is similar. This suggests that it was the material difference among the two (copper and silver coinage) rather than the functional differences of the excavations (intentionally saved coins vs coins unintentionally lost) that led to the different patterns of monetary circulation. Again, remember that the two coinages worked as mono-unit currencies or were basically indivisible into smaller coins. However, reflecting the differences of their metallic content, the values of the two coins varied considerably. It is difficult to estimate the difference in value between Chinese copper coins and silver pennies, since there was a considerable difference in the rate of silver to copper in Europe and in China. We may take the English penny to be 1.7 grams in weight and 95 percent in fineness and assume that silver was valued at 25 times as much as copper in contemporary Europe (Stahl 1985, 13). Chinese descriptions often listed a silver-to-copper ratio of 1:100. Ignoring the values of lead and tin (though not so inexpensive themselves) we may assume that one copper coin contained three grams of copper. In the case of the 1:25 exchange rate of silver to copper, the metallic content of an English penny would have been equivalent to 13 in copper coins: in the case of a 1:100 exchange rate, then one English mono-unit coin would have been worth 54 Chinese copper coins. Thus, the smallest coin in one society may be expensive in another. In addition, the coins in the former society appear to be ready to assemble by issuers’ recall, while the coins in the latter seem less certain to return to centre. The differences in unearthed mono-unit coins between England and China reveal how widely the functions of exogenous currencies vary by institution. The big gap in the monetary system between both ends of the Eurasian landmass, basically, remained until the late 13th century. However, potentially, the increasing presence of silver in China, though not for monetary usage, paved the way for intercontinental transformation of the monetary system. During the Tang, silver began to assume silk’s monetary role to a limited extent. It was in the Northern Song period, however, that silver clearly became more than an ornamental item. During the Northern Song, silver ingots (sycee) were cast at a higher frequency than in previous periods, which suggests that silver had become a more popular form of currency (Wang WC. 2001, 38–46).

72  Global history of monetary delocalization

The amount of silver tribute offered by silver mining districts to the Song Dynasty indicates the total amount of silver produced during the Northern Song period. In 1075, 411,420 liang (about 16 metric tons) were collected across the Song Dynasty. Silver production peaked in the late 11th century, a period during which a single silver mine, Baorui, near Fuzhou prefecture, is recorded as offering a total of 440,000 liang in silver tribute. Since silver offered in tribute is believed to represent only one-fifth of the total silver production, Baorui likely produced 2.2 million liang of silver in that period (Wang LL. 2005, 27–8, 59–60). The 11th-century silver mining boom happened in conjunction with an increase in the minting of copper coins. Written records indicate that during the Yuanfeng reign (1078–1085), the Song state issued some six million guan (six billion wen) of copper coins every year, the largest quantity of copper coin minted at any point in Chinese history. This anecdotal evidence is supported by the high frequency of Yuanfeng coins found in hoards from this period. It is unlikely that the increased production of silver in the Northern Song period was planned. In general, copper mines active in this period also produced silver ore. Since silver coins were not minted at the time, this suggests that increased silver production was an unexpected by-product of the increased minting of copper coins. Casting six million guan in copper coin would have required 12,000 metric tons of copper, so even if copper ore was only 1/1000th silver, this would have meant the production of 12 metric tons of silver as a by-product. Though published in the late 16th century, Tiangong Kaiwu, a scientific encyclopaedia, described copper ore as containing silver, which suggests that silver was understood to be a by-product of copper production (Song 1978, 356). Although silver mines were mostly located in Southern China, a significant quantity of silver made it to the north and west. The power of northern nomadic dynasties relative to the Song caused silver to stream north. Under the treaty of Shanyan, established between the Liao and the Song in 1004, the Song made annual tribute payments of 200,000 bolts of silk and 100,000 liang of silver to the Liao. Similarly, in the Qingli treaty of 1044 the Song promised to make annual payments of 50,000 liang of silver, 130,000 bolts of silk and 20,000 jin of tea to the Xixia.Thus, under these treaties, the Song was obligated to send annual payments of six metric tons of silver to the two nomadic dynasties in the mid-11th century. As described earlier, though silver production began to increase in the 8th century, silver was rarely used as money in China proper prior to the 13th century. In contrast, Europe, especially north of the Alps, depended exclusively on silver coinage after the collapse of the Roman Empire, and in both taxation and large commercial transactions silver coin was the dominant medium. Though China produced silver between the 10th and 13th centuries, demand for it was thus greater in the western end of Eurasia than in the eastern end. At the end of the 19th century, a cache of uncoined, Chinese-style silver was unearthed from the remains of medieval Brandenburg (Friedel 1896, 5). The

The ignition of monetary delocalization  73

dates on European coins found with the silver suggest that the cache may have been buried as late as the end of the 11th century.4 Considering the big gap in the monetary demand for silver between East and West, it is entirely possible that some ingots from the Song’s six metric tons of silver tribute could have made their way to eastern Germany. Prior to the Mongol Yuan Dynasty, Northern China under the Jin Dynasty (1115–1234) was the first state in Chinese history to use silver as a unit of account. However, the introduction of silver as a unit of account did not mean that actual silver was commonly used to collect taxes or make purchases. At the start of the dynasty, the Jin did not have much copper coin in circulation and so, like the Southern Song, its neighbour and rival, it issued paper currency denominated in terms of copper coin. Unlike the Song, the Jin also issued coined silver, but the dynasty failed to establish silver in circulation and soon discontinued its use. Silver worked in tandem with silk as the main medium for tax payments, and even paper money denominated in terms of silk was issued during the transition between the Jin and the Yuan. Recently a metal format for printing local paper currency denominated in terms of silk with the date 1250 was found (Gao 2016). It proves that silk still worked as money in mid13th century China and that the Mongols did not stick to the exclusive use of silver at that time. All in all, until the late 13th century there was no commonly accepted unit of exchange across Eurasia. The use of silver as currency was geographically limited, although it was more widely used than gold. It is worth noting that in 1221 the Mongols minted a coin called ‘Khani Dirham,’ the first Mongolissued coinage in Samarqand, but despite being called a dirham, it consisted mostly of copper (Kolbas 2006, 34). Until the late 13th century the Mongols brought few significant changes to previously existing monetary systems in regions under their rule. Although significant quantities of Chinese silver might already have been moving along Eurasian trade routes, payment in silver was far from common.

2 The Eurasian silver century: emergence and collapse In the last quarter of the 13th century, the use of silver abruptly became common across the entire Eurasian landmass. Let us examine developments in silver usage from west to east. In Tunis, silver became cheaper than gold in 1278, a drop in silver price that seems to have occurred at the same time as a similar shift in Genoa (Spufford 1988, 178). Egypt seems to have emerged from a silver shortage at just this time. A hoard from Moldavia, Prajesi, contained silver Kipchak coins from the last two decades of the 13th century as well as Byzantine gold coins (Boldureanu 2007). In Caucasian hoards from the 13th and 14th centuries, silver is dominant, while in other periods only gold coinage from Byzantium is found (Watson 1967, 18). Under the Ilkhanate, silver output appears to have begun to increase in the 1280s.The Delhi sultans began to issue silver rupees on a large scale in 1295 (Deyell 1983). In Transoxania there was an

74  Global history of monetary delocalization

increase in the number of silver mints in the 1280s (Davidovich and Dani 1998, 406). Documents unearthed from Turkestan reveal that, during the latter half of the 13th century, the most common form of payment shifted dramatically away from cloth and copper coins towards silver. The official history of Korea described the circulation of uncoined silver beginning in 1287, following the Mongol expedition to the peninsula (Jeong 1972, 736–9). The exchange rate of silver to gold in China dropped to 10:1 in the latter half of the 13th century from 5:1 in the first half of the 13th century (von Glahn 1996, 60–1). An even more important shift was in the quality rather than the quantity of silver coinage in parts of West Asia where the silver dirham had been dominant. Silver coins of higher fineness with whiter appearance rapidly replaced blacker ones of lower fineness during the late 13th century. In the case of Transoxania, the transformation can be clearly dated to a distinct starting point. In 1271–2, the governor Masud Beg banned the use of silver-coated copper dirhams and thereafter established a high standard of high (80 percent) for silver dirhams. As mentioned earlier, more new silver mints opened in the 1280s in the district. Prior to the mid-13th century we have no reliable data on the abundance or scarcity of silver. Fortunately, after 1273, the Royal Mint in London maintained annual records of the silver coinage produced. As shown in Figure 3.1, large peaks in the issuance of silver currency appeared in 1278, 1300 and 1340. The quantity of silver minted annually in this period was larger than in any other period prior to the Napoleonic Wars. Since the coin hoards in Bruges from the late 13th century to the early 14th century mainly consisted of silver sterling and its imitations, the surges likely increased silver supply not only within England but also in the Low Countries. Though they kept far less consistent data, other mints in Western Europe, such as those in Paris and Flanders, 70000 60000 50000 40000 30000 20000 10000

1275 1280 1285 1290 1295 1300 1305 1310 1315 1320 1325 1330 1335 1340 1345 1350 1355 1360 1365 1370 1375 1380 1385 1390 1395 1400


Figure 3.1  Minting of silver coinage in London (kg) Source: Miskimin 1983, 90–2.

The ignition of monetary delocalization  75

appeared to have had similar surges in the production of silver coinage around 1350 (Munro 1983, 131). Some may suspect that the first two peaks of silver minting by the Royal Mint resulted from the decrees of recoinage in 1279 and in 1299. According to popular understanding, circulations of inferior silver coinages in England caused rulers to order recoinages for the purpose of restoring the unification of currency through new coinage. Indeed, in the late 13th century large quantities of base silver coins known as pollards and crockards were introduced, probably by Italian merchants who exported English wool to the Low Countries. In 1299, £250,000 worth of pollards and crockards was brought into England and recoined in a few months. The Exchequer in the period stored enough silver to coin sufficient good money in advance to replace the pollards as they were withdrawn. Otherwise, why would the holders of pollards and crockards have followed the decree? The basic principle in recoinage should have put the loss upon the holders of light or counterfeit coins. The circulation of those unofficial coinages continued in later periods, but England never issued such a drastic decree after 1299. Thus, the fact was not that widespread circulation of foreign inferior coins caused the English authorities under Edward I to order recoinages, but that affluent silver inflows enabled the authority to carry out the policy (Feavearyear 1931, 10–12). Though, due to the lack of data, we cannot deny the possibility that the surge in the minting of silver might have begun prior to the 1270s,5 the circumstances across Eurasia previously described suggest to us that it is unlikely that any significant growth in silver usage would have occurred before the late 13th century. The conclusion of this Silver Age can be identified more precisely than its beginning. In Bengal, the issuance of silver rupees stagnated after the 1360s, and Delhi also began to suffer from a scarcity of silver in the late 14th century (Haider 2007, 191). In 1359, the basic unit of silver in Aden decreased in weight; in the 1360s, Egypt suffered a silver shortage severe enough to force it to switch to a reliance on minted copper; Italian cities appeared to experience an outflow of silver on account of the Levant trade; and the minting of silver in London decreased sharply. Allen estimates that silver circulation in England fell from between £700,000 and £900,000 in 1351 to between £150,000 and £200,000 in 1422. (Lopez et al. 1970, 124; Bacharach 1983, 168–9; Allen 2001, 607). A consistent source of data for the change in currency supply in this period is the distribution of dates of Venetian tornesello coins unearthed in Greece, which we saw already in Chapter Two. Containing little silver, tornesello coins were basically a bronze alloy, though it was denominated as silver. Mid-14th century Venice issued the tornesello exclusively for use in its Greek colonies and increased its production towards the end of the 14th century (Stahl 1985, 24–5). In this period, then, the Aegean Sea region appears to have relied predominately on copper alloy coinage with little silver available. Prior to 1350, as Pegolotti has noted, there were a number of types of silver coins in circulation in Cyprus

76  Global history of monetary delocalization

(Grierson 1979, 491). It is thus unlikely to be a mere coincidence that silver rapidly retreated at the same time that Egypt was forced to switch from silver to copper currency. In Caffa, the inferiority of the aspers, a type of silver coin, produced after 1380 caused local authorities to mark coins produced after that year with a special stamp (Di Cosmo 2005, 415). Thus, the Eurasian continent appears to have experienced a simultaneous and sudden collapse in the availability of silver after 1360. Silver would not be found in such abundance again until the end of the 15th century. Where did silver come from in such abundance in the last quarter of the 13th century? Why did shortages of silver strike so suddenly and across such a large area in the third quarter of the 14th century? These unprecedented monetary transformations actually originated in China, where silver, until the late 13th century, had functioned as money only in limited circumstances.6 However, before we uncover how Chinese silver reached the west end of Eurasia, we must confirm what roles silver had played in the fiscal system under the Mongol prior to 1276.

3 The Mongolian taxation system depended on commerce In three respects, the Mongolian regime was unlike any previous dynasty in China: (1) it depended on revenues raised through taxes imposed on commercial activity, (2) it instituted a tax in silver across the entire empire, initially in the service of military requisitions and (3) it established a postal relay system across the entire Eurasian continent. Unlike other Chinese dynasties which relied on the land tax as their primary source of revenue, the Mongol states depended mainly on commercial taxes and tribute, although, in agrarian regions, the Mongols also collected taxes on land, either in kind or in money. The Yuan relied on the revenue of the salt monopoly more than the other Mongol states in Eurasia, and more than any other previous Chinese dynasty. For example, in 1309 the annual expenditure of the Yuan was approximately five million ding in zhongtong chao, the paper currency then in use (equivalent to 25 million liang in real silver), while in 1307, its entire revenue was four million ding and its issuance of paper money was one million ding. In 1308,Yuan revenue from the salt monopoly was 3.25 million ding, which likely amounted to more than 80 percent of total revenue (Miyazawa 2012, 58, 60). The Yuan thus relied on a unique structure of public finance. Even the Qing, a conquest dynasty established four centuries later, depended mainly on land taxes for its revenue, as had most other Chinese dynasties. In 1766, Qing revenues derived from a land tax totalling 32.91 million liang, a salt tax of 5.74 million liang, custom taxes of 5.4 million liang, miscellaneous taxes of 1.42 million liang and contributions of three million liang (Momose 1943, 79). The tamga, a tax on merchants, was common across the Mongol Empire.The term tamga may have referred to different forms of taxation in different regions of the empire: it appears as a sales tax in Uigur documents from Turfan, a tax

The ignition of monetary delocalization  77

on city dwellers under the Kipchaks and a customs tax in Novgorod. Regardless of local variations, however, the basic principle was consistent throughout the empire: commercial profits were to be taxed in any form (Matsui 2005, 75; Allsen 1987, 158–62; Michell and Nevill 1914, 95). This made it similar to the yashui tax in China, a sales tax that is discussed later. As one might expect, commercial taxes were collected in currency. As will be explained later, tamga was also the name of the unit of currency that was the most popular in the Kipchak Khanate and other western Khanates. This commercial taxation under the Mongol regime encouraged monetization. However, across the empire this monetization consistently occurred as a silverization. This was due to a tax imposed on all subjects in order to support military requisition – rather than requisitioning materials in kind, the Mongol regime collected a tax in silver used to purchase those materials. This tax was called baoyin in China proper, qubchir in the western Khanates and zapros in Novgorod (Vernadsky 1953, 222). What Mongols did in organizing taxation at Samarkand soon after the conquest suggested their principle of administration. They kept the existing exchange system (merchants conducted businesses in terms of gold, while people depended on copper-based currencies) and imposed taxes respectively in gold on merchants and in copper coinage on commoners. However, importantly, they introduced taxation in terms of silver for Mongols and their trade circuits (Kolbas 2006, 68). Establishing a prevailing universal unit across the empire accompanied practical flexibility in keeping existing local systems with some modification in order to interface with the universal system. As we will see, the empire-wide imposition of qubchir in 1255 followed the same principle. The baoyin system originated with the Jin Dynasty and was inherited by the Yuan, although at the start of Yuan rule over Northern China the Mongols levied a requisition tax in terms of silk. ‘Yin’ referred simply to silver. ‘Bao’ meant that ordinary households responded to one-time requisitions by the ruler designed to support soldiers and artisans (Abe T. 1972, 148–9). Baoyin originally was a contribution in silver paid to the dynasty once a year to cover all requisitions. Based on its name, the baoyin system likely originated from occasional or additional (but by no means minor or merely supplemental) requisitions by the nomad army, but later it became a fixed tax. There seems to have been local variation until 1255, when the Yuan issued a decree that introduced baoyin throughout the empire and required each household to pay four silver liang (Abe T. 1972, 131). This decree may have had an effect as far as Russia. In 1257, the Mongol census officials counted all the Suzdalian land and the land of Ryazan’ and Muron’ and divided the country into tens, hundreds and thousands for the purpose of determining the number of taxpayers and recruits for the Mongol armies (Fennel 1983, 113).The chronicle of Novgorod records that in 1259, the Mongols required Novgorod to count households in preparation for payment of the tamga and a one-tenth tax on everything. In spite of strong opposition by the people of Novgorod, the Russian ruler Aleksandr Nevskiy succeeded

78  Global history of monetary delocalization

in imposing the will of the Mongols on them. The tax was collected, and the census was completed (Fennel 1983, 118).This likely indicates that the Kipchak were also affected by the baoyin decree. Novgorod had at first paid tribute to the Mongols in fur but later began to pay in silver as a substitute. Perhaps not coincidentally, the earliest coinage in Kipchak Bulgar appeared soon after this change. In the Chagatai Khanate, meanwhile, silver-coated copper coins were minted from 1253–54. As late as 1259, the Mongols cast a Chinesestyle silver coin, dachao tongbao, which may have circulated under the name of tamga in Central Asia. Requisitions levied as a tax in silver and the issuance of currency were two components of a single system (Allsen 1987, 175, 179, 185). Why did requisitions in terms of silver prevail across the empire? The diffusion of silver apparently related to military activities. For example, a find of dachao tongbao, a Mongol silver coinage, in the Tianshui hoard, which contained no other coin, suggests that it was used in military camps rather than in commercial caravans (Whaley 2001, 53; Sheng 1995, 18, 21). Documents from Turfan suggest that the requisition tax was used to support the cost of military expeditions and the postal relay system (Matsui 2005, 78). Meanwhile, a unified system of measurement used in the postal relay system across the entirety of the Mongol world, from the Korean peninsula to Eastern Europe, supported the commensurability of various local currencies. As was common in the preindustrial era, even under a single political administration there was a plethora of measures, scales and weights used side by side, despite the standardization of the unit of taxation. For example, a Venetian merchant visiting Sarai in 1345 observed that merchants used different measures for different commodities (Lopez and Raymond 1955, 152–3). However, because it crossed multiple administrative regions, the postal relay system likely used a single set of measurement units, to avoid substantial variation in the amounts of grain and liquor supplied to messengers in each station. The standard established under Ögedei Khan was that a certain amount of grain and wine should be provided to messengers stopping at postal relay stations (Matsui 2004). One of the purposes of baoyin (or qubchir) style requisitions was to support the postal relay system across the entire Mongol Empire. Thus, even if measurements continued to vary in local markets, there must have been a standard measurement system across the entire Mongol Empire for use in official business that crossed regional administrative boundaries, such as the postal relay system. It appears that the Mongol rulers encountered few difficulties in preserving the link between the fiscal system and issuances of silver coinage, or in maintaining a single measure throughout the Eurasian communication system, until they conquered a populous empire with its own, quite different monetary system.

4 Paper monies drove out Chinese silver ingots to the west We should keep in mind that there is an important difference between the use of coined silver and the use of uncoined silver: the value of the former

The ignition of monetary delocalization  79

is determined by the count, while that of the latter is determined by weight. Archaeological remains confirm that from the beginning the Mongols issued silver coins. The western Khanates continued to use coined silver in later periods.The Yuan thus originally stood in stark contrast to other Chinese dynasties, most of which had been reluctant to issue silver coins and instead used silver only by weight. However, after the establishment of the baoyin system in 1255, the Mongols’s silver policy was transformed. In 1260, the Mongols began to issue paper money, the zhongtong chao. A 1263 decree permitted households to pay the baoyin with paper money in place of four silver liang (Abe T. 1972, 168). The face value of paper money, zhongtong chao, was denominated in copper coins but also had an official value in terms of silver liang: two guan of zhongtong chao could be used to purchase one liang of uncoined silver, but, when used in tax payments, two guan was considered equivalent to two liang of silver (Abe T. 1972, 167). The Yuan did not abandon the principle that the baoyin should be paid in silver, but because they allowed the substitution of paper money for uncoined silver, zhongtong chao began to circulate as a standard currency and became the currency with the highest total value (in terms of silver) in circulation in Mongol territory. At this stage, the Mongols had already deviated from their original preference for silver coinage. However, the Yuan only issued small amounts of paper money until 1276. After the conquest of the Southern Song, the Yuan’s deviation from its initial silver policy accelerated considerably. In 1276, the capture of the Lower Yangzi, the most prosperous region of China, where Hangzhou, the capital of Southern Song was located, must have brought the Yuan court a large amount of confiscated silver ingots. The Yuan minister Dayan is said to have ordered that the soldiers’ bags be searched for silver, which, if found, was to be taken for the use of emperor Khubilai (Tao 1959, 377). The discovery of a silver ingot inscribed with the word ‘confiscation’ and the year 1276 seems to support this story (Whaley 2001, 59). So far, at least 14 Chinese-style silver ingots (sycee) have been found with the same inscription: Yangzhou, the 14th year of Zhiyuan. Numismatists call them Yangzhou yuanbao. Yangzhou was a city located along the Grand Canal in the Lower Yangzi region and was the centre of China’s salt trade networks. The inscription shows that it was cast in 1277. Concrete information about the excavation is available for four of the ingots. Two were unearthed from Inner Mongolia, one from Liaoning and one from Jurong, Jiangsu, along the Grand Canal (Sun 2015, 13–14). Most of them weigh 50 liang (nearly 2 kg). The discovery of so many silver sycees sharing the same place name and year of casting is rare, even in the case of sycee cast in later periods, although there is no way to estimate how many silver ingots were taken by the Mongols during their conquest of Southern China. After the defeat of the Southern Song, the Yuan abandoned the minting of silver coins and, throughout the territory of the Yuan, silver was used by weight in the form of sycees. This transition coincided with a rapid increase in the Yuan’s printing of paper money, which began precisely in 1276, the same year

80  Global history of monetary delocalization

as the conquest. At this time, the Yuan also began to use copper plates in place of woodblocks to print paper money, which enabled the dynasty to issue far more paper currency (Maeda 1973, 49–50). Like the Song, the Yuan administration continued to denominate the face value of paper money in terms of copper coins, but the Yuan calculated these values in terms of silver by weight. In short, given that the Yuan’s main use for silver was to serve as backing for an increased quantity of paper money, it had no particular reason to melt silver ingots and reissue them in the form of silver coins, which, in any case, had never been very popular in China proper, particularly in the territory that had belonged to the Southern Song. The year 1276 was a turning point: a year in which a set of institutions built on taxation in terms of paper money helped bring about the continentwide circulation of silver. From 1263 to 1311 (except for 1284 through 1287), the Yuan prohibited the use of silver in private commerce, though Mongolian nobles were allowed to use it both in commerce and religious activities. In order to effect such a monopolization of the use of precious metals, the Yuan also instituted a policy in 1282 to register gold and silver craftsmen, although it is unlikely that all craftsmen actually ended up registered (Vogel 2013, 162). This transition from payment in real silver to payment in paper money also appeared in the prices of salt vouchers. A price quoted in 1263 was in real silver, while one in 1277 was already in paper money (Miyazawa 2012, 48). Considering the importance of salt revenue to Yuan public finances, this change should have affected the acceptability of paper money more generally. Undoubtedly, the significant increase in the issuance of zhongtong chao after 1276 accelerated this transition. Figure 3.2 shows the changes in paper money issuance by the Yuan in terms of real silver. We should ignore the isolated peak of 1311, since that number reflects extraordinary changes in the monetary system that were discontinued within two years. The first peak begins in 1276, coinciding with the conquest of the Lower Yangzi region, as already discussed. The peaks beginning in 1276 and 1302 were followed by increases in the amount of coined silver issued by the London mint. Was there any connection between those movements? Actually, the two peaks also overlapped with the sudden establishment of new appanages, touxia.This was the key element generating synchronized dynamism across Eurasia. The Mongolian regime had an unprecedented institution called touxia, an appanage within China proper possessed by princes and nobles who also kept basic territories mostly outside China proper. Since their occupation of Northern China and defeat of the Jin, the Mongols had allotted estates to princes and nobles of the regime, who used the estates to raise horses and collect labour and grain. Likewise, the Yuan possessed appanages in Iran.The nobility administered these estates independently from the central Yuan state. After 1276, the Yuan also permitted the nomadic nobility to establish this type of estate in Southern China, and the late 13th century through the mid-14th century was thus the only period in Chinese history during which nomad lords based in northern

The ignition of monetary delocalization  81 80000000 70000000 60000000 50000000 40000000 30000000 20000000 10000000 0 1260 1265 1270 1275 1280 1285 1290 1295 1300 1305 1310 1315 1320 1325

Figure 3.2  Issuance of paper money by the Mongol Empire (in silver liang) Source: Song 1976, 2371–3.

and central Asia directly collected revenues from estates in wealthy Southern China.7 There are a few cases concretely demonstrating that tribute was brought from Chinese appanages to the western Khanates. For example, in 1281, a tribute called wuhusi was sent from an estate in Taiyuan, Shanxi, controlled by the Yongning King, to Bishbalik, Turkistan.8 This was the same year in which the postal relay system was extended from Bishbalik to the Caucasus region.9 It is perhaps not coincidental that many touxia were established there that year. The same year the Yuan granted appanages in South China containing a total of 1,048,107 households. In 1281, Qaidu received a fief of 60,000 households in Xinzhou, which was known for its silver production during the Song period. If each household had paid 0.5 liang in paper money (0.25 liang in silver) according to the decree (Song 1976, 2411), Qaidu would have received 15,000 liang of silver (600 kg). In the same year. Chagatai received 67,330 households in Lizhou (Li 2007, 134). As discussed earlier, it was permissible to pay the levy in paper money. However, outside Yuan territory, paper money was not in circulation. Thus, the revenue from these fiefs in the Lower Yangzi must have been remitted to the western Khanates in the form of silver ingots after conversion from paper money. Otherwise, the princes would not have been able to use the revenue they received.

82  Global history of monetary delocalization

The major difference between developments under the Yuan and in the western Khanates is that, while silverization in taxation was accompanied by actual payment in paper money under the Yuan, the western Khanates did not issue paper money, and therefore collected taxes in real silver.The Ilkhanate had tried to introduce a paper money system in 1294, but in vain.10 This contrast in currency systems affected cross-regional trade. Typically, Italian merchants brought linen to Sarai or Urjench, which they exchanged for silver, specifically the somo silver ingots of the Kipchak khanate and then advanced to Hangzhou where they converted silver for paper money with which they could purchase silk and porcelain, a process described by Pegolotti (Lopez and Raymond 1955, 358). Visitors from Europe, including Marco Polo, almost without exception observed that paper money circulated stably in China.Where did the silver used in Sarai or Urjench come from11? Now we can provide an answer: silver came from China proper through the channel of touxia. Unlike in the late 13th century, when confiscated silver played a crucial role in determining the Yuan’s policy towards the metal, in the first half of the 14th century newly mined silver from Yunnan was likely the more important factor. In 1328,Yunnan is recorded to have paid a tribute of 735 silver ding (about 1.4 metric tons) to the Yuan government, making up half of the entire empire’s silver tribute for that year (Song 1976, 2383–4). By 1340, the Yuan’s influence stretched deep into Burma via Yunnan, and the increased silver production there coincided with a surge in the minting of silver rupees by the sultan of Bengal, which began soon after the independence of the Bengali sultan from the Delhi dynasty (Jackson 2009, 26). Since silver was not mined in Bengal, it is likely that the silver used by the sultan originated in Yunnan. The peaks in the issuance of paper money by the Yuan that began in 1276 and in 1300 overlapped with the sudden creation of new touxia appanages in Southern China, and these peaks were also followed by increases in the amount of coined silver issued by the London mint. The establishment must have increased the issuance of paper money for collection of tribute and the remittance of the tribute to the western Khanates must have increased the transfer of silver ingots to the west. The first peak of creating new touxia must have resulted from the conquest of Southern Song territory.12 Between 1276 and 1285, new appanages with a total of 1,242 thousand households were established in the Lower Yangzi. According to the rate of tribute, 0.5 liang in zhongtong chao per household, the total tribute amounted to 310,000 liang in silver (more than 12 metric tons).13 The second peak might have been influenced by more peaceful conditions than before due to the reconciliation with western Khanates in 1296, and the end of Qaidu’s rebellion against the Yuan.14 The establishment of new fiefs was suspended between 1285 and 1298. Its resumption followed the surrenders by Yobuqur and Ulus Buqa, important allies with Qaidu, to the Yuan court in 1296 (Matsuda 1983). The next year the Yuan increased their appanages in order to encourage further surrender from the Qaidu camp (Song 1976, 408–9). Importantly, Temür Qa’an already raised the rate of tribute per household from

The ignition of monetary delocalization  83

0.5 liang to 2 liang in paper money (1 liang in silver) in 1294 (Song 1976, 382) and resumed the establishment of new touxia in the Lower Yangzi in 1298. It is important that remittances from appanages in China to the western Khanates seem to have been discontinued soon after their creation. That is why western Khanates sometimes requested the Yuan to collect their tributes by themselves. For example, Ghazan of the Ilkhanate dispatched envoys with gifts to the Yuan in 1297–98, and they returned with silks from Hulegu’s holdings in China (Allsen 2001, 49–50). Interestingly, coin specimens in museums show that Ilkhanate silver minting underwent a sudden increase around 1300 (Martinez 1995–1997, 153). The lift on banning the usage of silver for commercial purposes in 1311 might have caused silver ingots to stay in the Yuan’s territory and consequently discouraged the remittances in silver to western Khanates. However, the remittance through the touxia system was resumed in 1339 after a previous suspension. As far as the Kipchap Khanate was concerned, the probability that they were receiving resumed remittances is very high.The Khanate had apportioned territories in Pingyang, Jinzhou and Yongzhou. The official history of the Yuan records that, from 1339, 2,400 ding in zhongtong chao (paper money) was given to Kipchap annually (Song 1976, 2906). As we have seen the remittances had to be made in silver ingots, since the paper monies from the Yuan were not available beyond its territory.The amount is equivalent to 2,300 kg in silver. In 2007, a pot containing 65 silver bars was excavated from Orheiul Vechi, a fortress town of the Kipchap Khanate. The pot is thought to have been left between 1340 and 1360 (Boldureanu 2007). The bars had two features showing Chinese connections. One is the weight. They were apparently cast as 200 grams, equivalent to five liang, and one-tenth weight of silver yuanbao. They must be the silver bars contemporaries called somos. The other feature is the bubbled surfaces which appear to be similar to the common appearance of Chinese silver.We can find a bubbled surface in the Yangzhou yuanbao with a weight of 50 liang, as mentioned earlier. The Orheiul Vechi excavation clearly demonstrates that the tribute in silver from China must have reached the Kipchap Khanate. Since 1335 when the Ilkhanate, the most important alliance for the Yuan, politically fragmented. The silver coins, called asper, used in Iran, became much lighter after that (Martinez 1995–1997, 190). The resumption of remittance to the Kipchap must have been important for the Yuan in order to keep an ally in the western Eurasia. Meanwhile, Venice succeeded in partially lifting the papal ban on trade with Mamluk in 1345 and caused the trades connecting the Golden Horde and Egypt through Venetian merchants to increase. Venetians had wanted a safer trade route since the collapse of the Ilkhanate (Martinez 2011, 102). During the 1340s, imitation gigliate, a type of coin minted by the king of Naples using Sardinian silver, circulated widely in the Eastern Mediterranean region. Unlike genuine gigliate, imitated ones were mainly minted in towns located along the eastern end of the Mediterranean (Spufford 1988, 155, 284–5). It is also worth noting that between 1337 and 1353, the Byzantine Empire issued a particularly large type of silver coin, weighing between 8 and

84  Global history of monetary delocalization

10 grams, called the stavraton (Grierson 1982, 315). As shown in Figure 3.1 at a continental level, London mint records suggest that the 1340s through 1350s represented the last medieval peak in the issuance of silver coin. Thus, all three peaks of silver mintage in Europe from the late 13th century to the mid-14th century coincided with the peaks of remittance from the touxia, the appanages of the western Khanates in China.

5  Commensurability prevailed across Eurasia The unprecedented scale of tribute transfers through touxia channels caused silver bars of large denominations to flow across the Eurasian continent. Through these abundant movements of silver bars across Eurasia, for the first time, humans had a unit of account for interregional settlement prevailing far beyond the boundaries of civilization. They performed a function quite different from that of silver coinages previously circulated such as the dirham and denarius. Unlike silver coins available to both interregional settlements and local transactions, uncoined silver such as somo worked exclusively for distant exchange. Silver bars of around 150 grams had already existed, such as grivna in Kievian Russia.15 According to a late version of Russian Law one grivna was paid for the annual wage of one female labourer with one daughter (Vernadsky 1953, 126). The grivna must have been inappropriate to mediate the exchange in the marketplace which was typically held on Fridays and in which farmers brought their produce for sale (Vernadsky 1953, 117). Fractions of grivna were known as nogata and rezana: there were 20 nogata and 50 rezana in one grivna of kuna. The lowest unit was known as veksha (literally,‘squirrel’); in Smolensk one nogata was equal to 24 veksha (Vernadsky 1953, 122). To facilitate the exchanges of peasant products in the marketplaces, small currencies of furs (or low fineness silver as fur substitutes) whose value was sufficiently fractional like veksha must have been indispensable.16 A silver bar of around 200 grams, however, played a significant role in this period, to the highest degree yet in history. Consequently, between the late 13th and early 14th centuries, moving from west to east across Eurasia there was a chain of commensurable currencies: the gold florin or large silver gross in Western Europe, the silver somo in the western Khanates and the silver yuanbao (counted by ding) in China proper. Five florins were equivalent to one somo.The gold coin reappeared in Western Europe for the first time since the collapse of the western Roman Empire and was part of this chain.17 Meanwhile, ten somi were equivalent to one silver ding. Thus, along a major route of interregional trade from Western Europe to the Korean Peninsula, there was a set of currencies that could be converted easily. Of the three types of silver currency just mentioned – the gross, the somo and the yuanbao – the somo played the most important role in continent-wide trade, as described by Pegolotti (Lopez and Raymond 1955, 358).The silver bar weighed around 200 grams (5 liang) and circulated widely for a long period of time. The silver unit called the grivna was used in Novgorod. When the

The ignition of monetary delocalization  85

tribute that Novgorod paid to the Kipchak shifted from fur to silver, the grivna became popular (Martin 1978, 406). The grivna was, like the somo, equivalent to uncoined silver weighing five liang and was the original form of the Ruble, a term derived from the word rubit (‘to cut’), which designated a silver bar. Importantly, a silver bar held in the British Museum which is thought to be a grivna from 14th-century Novgorod has bubble-like small holes on the surface, the same as on the silver bars from Orheiul Vechi.18 Here we should remember that in Central Asia silver-coated copper coins dominated until the 1270s. The Chagatai Khanate had also minted them, but Masud Beg banned them, and high-fineness silver issuance increased in the 1280s (Davidovich and Dani 1998, 406). It is highly probable that the sudden prolific silver supply in silver-coated copper dominating Central Asia led to the development of the bubbled-surface silver bars in order to prove purity. The way of casting silver ingots had already become a standard in China and continued until the abolishment of the silver liang system in 1933, which we will see in Chapter Five. Considering that, as the confiscation records shown, silver bars worked as the most popular form of liquid wealth in Bruges (Murray 2005, 132) until the late 14th century, when they were replaced by gold among Italian merchants, somo might have acted as a catalyst in the silverization of the western part of Eurasia. Needless to say, a silver bar of 200 grams was convenient for a long-distance trader engaging in large-scale trade in valuable goods but was far from practical for the daily transactions of ordinary people. Even in the western part of Eurasia, in lower-level markets, as we will see later, there were many different local currencies in circulation. We have already seen that in the eastern part of Eurasia real silver was seldom used. Transactions were so dependent on paper money, especially in Northern China, that prices were often given in terms of zhongtong chao. In Southern China, the former territory of the Southern Song, paper money appears to have been less popular. In this period, as we saw in the beginning of this chapter, peasants in villages of the Lower Yangzi would bring sacks of grain to town to purchase basic necessities. Thus, in lower-level markets, grains continued to function as money. Silverization during this period thus did not always involve the use of real silver; rather it meant the use of silver as a unit of account, accompanied by the circulation of paper money and other local currencies. The Mongol Empire introduced silver as a unit of account to societies along the continental trade routes. As shown in Figure 3.3, while the commensurability among silvers prevailed across Eurasia, local exchanges remained independent from interregional exchange. This structure enhanced interregional exchange substantially.

6  Copper coins driven East and South We have already seen that paper money substituted for silver in China. However, as paper money grew in popularity, copper coins were also driven out of use in local markets. This second type of substitution caused unprecedented

86  Global history of monetary delocalization

Europe 50 florin 1 somo = 5 florin 100 gigliati deniers


Kipchak 10 somi 1somo = 5liang

grains linen

1 silver ding, (=50 liang)




money of account

200 asper 1gigliat=2aspers

paper monies

copper coin cloths

copper cash grains

Figure 3.3  Multiple strata of monies

transformations in societies across the South and East China Seas.Without paying attention to flows of copper coins, it is impossible to fully recognize the level of prosperity of overseas trade in the China Seas during this period. After incorporating the territory of the Southern Song, the Yuan was faced with an overabundance of copper coins. Generally speaking, the Yuan engaged in the demonetization of copper coins to an even greater extent than the Jin, which had followed the same policy. However, the extent of demonetization varied widely by region. In Fujian and Guangdong, for instance, the Yuan permitted the continued use of copper coins and made little effort to circulate paper money. This compromise was probably a result of the huge stock of copper coins that circulated in former copper-producing regions as well as the high levels of maritime trade in south-eastern China (Maeda 1973, 72–5). Quanzhou, for instance, was a major porcelain-exporting port, where many Arab merchants, including the famous Pu Shougeng, conducted business (Kuwabara 1989). In the Lower Yangzi, however, the Yuan actively worked to demonetize copper coins, which almost certainly resulted in a decrease in their use. The Lower Yangzi was not just the most commercialized region of China but also the largest source of salt revenue for the Yuan. Thus, the heavy use of paper money was necessary in the Lower Yangzi, and one contemporary source suggests that the circulation of copper cash dropped 90 percent in the region through the transitional period from the Song to the Yuan (Maeda 1973, 85). Another contemporary source, meanwhile, suggests that under the Yuan, about half the currency in circulation was copper coins and half was paper money, and in the former territories of the Southern Song, copper coins retained an important role (Maeda 1973, 76). It is hard to judge these claims because no statistics exist

The ignition of monetary delocalization  87 Table 3.3  Casting of bronze bell in Japan, Korea and China





0 0 0 0 0 0 4 6 9 9 22 31 33 36 36 42 26 28 26 25 16 17 18 15

3 4 3 1 1 0 1 2 7 9 2 3 6 0 0 1 0 0 0 0 0 0 1 2

9 2 5 9 13 18 11 12 10 5 18 20 20 30 8 9 5 11 9 16 9 4 4 5

Sources: Japan: Tsuboi 1970, 305–55; Korea: Tsuboi 1974, 265–8; China: Tsuboi 1984, 462–550.

to show how many coins were taken out of circulation, but the casting of temple bells potentially provides a clue. As shown in Table 3.3, bronze bell casting peaked in China at the same time that paper money became popular and then fell dramatically when the Yuan collapsed, and copper currency again became dominant.19 The demonetization of copper coins in China increased their export abroad, especially to Japan. During the 10th through the 14th centuries, Japan is not believed to have produced any of its own copper. The high rate of bronze bell casting in Japanese temples during this period (Table 3.1) suggests that a large amount of old coins were likely imported from China, and it is not coincidental that the proportions of copper, lead and tin in the Kamakura Big Buddha perfectly coincide with the proportions in Northern Song coins like the yuanfeng qian.20 Isotope analysis of the lead in the Buddha further indicates that it originated in Southern China (Hirao 2008a). In addition, analysis of lead isotopes in bronze items produced in medieval Japan demonstrates that lead of southern Chinese origin had been used since the late 12th century, when Chinese coins began to appear in large numbers in Japan (Hirao 2008b, 29).

88  Global history of monetary delocalization

Thus, the demonetization of copper coins in China supplied the material needed to meet religious demands in Japan, where rice and cloth had functioned as money since the 10th century. In the same period, copper coins gradually came to be used as currency in many regions of Japan, as well. Particularly in areas far from the capital, tribute to aristocrats started to be denominated in copper cash, replacing payment in kind, a process called daisenno, which spread across Japan, and, as a result, it became popular to set prices in terms of copper coins. In conjunction with the spread of copper currency, local markets, typically held every ten days, also proliferated, and as shown in the next section, the emergence of new marketplaces peaked in the first half of the 14th century. We can thus see that the growth of marketplaces followed the rise of copperdenominated tribute payments, which itself followed the increased production of cast bronze bells.21 In Japan, there was always strong demand for Chinese goods, called karamono, such as silk and porcelain. However, from the late 12th century through the 14th century, copper coins seem to have been the largest import from China. In exchange for copper, timber for temple building, sulphur for gunpowder and decorative gold for ornaments (particularly for ornamental cloth as we will see in the next section) were all exported to China.22 Though the two attempted Mongol invasions of western Japan in the late 13th century had resulted in political hostility between the Yuan and Japanese authorities, Sino-Japanese trade nevertheless rose to unprecedented levels. A snapshot of the thriving East China Sea trade can be provided by the Sin’an wreck off southern Korea, which contained remains of the cargo of a ship that left Ningbo in 1323 carrying 28 tons of copper coins, tin bars and ceramics (National Maritime Museum of Korea 2006, 204). The sudden abundance of copper coins was not limited to Japan. A copper epitaph inscribed around 1350 reveals that land contracts in Java began to be denominated in terms of copper coins (Reid 1993, 96). The circulation of ganza (tin–copper alloy) in Malay and Indonesia also began in the 13th century, usually in the form of uncoined bars with boat-like shapes reminiscent of the somo silver bars of the western Khanates.23 In Korea, however, the importation of copper coins was much more limited and, as shown in Table 3.3, trends in temple bell casting in Korea suggest that conditions in the peninsula differed from those in China and Japan.

7 The Eurasian age of commerce: synchrony built on stratified markets The Mongol period is known as an age in which both caravan and maritime trade across and around the Eurasian continent reached unprecedented heights.24 We can call the period between the mid-13th and the mid-14th centuries the Eurasian age of commerce. A number of famous travellers, most notably Marco Polo, made their way along these trade routes and recorded their observations.25 Flourishing long-distance trade between the western and

The ignition of monetary delocalization  89

eastern ends of Eurasia was accompanied by the rise of regional maritime trade, including in the Baltic Sea and the China Sea. Over the course of the 13th century, the number of merchants in Novgorod doubled in conjunction with the development of the fur trade with the Kipchak Khanate (Moshenskyi 2008, 182).The German Hanse trade monopolies connected trade in Novgorod with that in Bruges (Ogilvie 2011, 97).26 And these examples represent just a small part of developments occurring across the Eurasian continent. There are illuminating stories which illustrate unprecedented trade connections ranging between both ends of Eurasia. Despite wars between the Yuan and Japan, the trade crossing the East China Sea enjoyed a prosperity never seen before. A most important export from Japan in this period was gold. The Mongol elites were very fond of cloth decorated with gold threads.This fashion was brought to the west. Simone Martini in Siena painted the Annunciation around 1333 in which the angel Gabriel wears a Mongol-style white-and-gold robe. The Sienese textile industry was influenced by Mongol fashion, subsequently Martini incorporated this into his painting (Finlay 2010, 155). In contrast with gold, cobalt has left us a story of movement in the opposite direction. It is well known that the white-blue porcelains produced in Jingdezhen in China used cobalt for a beautiful blue. Chemical analysis has identified the cobalt used in such porcelain as being the same as ores from Germany and Italy, as well as from Xinjiang and Gansu in northwestern China. Cobalt of the same chemical characteristics has been detected in Chinese porcelain found in Delhi, and the use of finer cobalt in Jingdezhen seems to have begun only after 1325, peaking around 1350 (Chen et al. 1994, 14–19). Silk from China appears to have been the most important and profitable product moving west (mainly through the overland caravan trade), though we have no way to measure its volume and thus cannot identify any temporal trends.27 In contrast, enough records remain from the trade in porcelain, another major commodity from China that travelled west (in this case mainly along maritime routes), that we can say something about its development. Regarding the trade in porcelain bound for western Eurasia, peaks in volume coincided with periods of abundant silver. Tax records from the treasury of the Rasulid Dynasty of Yemen, which show the collection of taxes on commodities from China, reveal that during the period in question, Chinese ships seem to have voyaged at least as far as Aden.28 A large quantity of porcelain was brought via Aden to Egypt, where a number of blue-and-white ceramics have been unearthed.29 Porcelains excavated from Egypt and Iran demonstrate that porcelain imports around the Red Sea and the Persian Gulf increased in the late 13th century and declined in the late 14th century after maintaining a consistently high volume in the first half of the 14th century.30 In addition to high-quality blue and white porcelain from Jingdezhen, a large quantity of blue porcelain from Longquan and white porcelain from Fujian made their way west during this period (Mori 2012, 2–4). Though we have some evidence of the westbound flow of porcelain, the movement of eastbound commodities is more difficult to track. Merchants

90  Global history of monetary delocalization

from Italy and Novgorod brought linen and fur east to Kipchak cities like Sarai, but there was little demand for these products in China proper. As Ibn Battuta observed, cowries from the Maldives transported by Arabian merchants to exchange for Bengali rice reached Yunnan and circulated there but were unlikely to have been of great value (Gibb 1962, 827). Cobalt, some of which was of European origin, was brought to Southern China for use in decorating blue and white porcelain and in other industries, but it is unlikely that the value of cobalt was high enough to equal the value of westbound porcelain. It is possible that horses raised in the steppe regions controlled by the western Khanates were an important part of eastbound trade, and it seems likely that high-quality horses from the Kipchak Khanate made their way to India via the Arabian Peninsula to meet an increasing demand driven by wars on the subcontinent. Ibn Battuta also witnessed the export of horses from the Black Sea region to India via Arabia as well as the same trade through Kish along the Persian Gulf (Chakravarti 1991, 170–4, 17; Digby 1971, 30). However, there is no clear evidence to indicate that any significant number of horses from the western Khanates reached China proper during this period. Given the dates of the trans-Eurasian trade, it seems likely that there was an imbalanced structure in which unidirectional silver tribute payments from estates in China proper, the touxia, to the western Khanates created a silver surplus in the west that was balanced through the importation of silk and porcelain from Southern China. Given the use of paper money in the Yuan and of silver currency in the western Khanates, silver that reached China would have been likely to quickly flow back out. However, should the paper money become invalid, the entire structure, as shown in Figure 3.3, was liable to collapse – as it did around 1360. Marketplaces that attracted neighbouring peasants for exchanges likely proliferated across the world far before industrialization. However, it is difficult to trace their development. Only after the 13th century can we identify the chronologies of marketplace emergence.Table 3.4 shows the emergence of new marketplaces in Holland.The establishment of new markets increased from the 13th century and peaked in the 14th century before stagnating in the 15th century.31

Table 3.4  New markets in Holland and Japan

before 1200 1201–50 1251–1300 1301–50 1351–1400 1401–1450 1451–1500



1 7 13 29 27 11 8

1 7 16 29 26 15 2

Sources: Holland: Dijkman 2011, 43; Japan: Toyoda 1952, 112–18.

The ignition of monetary delocalization  91

During the same period, marketplaces also emerged in Japan parallel to the establishment of markets on the western end of Eurasia. As shown in Table 3.4, the emergence of new marketplaces peaked in the first half of the 14th century in Japan. In conventional Japanese historiography, this development has been considered in conjunction with the transformation of the tribute system from payment in kind to payment in copper coins, which was called daisenno, as mentioned in the previous section. Since historians have always evaluated these changes through a national framework, both the emergence of new markets and the transformation of the tribute system have generally been interpreted as the result of the development of market activities in medieval Japan. Similarly, medieval European historiography has, in general, argued that developments within Europe conditioned patterns of market emergence there. From this perspective, markets emerged as a result of intensified commercialization that began in the 13th century, which was the product of the establishment of new cities. This commercialization, and the resulting emergence of markets, was then suddenly interrupted by the Black Death in the second half of the 14th century.32 The emergence of large numbers of new marketplaces in early 14th-century Holland, followed by late 14th-century stagnancy, also follows this chronology and seems to support this general interpretation. However, the general trends in the emergence of new marketplaces in Holland and Japan appear to be synchronous. Did these markets really emerge independently with no relation to each other? As we saw, throughout this period similar transformations in the spheres of money, trade and taxation occurred throughout Eurasia. As a result of these transformations, similar developments took place in different, distantly separated regions of the continent; the parallels in the trends of new market emergence in Holland and Japan are but one example of this. We can only understand these parallel phenomena by finding imperial frameworks that operated behind the scenes, and across the continent. Many thinkers, such as Hicks, believe that markets create money (Hicks 1969, 63). However, as these cases show, in some situations, this relationship of causality can be reversed: monetary supply can create markets.There were cases when increasing currency supplies stimulated the emergence of marketplaces. In the case of the domains in Japan, though in later period, the rulers appeared to believe that a stable currency supply was a precondition for successfully establishing a marketplace (Abe Y. 1965, 52–3). A marketplace provides people with opportunities for anonymous exchange. Even if a seller and a buyer are acquainted, in markets they can conduct one-time transactions. The principle of one-time transactions – i.e., a transaction settled without any expectation of a subsequent transaction – makes cash indispensable in the marketplace. The Quadrant IV (anonymous-proximate) pattern of exchange (in Figure I.1) cannot develop without a currency supply. When potential demands for anonymous exchanges already existed among locals, an exogenous supply of currency might have ignited the emergence of a new marketplace. Throughout this period, the spread of local marketplaces and the unprecedented rise in long-distance trade occurred in tandem, but the two trends

92  Global history of monetary delocalization

11 60 11 –79 80 12 –99 00 12 –19 20 12 –39 40 12 –59 60 12 –79 80 13 –99 00 13 –19 20 13 –39 40 13 –59 60 13 –79 80 14 –99 00 14 –19 20 14 –39 40 14 –59 60 14 –79 80 15 –99 00 –1 9

2000 1800 1600 1400 1200 1000 800 600 400 200 0

Figure 3.4 Wheat price in grain of silver, England Source: Postan 1987, 215, 255.

occurred under different organizational frameworks and thus did not converge perfectly. As Figures 3.1 and 3.4 both show, long-term trends in the price of grain in terms of silver in England and in the issuance of silver coinage in London appear similar.The general trend in the emergence of new marketplaces in Holland, as discussed earlier, also appears to follow the trend of silver coinage in London. Considering that English silver, genuine or imitated, dominated in the Low Countries in this period, it might not be wrong to assume that an abundant silver supply stimulated local exchanges enough to establish new marketplaces and that its suspension discouraged local trades. However, on shorter time scales, as Harry Miskimin has found, the increase and decrease in the supply of silver coins in both London and Paris had little relationship with wheat prices in the 14th century (Miskimin 1963, 118). In other words, during this period, the proliferation of commercial activity likely occurred on multiple levels. Let us now examine how markets were stratified from top to bottom in accordance with the type of money used for transactions. As mentioned earlier, at the highest levels of transcontinental trade uncoined silver was the main unit of exchange. Yuanbao ingots from China proper and its surrounding regions and the somo, a silver bar from the region stretching from Turkestan to Eastern Europe both seem to have circulated widely. However, the uncoined forms did not contain equal weights of silver. The weight of 65 silver bars unearthed from remains at Orheiul Vechi in Moldavia, mentioned in Section Four, range from 173 to 227 grams.33 These weights appear to be distributed around 200 grams, but with a large variance. Similar variance is found in yuanbao ingots, which had a mean weight of 50 liang.34 The irregular distribution of weight among 72 silver ingots found in a

The ignition of monetary delocalization  93

10th-century Intan shipwreck near Java suggests that uncoined silver had never circulated in ingots of uniform weight in China (Flecker 2002, 84–5). It is important to remember that, as argued earlier, silver was originally a substitute for silk in Chinese monetary history. Small differences in the length or weight of a bolt of silk were likely of little concern in transactions, even if we assume that dealers might have argued about the quality of the silk. Similarly, neither the yuanbao in the east nor the somo in the west was defined by any precise standard. Unlike transactions conducted according to fixed measures adopted by mercantile states in later periods, economic activities under the Mongol regime did not depend on the intrinsic properties of silver. Rather, silver served primarily in a conceptual role as the unit of account. Gold was also used in a similar way in mid-14th-century Western Europe. Items listed in terms of gold in merchants’ ledgers were not actually purchased with real gold.35 The conceptual function of silver as a unit of account was the basis of currency commensurability across the continent, and it developed in tandem with the emergence of a division between currency used for interregional settlement and that used in local transactions. Even if we can confirm that late 13th-century bills of exchange could enable transactions between some cities, like Genoa and Tavriz (Moshenskyi 2008, 197), in western Eurasia the somo must have played a significant role as a means of bridging gaps between different local currencies. At the same time, such a link did not result in price convergences across different local markets connected to each other through silver-denominated trade. The connections between local markets were quite loose and the prices of most goods moved independently in different places. In the Kipchak Khanate, a type of silver coin called the tamga was already in wide circulation, with one somo theoretically equivalent to 120 tamga. As mentioned earlier, in the Kipchak and Chagatai khanates, tamga was also the name of the commercial tax. The term was also used to refer to customs taxes in Anatolia under the Ilkhanate (Köprülü 1992, 67). However, the tamga was not the sole currency used in local transactions. Novgorod also used den’gi and grivna (worth 100 den’gi) and later the ruble. Silver coins with inscriptions from the late 13th century found in the Chagatai Khanate and weighing 1.8–2.0 grams (almost half a gigliat, or one twentieth of a liang) thus may have been seen as equivalent to den’gi, rather than tamga (Wang and Zhong 2007, 17–18). The Chagatai silver coins must have been minted after the Masud Beg reformation which unified the weight at 2.1 grams (Davidovich and Dani 1998, 406). The reformation showed a divergence from the dirham system for which minting centred on three grams. In the case of Tana on the Sea of Azov, the somo was used to purchase goods in long-distance trade, while the mint used the silver from a single somo to coin about 200 aspers for local use. A Genoese statute of 1304 treated the Kipchak asper as equivalent to ten Genoese deniers. The Byzantine folleri, a copper coin valued at 1/16th of an asper, was used to purchase vegetables and small items for daily use (Lopez and Raymond 1955, 356;Yule 1914, 159).

94  Global history of monetary delocalization

As the case of Tana demonstrates, commercial exchange tended to be divided into three layers, each with its own distinctive currency. Uncoined silver like the somo was used in the top layer to conduct interregional trade. A local mint would issue silver coins, like the asper, made from melted-down ingots, for local uses such as tax payments. Meanwhile, even the smallest denominations of silver coins were too valuable to be used by most people in daily life, so various coins made of other metals, like the copper folleri, were used in small transactions.The three currencies thus worked in concert to handle all levels of commerce; their relationship was not substitutive but complementary. Thus, transactions of daily goods remained largely independent from transactions mediated by officially recognized currencies.The boundaries between the layers, however, were not clear and could shift in different situations. A guidebook to Northern China in Korean, the Nogeoldae (Laoqida in Chinese) published in 1340, provides us with snapshots of a variety of transactions, such as payment for a night at an inn and the sale of Korean ginseng, in which all payments were denominated in terms of the paper zhongtong chao (Funada 2001: 17–20). Zhongtong chao continued to be issued after the Zhongtong Era (1260–1264), in association with zhiyuan chao, and remained a major unit of account in transactions through the Yuan period (Miyazawa 2012). The guidebook describes in detail how traders calculated the yashui, or sales tax, and made payment in paper money, confirming that paper money was used to collect taxes. Meanwhile, the smallest transaction described was the purchase of one jin of pork, the value of which was given as one liang. From this it is possible to determine that even if paper money was common in Northern China, it could not have handled most daily transactions. The popularity of paper money thus did not mean that officially produced currency completely dominated private transactions. The Nogeoldae shows that merchants often ignored the face value of a bill, negotiating the value of paper money depending on its appearance, and local practice usually trumped official regulations (Funada 2001: 6–11). Transactions described in Yuan stelae further reveal that even people in Northern China frequently carried out business in terms of copper coins (Ichimaru 2008, 97–9). Thus, not only in small-scale transactions, but even at higher levels of commerce, private trade often worked according to its own principles, independent of the principles of taxation and administration. This is why, for instance, following the collapse of Yuan rule, the use of copper coins quickly revived in the Lower Yangzi, in spite of official prohibitions lasting for more than half a century (Kong 1987, 25). Multiple layers of commercial activity mediated by various currencies thus appeared across Eurasia during the Mongol period. In Yuan territory, paper money appears to have circulated commonly, as required by regulation. However, a lower layer of transactions also existed, making use of several local currencies. As mentioned earlier, copper coins continued to play a significant role in former Southern Song territories and in towns across central and west Asia, while in Yunnan, cowries, shell money from the Maldives, were more important for local commerce. Meanwhile, in regions like Northern China, which lacked

The ignition of monetary delocalization  95

currency in denominations small enough for use in daily life, wooden tablets, wrapping paper and other monetary substitutes provided by merchants served to supplement official paper money (Fang LG. 2001, 437). In addition to these examples, in rural areas, grain continued to be used as currency.

8 Sprouting of credit-oriented systems in Europe: an aftermath As we saw, the growth of new marketplaces in Holland followed patterns so similar to that of Japan that it appears almost perfectly synchronous. In both places, the establishment of new markets peaked in the 14th century and then stagnated in the 15th century. England also experienced a peak in market establishment in the 14th century (Poos 1991, 34-5; Hilton 1983, 172), and in all three countries, market establishment thus closely tracked patterns of currency issuance that we have seen earlier. Unfortunately, we do not have the same sort of statistical data for marketplaces in regions under direct Mongol rule. All we can say for sure is that both ends of Eurasia simultaneously experienced a rapid increase in the number of rural markets, and this trend came to an end at around the same time. In Essex, England, grain prices in terms of silver rose in the 1360s before falling and remaining low until the mid-15th century. Rent in Essex fell after the 1350s and also remained low until the mid-15th century. An increase and subsequent stagnation in the establishment of rural markets likely followed the same trend. Currencies, or tangible means of exchange, rapidly became scarce in the western end of Eurasia in the 1360s, and the shortage of currency continued through the 15th century. It is possible to interpret the drop in grain prices and rent as having been caused by the population collapse that occurred in the wake of the Black Death (Poos 1991, 51, 212). However, based on Chinese sources, it seems that low grain prices in terms of silver appeared more broadly across Eurasia from the late 14th century through the 15th century (Peng 1965, 455–61). As we will see in the next section, after one century of Eurasian-wide silver proliferation, a number of autonomous local currency zones using copper coins appeared and finally covered almost the entire region surrounding the China Sea. In contrast, responding to the contraction of currency circulation, forms of credit became a more important device to mediate exchanges in Western Europe. In the region, proximate credit and distant credit, or following the concept of the four Quadrants of exchange, Quadrant III and Quadrant II increased their importance along different trajectories. At the local level, annuity contracts made privately as well as supplied by municipalities developed through the 14th and 15th centuries. At the interregional level, there were offsetting drafts at fairs for payment such as the one Lyons established. Meanwhile, states with the right to mint tended to discourage merchants from using credit devices such as drafts, in order to secure seigniorage for themselves.

96  Global history of monetary delocalization

Before the 13th century, some European cities which suffered from serious shortages of revenue had already forced citizens to lend their money to the cities, but these forced loans were still sporadic. However, as late as the 1260s, some cities in northern France such as Calais and the Low Countries began to borrow money from citizens regularly. The cities commenced selling annuities to citizens. Loans from citizens in compensation for future pensions spread among other cities. It was also in the second half of the 13th century that some cities in Italy, such as Venice, began to require senior class citizens to regularly lend money to the city-states. Contracts for annuities were made privately, as well. For example, being secured by agricultural lands, life-long annual charges were contracted between individuals. In order to secure the best utilization of land, full title to an estate would be conveyed to the son or grandson subject to payment of an annuity sufficient for the support of a parent or grandparent. In cases where there was no direct descendent, the property was transferred to an adopted person subject to provision for the needs of the original owner. As early as the 13th century it was usual to define the amount of the annuity, so that non-kinship relationships began to creep in more easily (Usher 1943, 145). Even so, as the cases of northern German towns show, full commercialization of the life annuity did not appear explicitly until the beginning of the 14th century (Usher 1943, 148). On the other hand, European cities increased their direct influence over the properties possessed by citizens. Repurchase of rents appeared at Caen in the second half of the 13th century and became common in the early 14th century. The timing of this development was about the same at Lübeck, but in Lübeck the rule assumed a more general form. An ordinance of the city council passed soon after a great fire (presumed to be the fire of 1276) provided that all ground rents thereafter established should be repurchasable at the price originally paid. All rents created prior to the fire should be repurchasable at rates to be determined by the city council. This general right was not established in northern France and Flanders until the early 16th century. Relying on the increasing power of cities, pensions for life granted by temporal rulers and by towns appear in accounts of the 13th and 14th centuries (Usher 1943, 145). In the second half of the 14th century, some municipal governments were already depending heavily on funded debt, including annuities. In the case of Barcelona, in 1360–61, the annual charges for funded debts amounted to 57 percent of tax revenues (Usher 1943, 356). Life annuities generally became a more important resource for cities in the 13th and 14th centuries. Perpetual annuities became more important during the course of the 15th century but did not displace the life annuity entirely (Usher 1943, 157). Table 3.5 shows the case of Cologne. There appears to have been a presumption that the income from a life annuity which was not ordinarily repayable should be greater than the rate of return on a perpetual annuity which was, in fact, redeemable. Thus, for example, in 1449, Nurnberg offered annuities in connection with a forced loan at eight years’ purchase for one lifetime, ten years’ purchase for two lifetimes and 18 years’

The ignition of monetary delocalization  97 Table 3.5 The funded debt of Cologne (marks) Life annuities Capital 1351 1370 1380 1382 1432 1454 1477 1482 1506 1512

annual charge

30920 102560 168950 76820

3092 10256 16895 7682

935635 1109311 840400 819100

84200 99838 75636 73719

Perpetual annuities





30920 102560 168950 76820 818911 899755 1898955 2424444

36851 40489 85453 109100

1754566 2009066 2739355 3243544

annual charge 3092 10256 16895 7682 35135 21629 121061 140327 161089 182820

Source: Usher 1943, 160.

purchase for a perpetual annuity (Usher 1943, 150). The increasing proportion of perpetual annuities was apparently of great help to cities suffering heavily from mounting debts. In the case of Barcelona, the annual charge on the city budget did not increase so much over the course of the next half century, but the capital amount of the debt increased rapidly, as it was possible to substitute perpetual for life annuities and the rate of interest on perpetual annuities was gradually reduced to 4 or 5 percent (Usher 1943, 357). Importantly, the interest rate in private annuities tended to fall from the 14th century to the 16th century. Table 3.6 shows the rate of interest on sales of rent charge against agricultural land in Germany. In the 14th century through the first half of the 15th century, the rate in most cases was 10 percent, in the second half of the 15th century 6–7 percent and in the first half of the 16th century 5 percent. The rates of annuities issued by authorities were apparently influenced by the rates prevalent in private annuities. For example, after the accession of the Spanish monarch Phillip II, the interest on the annuities issued were reduced from a 10 percent basis to 7.15 percent, and in 1575, a significant amount of annuities were sold at rates up to 6 percent. The archives of Castile contain materials on the loans secured by real estate kept by private individuals. Among them, in 1563, rates ranging from 7.7 to 10 percent are found, but in 1575, rates were commonly under 7.15 percent, though rarely less than 5 percent (Usher 1943, 174). The growth of funded debt in private borrowing as well as public lowered the interest rates in Europe in the period from the 14th century to the 16th century. In this sense, the municipal annuities represented a part of the custom of life-long credit which had already taken root in the Euro-Mediterranean world in this time. However, the engagement of long-term borrowing by local authorities eventually paved the way to establishing a national public debt

98  Global history of monetary delocalization Table 3.6 Rate of interest on sales of rent charge against agricultural land, Germany (percentage of distribution for each period) Number of over 10 percent 8–9 percent 7–6 percent 5 percent under items 10 percent 5 percent 1200–1300 1301–1350 1351–1400 1401–1450 1451–1475 1478–1500 1501–1525 1526–1550

5 56 304 108 74 236 237 194

40 16 5.5 2.7 0 0 0 0

40 63 72.6 41.8 5.4 6.4 0.8 0

20 16 15.1 19.5 6.8 7.2 3 0

0 5 6.2 28.8 67.6 57.6 35.7 5

0 0 0.6 4.5 14.8 26.2 52.2 84

0 0 0 2.7 5.4 2.6 8.4 11

Source: Usher 1943, 172.

system connecting the state and households across a wide range of people two centuries later. Compared with raising tax rates or levying a new tax, loans from citizens was a quicker way of collecting money to manage emergent expenditures.Wars especially caused cities to be inclined to establish devices for forcing loans from citizens. The case of Bruges in the 15th century is a clear example. In peaceful times public debt constituted less than half of annual income, but it reached two and a half times the amount of annual revenues in 1491–1496 when the city was at war. Throughout the 14th and 15th centuries, municipal debts to residents spread across Europe and many cities suffered from mounting debts. It is striking that the lower middle classes also took an active part in investment in annuities, certainly during boom periods such as the 16th century. In Antwerp, of the purchasers of annuities who had their transactions officially entered in the aldermen’s register in 1545, 25.15 percent were craftsmen, 21.19 percent administrative officials, 16.8 percent widows, 16.24 percent merchants (this group also included many small peddlers and hawkers) and 20.62 percent miscellaneous (van den Wee 1977, 306). A municipal loan in Genoa, luoghi di San Giorgio was highly popular in the 15th century. More than 11,000 names appear in the registers, including those of many quite humble people – widows, orphans and religious communities (van der Wee 1977, 363). In 1433, the Book of Deposits in Barcelona carried the names of 470 creditors and of seven debtors. With allowances for repeated names, there were accounts with 1,494 individuals. The population of Barcelona at this time may be estimated at 30,000 persons, which means that about one-fourth of all the families had an account with the Bank of Deposit (Usher 1943, 181). The contraction of silver circulation in the second half of the 14th century affected the settlements of distant trade more quickly than it did those of proximate trade, since increasing dependence on book transfers among merchants in the same town could compensate for the decrease in currencies’ circulation

The ignition of monetary delocalization  99

more easily than in the case of distant transactions.Then merchants engaging in long-distance trade found a way to make settlements without the use of specie. For example, huge quantities of English wool were purchased by Tuscan houses. Without shipping silver from Italy, they paid for it with the taxes they collected for the Apostolic Chamber in England and northern Europe. They paid the equivalent amounts to the Papal treasury from funds they got through the resale of finished goods in southern Europe (Körner 1994, 509). In this way, public transfers from English churches to the Pope and the payment for English wool by Italian merchants happened to be combined. The unidirectional flow of money due to this religious factor provided long-distant merchants with chances for offsetting private payments. Thereafter, Italian financiers increased their influence in England. As typically shown in the case of Bruges, merchants in cities settled transactions with book transfers rather than with coins or specie.The book transfer system appears to have been difficult to use in mediating the transactions beyond a city. However, an account book from a moneychanger in Bruges showed that they made settlements through letters issued by moneychangers in proximate cities, such as Ghent. Through the second half of the 14th century, transfers through letters among moneychangers’ sharing book account system probably became more popular. However, as the account book of another moneychanger in Bruges did not show any such cases, the extension of bookkeeping beyond a town must have been subject to certain limitations (Murray 2005, 252-3). In this way, transfers of public money had the potential to weave together a network of distant settlements. Mutual offsets of distant settlements became more popular in Europe after the mid-14th century as distant settlement by specie decreased. Representing this tendency, Lyon became the centre for multi-offsets of distant settlements. From a short-term viewpoint, the development of fairs as hubs for the offsetting payments, like that in Lyon, resulted from the upset of a fundamental balance during the 15th century when the exports of products from northwest Europe to the south dropped sharply, whereas imports of luxury goods from Italy and the Middle East to the north increased. The negative balance of payments in the north relative to the south was initially financed through imports of capital by the big Italian firms, which extended their financial investments in the north through local branches, and increasingly by exporting precious metals. The result was a gradual depletion of the reserves of precious metals in northwest Europe, which undermined international trade and hampered international payments (van der Wee 1977, 307). Although there was variety in the degree of activities according to country, generally through the 15th century, minting activities in Western Europe became inactive. For example, from 1440 to 1466, mints in the Low Countries issued only very small quantities or became idle (Spufford 1970, 47, 200). In particular, there was a considerable lack of official small change, since coins of so low a silver content gave practically no profit to the duke, whose seigniorage was calculated on the amount of fine silver used (Spufford 1970, 44). It was not

100  Global history of monetary delocalization

by accident that a significant proportion of inferior silver coins, called monnaie noire, seems to have been issued during this period.When the state failed to supply the everyday needs of its subjects, informal enterprises, domestic or foreign, provided alternative currencies. However, they were not in sufficient quantities to meet demand either (Spufford 1970, 202). People hoarded even the smallest denominations (Spufford 1970, 47). Between 1421 and 1461, the Royal Mint coined a little gold and a little silver. All the mints of the Burgundian Netherlands were idle from 1440 to 1466. Wool purchases in the Low Countries, mainly from England, dropped sharply by almost half after 1430, never to recover (Munro 1966, 1141).The Casa di San Giorgio in Genoa ceased functioning as a public bank in 1444 (van der Wee 1977, 312). Barcelona suffered from the devaluation of vellon pennies against fine silver croat during 1454 and 1456 (Usher 1943, 377–80). The croat was rated at 18d., instead of 15d., and the gold florin was to circulate at the values established in the open market. Basically, a state which issues a currency and collects taxes in terms of that currency tends to prioritize its stable circulation. Even if credit instruments enhanced transactions, as far as the devices possibly might have brought disruption to the formal currencies, the state often prohibited credit instruments which were beyond official control. In particular, when in need of revenue for urgent expenses such as in war, states preferred bullion to any form of credit like drafts in their collection, since they required a means for instant payment to anyone. For example, in the 15th century, during the Hundred Years War, English monarchs discouraged merchants from using drafts in their settlements. Kings needed cash for war. The dependence on credit among merchants made the Royal Mints idle and caused the state to receive less cash revenue. Thus, Henry VI’s Council required wool merchants to exact full, immediate payment in coin and bullion, and subsequently this taxation overthrew the existing system of sales-credit which was essential to the wool trade (Munro 1966, 1141). The policies worsened the stagnant situations of interregional trade. Through the 14th century and the 16th century, contracts of life insurance among people increased in importance and cities’ finances became more dependent on funded debt. Both trends influenced each other, subsequently the interest rate in Europe tended to fall. However, until the 17th century, European states often announced bans on credit devices such as assignment (van der Wee 1977, 336). Meanwhile, states usually borrowed monies in the short term with high interest rates. Quadrant III became important in European private activities and municipal administrations, but it did not tie up with the development of Quadrant II.

9 Currency circuits with ‘Old’ coins flourished across the China Sea: another aftermath As mentioned, according to excavations, imports of Chinese porcelain to Egypt and the western part of the Indian Ocean increased significantly in the late 13th century and seriously declined in the late 14th century. Throughout the

The ignition of monetary delocalization  101

15th century, Chinese porcelains almost vanished from these regions. In contrast, there is no evidence to clearly suggest that the maritime porcelain trade in the eastern part of the Indian Ocean and China experienced a significant downturn from the late 14th century to the 15th century (Mikami 1969). The end of the silver boom thus clearly brought the Eurasian age of commerce to a close, but the extent to which trade contracted in different regions appears to have varied considerably. The transformation of exchange after the collapse of the Mongol regime thus differed by region, and the asymmetric paths that resulted laid the foundation for a later, 16th-century boom in interregional exchange. We must now confirm the extent of the variations that developed before the subsequent global synchronization. As we saw in the previous section, in Europe, currency issuances stagnated through the 15th century and the price of grains remained low. Those phenomena are in accordance with slumping interregional trades. How about the situation at the eastern end of Eurasia which absorbed huge quantities of copper coins due to the demonetization policy by the Yuan? Although there is only limited data for rice prices in Japan, materials from estates owned by the Toji temple do not show any tendency towards a decline in the price of rice in terms of copper during the same period (Momose 1957, 64–7). If this is true, it would appear that the stagnation of grain prices in Eurasia occurred only in regions where silver was the principal unit of account. Importantly, however, official copper coinages in China almost suspended through the 15th century. No other states surrounding China minted copper coins either in the period. If both ends of Eurasia shared almost no official minting through the 15th century, why did the price movements diversify? In order to properly understand what happened to the transactions in the eastern part of Eurasia after the collapse of the Mongol regime, we need to know the transactions that depended on copper coins at the ground level in more detail. As mentioned earlier, throughout the 11th century the Northern Song issued an astronomical quantity of copper coins.The intrinsic value of these coins was high enough that dealers were induced to melt them illegally. With some variations, we should note that Chinese dynasties basically supplied currencies in order to enhance the collection of taxes and to facilitate administrative and military expenditures. Profiting through seigniorage was not the main purpose of coin issuance. Due to high costs, however, no authority after the Northern Song again issued such large quantities of copper coins until the Qing in the mid-18th century (Kuroda 2008, 38, 49—50). While the government instituted standards for the use of these copper coins, in private transactions, people actually used them in a variety of ways. The government distributed copper coins and collected them as taxes in terms of mo, which literally meant a string of copper coins denominated 100 wen. The government’s official mo was 77 coins, but it could not force people to make transactions in this amount. One example is the people in Kaifeng, the capital of the Northern Song, where rice sellers used a mo of 70 cash while fish retailers used a mo of 67 cash. We should note that such variable usage of the mo occurred soon after the peak of copper coin issuance in the 11th century. Thus,

102  Global history of monetary delocalization

in private transactions, people used smaller units of account deliberately rather than due to any shortages of copper coins (Meng 1936, 69). This combination of a unified unit of money used in official transactions and various local units used in private transactions dominated throughout the history of the Chinese empires, in the case of both copper and silver. In other words, making prices in local exchanges in China did not always depend on the currency supply by the dynasty. Interestingly, recent excavations reveal that in 14th-century Vietnam, under the rule of the Yuan, strings of 67 copper coins existed, though the official unit was 70 coins (Sakuraki 2009, 258). Basically, Chinese dynasties did not demonetize copper coinages issued by former rulers. At the same time, insufficient supplies of official coins induced the production of private coinages, even though this could incur capital punishment. Eventually, except for periods of large issuances, such as the mid11th century and the mid-18th century, marketplaces were usually full of old coins and unofficial coins. Their use depended on local custom, which was beyond governmental concern. Even when the authorities provided sufficient currencies of a high enough standard, people basically still followed local agreements when actually using them in transactions. However, when the government failed to supply adequate official currencies, private issuance of currency (though illegal) and local customs for monetary usage flourished across the empire. Replacing the Yuan, whose revenues depended heavily on commercial taxation, the Ming began its reign with policies intentionally based on ancient practices and placed greater importance on taxation in kind and mandatory services by peasants. With the transition in taxation, the official monetary system also transformed. After the collapse of the Mongol regime, silver mostly retreated from use in China for more than half a century until the Ming introduced silver taxation, called jinhuayin, in 1436. The Ming initially issued copper coins, but their quantity was never sufficient to meet shortages of copper coinage. In addition, the Ming almost completely ceased to issue copper coins between 1436 and 1503. Bureaucrats detested the idea of hard currency, since it was believed to hamper the circulation of the paper money issued by the dynasty. However, paper money (baochao) had already lost its credibility in markets. Unlike the paper money issued by the Yuan, Ming paper money had no serial number or expiry date. The dynasty also did not establish places to exchange new notes for old ones. Put simply, paper money under the Yuan was designed to return to the issuer, but under the Ming it was distributed unilaterally. Official paper money then disappeared from China until the end of the 19th century. Thus, throughout the 15th century, China’s society suffered from a serious dearth of official currency, and each region created different devices of exchange. A memorial from 1503 states that Yunnan used shells, Jiangxi and Huguang used rice, silver and cloth, and Shanxi and Shaanxi used hides, but that few regions still used copper coins36. These conditions accelerated the localization of money even in regions where copper coins were in use, as locals

The ignition of monetary delocalization  103

practiced the custom of discriminating between coins. Here we will call this practice of discriminating between coins – in other words, removing some coins and selecting particular coins – ‘shroffing,’ which authorities tried in vain to prohibit. Between 1460 and 1480, the authorities attempted the most prominent antishroffing measures. During these years, ‘new coins’ suddenly began to pour into Beijing and they soon came to dominate transactions in the capital. At the time, people tended to hoard Hongwu and Yongle coins (issued in the early Ming), or use them at several times the value of the ‘new coins.’ It was said that the silver/copper coin exchange ratio fell from 1:800 to 1:1,300. Confronted with such a drastic change, the government issued an order placing greater emphasis on preventing shroffing than on the use of counterfeit coins. In other words, the authorities appeared to think that shroffing, rather than the increase of coins, caused the rapid price hikes. Around 1500, ‘new coins’ further flooded into Beijing and ended up with half the value of good coins. During the early 16th century, the custom of taking two ‘new coins’ for one good coin spread outside Beijing. We may assume that the ‘new coins’ would have borne Song-era names like the old coins, but, though they may have appeared to be real Song coins, it was impossible to conceal their difference from the actual 300-year-old coins. That is why people called them daohao (‘imitated well’). In Shanghai, Song coins dominated around the turn of the 16th century, but ‘new coins’ began to replace the good coins in the 1510s. As time went by, cheap coins became more popular. Throughout the first half of the 16th century, under Ming rule, Song coins, genuine and imitated, dominated transactions (Kuroda 2005, 67–8). Both ‘new’ and old coins circulated side by side, although they tended to be associated with a particular exchange ratio. Importantly, the ratio differed by region and changed as time passed. People who practiced discrimination between copper coins did so not only to remove particular coins but also in order to sort them by function. People hoarded good old coins to attain wealth, while using bad new coins for daily transactions. Here we shall refer to the multivalent coins as ‘standard coins’ and the single-value (one wen) coins as ‘operational coins.’ In 1511, the use of ‘newly minted coins’ was banned by the emperor. Historical sources do not refer to the provenance of the ‘new’ coins. However, it was likely no coincidence that in about the same year Putian County, Fujian, began to suffer from the circulation of ‘privately minted new coins.’ The Prefect of the county banned the use of ‘new’ coins, but these orders were only followed in public offices. For ‘woodcutters and female peddlers’ the ‘new’ coins were indispensable for buying daily rice. The ‘new’ coins were recognized as having been brought from the southern part of Zhangzhou prefecture, Fujian. Around the turn of the 16th century, the Zhangzhou region became a source for large quantities of forged copper coins destined mainly for the cash-starved northern capital of Beijing and partly also for coastal areas along the northbound route, including Putian (Kuroda 2005, 70–1).

104  Global history of monetary delocalization

Zhangzhou was known for foreign trade and therefore needed to retain larger amounts of Song coins than other regions. The large and heavy Song coins could be used to forge lighter and smaller coins. Coastal Fujian’s location in the south, moreover, gave it advantages in securing the metals necessary for casting copper coins and other items made of bronze, which is an alloy of copper, tin and lead. At this time, Japan was the largest supplier of copper in Asia, and Japanese traders frequently engaged in illegal trade with Fujian. Despite the geographic limits on sources of tin, ample supplies could be obtained from nearby Hezhou, Guangxi, as could a large amount of lead brought to Fujian from Song Toh, Thailand. Isotope analysis of lead in bullets left on the battlefields of the late 16th century in western Japan strongly suggests that the lead was produced in Song Toh. For example, the isotope proportion of lead bullets found at the site of Todaiji in 1563 coincides with lead from Song Toh (Hirao et al. 2014). Zhangzhou was thus a port ideally located to import all of these metals. The gazetteer of Zhangpu, a county in southern Zhangzhou, proves that people in the region minted ‘new coins’ with old-era names. It fearlessly noted that the people there did not use official Ming coins but illegally minted ‘Song coins.’ The account adds that the Song reign-mark on the forged coins was changed every few years and some abolished coins bearing a Song reign-mark left the county. The source also tells us of differences in the quality of counterfeit coins: the coins of Zhao’an County were the best quality, those of Zhangpu County were next and those of Longxi County were the worst.37 Seen from Fujian, Beijing was not the only place in the north that valued copper coins. Ships sailing north from coastal Fujian could reach western Japan more quickly than they could Beijing, and some may have profited from unloading their ‘new’ coins in the archipelago. Imitating China, the Japanese imperial court tried to establish its own copper coinage in the 8th century but discontinued the coins no later than the 10th century due to difficulties accessing sufficient supplies of copper. In Japan, metallic coins were not popular, but, instead, rice and cloth worked as money until the 12th century. However, as shown earlier, the monetary policy of the Yuan, with its dependence on paper monies, encouraged merchants to export large quantities of copper coins, especially Song coins, to surrounding societies. In the first stage, Chinese coins met Japanese demand for materials to use in the production of religious wares like bronze bells for Buddhist temples, but later they gradually took root as currencies in regions across the China Sea. Surging imports of Chinese copper coins drastically changed the usage of money in Japan, but, though inferior coins were mixed in with those circulating in Japan, each coin was basically counted as one wen until the mid-15th century. In 1485, the Ouchi clan issued Japan’s first anti-shroffing ordinance, which is well known for its particular reference to Eiraku (Yongle) Chinese coins. The Ouchi clan had been involved in the tribute trade with the Ming, while, beginning in 1480, Yongle coins began to disappear from Beijing markets due to shroffing. Although there is no direct proof, we can perhaps assume that the

The ignition of monetary delocalization  105

same ‘new coins’ that had been pouring into Beijing from Fujian suddenly arrived in the Ouchi clan domain (Honda 2007, 231–2). Moreover, similar historical processes occurred on both sides of the East China Sea following the arrival of such coins. Further, interestingly, in 1486, the Vietnam king issued a prohibition against shroffing copper coins (Chen 1985, 729). Generally speaking, leviers tended to discourage payers from using inferior coins in payments and referred to bad coins by various terms, such as akusen or akuryo. However, it was not until 1500 that bad coins became a serious issue. In 1500, Toji, a powerful temple in Kyoto, for the first time ordered its Yano estate to check whether the taxes that had been collected included bad coins. This was also the year that the Muromachi bakufu, the central government in Kyoto, issued its first anti-shroffing edict. Over the next 20 years, the central government frequently issued similar edicts, which reflected the increasing circulation of bad coins. In fact, during this 20-year period, tax monies sent from local estates to Toji included a higher proportion of bad coins than in any period before or after. In the early 16th century, in Bungo (present-day Oita prefecture), copper coins with various names that appeared for the first time began to circulate (Kuroda 2005, 76–7). Thus, as in the case of China, around 1500 something changed in the circulation of copper coins in Japan. Interestingly, again, in 1497, the kingdom of Vietnam issued a proclamation against shroffing copper coins (Chen 1985, 748). Let us consider a local case related to shroffing. People of Sasai district (in present-day Hiroshima prefecture) had long been affiliated with the Itsukushima Shrine and paid tribute to it in copper coins obtained by selling wood in a local marketplace called Hatsuka-ichi. In the early 16th century, a conflict broke out between the people and the shrine when the shrine began to shroff the copper coins sent from Sasai and refused to accept inferior coins. Since the shrine insisted on payment in good coins, the leaders of Sasai district appealed to the Ouchi clan, the daimyo family ruling the local domain, to ask that the shrine stop rejecting copper coins circulating within the region. As a result, the Ouchi ordered the shrine to accept all copper coins (Kuroda 2005, 76). This incident indicates that there were two kinds of copper coins, those of good quality preferred for official payment and those of inferior quality but in use among locals. In addition to functioning as a medium of official payment, good quality coins also served as an interregional currency. A letter written in 1515 to a local landlord in Buzen (present-day Oita prefecture) by the Ouchi clan clearly shows why they chose good copper coins when receiving interest from official money borrowers. The Ouchi clan had to pay in good coins when buying wood from distant places, such as Aki, Iwami and Tosa, and so they needed to accumulate good coins by identifying and rejecting inferior coins (Honda 2007, 233). Good copper coins were called seisen (‘pure coins’), while inferior coins were called namiasen (‘ordinary coins’). In other regions such as Nagato (present-day Yamaguchi prefecture), the former were called seiryo and the latter toryo. Although the names varied from region to region, in the early 16th

106  Global history of monetary delocalization

century it was common for locals in western Japan to distinguish the quality of copper coins. Importantly, locals tended to set an exchange ratio between standard coins and operational coins. For example, a receipt for tribute sent to a shrine in Buzen in 1514 makes it clear that, while the donor sent 2,000 coins of inferior quality, the shrine accepted them as equivalent to 500 coins of good quality. These ratios, moreover, varied by region and period. Sometimes several ratios appear in the sources. For example, in the case of a donation of 45,000 operational coins to a shrine, Sawa district paid 36,000 operational coins counted as 6,000 ‘old’ coins and 9,000 operational coins counted as 2,250 ‘old’ coins. In other words, two ratios, 1:6 and 1:4, were used concurrently (Kuroda 2005, 75). These exchange ratios between two types of coins often led to disputes between payers and leviers. We have already seen the case of Putian County, Fujian, in which locals discriminated between standard coins and operational coins. A later source reveals that, in Putian, old Song coins were traded at three times as expensive as official coins (Kuroda 2005, 71–2). There are parallels in how the circulation of copper coins changed in China and Japan in the first half of the 16th century. Initially the exchange ratio between standard coins and operational coins was not so great on either side of the East China Sea. In the case of China, all we can find is 1:2 or 1:3, while in western Japan, possibly because of a relative abundance of local sources, the ratio ranged from 2:3 to 1:4. In contrast, towards the middle of the century a ratio of 1:10 appeared in both countries. In 1554, the ratio of silver to operational coins in China was said to be 1: 6,000–7,000, while in the case of standard coins it was 1:700 – that is, the exchange ratio for operational coins was nearly ten times that of standard coins. Sources indicate that a ratio of 1:3,000–4,000 also applied to exchanges of silver for ‘bad coins’ until coins of even worse quality appeared. In other words, there were three categories of coins in China. Likewise, we can find evidence of four categories of coins in Japan in a proclamation issued in 1569 by Oda Nobunaga, who advanced into Kyoto with the last shogun of the Muromachi bakufu, Ashikaga Yoshiaki, that year. The decree set the ratio between the four kinds of coins at 1:2:5:10 (Kuroda 2005, 78–9). Much as the Zhangpu gazetteer distinguished between coins of different quality, in the middle of the 16th century both China and Japan stratified forged coins into categories ranging from high quality coins that appeared as fine as standard coins to inferior coins, including exceptionally poor forgeries that lacked an era name. This synchronization across the sea was undoubtedly centred on Zhangzhou. For example, Zheng Jungong, who visited Kyushu in 1556 to investigate the background of Japanese pirates, wokou, wrote in the Riben yijian, that, ‘The region of Longxi, Fujian, is secretly minting coins to sell them. The Japanese appreciate Chinese coins so much that they do not mind the falseness of Longxi coinages’. Longxi, located in the southern part of Zhangzhou, was also called Yuegang, Moon Port, which was a hub port for the illegal trade networks of the wokou (Kuroda 2003, 128–9).

The ignition of monetary delocalization  107

Hoards of copper coins unearthed at medieval sites in Japan show that they consisted mostly of Song coins, and the second, third and fourth largest quantities of coins bear reign names mentioned in the Zhangpu gazetteer:Yuangfeng, Yuanyu and Xining (Suzuki 1999, 80). Interestingly, these era names are also found in moulds for imitating Chinese coins that have been discovered at late 16th century sites in Sakai, the largest commercial city in medieval Japan (Shimatani 1994).Thus, archaeological discoveries strongly suggest that 16th-century Southern China and western Japan, though separated by the East China Sea, shared the use of Song coins, including a significant quantity forged by contemporaries. Crossing the South China Sea, with a high probability, forged Song coins reached Vietnam as well.

10  Linkage to the second silver century As forged copper coinage came to predominate across the China Sea through the processes shown in the previous section, people tended to become reluctant to keep assets in the form of copper coins. For example, the gazetteer of Changshu, Jiangnan, issued in 1540, observed that, though in the past people had accumulated copper coins to attain wealth, after counterfeiting flourished, people stopped building up stocks of coin (Deng 1965, vol. 4, 15a). While copper coins were still used for buying and selling goods, local people had already found other ways to keep their assets.The gazetteer of Wujiang County, Jiangnan, in 1561 described how peasants now obtained loans either in silver or rice, rather than in copper coins as before.38 This was a precondition of the next silver boom, which occurred in the late 16th century, as we will examine in Chapter Four. The combination of two factors, the introduction of taxation in terms of silver by the dynasty and the potential utility of silver as a device of interregional settlement replacing standard copper coins, increased demand for silver in East Asia. Actually, the supply of Japanese silver met the demand and forms of silver popularly circulated in the region as early as 1540. However, Japanese silver worked only within the China Sea region. Why did silver suddenly make a global march in the late 16th century? A global configuration of different demand/supply of silver happened to create the conditions for a greater silver flow than ever before. Potosí produced silver on an unprecedented scale, but little silver had ever circulated in the Americas where other materials such as cacao beans mediated transactions. It was the Spanish, Portuguese and Dutch who distributed the most silver in the intercontinental trade. However, most of the silver passed through their homelands and made its way to Asia. Silver had demand as a devise for interregional settlement but no strong demand as a means for local exchange in Western Europe where a local credit system substituted currencies to some extent. In contrast, silver had the potential to provide the ground-level markets in East Asia with a tool bridging local currency circuits formed by old and imitated copper coinages.

108  Global history of monetary delocalization

The difference between Europe, with increasing prices, and Asia, with fluctuating prices, throughout the period of the silver boom after the late 16th century has led some scholars to argue that precious metals worked as money in Europe, while in Asia they formed hoards, assuming even the form of artistic decoration. However, the last part of this chapter provides us with a different viewpoint, revealing that the two paths as remedies for the shortage of currencies appeared after the collapse of the Mongol regime. In Western Europe, municipalities created local credit institutions, while in East Asia local traders formed endogenous currency circuits consisting of multiple devices. According to the concept of the Four Quadrants of Exchange, the increased supply of silver must have worked to make Quadrant I particularly prolific. However, a divergence between a local society orienting towards Quadrant III and another society strengthening Quadrant IV would have responded in quite different ways to an excessive Quadrant I.We will see the diversities in the next chapter.

Notes 1 For Europe, see Spufford (1988) and Depeyrot (2018); for central Asia, Moriyasu (2004) and Davidovich and Dani (1998); for West Asia, Kolbas (2006); for China, Kuroda (2008). Throughout this volume, I use the term copper coin instead of bronze coin to avoid the confusion of referring to Chinese coinage as bronze prior to the mid-16th century and brass (as used in the jiaqing tongbao) after that point. 2 In the Netherlands, believed to be the most commercialized country in 17th-century Europe, the denomination of coin most commonly minted had been almost the same as the level of daily wages. Lucassen 2007, 261. 3 This is the diary of a Chinese official visiting southern Hebei, which suffered from famine in 1870. 4 Silver lumps were found in small linen bags as seen in late 19th-century China. 5 Coins were gradually replaced by grivnas and finally disappeared from Kiev hoards in the 12th–14th centuries period. It suggests that the usage of silver bars might have become popular earlier than late 13th century (Noonan 1987). Grivnas with Cyrillic characters were found from the Burge hoard, Gotland, buried in the 1140s. Lindberger 1998, 42, 49. 6 This chapter, in part, aims at offering corrections to Kuroda 2009, and supporting its main hypothesis with new facts, including data derived from archaeological excavations. In the previous paper, I failed to notice the importance of remittance from touxia estates and the unified system of measurement in postal relay stations. von Glahn (2010) has already argued that the Yuan establishment of silver as principal unit of account eventually led to increased use of silver in China in later periods. 7 The most detailed work on the touxia system is found in Sugiyama 1993. According to his analysis, based on evidence from stelae, the administration of these estates was almost entirely independent of the central Yuan state. 8 Muraoka 2002, 158, 160. In 1283, the office for issuing paper money was established there. Maeda 1973, 77. 9 Song 1976, 231. 10 The use of metal currencies was prohibited in the same way as under the Yuan. The compulsory use of paper money, called Chao, led to a collapse of trade in the Ilkhanate, leading the Ilkhanate to abandon this policy after only two months (Makhdumi 1988, 52–3). In the same year, the output of silver decreased. Martinez 1984, 165. 11 Halperin suggests that the Russians paid between 3,000 and 5,000 silver rubles a year. However, it is not certain if these numbers excluded payments in commodities like fur that were denominated in terms of silver. Halperin 1985, 77.

The ignition of monetary delocalization  109 12 Between 1276 and 1285, new appanages totalling 1,242 thousand households were established in the Lower Yangzi, including 1,048 thousand households established only in 1281. According the rate 0.5 liang in zhongtong chao per household, total tribute amounted to 310,000 liang in silver (more than 12 metric tons). Song 1976, 2411–44. 13 Song 1976, 2411–44. 14 Importantly, Temür Khan increased the rate of tribute per household from 500 wen to 2,000 wen in paper money (1 liang in silver) in 1294 and resumed the establishment of new touxia in the Lower Yangzi in 1298. Song 1976, 382. 15 The weight of grivna used in Kiev was 140–160 grams, and their shapes were hectagonal. The weight of grivna circulated in Novgorod was around 200 grams, and they had shapes of long bars. Zubko 1999, 44–5. 16 An Arab merchant from Spain who visited Kiev in the 12th century described that sheaves of squirrel fur were used as money. Zubko 1999, 43. 17 Gold became significant in Bruges, the financial centre of medieval Europe, only after 1340. Murray 2005, 295–6. 18 The bars are registered with Museum number 1906.1103.2502. 19 Tsuboi 1984 relied not on the bells themselves but on descriptions found in local gazetteers, making accurate comparisons between China and Japan impossible. 20 Mabuchi 1998, 16. On the composition of yuanfeng tongbao, see Zhou 2004, 71–3. 21 The causality appears to be in opposition to the common presumption that the development of markets was followed by an increase in the supply of money.Toyoda 1952; Segal 2011. 22 Mongol use of gold cloth in their imperial ceremonies accelerated the demand for gold from Japan. Allsen 1997, 37, 65, 97, 101. 23 Mitchiner 1979, 397–8. The weight is 30 grams, lighter than somo. 24 Merchants could use the postal relay system unless they impeded the flow of military orders. Allsen 1989, 97. 25 Vogel (2013) has persuasively confirmed that the observations of Polo largely reflect reality. 26 From 1299, on, the German Hanse also had a special relationship with English monarch. For example, in 1327, Hanse loans helped Edward III to the throne and, in return, the king granted Hanse wool-exporting privileges, tax cuts and rights to collect customs dues from other merchants. Dollinger 1970, 56–8. 27 Pegolotti showed that merchants from Genoa could get 19 to 20 Genoese pounds of Chinese silk in exchange for one silver somo. Lopez and Raymond 1955, 358. 28 In 1279, an embassy with gifts was sent to the Chinese emperor. Shamrookh 1996, 256–7, 321, 329–31. 29 Mikami observed several hundred pieces of high quality blue-and-white porcelain in the Foustat remains. Mikami 1969, 25. 30 Mikami 1969; Mori 2012. On Chinese porcelain of the 13th and 14th centuries found around Aden, see Margariti 2007, 65, 137. 31 As dense distribution of marketplaces in West Midland shows, local markets in England appeared to be the most proliferated in the early 14th century (Hilton 1983, 172–3). The building of stone walls in medieval towns also had a boom in the late 13th and early 14th centuries (Dobson 1990, 271). However, the issuance of license to local markets in England had a peak in earlier period, the third quarter of 13th century. Britnell 1981, 210. 32 Two hundred fifty years between the Doomsday Book and the Black Death was also a period of great agricultural improvement and expansion. The number of agricultural holdings more than doubled. Postan 1973, 340. 33 The hoard is preserved at the National History Museum of Moldavia, Inventory number 24875 (1–65). 34 The weights of silver ingots made under the Yuan, and marked as weighing 50 liang, show substantial variation. This is in striking contrast to four ingots from the same

110  Global history of monetary delocalization county, Jurong, cast in the same year 1277, which differed in weight by only three grams. Vogel 2013, 482–4. 35 This can be seen in account books by a merchant in Montauban from between 1339 and 1369. Lopez and Raymond 1955, 138. 36 Ming shilu. vol. 197, 8a. 37 ‘Zhanpuzhi’ in Gu 1975, vol. 26, 120a–21a. 38 Cao 1987, vol. 13, 10a–b; von Glahn 1996, 103.

References Abe, T. 1972. Gendai shi no kenkyu. Tokyo: Sobunsha. Abe,Y. 1965. Kakeiri nominshi. Tokyo: Shibundo. Abu-Lughod, JL. 1989. Before European Hegemony:The World System A.D. 1250–1350. Oxford: Oxford University Press. Allen, M. 2001. ‘The volume of the English currency, 1158–1470,’ Economic History Review 54(4): 595–611. https://doi.org/10.1111/1468-0289.00204 Allsen, TT. 1987. Mongol Imperialism: The Policies of the Grand Qan Mongke in China, Russia, and the Islamic Lands, 1251–1259. Berkeley: University of California Press. Allsen, TT. 1989. ‘Mongolian princes and their merchant partners, 1200–1260,’ Asia Major 2: 83–126. Allsen, TT. 1997. Commodity and Exchange in the Mongol Empire: A Cultural History of Islamic Textiles. Cambridge: Cambridge University Press. Allsen, TT. 2001. Culture and Conquest of Mongol Eurasia. Cambridge: Cambridge University Press. al-Shamrookh, NA. 1996. The Commerce and Trade of the Rasulids in theYemen, 630–858/1231– 1454. Kuwait: State of Kuwait. Bacharach, JL. 1983. ‘Monetary Movements in Medieval Egypt, 1171–1517,’ in Richards, JF. (ed.) Precious Metals in the Later Medieval and Early Modern World, 159–81. Durham: Carolina Academic Press. Boldureanu, A. 2007. ‘Some remarks on the metrology of the Golden Horde silver ingots in the light of a hoard found at the Orheiul Vechi,’ www.arheomet.ro/ro/evenimente/ simpozion2007/rezumate/boldureanu.html, accessed on 21 April 2015. Britnell, RH. 1981. ‘The proliferation of markets in England, 1200–1349,’ Economic History Review, second series 34(2): 209–21. https://doi.org/10.1111/j.1468-0289.1981. tb02063.x Cao,YL. 1987. Wujiang xianxhi. Taipei: Xuesheng shuju. Chakravarti, R. 1991. ‘Horse trade and piracy at Tana (Thana, Maharashtra India),’ Journal of the Economic and Social History of the Orient 34(2): 159–82. doi:10.2307/3632243 Chen, JH. 1985. Dayue shiji quanshu.Tokyo: University of Tokyo, Institute of Oriental Culture. Chen,YC., Guo,YY. and Chen, H. 1994. ‘Sources of cobalt pigment used on Yuan blue and white porcelain wares, Shanghai institute of ceramics, Chinese academy of sciences,’ Oriental Art 40(1): 14–19. Davidovich, EA. and Dani, AH. 1998.‘Coinage and the monetary system,’ in Asimov, MS. and Bosworth, CE. (eds.) History of Civilizations of Central Asia, vol. 4, part 1, 391–419. Paris: UNESCO Publishing. Deng, F. 1965. Changshu xianzhi. Taipei: Xuesheng shuju. Depeyrot, G. 2018. ‘Monetary system of the “ancient régime” (3rd to 18th centuries),’ in Battilossi, S., Cassis, Y. and Yago, K. (eds.) Handbook of the History of Money and Currency, 1–52. Singapore: Springer.

The ignition of monetary delocalization  111 Deyell, J. 1983. ‘The China connection: Problems of silver supply in medieval Bengal,’ in Richards, JF. (ed.) Precious Metals in the Later Medieval and Early Modern World, 207–27. Durham: Carolina Academic Press. Di Cosmo, N. 2005. ‘Mongols and merchants on the Black Sea frontier in the thirteenth and fourteenth centuries: Convergences and conflicts,’ in Amitai, R. and Biran, M. (eds.) Mongols,Turks, and Others: Eurasian Nomads and the Sedentary World, 391–424. Leiden: Brill. Digby, S. 1971. War-Horse and Elephant in the Delhi Sultanate: A Study of Military Supplies. Oxford: Orient Monographs. Dijkman, J. 2011. Shaping Medieval Markets:The Organisation of Commodity Markets in Holland, c. 1200–c. 1450. Leiden: Brill. Dobson, RB. 1990. ‘Urban decline in late medieval England,’ in Holt, R. and Rosser, G. (eds.) The Medieval Town: A Reader in English Urban History 1200–1540, 265–8. London: Longman. Dollinger, P. 1970. The German Hansa. Ault, DS. and Steinberg, SH. (trans. and ed.). London: Macmillan. Fang, H. 1971. Gujinkao. Taipei: Xuesheng shuju. Fang, LG. 2001. Tongzhi tiaoge xiaozhu. Beijing: Zhonghua shuju. Feavearyear, AE. 1931. The Pound Sterling: A History of English Money. Oxford: Clarendon Press. Fennel, J. 1983. The Crisis of Medieval Russia, 1200–1304. London: Longman. Finlay, R. 2010. The Pilgrim Art: Cultures of Porcelain in World History. Berkeley and Los Angeles, CA: University of California Press. Flecker, M. 2002. The Archaeological Excavation of the 10th Century Intan Shipwreck. Oxford: Archaeopress. Friedel, E. 1896. Hervorragende kunst- und alterthums-gegenstände des Märkischen provinzial-­ museums in Berlin. heft 1: die hacksilberfunde. Berlin:Verlag von Dr. E. Mertens & Cie. Funada, Y. 2001. ‘Gendai shiryo toshite no kyuhon Rokitsudai: sho to bukka no kisai wo chushin to shite,’ Toyo gakuho 83: 101–30. Gao, CM. 2016. ‘Zizhou jiaohui kao,’ Zhongguo qianbi (1): 9–12. Gibb, HAR. 1962. The Travels of Ibn Battuta A.D. 1325–1354. Cambridge: The Hakluyt Society. Grierson, P. 1975. Numismatics. London and New York: Oxford University Press. Grierson, P. 1979. ‘The coin list of Pegolotti,’ in Grierson, P. (collect.) Later Medieval Numismatics (11th–16th Centuries), vol. XI, 485–92. London:Variorum. Grierson, P. 1982. Byzantine Coin. Berkeley: University of California Press. Gu,YW. 1975. Tainxia qunguo libing shu. Taipei: Zhongwen chuban. Haider, N. 2007. ‘The network of monetary exchange in the Indian Ocean trade, 1200– 1700,’ in Ray, HP. and Alpers, E. (eds.) Narratives of the Sea: Encapsulating the Indian Ocean World, 181–205. Oxford: Oxford University Press. Halperin, CJ. 1985. Russia and the Golden Horde. Bloomington: Indiana University Press. Hicks, JR. 1969. A Theory of Economic History. Oxford: Clarendon Press. Hilton, R. 1983. A Medieval Society: The West Midlands at the End of the Thirteenth Century. Cambridge: Cambridge University Press. Hirao,Y. 2008a. ‘Kamakura daibutsu no sozai wa chugokusen,’ Isotope News 656: 22–7. Hirao, Y. 2008b. ‘Zairyo ga kataru chusei: namari doitaihi sokutei kara mita kyozutsu,’ in Oda, F., Hirao, Y. and Iinuma, K. (eds.) Kyozutsu ga kataru chusei no sekai, 21–34. Kyoto: Shibunkaku. Hirao,Y., Iinuma, K. and Murai, S. (eds.). 2014. Daikokai jidai no Nihon to kinzoku koeki. Kyoto: Shibunnkaku.

112  Global history of monetary delocalization Honda, H. 2007. ‘Copper coinage, ruling power and local society in medieval Japan,’ International Journal of Asian Studies 4(2): 225–40. https://doi.org/10.1017/S1479591407000745 Ichimaru, T. 2008. ‘Sodai ni okeru gin, sho, dosen no sogo kankei ni tsuite: shiyo tan’i no bunseki wo chushin ni,’ Kyushu daigaku toyoshi ronshu 36: 88–122. Jackson, P. 2009. ‘Delhi: The problem of a vast military encampment,’ in Jackson, P. (collect.) Studies on the Mongol Empire and Early Muslim India, vol. XII, 18–33. London:Variorum. Jeong, RJ. 1972. Goryeosa, vol. 2, 2nd ed. Seoul:Yonsei University. Kolbas, J. 2006. The Mongols in Iran: Chingiz Khan to Uljaytu, 1220–1309. London and New York: Routledge. Kong, Q. 1987. Zhizhengzhiji. Shanghai: Shanghai guji. Köprülü, MF. 1992. The Origins of the Ottoman Empire. Albany, GL. (trans. and ed.). New York: State University of New York Press. Körner, M. 1994. ‘Public credit,’ in Bonney, R. (ed.) Economic Systems and State Finance, 507– 38. Oxford: Clarendon. Kuroda, A. 2000. ‘Another monetary economy: The case of traditional China,’ in Latham, AJH. and Kawakatsu, H. (eds.) Asia Pacific Dynamism 1550–2000, 187–98. Routledge: London. Kuroda, A. 2003. Kahei shisutemu no sekaishi. Tokyo: Iwanami. Kuroda, A. 2005. ‘Copper-coins chosen and silver differentiated: Another aspect of “silver century” in East Asia,’ Acta Asiatica 88: 65–86. Kuroda, A. 2008. ‘Locating Chinese monetary history in global and theoretical contexts: From multiple and complementary viewpoints,’ in Huang, KZ. (ed.) Jidiao yu bianzou: 7 zhi 20 shiji de Zhongguo, vol. 2, 33–50.Taipei: Chengchi University, Department of History. Kuroda, A. 2009. ‘The Eurasian silver century, 1276–1359: Commensurability and multiplicity,’ Journal of Global History 4(2): 245–69. https://doi.org/10.1017/S1740022809003143 Kuroda, A. 2017. ‘Why and how did silver dominate across Eurasia late-13th through mid14th century? Historical backgrounds of the silver bars unearthed from Orheiul Vechi,’ Tyragetia, Archaeology 11(1): 23–34. Kuwabara, J. 1989. Hojuko no jiseki. Tokyo: Heibonsha. Li, XR. 1984. Li Xingrui riji. Beijing: Zhonghua shuju. Li, ZA. 2007. Yuandai fenfeng zhidu yanjiu. Beijing: Zhonghua shuju. Lindberger, E. 1998. ‘Almost like a fairy-tale,’ in Quadra, 42–9. Stockholm: Treasury in the Royal Coin Cabinet. Lopez, RS., Miskimin, HA. and Udovitch, A. 1970. ‘England to Egypt, 1350–1500: Longterm trends and long-distance trade,’ in Cook, MA. (ed.) Studies in the Economic History of the Middle East, from the Rise of Islam to the Present Day, 93–128. Oxford: Oxford University Press. Lopez, RS. and Raymond, IW. 1955. Medieval Trade in the Mediterranean World: Illustrative Documents. New York: Columbia University Press. Lucassen, J. 2007. ‘Wage payment and currency circulation in the Netherlands from 1200 to 2000,’ in Lucassen, J. (ed.) Wages and Currency: Global Comparisons from Antiquity to the Twentieth Century, 221–61. Bern: Peter Lang. Mabuchi, K. 1998. Kamakura daibutsu no chusei. Tokyo: Shin jinbutsu oraisha. Maeda, N. 1973. Gencho shi no kenkyu. Tokyo: Tokyo daigaku shuppankai. Makhdumi, R. 1988. ‘Mongol monetary system,’ in Sharma, RC. (ed.) Perspective on Mongolia, 49–54. New Delhi: Seema. Margariti, RE. 2007. Aden and the Indian Ocean Trade. Chapel Hill: University of North Carolina Press.

The ignition of monetary delocalization  113 Martin, J. 1978. ‘The land of darkness and the Golden Horde: The fur trade under the Mongols XIII–XIVth centuries,’ Cahiers du monde russe et soviétique 19(4): 401–21. Martinez, AP. 1984. ‘Regional mint outputs and the dynamics of bullion flows through the Il-Xanate,’ Journal of Turkish Studies 8: 121–73. Martinez, AP. 1995–97. ‘The wealth of Ormus and of Ind:The Levant trade in bullion, intergovernmental arbitrage, and currency manipulations in the Il-Xanate,’ Archivum eurasiae medii aevi 9: 123–252. Martinez, AP. 2011. ‘The Il-Khanid coinage: An essay in monetary and general history based largely on comparative numismatic metrology,’ Archivum eurasiae medii aevi 17: 59–164. Matsuda, K. 1983. ‘Yubukuru to no Gencho toko,’Ritsumeikan shigaku 4: 28–62. Matsui, D. 2004. ‘Mongoru jidai no doryoko: higashi torukisutan shutsudo bunken kara no sai-kento,’ Tohogaku 107: 153–66. Matsui, D. 2005. ‘Taxation systems as seen in the Uigur and Mongol documents from Turfan: An overview,’ Transactions of the International Conference of Eastern Studies 50: 67–82. Mayhew, N. 2007. ‘Wage and currency:The case in Britain up to c. 1600,’ in Lucassen, J. (ed.) Wages and Currency: Global Comparisons from Antiquity to the Twentieth Century, 211–20. Bern: Peter Lang. Meng,YL. 1936. Donging menghua lu. Shanghai: Shangwu yinshuguan. Michell, R. and Nevill, F. 1914. The Chronicle of Novgorod, 1016–1471. London: Offices of the Society. Mikami, T. 1969. Toji no michi. Tokyo: Iwanami. Miskimin, HA. 1963. Money, Prices, and Foreign Exchange in Fourteenth-Century France. New Haven:Yale University Press. Miskimin, HA. 1983. ‘Money and money movements in France and England at the end of the Middle Ages,’ in Richards, JF. (ed.) Precious Metals in the Later Medieval and Early Modern World, 79–96. Durham: Carolina Academic Press. Mitchiner, M. 1979. Oriental Coins and Their Values. London: Hawkins. Miyazawa, T. 2012. ‘Gencho no zaisei to sho,’ Bukkyo daigaku rekishigaku ronshu 2: 43–64. Momose, H. 1943. Shincho no iminzoku tochi ni okeru zaisei keizai seisaku.Tokyo:Toa Kenkyujo. Momose, K. 1957. ‘Muromachi jidai ni okeru beika hyo: Toji kankei no baai,’ Shigaku zasshi 66(1): 58–70. Mori, T. 2012. ‘Perusyawan hokugan hakken no chugoku toji,’Aichiken toji shiryokan kenkyu kiyo 17: 1–18. Moriyasu, T. 2004. ‘Sirukurodo tobu ni okeru tsuka,’ in Moriyasu, T. (ed.) Chuo Ajia shutsudo bunbutsu ronso, 1–40. Kyoto: Hoyu. Moshenskyi, S. 2008. History of Weksel: Bill of Exchange and Promissory Note. New York: Xlibris. Munro, JH. 1966. ‘Bruges and the abortive staple in English cloth: An incident in the shift of commerce from Bruges to Antwerp in the late fifteenth century,’ Revue belge de philologie et d’histoire 44(4): 1137–59. Munro, JH. 1983. ‘Bullion flows and monetary contraction in late-medieval England and the Low Countries,’ in Richards, JF. (ed.) Precious Metals in the Later Medieval and Early Modern World, 97–158. Durham: Carolina Academic Press. Muraoka, H. 2002. ‘Mongoru jidai no uyoku urusu to Sansei chiho,’ in Matsuda, K. (ed.) Hikoku to shiryo no sogo bunseki ni yoru Mongoru jidai, Gencho no seiji keizai sisutemu no kisoteki kenkyu, 151–70. Tokyo: Kagaku kenkyuhi hojokin kenkyu seika hokokusyo, no 12410096. Murray, JM. 2005. Bruges, Cradle of Capitalism, 1280–1390. Cambridge: Cambridge University Press.

114  Global history of monetary delocalization National Maritime Museum of Korea. 2006. The Shin’an Wreck. Mokpo: National Maritime Museum of Korea. Noonan, TS. 1987. ‘The monetary history of Kiev in the pre-Mongol period,’ Harvard Ukrainian Studies 11(3–4): 384–443. Ogilvie, SC. 2011. Institutions and European Trade: Merchant Guilds, 1000–1800. Cambridge: Cambridge University Press. Peng, XW. 1965. Zhongguo huobi shi, 2nd ed. Shanghai: Shanghai renmin. Poos, LR. 1991. A Rural Society After the Black Death: Essex, 1350–1525. Cambridge: Cambridge University Press. Postan, M. 1973. Medieval Trade and Finance. Cambridge: Cambridge University Press Postan, M. 1987.‘The trade of medieval Europe:The north,’ in Postan, M. and Miller, E. (eds.) Cambridge Economic History of Europe, vol. 2, Cambridge: Cambridge University Press. https://doi.org/10.1017/CHOL9780521087094.006 Reid, A. 1993. Southeast Asia in the Age of Commerce 1450–1680, vol. 2. New Haven: Yale University Press. Sakuraki, S. 2009. Kahei kokogaku josetsu. Tokyo: Keio daigaku shuppankai. Sargent, TJ. and Velde, FR. 2002. The Big Problem of Small Change. Princeton: Princeton University Press. Segal, EI. 2011. Coins,Trade and the State: Economic Growth in Early Medieval Japan, Cambridge, MA: Harvard University Asia Center. Sheng, GX. 1995. ‘Dachao tongbao yinqian kaolü,’ Zhongguo qianbi (3): 18–22. Shimatani, K. 1994. ‘Chusei no mochusen seisan: Sakai shutsudo no zeniigata wo chushin ni,’ Kokogaku Journal 372: 26–34. Song, L. 1976. Yuan shi. Beijing: Zhonghua shuju. Song,YX. 1978. Tiangong kaiwu. Hong Kong: Zhonghua shuju, Xiang Gang fenju. Spufford, P. 1970. Monetary Problems and Policies in the Burgundian Netherlands, 1433–1496. Leiden: Brill. Spufford, P. 1988. Money and Its Use in Medieval Europe. Cambridge: Cambridge University Press. Stahl, A. 1985. The Venetian Tornesello: A Medieval Colonial Coinage. New York: American Numismatic Society. Stahl, A. 2000. Zecca:The Mint of Venice in the Middle Ages. Baltimore: Johns Hopkins University Press. Sugiyama, M. 1993. ‘Babusa daio no reishi hai yori,’ Toyoshi kenkyu 52(3): 435–84. Sun, JL. 2015. ‘Zhongguo caishui bowuguan cang yuandai yangzhou yuanbao kao,’ Zhongguo qianbi (6): 12–19. Suzuki, K. 1999. Shutsudo senka no kenkyo. Tokyo: Tokyo daigaku shuppankai. Tao, ZY. 1959. Nancun chuogeng lu. Beijing: Zhonghua shuju. Toyoda, T. 1952. Chusei Nihon shogyo shi no kenkyu. Tokyo: Iwanami. Tsuboi, R. 1970. Nihon no Bonsho. Tokyo: Kadokawa. Tsuboi, R. 1974. Chosen sho. Tokyo: Kadokawa. Tsuboi, R. 1984. Rekishi kokogaku no kenkyo. Tokyo: Bijinesu kyoiku shuppansha. Usher, AP. 1943. The Early History of Deposit Banking in Mediterranean Europe. Cambridge, MA: Harvard University Press. van der Wee, H. 1977. ‘Monetary, credit and banking systems,’ in Rich, EE. and Wilson, CH. (eds.) Cambridge Economic History of Europe, vol. 5, 290–392. Cambridge: Cambridge University Press. https://doi.org/10.1017/CHOL9780521087100.006 Vernadsky, G. 1953. The Mongols and Russia. New Haven:Yale University Press. Vilar, P. 1976. A History of Gold and Money, 1450–1920. White, J. (trans.). London: NLB.

The ignition of monetary delocalization  115 Vogel, HU. 2013. Marco Polo Was in China: New Evidence from Currencies, Salts and Revenues. Leiden: Brill. Von Glahn, R. 1996. Fountain of Fortune: Money and Monetary Policy in China, 1000–1700. Berkeley, CA: University of California Press. Von Glahn, R. 2010. ‘Monies of account and monetary transition in China, twelfth to fourteenth centuries,’ Journal of the Economic and Social History of the Orient 53(3): 463–505. Wang, HL. and Zhong, CW. 2007. ‘Dui Shufu xian faxian Chahetai hanguo jiaocang yinbi de yanjiu,’ Neimenggu jinrong yanjiu qianbi zengkan 1: 16–23. Wang, LL. 2005. Sondai kuangye yanjiu. Baoding: Hebei daxue chubanshe. Wang, WC. 2001. Songdai baiyin huobihua yanjiu. Kunming:Yunnan daxue chubanshe. Watson, AM. 1967. ‘Back to gold—and silver,’ Economic History Review, second series 20(2): 1–34. doi:10.2307/2592033 Whaley, MA. 2001. ‘An account of 13th century Qubchir of the Mongol “great courts”,’ Acta orientalia academiae scientiarum hungaricae 54(1): 1–84. Yule, H. 1914. Cathay and the Way Thither, vol. 3. London: Hakluyt Society. Zhou, WR. 2004. Zhongguo gudai qianbi hejin chengfen yanjiu. Beijing: Zhonghua shuju. Zubko, A. 1999. ‘The Mystery of the Kyiv hryvnia emergence,’Ukrainska numizmatyka i bonistyka 99(2): 42–52.

4 The world diversified and stratified Three paths after the global silver march, c.1600

1 A breakthrough with large silver coins This chapter will demonstrate that, unlike the idea of a world system with the hierarchy between core and periphery as envisaged by I. Wallerstein, the global silver march after the late 16th century brought about a mutual interdependence among different types of monetary economies (Wallerstein 1974). Unlike A. Frank, who focusses only on the flow of precious metals, the current analysis reveals the diversified responses at the ground level by paying attention to basematerial currencies, local paper money and local credit (Frank 1998). Through the Eurasian silver century, from the late 13th century to the mid14th century, no particular monetary item representing units of silver circulated across the whole of the Eurasian continent. Silver bars of around 200 grams, somo, played a significant role in interregional settlements in the western part of Eurasia, but actual silver items circulated to a far lesser extent in eastern Eurasia. More importantly, a monetary system depending on silver by weight could not establish vertical conduits between interregional settlements in the upper-level market and local exchanges in lower-level markets. These features were quite different from the next silver century, the late 16th century through the mid17th century, which we are now going to examine in this chapter. It was the contraction of currency supplies rather than their expansion that forced societies to make profound transformations in the ways they conducted exchanges. As we saw in the previous chapter, throughout the 15th century both ends of the Eurasian landmass responded to silver shortages in quite different ways. In East Asia currency circuits using a variety of copper coinages flourished locally, while the Chinese government introduced silver as the unit of taxation in 1436 after the collapse of official paper money. In Western Europe, at the local level, the proportion of transactions concluded with credit increased; meanwhile, at the interregional level, payment fairs grew in importance. But at the same time, states required bullion for military expenditure, and this often hampered interregional settlements through credit instruments. Compared with transporting old copper coins of fine quality but great bulk over long distances, silver had a potential advantage in facilitating interregional trade across the China Sea. In 1540, at the latest, silver from the Iwami mine

The world diversified and stratified  117

in Japan began to provide maritime traders with silver to exchange for silk and other Chinese products (Wakita 2002). This profitable business stimulated unlicensed trade across the China Sea and attracted Portuguese merchants, who laid the ground for the global silver flow in cooperation with Spanish connections across the Pacific three decades later when the silver production at Potosí increased. The Ming welcomed the inflow of silver which the dynasty was using for the reform of her taxation system, even at the same time that they were fighting hard to suppress clandestine maritime trade (Kuroda 2005b). In the West, Spanish law in 1497 first established a coinage of silver pieces of eight reals. This was the origin of the silver dollar. As German silver mines increased production, the silver Thaler (or dollar) began to substitute for the gold florin in early 16th-century Central Europe but was not yet the dominant currency. At the same time, since the accession of Charles V, merchant bankers had sold to the Habsburg rulers bills of exchange issued in Medina del Campo or Madrid and payable outside Spain. Repayment of these advances was secured, increasingly, on claims against future silver expected from the Indies as well as on Spanish public income (van der Wee 1977, 372). European monarchs’ heavy dependence on floating debts from merchants accelerated towards the middle of the 16th century. Ceaseless military conflicts pushed official borrowings to unprecedented heights. The size of these debts finally became large enough that governments declared bankruptcy. In 1557, Phillip II issued his first bankruptcy decree. France followed suit in 1558 and Portugal in 1560 (van der Wee 1977, 370–1). South German financiers, such as the Fuggers, could not survive these consecutive state bankruptcies. However, the mutual dependence between mercantile states and merchant bankers continued through repeated debt moratoria. Simultaneously, heavily debt-dependent European states fatally affected the interregional settlement system. From the second third of the 16th century onwards, the French kings increasingly subjected the Lyons money and exchange market to inexhaustible demands for credit. The wars of religion finalized the decline of the Lyons payment fair. The quarterly periods of payment became extremely irregular after 1562 (van der Wee 1977, 319).This payment fair diminished rapidly in importance after 1575 (Usher 1943, 134). This was when a large quantity of silver began to flow in from Potosí, which finally resolved the interregional settlements crisis. From 1557–60 the Genoese had dominated the asiento contracts in Spain (van der Wee 1977, 372) and substituted themselves as the main lenders for the Germans. From 1579, when the Genoese had gained indisputable control over the redistribution of silver from Latin America throughout Europe, these fairs were held at Piacenza, not at Lyons (van der Wee 1977, 320).Then, in European long-distance trade, settlements in silver dollars became more dominant than those by exchange bills. The rise of Dutch merchants increased the importance of silver dollars still more. Cashkeepers and money-changers had been given official permission in 1595 to sell good specie in Amsterdam at a premium of not more than

118  Global history of monetary delocalization

1.25 percent, but the premium regularly exceeded this maximum (van der Wee 1977, 337). They provided credit instruments such as assignments for domestic transactions and saved specie for long-distance settlements. The premium on silver dollars showed the strength of demand for interregional payments, especially to Asia which absorbed a large quantity of silver in exchange for products such as silk, porcelain and cotton, among others. Thus, besides the diversity in silver demand among continentals both east and west which was formed through a long-term process of transformation as suggested in the previous chapter, there was a combination of three conditions over the shorter term – the demand for silver in Asia especially, in exchange for Asian products; the deterioration of the interregional settlement system in Europe; and the rapid increase of silver production in the Spanish colonies – which happened to generate a global stream of silver dollars from the late 16th century, from Latin America to Asia via Europe, or sometimes directly across the Pacific. Silver dollars, which were nearly 30 grams in weight, began to do what gold florins or silver somo could not. Silver dollars worked to mediate long-distance exchanges most effectively and over the widest range ever. An abundant supply of standardized large-denomination silver coins caused transactions in Quadrant I to intensify. Importantly, global circulation of silver dollars encouraged locals to increase tangible means of local exchange as well. Hereafter, the combination of standard large silver for interregional settlement (Quadrant I) and small currencies for local exchanges (Quadrant IV) continued to survive. The last survivor of this was the Maria Theresa Dollar in the early 20th century Red Sea region, which we will see in Chapter Five. However, silvers could not always work in the same form of coined dollars across the world. Ironically, it was a system depending on silver by weight, the protagonist of the first silver century, that became more deeply rooted in the east part of Eurasia after the 16th century. Silver dollars did not work as currency in East Asia until the turn of the 19th century, and instead were melted down into ingots with which the system based on silver by weight began to flourish. In this sense, the monetary system of silver by weight, silver liang, originating in the first silver century, continued to coexist with the system of silver by count, the silver dollar, created in the second silver century. The coexistence of remnants from both silver centuries lasted in China until as late as the 1930s, as we will see in Chapter Five. Owing to trade surpluses, India as well as China absorbed large quantities of silver from the late 16th century through the 18th century. However, silver dollars rarely circulated under the Mughal Empire either, since precious metals were required to re-mint into silver rupees (Prakash 2004). Though silver rupees were used by count, unlike by weight as in China, India also kept her original way of using silver in spite of a huge inflow of silver dollars. Unlike the contemporary Chinese, Japanese did not weigh silver but counted it in terms of mai units within the archipelago through the 16th century. When the Tokugawa government established gold coinage and its usage by count, by

The world diversified and stratified  119

contrast, the usage of silver by weight, monme (one tenth of a Chinese liang) became common. Japan was able to establish commensurability with Chinese silver usage through the weighing system. In other words, the country of the second largest silver production followed the practice of its neighbouring country, which had the largest silver usage in the world. However, it also kept its independence from the continental system through gold coinage which mainly circulated in the eastern part of the archipelago (Yasukuni 2016, 31–3). We should also note that silver dollars never became popular in Latin America where silver was mined until the 18th century. People used cacao beans as fractional money for local transactions (Young 1925, 12; Giraldez 2012). As the Spanish authority was reluctant to establish mints in colonies, silver of irregular shapes and bearing official stamps became common. These peculiar coins circulated as money and were known as macacas. Some have been unearthed near Quanzhou, China, on the opposite side of the Pacific Ocean. They continued to circulate even after silver pesos were minted in the colonies. It was not until 1873 that their circulation was prohibited in Guatemala (Young 1925, 14). The absence of the formal silver dollar in the region of the largest silver production in the world was closely bound up with the situation of the political economy in the Spanish colonies.1 We will see later that commercial oligarchies disconnected distant exchange and proximate exchanges there. Unlike the first silver century, the second silver century saw major silver items (silver dollars) circulate intercontinentally, and, throughout the period, silver items, regardless of whether they were coined or uncoined, disseminated sufficiently that they affected proximate exchanges as well as facilitating distant exchanges. Again, the contraction of silver circulation transformed societies to a larger degree than its expansion. Responding to liquidity shortages after the mid-17th century across the world, the discrepancy between currencyoriented and credit-oriented societies became more marked, and the organization between local and interregional markets varied more significantly country by country.

2 A silver century followed by a copper century: prosperity in a currency-oriented economy Silver extracted from Potosí flooded over the globe from the late 16th century to the mid-17th century. Some thinkers have identified this influx of precious metals as the cradle of the modern capitalist world, although this is debatable. For example, inspired by Hamilton’s findings and the idea of price revolution, Keynes thought that commodity prices rising faster than wages brought entrepreneurs unprecedented opportunities to accumulate assets. He called this the profit revolution (Keynes 1971, 135–45).2 Before arguing the causality between the flow of precious metals and the rise of the capitalistic economy, another issue remains unsolved. Examining region by region, there is no clear connection between the increase in the minting of precious metals and rises in commodity prices. It is clear that some Western

120  Global history of monetary delocalization

European countries experienced serious hikes in prices from the late 16th century to the mid-17th century, but other countries such as England did not increase the issuance of coinages proportionately. Meanwhile, China and India must have absorbed large quantities of silver, but they showed no clear rises in prices as occurred in Western Europe over the same period (Brennig 1983). In other words, it is not yet established that the increasing precious metal flow from the late 16th century brought about the price revolution. Unlike the idea of the quantity theory of money in which the size of the money supply and the price level must move together proportionately, the concept of the four quadrants of exchange provides us with a different view. The prices of daily necessities must reflect their demand/supply in proximate exchange, in Quadrants III and IV. However, the rapid increase of precious metals circulating in Quadrant I, in the late 16th century through the mid-17th century, must have enhanced distant exchanges, but it did not always directly affect proximate exchanges. The expansion of Quadrant I, to some extent, might have intruded into the sphere of Quadrant II. Until the mid-16th century the networks of exchange bills centred at Lyon supported long-distance settlements across Europe. However, the increasing flow of precious metals between major ports across Europe, including Seville, Piacenza, Antwerp and London, outweighed the development of credit settlements in the period (Boyer-Xambeu et al. 1994, 31–3). However, as far as proximate exchanges were concerned, credit still played the most significant role in expanding commercial activities in some countries. We will argue this point later. The abundant inflow of precious metals must have accelerated interregional trade. Since interregional trade is inevitably stimulated by differences in prices between regions, continuous development of interregional trade would have decreased the gap in prices. As far as global impact is concerned, prices in terms of silver are more suitable than those in gold if we wish to examine if metal flows affected the gap in prices between regions. Gold was used as money in more limited regions than silver until the 18th century. Especially, it is important that silk was the most profitable commodity in distant trade during the period and it was mostly supplied by China which exchanged silk for silver.3 Roughly speaking, in the beginning of the 17th century the price of Chinese silk in terms of silver in Europe was nearly three times as high as in East Asia. Thus, any trader could make large profits by bringing silver to coastal Southern China in exchange for silk and returning it back to Europe. The export of high quality silk from China destroyed the inferior silk industry in the regions along the silver flow route, such as Bursa in Turkey (Barkan 1975). This arbitrage trade flourished until the mid-17th century, when the gap in Chinese silk prices in terms of silver had shrunk.4 As shown in Table 4.1, the price of raw silk in Valencia declined after 1636. Sino-Spanish trade through Manila seriously decreased as late as 1635 (Atwell 2005, 475). Thus, from the viewpoint of profitability, it was not the fall in the production of silver but the decrease of demand for silver that brought an end to the global silver march.

The world diversified and stratified  121

Japan was the second largest supplier of silver in this period, next only to Peru.We may estimate that the exchange of Japanese silver for Chinese silk was so large that it was equivalent to ten percent of Japanese rice production.5 As we have seen, silk served as a currency in China as late as in the 13th century. The commodity had a high acceptability beyond its own utility as a fabric even outside China.When merchants happened to get excess silver, they had no hesitation in investing it in silk. The price movements of raw silk and linen in Valencia show the different influence of the silver flow between a fabric interregionally traded (silk) and another sold locally (linen). The price of linen rose steadily between 1551 and 1650. Meanwhile, the price of raw silk jumped at the end of the 16th century and gradually fell towards the middle of the 17th century (Table 4.1). As early as 1594 Chinese raw silk was imported to Spain by the New Spain fleet

Table 4.1 Prices in Valencia (dinar) silk raw/lb 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560




352 414 348 340 350 348 354 480 547.5 390 414 420 384 480 534

480 540 552

linen/alna 24 30 34.8 28.5 33.8 36 21 34 23 44 30 29.5 44.7 39 42 23.1 50


wage line weaver/alna

wheat (in Cahiz)

6.5 7 7 7 7 7 7.3 7 7.3 7.3 8.5 8 8.2 8.4 8.8 8.9 9 9.2 9.6 9.8 8.8 9.3 9 9 8.6 9.6 9.8

780 864.5 783.8 636.3 729 756 1333.3 1027.5 910.9 845.6 875.5 1016.3 951.5 853.5 770.9 1032.8 1047 950.3 968.6 1116 1471.5 1366.5 1090.9 1173 1179 1507.1 1429.3 1773.8 1729 (Continued)

Table 4.1 (Continued) silk raw/lb 1580











wage line weaver/alna

wheat (in Cahiz)

588 345 564




834 663

66.5 46.2

1579.9 1363.3 1715.1 1974.3 2154.5 1538 1359.8


65 54.2 69 56 52 54.8 54

10.2 10 10 10 11.5 12 11.5 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12.5 12.3 12 12 12 12 11.7 12 12 12 12.5 12.3 11.9 12 12.3 12 12 12 12 12 12 12 11.7 12 12 12 12 12 12

612 648 460 780 887 822 1171 1020 989 1000 943 782 736 759 644 552 862.5 943 931.8 828 862.5 831.8 782 575 618 744 870 840 894 912 966 816 672 816 816 720 684

54 60 69 60 58 72 42 50 53 53 64 60 48 63.7 48 45.9 45 48 53.3 57 50 48 56 59 48

1523.3 1620 1594 1621.5 1822.3 1957.5 1983.1 1486 1488.7 1800 1895 1684.1 1828.6 1347.2 1920 1462.5 1820.3 1512 2043.5 1738.6 1698.6 1581 1479 1560 1621.5 1900 1530 1500 1878 1755 1656 1272 1371.8 1756.5 1993.5 1521 1592 1714.5 1522 1896 2160

The world diversified and stratified  123 silk raw/lb 1630





708 672 828 840 852 744 864 696 672 576 480 576 417 636 720 696 600 405 588


wage line weaver/alna

wheat (in Cahiz)

66 55.3 55 62 62 57 78 66

12.3 12.9 13 13 13 13 12.7 12 13 13 13 14 13.7 13.4 13 12 12 12 12 12 12

2240 2700 2271 1964.5 1828.5 1935.5 2473.5 2522 2341.3 1796 1791 2087 1761.8 1524 2414 2310 2385 2826 2268 2159.5 2143.5

69 59.5 54 54 61 62 68 54 50 64.5

alna is equivalent to 91 cm From 1501 to 1602 the diner was equivalent to 0.1389 grams of pure silver Source: Hamilton 1934, 354–7, 384–9.

(Hamilton 1934, 37). It is clear that the period of significant silver imports to Seville overlaps with the period of silk price rises in Valencia (Table 4.2). Interregional silver flows did not affect prices in general but did affect prices of particular commodities such as Chinese silk. What commodities other than Chinese silk were there when the surge in the silver flow stimulated interregional trade? Table 4.1 shows that through the period of large silver imports the price of wheat also rose. The five-year average price of wheat rose from 1268 dinar between 1571 and 1575 to 1647 dinar between 1621 and 1625.6 However, importantly, the degree of fluctuation year by year was high during this period. For 50 years from 1576 when imported silver contributed most of the monetary supply in Spain, as shown in italics, the wheat price rose by more than 20 percent from the previous year on nine occasions. Although we should keep in mind less consistency in the data than in the case of wheat, during the same period only twice did the price of linen rise by more than 20 percent from the previous year, as shown in italics as well. It was the interregional trade in grains that most seriously impacted societies in this period.7 Since each civilization consumed a particular grain in own way, at this stage the trade in any one grain did not extend globally but developed only within

124  Global history of monetary delocalization Table 4.2 Total decennial imports of fine gold and silver to Seville (grams)

1503–10 1511–20 1521–30 1531–40 1541–50 1551–60 1561–70 1571–80 1581–90 1591–1600 1601–10 1611–20 1621–30 1631–40 1641–50 1651–60



148,730 86,193,876 177,573,164 303,121,174 942,858,792 1,118,591,954 2,103,027,689 2,707,626,528 2,213,631,245 2,192,255,993 2,145,339,043 1,396,759,594 1,056,430,966 443,256,546

4,965,180 9.153,220 4,889,050 14,466,360 24,057,130 42,620,080 11,530,940 9,429,140 12,101,650 19,451,420 11,764,090 8,855,940 3,889,760 1,240,400 1,549,390 469,430

Source: Hamilton 1934, 42.

the boundaries of a particular civilization. Western European consumption could not depend on wheat imported intercontinentally from Canada and Argentina until the 19th century (Adelman 1994). The range of the grain trade was not as global as that of the Chinese silk trade, but its impact on societies was greater. When merchants found a sudden abundance of silver available to increase distant exchanges, it was commonly grains which attracted the excess silver – the interregional currencies – to exchange for. However, the difference between grains and Chinese silk was that the former attracted a strong demand for proximate exchange, while the latter mainly depended on distant demand. The increasing export of grains sometimes clashed with local demand for grains, and instability in grain supply often brought panic to local societies. In varying periods during the late 16th century and again in the early 18th century, grain riots, which were targeted at intercepting grains being exported out of a region, emerged as a major issue across the world. Such riots were common occurrences in England in the 16th century and the early 17th century, as they were not in the late 17th and the early 18th centuries.8 As we saw, the 16th century was a period of price rises in England. However, the rises were not of the same degree as they were in commodities. The prices of grains rose more rapidly than others (Hammarstrom 1957, 144–5).9 These conditions must have caused merchants to concentrate on the grain trades. As far as Europe is concerned, profits from the distant trade in grains seem to have decreased gradually from the 16th century to the 18th century. The price gap between the Baltic regions and southern Europe shrank from six times to less than double (Spooner 1972, 72–3). We will touch on the situation of the grain trade in 18th-century Europe in the next section.

The world diversified and stratified  125

The Chinese case reveals that the problem of grain riots was closely related to problems in market activity generally in the same period. China experienced serious grain riots from the late 17th century through the mid-18th century. The lifting of the ban on maritime trade after the conquest of Taiwan by the Qing army in 1683 might have been what triggered the growth of the longdistance grain trade as well as continuous silver inflow from abroad. ‘The appreciation of the price of rice’ migui became a big social issue across the empire during this period. Local people intercepting grain shipments being exported out of regions occurred frequently. The Qing dynasty needed to punish locals’ violence and, at the same time, had to ease the instability in the local grain supply. The solution was to build granaries locally. According to the regulations, in ordinary harvest years, each granary annually sold a third of its grain stock and purchased the same amount in the local marketplace to maintain its initial reserves (Kuroda 1994, chap. 3). These local marketplaces were like the five-day market we saw in Chapters One and Three. An important point is that, when officials purchased grain in the local marketplaces, they paid in copper coins, not silver. The rapid establishment of local granaries by the Qing coincided with the issuance of Qianlong coins, which was the second largest quantity ever issued in Chinese history, behind only the 11th-century Northern Song. Before the harvest, local governments sold old grains for copper coins at a price cheaper than the current market price and, after the harvest, they purchased new grains with copper coins at a higher price. This system worked well for several decades and caused local traders to get accustomed to using copper coins (Kuroda 1994, chap. 3). Through the third quarter of the 18th century, copper cash replaced silver to dominate local usages of currency. Jiangnan or the Lower Yangzi region was typical in experiencing this change. In mid-18th-century Huizhou and Yangzhou, transactions in terms of silver by weight were so popular that people cut silver plate with scissors minutely enough to exchange even for vegetables. However, as copper coins circulated popularly, finally even land sales were made in terms of copper coins. It is no exaggeration to say that, for the first time in almost 600 years, copper coins officially cast became the most important device of proximate exchange (Kishimoto 1997, 359; Kuroda 1994, 37). An account book for the annual spring ritual by the Cheng clan in Huizhou, where silver usage in daily transactions lasted for the longest period in the Lower Yangzi, shows that in 1780 records of purchases of rice, beans and salt changed from being in terms of silver to copper coins.10 Setting prices in terms of copper, which was available for proximate exchanges, did not synchronize with prices in terms of silver for distant exchanges, as we saw in Chapter One. Grain prices maintained a moderate rise for a half-century after the 1730s, in contrast with the previous period when they showed large fluctuations. As the prices of most commodities rose proportionately, the particular appreciation of rice disappeared. Through this period, a complementary relationship between silver and copper coins firmly took root across the whole of China, in which silver mediated distant exchange and copper coins proximate exchange. The

126  Global history of monetary delocalization

example of Huizhou mentioned earlier is the last region in which locals made transactions of daily necessities in silver by weight. The exchange rate between silver and copper coins depended on demand and supply, and the differences between regions were substantial.This flexible system continued in China until the early 20th century, as we saw in Chapter One. The Chinese transformations in monetary usage and price movements from the 16th to the 18th century shows how endogenous exchanges reacted to an excess flow of interregional currency. A huge amount of silver was absorbed into the Chinese southern coastal regions in exchange for silk and other goods in the latter half of the 16th century. The silver influx rapidly transformed transactions into using mainly silver. One local gazetteer wrote that previously nobody ever saw silver, but now every household possessed scissors for cutting silver plate and a scale for weighing pieces of silver.11 The extraordinary amount of silver inflow ended for a time in the mid-17th century, but after the turmoil during the Ming-Qing transition, including the ban of maritime trade against the Zheng government in Taiwan, silver once again continued to flow into China. Besides Chinese silk, grains were the most attractive commodity for silver holders. Speculative businesses in the gain trade disrupted local supplies of grains. Grain price movements showed large fluctuations rather than a continuous rise. Locals had no other choice than to stop grain being exported from their regions by force. Only the introduction of the official granary system, combined with a large supply of copper coins working as local currency, finally induced the grain trade to confine itself within proximate exchanges. In other words, an increasing supply of small currency into Quadrant IV took the grain trade back from distant exchanges to proximate exchanges. Interestingly, contemporary India under the Mughals shared some characteristics with China. India must have absorbed a large amount of silver just as China did. However, though this is debated, through the 16th and 17th centuries there were no clear price rises; rather fluctuations characterized the price movements of the period. In contrast, throughout the 18th century prices rose continuously. Important to note is that small denomination currencies simultaneously increased in supply. Shell monies (cowries), which were imported from the Maldives, circulated in the largest quantities in Bengal and Orissa. Cowries served as a main device for proximate exchange at rural marketplaces, while forms of silver functioned as interregional currencies. Silver rupees were not available in rural markets. The Collector of the East Indian Company at Rajshahi was surprised that silver rupees were not accepted in the open bazaar (Mitra 1991, 176). The exchange rate between cowries and silver varied region by region. In addition, local currencies appreciated against silver after the harvest (Mahapatra 1969–1970). The money changers earned profits through exchanging these currencies, as we saw in Chapter One. These similarities between the two most populous countries suggest a global tendency for excess supplies of interregional currencies to bring unrest to local societies by disturbing proximate exchange, and then for increased supplies of

The world diversified and stratified  127

local currency to restore a stable relationship between upper-level and lowerlevel markets.12 We do not have sufficient data on grain prices and the issuance of small currencies to examine the causality mentioned earlier globally. At least, however, we can confirm that increasing issuance of small currencies commonly followed surges in silver flows after the late 16th century. The particular forms of small currency and the periods of their increasing supply varied region by region. Ever since the collapse of the Roman Empire, as mentioned repeatedly earlier, states in Western Europe had been reluctant to mint base metal coinages. Simply minting them incurred high cost with little profit for the authorities. However, the Spanish kings began to issue a large quantity of copper coinage early in the 17th century. Unlike the previous period, the government expected some seigniorage from the issuance (Motomura 1994). France also issued billon and copper coinages from the latter half of the 17th century. In 1654 through 1656 French mints issued 7,281,144 livre of copper coins which was equivalent to 38.9 percent of the entire coinage, 18,708,381 livre, through the period. From 1674 through 1676, they also minted 13,524,721 livre of billon which was equivalent to 49.4 percent of the entire coinage, 27,361,889 livre. Although to a lesser degree than the Spanish state, the French government for the first time issued substantial amounts of small denomination coins in the latter half of the 17th century (Spooner 1972, 339–40). Commonly, Western European countries experienced increasing circulation of small denomination coins in the 17th century while silver flows shrank. The differences were in whether a government issued coins or private providers made tokens. Supplies of copper followed the increasing demand for copper for the sake of minting coinages as well as manufacturing cannons. Sweden and Japan became the largest copper suppliers in the west and east respectively. The Dutch made a profit through trading copper internationally as well. Interestingly, it has been found that the import of Swedish and Japanese copper into Amsterdam had a complementary relationship (Glamann 1958: 167–182). Countries surrounding China also experienced their largest issuances of copper coins in the 17th through the 18th century. Japan under the Tokugawa regime issued copper coins, the kan’ei tsuho, in the 17th century, and Korea under the Yi dynasty issued sanpyon tonbo, whose issuances reached a peak in the 18th century. In the same way, Vietnam under the Le dynasty issued cảnhhưng thongbao in large quantities in the 18th century. These countries, for the first time, issued a large supply of copper coins with their own era names sufficient to substitute for existing Chinese coinages (Kuroda 2005b). In one sense, the copper century brought independence from China in monetary terms to Japan, Korea and Vietnam for the first time in East Asian history. Turning back to the global trend of issuing small currencies, we find that some states refused to do this. The English government kept to their policy of avoiding the official production of small currencies. Elizabeth I rejected the idea of minting a small denomination coin when it was proposed as a measure to relieve the liquidity shortage in daily transactions among ordinary people.

128  Global history of monetary delocalization

She insisted on the exclusive use of precious metals for royal coins. Her successors basically followed the same principle. Subsequently, people created what they needed for themselves. Early in the 17th century, a variety of tokens were issued by locals. However, these issuances of tokens appeared only in London and provincial towns. In the 1660s, the City of London, Westminster, and the suburbs of London boasted 3,543 tokeners who offered coins of various sizes, shapes and materials which included copper, tin, lead and even leather (Valenze 2006, 35, 37). Tradesmen, innkeepers, artisans and even the occasional poor widow joined in the issuance of ‘people’s money.’13 Currency circulation became so heterogeneous that shopkeepers sorted them into compartmentalized boxes and regularly exchanged them for other currency at the site of their issuers. ‘People’s money’ did not emerge solely in England. In 1614, 400 types of tokens circulated in the Low Countries and 82 in France (Supple 1957, 240n). Probably, due to official minting of small denomination coins, local tokens had less popularity in France than in contemporary England. Though there are differences between coins and notes, the heterogeneous situation in the urban districts of 17th-century England appears to be similar to the situation of early 20th-century China in which plenty of native notes circulated locally, as we have seen. Importantly, however, at the same time that heterogeneous tokens circulated in cities, dependence on mutual credit in proximate exchanges became stronger in small towns in England as we saw in Chapter One. The frequency of lawsuits brought in the local court per household in England reached a peak in the early 17th century (Muldrew 1998, 238). Some societies responded to excess supply in Quadrant I by expanding Quadrant III at the local level, not Quadrant IV. Regardless of whether it was generating money for the people in Quadrant IV, or increasing dependence on mutual credit in Quadrant III, after the reassociation of the four Quadrants through the 17th century, grain riots became unusual in England (Kussmaul 1990, 168). The change coincided with the disappearance of local markets and the development of long-distance settlement through credit in Quadrant II. In the middle of the 17th century a banking network through inland bills of exchange covered most cities across England and in 1697 the negotiability of inland bills of exchange was established by law (Kerridge 1988, 65–6, 74).

3 States organize debts nationwide: formation of a credit-oriented economy Western Europe passed a turning point in the 1550s. As we have seen, in 1557, Phillip II declared bankruptcy, and all payments of assignment against official income were suspended and replaced by 5 percent Spanish state annuities. In the Netherlands, the Provincial States of Flanders, Braband and Holland sold a flood of annuities on behalf of the central government. France also took the same route. For example, the capital value of annuities issued by Paris increased

The world diversified and stratified  129

from 600,000 livre tournoir in 1543 to 9600000 livre tournoir in the 1560s, through 3,765,000 livre tournoir in the early 1550s (Usher 1943, 167). As we saw in Chapter Three, annuities had a long history of financing municipal budgets in Western Europe. In the middle of the 16th century, states began to incorporate funded debts through the operation of municipal governments. Importantly, despite consecutive bankruptcies by the leading European monarchs, the interest rates did not return to the erratic heights of the 15th and early 16th centuries. The interest rate for short-term public credit on the Antwerp exchange dropped to 12 percent per year in 1524 and never rose again after the 1530s (van der Wee 1977, 368). From a long-term viewpoint, the substitution of funded debt for floating debt stabilized European states’ finances. Transformations in public finance went in tandem with the development of private finances. A jewelry merchant from Antwerp, Hans Thijs, originally started his business with a hide trade in Danzig. Before moving to Bruges in 1598, he used to take out loans from his relatives when he was short of money in Danzig. At that time, the interest rate of credit through family connections was far cheaper than that of credit available openly, such as IOUs in Danzig. In contrast, after moving to Bruges, his transactions using IOUs increased substantially, while his borrowing from relatives remained at around the same level. The interest rates of IOUs in Bruges at that point were no higher than credit through family connections (Gelderblom 2013, 62–70). The case of Bruges was not isolated. The increasing popularity of using credit was quite apparent in England during the same period. As we have seen, inventories of merchants in Liverpool and London from the 16th through the 17th century mostly consisted of items acquired through credit. We should pay attention to the argument that the development of credit caused prices to rise in England through the period.14 In other words, credit dependence in Western Europe multiplied the impact of the silver flow from the late 16th century to the mid-17th century. Following the contraction in silver circulation from the late 17th to the mid18th century, the system for long-distance settlements among private traders developed substantially in Europe. For example, in England the acceptance of checks in payment became popular enough among merchants to establish the clearance system for checks in London around 1750. In 18th-century France, the settlement of distant trade mainly depended on bills of exchange. Paris also played a significant role in mediating transactions between provincial traders. The centralization of clearance must have advanced in tandem with the development of interregional trade. It was common practice in France and England to pay rent in kind on agricultural holdings, though the rent was set in terms of money. Grain, or other produce, but most commonly grain, was delivered to the landlord on the basis of the official market price. In both England and France official market records were kept in order to furnish an objective basis for the payment of rents. In France, at least, many receivers of rents in grain actually shipped grain from their rural estates to Paris, notably after the migration of the nobles and the

130  Global history of monetary delocalization

clergy to Paris in the 17th century.Thus, much of the grain received as rent was never sold locally (Usher 1943, 178). However, the grain paid for rent did not always include all grain for selling. Besides turning over the rent in kind to proprietors’ granaries, tenants could also take their grain to the nearest market town or to any other place within a radius of three leagues. Petite dealers, called blatier, went around from farm to farm buying small amounts of grain and transporting it to the market. However, they did not engage in export trade (Le Goff 1973, 98–9). Here, we can find a division of labor between upper-level and lower-level markets. In proximate exchanges, cash decreased in importance. Peddlers, or petite dealers, tended to engage in cash transactions. However, through the 18th century these decreased. In villages, people exchanged land for animals, for labour, and for other items, but less frequently for cash. Accumulating letters of credit became a better strategy than accumulating land. In towns, merchants accumulated most of their assets in commodities and credit, very little in cash (Fontaine 1996, 101–3). Grain merchants engaging in interregional trade depended heavily on payment through bills of exchange. They rarely used cash. In the case of the merchant Desruisseaux, French correspondents assigned credit on Paris bankers, while foreign customers authorized them to write drafts on bankers and clients in London or Amsterdam (Le Goff 1973, 107). How to cash the drafts was crucial for such a business. Landed proprietors who sold grain did not usually accept drafts and demanded hard cash. Desruisseaux relied on men holding positions in government in charge of tax collection and the bulk sale of tobacco to cash his drafts (Le Goff 1973, 108). These men could dispose of large amounts of money freely between the collection of funds and forwarding them to the government. Here is a point at which interregional settlements in private trade could be enhanced by the transfer system of official money. Most merchants sometimes needed drafts through Paris for ordinary credit transfer even within a single region. When Desruisseaux sold commodities to foreign firms, few local financiers or merchants could accept their drafts. Desruisseaux held an account with a banker in Paris. He sold the drafts to the bank and asked it to write a draft which could then be sold to local financiers at the other end. Here is another point showing that credit dependence needed a clearance centre to develop further. As Paris was the destination for most rents and taxes, the two points overlapped. After 1685 the Swiss and Geneva bankers’ colonies in Lyons opened branches in Paris. After the depression in 1709, Lyons, which had suffered from the disasters of the French monarch’s war policies, lost its importance as a hub for continental-wide settlement. Unlike Lyons, the Paris banking system evolved from a nation-based system into an internationally oriented financial centre through the 18th century. The presence of public debt increased in importance across the whole of economic activity from the late 17th century to the 18th century in Europe. In addition to funded debts such as annuities, new devices for collecting money

The world diversified and stratified  131

in the short term to meet urgent demand, especially for military purposes, were created. From the very end of the 17th century, the English government began to pay her debts by tallies. Tallies were the Exchequer’s formal wooden receipts. In 1693, the Board of Ordnance, in paying the gunmakers of the City of London, offered them tallies for their work instead of cash. It seemed to the gunmakers that they could raise no money with the wooden pieces (Dickson 1967, 344). However, this payment became popular over the course of the 18th century. From the 1720s, the Bank of England became the government’s agent for underwriting bills on the annual land and malt taxes and paid the interest itself, without any right to reimbursement from the Exchequer (Dickson 1967, 378). Exchequer bills reigned supreme until their supersession by Northcote’s Treasury Bill of 1877 (Dickson 1967, 365). The broad range of public creditors at city level in Western Europe, which we saw in Chapter Three, inherited public debts organized at the national level. Let us take a look at the case of the Tontine, which was the first English longterm loan to be floated. Between 3 February and 1 May 1693, £377,600 was subscribed to in minimum units of £100. The average contribution was only £236 (Dickson 1967, 254–5). Of course, £100 was far beyond ordinary people’s reach. However, compared to the average size of the subscription of the stock of the Bank of England, £946, it was a moderate amount. Following the trend mentioned earlier, through the first half of the 18th century, specialized financial institutions with a professional interest in government securities arose. In an earlier period, Child’s Bank, like other London bankers, took diamonds, jewels and plate as security for loans. However, as early as 1689 they were also lending at the Exchequer, discounting shortterm government tallies and making loans using the tallies as security (Dickson 1967, 437). Among his total assets £207,743, tallies and securities occupied £120,750, mortgages £75,493 and cash £11,500 when Samuel Child died in 1752 (Dickson 1967, 445). Public debts by the English government amounted to £245 million in 1783. The amount was 20 times her annual tax revenue, £12 million (Brewer 1989, 114–26). The proportion clearly shows how deeply British state activity depended on debt. In contrast, in the same year, silver stock in the Board of Revenue in Beijing under the Qing had accumulated 84 million liang. The stock was more than 6 times as high as the annual receipt of silver by the Board, 13 million liang (Shi 2008, 141, 174). The Qing’s entire annual expenditure is thought to have been around 35 million liang (Matsui 1935, 71). It would be ideal if we could extend this by indexing stocks kept across the entire empire, but no reliable statistics for this exist. Even if the records by the Board represented only a part of the silver stocks, it is sufficient to reveal how large a fiscal surplus the Qing government enjoyed. Without knowing the historical background, some might seriously wonder why the English government could continuously issue bonds in spite of huge existing unpaid debt. From the opposite viewpoint, it also seems quite strange

132  Global history of monetary delocalization

that anyone would dare to purchase bonds issued by the English government. And in fact, governmental bonds became less popular in the market as their issuance increased. For example, the discount rates of Navy bills rose to 22 percent in the period of the American Independence War (Binney 1958, 141). However, merchants in fact continued to buy public bonds, since they could use them as collateral when borrowing money.Thus, a debt-dependent government enabled merchants to depend further on debt in their own businesses. Public activity and private business were closely connected through debt. It was not only British subjects who purchased English public bonds; Dutch merchants were also among the best customers of British bonds. The Netherlands government also accumulated public debt to the extent that it amounted to about 2.5 times the Dutch Gross National Product (Fritschy et al. 2012, 47).15 Thus, the British case of public-debt dependence was far from isolated.We may locate the mutual relationship between a government and merchants during the period of mercantilism along the same line of development of public debt after the emergence of annuities between a city state and citizens in the 13th century, which we saw in the previous chapter. The debts accumulated by mercantile states were convertible with the debts made between private traders. In England they were made in terms of pounds sterling. However, in China the silver bullion accumulated by the Qing government was not convertible with silver circulated among private traders. Both forms of silver were exchangeable, but they were used in different units of account. In spite of using the same name for the unit, liang (or tael), the governmental liang, kuping liang and a number of different merchants’ liang were exchanged at fluctuating rates reflecting demand and supply. The dynasty accumulated huge quantities of silver, but it was restricted in being unable to flow out directly into private circuits. Chinese merchants accumulated silver through favourable trade with foreign merchants as long as Chinese products such as tea, porcelain and silk attracted silver in exchange. The dynasty could collect silver as long as its subjects could make these trade surpluses. However, the actual movable property, silver, was stored in inconvertible forms.

4 Local paper monies reveal differences among institutions: a comparison between England, China and Japan In early 19th-century England, banknotes issued by country banks circulated widely among local populations.16 In the same period in Japan, domain notes (hansatsu) also increasingly circulated in the western regions; in China as well, native notes (qianpiao, literally copper cash notes, also referred to as tupiao) also circulated widely. At that time, England was in the midst of the Napoleonic War, Japan was still relatively isolated due to the seclusion policy, and China would not freely open its ports until the 1842 Treaty of Nanjing. While the countries thus had quite different economic conditions, paper monies played a significant role in proximate exchanges in all three.17 In the following, we will evaluate

The world diversified and stratified  133

cases from all three countries in the 19th century, though, due to the scarcity of materials related to 19th-century native notes, we will often rely on 20thcentury sources to discuss China’s paper monies. From 1797 to 1825, the face value of the most commonly used country bank note was one pound, though throughout this period both the Bank of England and country banks suspended conversion of all notes. At this time, the government also lifted its ban on issuing notes with a face value of less than five pounds. Roughly speaking, one pound was equivalent to the wages for one week of labour in this period.18 In the case of Japan, it is difficult to judge which denomination was the most popular among domain notes, though the majority appear to have had face values of around one monme. At the time, wages for ten days of labour were probably around ten monme, and thus local paper notes in Japan provided a means of exchange at more fragmental values than those in England.19 It is far more difficult to estimate the real value of native notes in China. We have better information regarding the situation of native notes around 1900, when the most popular face value was the 1,000-wen note, or wages for around 10 days of labour.20 Despite these differences, we will work from the idea that local paper monies worked to facilitate exchanges among ordinary people in all three countries. A country bank in England usually had a corresponding financier in London, but its bank notes rarely circulated beyond the county in which it was located. Domain notes in Japan sometimes circulated in neighbouring domains but were basically accepted only within the domain of the issuer. Native notes in China mostly circulated only within a town and were quite rarely accepted beyond a county.Thus, none of the three paper monies circulated among populations larger than several hundred thousand people. While the three paper monies shared similarities in that they mediated smallsized exchanges among locals, there were also several differences among them. First, the characteristics of the issuers differed. Needless to say, country bank notes were issued by local banks in England. These were private firms, though the range of the denominations of notes was regulated by the state. Domain notes in Japan were formally issued by the domains, though their issuance required permission from the central government. In most cases it was merchants who issued the notes on behalf of the domains. These merchants engaged in businesses related to products (rice, cotton cloth, sugar, paper, etc.) exported from the domains to distant markets such as Osaka, the commercial centre of Japan. Native notes in China were also issued by merchants. In most cases, money exchangers issued local notes, especially when local markets suffered from shortages of currency. However, most exchangers were not specialized in financial activities and often engaged in other trades as well. When locals suffered from serious shortages of currency, a variety of merchants besides financiers issued local notes. Even baker or butcher would issue notes as we saw in Chapter Two and will see again later. Thus, in all three cases merchants commonly played crucial roles, but the role of the authorities differed. In the case of domain notes in Japan the finances and

134  Global history of monetary delocalization

administration of domains always affected the issuance and circulation of their notes. In England, authorities did not intervene in the daily business of country banks, but laws regulated the denomination and convertibility of country bank notes. In the case of China, by contrast, the government instead tried to prohibit the issuance of private notes. In the beginning of the 20th century, when chambers of commerce were established across China, some tried to discourage the circulation of notes without their seals, but such regulations rarely worked (Kuroda 2005a, 121). The different degrees of official commitment in the three countries was related to their different levels of acceptance for tax payments. In Japan, though some proportion of tax was still required to be paid in kind, especially in rice, domains basically accepted notes of their own issuance for tax payments. In England local tax bureaus mostly accepted country bank notes and transferred official revenues to London through Bank of England Notes or gold coins. In contrast, in China local notes could not be used to pay taxes, even at the lowest level of government administration, the county. Between 1797 and 1825 neither country bank notes nor Bank of England Notes were convertible. Bearers could not request banks to convert the notes into any hard currency. In other words, England was on a genuine ‘paper money standard.’ The relationship between country bank notes and the Bank of ­England Notes was stable – the exchanges between them were done at par. In Japan, domains generally declared that their notes could be converted into hard currencies (or sometimes in commodities such as rice) at a small fee, such as 2 percent of the amount converted. Needless to say, convertibility was not always perfectly maintained.Trouble occasionally occurred due to the failure of domains to convert their notes on demand. However, considering that in many cases the actual discount rates of domain notes against cash by merchants were not so high, it might be safe to assume that there was little trouble caused by the conversion of notes in local markets. In many cases the face value of monme (literally, ‘unit of silver’) had a fixed exchange rate with copper cash. In reality, in many domains in western Japan, the payment of one monme meant that a certain quantity of copper cash was paid. The custom was called monmesen (Fujimoto 1991).The copper cash that mediated local transactions and the silver that was used for settlement of distant trades were actually in a fixed relationship, though the rates differed domain by domain. In China, the convertibility of native notes varied. Importantly, the degree to which there was a shortage of currency used in proximate exchanges appears to have determined the degree of convertibility. When the shortage of currency was not serious, issuers could not make native notes circulate without preserving their convertibility. At times of intense shortages, however, convertibility was less of a concern. Observers in the early 20th century recorded that when a bearer actually requested the conversion, the issuer would simply exchange it for another note issued by different shop.21 In some cases the issuers declared that they would only convert notes into coins on the day of the fiveday market, the market held every five days.22 On other days, they would only

The world diversified and stratified  135

exchange one note for another at the demand of the bearer.23 As the exchange rate between copper cash and silver fluctuated daily at such times, the exchange rates of notes against silver were also unfixed. If convertibility into hard currencies was not actually guaranteed, how could those paper monies circulate? Domain notes in Japan were issued by domains which imposed taxes on their subjects. The use of these notes in tax payments was a strong incentive for subjects under the domain to accept them. However, there must also have been other factors, since sometimes demand for domain notes increased to the extent that traders suffered from serious shortages of them. As domain incomes became more dependent on exporting commodities such as cotton, sugar, paper, etc., domain notes became more important for their role as devices for collecting commodities from peasants and artisans. Often the notes were issued by merchants who engaged in the trade of particular commodities that sold well in Osaka or Edo.24 These merchants accumulated hard currencies on behalf of the domains by selling the commodities in these distant markets, and thus could maintain the convertibility of the domain notes. Thus, commodities sold in distant trades supported the acceptability of domain notes in local exchanges. In other words, commodities which were expected to be sold in distant trades worked as collateral. Viewed in the four quadrants of exchange, this is a case in which Quadrant IV was supported by either Quadrant I (in the case that trade in commodities was settled by hard currencies) or Quadrant II (in the case of bills of exchange). Country bank notes in England could also be used in tax payments. However, unlike domain notes in Japan, from 1797 to 1825 these notes were not convertible either in reality or formally. In order to precisely understand how a ‘paper money standard’ could emerge, we must remember some background already seen. Throughout the 18th century, the English authorities did not adopt any policy of supplying enough currency to match the demand of ordinary people. The total amount of silver coins minted through the century was equivalent to just 6 percent of the total amount of silver minted between 1558 and 1694. Thus, locals privately minted various forms of tokens with low denominations to overcome the shortage of currency. The Bank of England, which was established in 1694, also did not adopt any policy of providing notes to cover transactions among ordinary people. The Bank issued very few notes in denominations below five pounds – wages for five weeks of labour. A statics shows that £867,000 of below five pounds of Bank of England notes was in circulation on the last day of August 1797, £4,245,000 in 1807, while £10,246,000 of five pounds and above, £15,432,000 in 1807 and £21,550,000 in 1817.25 Under such a situation, employers paid employees jointly, sometimes breaking a single note up among themselves on their own authority (Pressnell 1956, 140–4). After 1797, new country banks proliferated. For example, even in 1840 (15 years after the resumption of conversion), among £34,420 of total assets by Banbury Bank, a country bank in Oxfordshire, £13,325 was made up by notes

136  Global history of monetary delocalization

issued by other country banks, while £3,938 was gold and silver (Pressnell 1956, 198). The majority of notes issued by country banks between 1797 and 1825 was the one-pound note, which supplied what local people demanded. Country bank notes tended to circulate more in agricultural districts. The circulation of notes had a seasonality that increased in winter (or after harvest) and decreased in summer (the slack season). For example, the banknote issuance by Bacon, Cobbhold & Co., a country bank in Ipswich, during October and December in 1834 increased by 10.6 percent from during July and September in 1834, 20.0 percent in 1835 and 22. 3 percent in 1836 (Yoji 1982, 424). In local markets, country bank notes were so popular that peasants preferred them to Bank of England notes or gold coins in their daily transactions.26 Although the proportion of assets made up by country bank notes appears to have varied, we can probably estimate that the number of notes in circulation was equivalent to about a third of the number of deposits. In the first quarter of the 19th century, overdrafts on bank accounts were already a popular form of lending in agricultural districts. The resumption of the prohibition against the issuance of notes less than five pounds in 1826 increased peasants’ use of bank accounts when they settled transactions. In 1840, a country banker noted that two-thirds of settlements were made with checks. Even if the importance of notes had already decreased, the possession of notes by Banbury Bank in 1840 amounted to £18,041, surpassing £9,775 in notes and £3,938 in gold and silver (Pressnell 1956, 198). Since, as we saw in Chapter One, England had already been dependent on credit in exchanges before the 18th century, people had become accustomed to settling exchanges through credit/debt. We can assume that, as a consequence, the presence of deposits in banks likely encouraged locals to accept the notes they issued.Viewed in the four quadrants of exchange, in this case Quadrant IV was supported by Quadrant III. In contrast to the local paper monies used in Japan and England, it is not easy to explain why people accepted the native notes in China. Unlike domain notes and country bank notes, native notes could not be used in tax payments. The native notes tended to mediate proximate exchanges, unlike domain notes, which facilitated distant trades of local products. Unlike country banks, the issuers of native notes had no deposit function. According to the characteristics of exchange represented by the four Quadrants, this case appears to be one in which Quadrant IV stood independently from the other three Quadrants. The report from an unidentified rural town in Julu County, Hebei, in 1936 gives us information on the reality of native notes’ circulation. In the year, native notes issued by 13 shops circulated in the town, which had only around 300 households; thirteen of the 26 shops in the town issued notes and included an inn, a grocery shop, a grain dealer and a wheat cake shop.The same transformation could happen in other towns at any time, as we saw the case of Xianrendu in Chapter Two. In the beginning, only merchants with large assets, such as grain dealers, issued the notes. However, as they became aware of the profitability of note issuance, many smaller-scale merchants soon followed. Even a wheat

The world diversified and stratified  137

cake baker with assets of only 200 yuan issued notes whose values amounted to 500 yuan (Chen TS. 1936). The reporter estimated the total value of native notes at 8000 to 9000 yuan. If we assume that 400 households likely used 100 yuan in cash each year, native notes totalling 8000 yuan must have circulated through the town five times. Although the reporter lamented the irregularity of the situation, native notes were not always over-issued. Thus, though no one had deliberately designed them to do so, the shops in the town generated small denomination currencies in the quantity appropriate to meet the demand for local transactions. It is difficult to measure how far native notes dominated across China. One index is a questionnaire on monetary conditions distributed to all counties in Shanxi in 1915. Its merit is that the investigation covered all of 102 counties, and thus we can measure the distribution of native notes. More than half stated that native notes circulated in their jurisdiction. Native notes tended to circulate in counties in which there was insufficient circulation of hard currencies, and vice versa. For example, more than 200,000 qian of copper cash and 20,000 to 30,000 silver dollars circulated in Linfen County, where there was no record of native notes (Anon 1915, 9811–12). However, the absence of a record of native notes in the questionnaire responses did not mean that such notes did not circulate in the county. For example, though the investigation did not indicate the presence of native notes in Hongtong County, a numismatic catalogue includes native notes issued there before 1915 (Wang and Liu 2001). In addition, a report by Japanese students from a Shanghai-based research association (the Toa Dobunkai) reveals that some exchangers issued native notes in Hongtong in 1917 (Toa Dobunkai 1920, 816–17).Thus, the 1915 questionnaire only indicates a minimum number of counties in which native notes circulated. There is no reliable data to show the extent to which native notes played a role in 19th-century transactions in China. However, we can at least confirm that as late as the 1830s, native notes were in common use throughout China. In this period, China seriously suffered from a shortage of silver that affected the collection of taxes. At the time, many bureaucrats investigating the monetary situation of each region noticed that native notes were quite common. A memorial by Lin Zexu is a typical example of bureaucratic views of the notes. Bureaucrats tended to find the circulation of native notes problematic and criticized note issuers who did not prepare cash for conversion. However, they also realized that prohibiting the notes would create a panic in trade, and thus Lin, for example, proposed making issuers form groups of five to mutually guarantee the conversion of native notes (Lin 2002, vol. 3, 1200–3). As late as the late 18th century, contemporaries described native notes and their circulation, but it was only after silver seriously began to drain from China due to opium imports in the early 19th century that bureaucrats began to pay attention to the popular circulation of native notes. With no formal support, local economies in China autonomously created this device to facilitate exchanges and adjusted themselves to fluctuating external circumstances such as the drain of silver.

138  Global history of monetary delocalization

The wide proliferation and long-term presence of native notes in China strongly suggests that demand for money alone can generate a means of exchange even without a formal guarantee or intrinsic value.Viewed according to the ‘common sense’ that either legal guarantees or a currency’s intrinsic value are necessary to make a currency acceptable, the popularity of native notes is difficult to imagine. However, once we recognize that native notes did not depend on such factors, we must reconsider if the domain notes in Japan and country bank notes in England did depend on, respectively, commodity exports and bank deposits. Commodities and deposits must have enhanced the acceptability of these paper monies but might not have been necessary preconditions. We should pay more attention to formless social institutions, such as in the case of China’s native notes. And, this system depending on native notes was too autonomous to create a system incorporating proximate exchanges into distant exchanges. Transactions with a strong preference for cash settlement cannot endogenously lead to the establishment of a long-term capital market, which requires investors to wait a long time for a return. The differences between the three local paper monies reflected the different economic transformations of the three countries that would occur after the 19th century. Focussing on the disparity between China and Japan, we will examine how differences in the individual activities of households and the organization of proximate exchanges was related to the differences between monetary systems.

5 Currencies, marketplaces and early industrialization: what happened to cement currency circuits? As we already saw in Chapter One, the popularity of local notes in early 20th century China was related to the proliferation of local marketplaces which ordinary peasants could access. The density of local marketplaces in mid-17th century western Kanto in Japan was no less than in China. As already mentioned, late 16th-century domain lords issued proclamations in their territories prohibiting new marketplaces located within 12 km from the nearest existing marketplace. This means that periodic markets such as five-day market had proliferated to such an extent that many peasants in late 16th-century and early 17th-century Japan could access them to do their business within half a day. The prosperity of autonomous local currency zones in late medieval Japan, as well as in China that we saw in Chapter Three, must have a relationship with the dense distribution of marketplaces. However, in contrast with China, such a proliferation of marketplaces across rural regions in Japan had already faded by the time trade ports were opened in the mid-19th century. This decrease of periodic markets suggests that villagers in Japan should have gradually lost their independence in commercial activities. As mentioned in Chapter One, a study of the rice trade with a village in Shinshu, present-day Nagano prefecture, shows that selling rice in the marketplace became unpopular and fixed

The world diversified and stratified  139

transactions with best customers became dominant from the 18th to 19th centuries (Tawada 2007). It might be safe to say that in the early 19th century, rarely did Japanese peasants bring their main products such as rice or raw cotton to a marketplace to negotiate prices by themselves. We should remind ourselves that things were not so simple. Between the 17th and early 19th centuries, new periodic marketplaces had emerged. As far as the case of Echigo, the present Niigata prefecture, was concerned, interestingly, the establishment of new marketplaces was led by village leaders, not merchants in towns (Kuwahara 1943).Villagers wanted access to marketplaces, while throughout the period merchants’ business became less dependent on rural marketplaces. Rather they came directly to visit peasants’ homes to collect commercial crops and products. The merchants sometimes made advances to peasants to secure commodities. We already saw a typical case of a putting-out system with the Yamawaki family in Chapter One. This transformation accompanied a structural change in handicraft manufacture, which most typically appeared in the cotton industry. Until the 18th century, the majority of cotton threads seems to have been made and used within the same region in Japan, just as in China. However, the situation changed towards the turn of the century. A typical case was merchants in Osaka who purchased raw cotton directly from cotton-growing peasants in Kawachi and transported it to Awaji or Kii, where peasants could not plant cotton and had idle time in the slack season to do spinning. Manually spun cotton threads, kaseito, were traded in large quantities by merchants in Osaka to distant regions like Ryukyu and Akita.27 We should not easily make a comparison depending on a limited case, but the variation over a short time in the case of the Tongtaihao in China was quite in contrast with the case of the Yamawaki family in Japan in which the quantities and prices did not vary (Table 1.6). For example, on the 27th day of 2nd month, 1841, the Tongtaihao purchased 20 kuai of cotton thread for 32,500 wen from the Hongtaihao, on the 2nd day of the 3rd month, 32.5 kuai for 53,800 wen and on the 3rd day of the 3rd month, 11 kuai for 17,500 wen. Within one period of less than ten days, they traded three times. The quantities and price per unit (kuai/wen) varied: 1,625 wen, 1,655 and 1,591.28 It is common across the world that textile industries were the most important economic sector besides crop farming. It is also common that the spinning process was the most consuming of time and cost, for example the wool industry of 17th-century England (Muldrew 2012). In some societies, merchants organized a division of labour among villages. For example, in early 18th-century southern Germany, although three villages maintained similar activities, Neckarhausen relatively specialized in the production of raw flax, Necktailfinger in fine yarn and Wolfschlugen in weaving (Sabean 1990, 156–7). This is similar to the situation of the cotton industry in early 19th-century western Japan as mentioned earlier. Table 4.3 shows a contrast: after opening their ports, China soon increased the import of cotton cloth while Japan increased the import of cotton yarn.

140  Global history of monetary delocalization Table 4.3  Imports of cotton cloths and yarns after opening ports, China and Japan cotton cloths (100000 yard)

1850 1860 1870 1880 1890 1900 1910 1920

cotton yarns (100000 kin)





732 2230 4173 5424 6224 6386 6805 8021

93 395 826 654 1423 1005 204

23 66 156 151 1082 1488 2283 1325

3 89 286 319 91 6 18

Source: Nakamura 1968, 224.

Unlike China, machine-made yarns substituted for manually spun yarn soon after the Japanese opened their ports. In this way, machine-made yarns deprived Japanese peasant households of their most longstanding and valuable side business. As Table 1.5 shows, raw cotton cultivation had already become less profitable than rice cultivation in 1880s Japan. The raw cotton cultivation which had thrived for three centuries was soon to disappear from the Japanese archipelago entirely. By contrast, in China, cotton cultivation continued all the while the cotton mill industry developed. Investigators in the Lower Yangzi delta observed in the 1930s that raw cotton growers still manually spun thread in spite of the fact that a cotton mill stood nearby (MSJC 1940, 66, 138). Many peasants across China engaged in selling manually spun cotton thread as well as manually woven cotton cloths to the marketplaces, even in the 1930s.29 Some written materials and investigations suggest that local women would have been expected to spin cotton thread to sell in the local market, but only a very small amount of cotton thread was sold beyond a county.30 In short, cotton spinning, the most timeconsuming and the most profit-making side-business for peasant households, had in China not been exposed to competition from machine-made thread being traded along distant trade circuits. These differences between China and Japan did not result from relative progress or backwardness. It is a false teleology to explain such differences according to a narrative of evolution or devolution. It is likely that the development of the market caused wholesale merchants to dominate and discouraged periodic markets from decreasing. However, this argument overlooks the fact that the degree of freedom of choice for the majority of ordinary people had diminished substantially as the transformations took place. Instead of adopting a linear evolutionary idea to endogenous exchanges, we should recognize that the differences in the relationship between distant trade and proximate exchange in China and Japan were in tandem with those in currencies and finances which were exogenously given to locals.

The world diversified and stratified  141

In contrast with Tokugawa Japan, in which public finances and private remittances were closely related to each other and formed a network of exchange bills, the Qing government in China favoured transporting silver ingots from local governments to the central government and between local governments. An account book of a financier, piaohao, shows that the amount of exchange bills drawn by bureaucrats in 1853 accounted for less than 1 percent of all of the bills issued by the financier (Zhang 1989, 36). Though bills of exchange had a long history in China, well-centred financial networks could not be established without coordination between private and official monetary flows. Interregional settlements by merchants were closely tied up with financial activities by the domains in Tokugawa Japan. Consider the case of Obama domain which was located north of Kyoto. Around 1871, when all the domains were abolished by the Meiji government, the debts of Obama domain amounted to 519,300 ryo, while the annual revenue of the domain was estimated at 381,600 ryo. Among these debts, borrowing from merchants in Tokyo accounted for 173,700 ryo, from those in Osaka to 59,000 ryo and from those in Kyoto to 39,000. Clearly the Obama domain’s administration depended heavily on nationwide financial networks.The Meiji government allowed the creditors to receive national bonds equivalent to almost half of the debt in compensation for clearing debts (Kagawa 1996, 199, 206). Building on the foundation of a peculiar feudal system with a centralized market hierarchy (or local markets which had lost autonomy), major banks connected local banks through correspondent accounts, established a nationwide network of documentary bills and subsequently eased seasonal fluctuations of interest rates soon after the establishment of the Bank of Japan in 1882, though local demand and interregional demand for a monetary supply continued to come into conflict. These financial conditions mattered for the establishment of some mechanized industries. Let us take a look at the situation of the silk industry after the opening of the ports. In 1891, in front of an audience including bankers and entrepreneurs from a silk-producing region, Tajiri Inajiro, the former vice minister of the Japanese Treasury, stressed the importance of introducing bills discounted by local banks in payments to sericulturists (Tsurumi 1991, 283–4). At that time, the Japanese economy was heavily dependent on silk exports, but financing the production and trade of silk was not an easy business. One of the difficulties for silk manufacturers was to manage the high degree of seasonality in the demand for money. Purchase of a large quantity of cocoons made it necessary to supply a large amount of currency over a very short period, while during the rest of the year the region did not require such quantities. Money always needed to be supplied from outside, since retaining currency reserves in the region would have made the currency inactive and wasted for much of the year. The most substantial supplier of currency at the time was the Bank of Japan (BOJ). Accepting silk in warehouses as collateral, the BOJ discounted bills issued by exporting firms in Yokohama, which then brought cash to cocoon-raising regions in order to pay sericulturists. However, these strong seasonal shifts in monetary demand made

142  Global history of monetary delocalization

it difficult to keep the reserves of the BOJ stable. According to Tajiri, 10 million yen in cash was necessary to meet the extra demand that arose during the silk season (Tajiri 1892, 160–1), an amount no less than one tenth of the annual revenue of the Japanese government. This is why Tajiri stressed the importance of discounting bills by local banks to ease monetary tensions region by region. Our aim here is not to examine the transformation of the Japanese financial system in itself but to contrast it with the case of China. China’s silk industry was in a situation similar to Japan’s, though the main harvest season for cocoons was only in spring rather than in both autumn and spring as in Japan. The export of silk in China also played an important role in balancing foreign trade. However, the sericultural regions seasonally required significant amounts of cash, even in the 1920s. During the five years between 1922 and 1926 Shanghai annually exported an average of 63.332 million yuan to other Chinese ports. During this period, exports in May amounted to 13.850 million yuan on average, which accounted for nearly 22 percent of a year’s exports, while exports in April and June only amounted to 3.326 and 2.458 million yuan respectively. This surge in May resulted from the collection of silk cocoons in the sericultural areas of Jiangsu and Zhejiang provinces. Another annual surge occurred in winter when grains and raw cotton were collected, though this fluctuation was not so drastic as in the spring: 4.813 million yuan in October, 9.962 million yuan in November and 8.798 million yuan in December (Shanghai Yinhang Diaochabu 1932). The big difference between Japan and China was that, while the Japanese side financed businesses in terms of a single currency, the Chinese side did so with plural monies. The BOJ notes were used for settlement by all traders in Japan and actually paid to sericulturists for their cocoons. In contrast, while silver dollars (yuan) were brought to silk-producing regions to make payments, a unit of silver account by weight, liang, was used among traders in Shanghai. As mentioned earlier, the exchange rate between the two forms of silver incessantly fluctuated according to supply and demand. Another apparent difference was that, while local banks engaged in the silk trade in Japan, money exchangers were dominant in China. Even more than money exchangers, in Japan, banks holding deposits could flexibly supply credit through overdrafts. More importantly, local banks in Japan had correspondent accounts among them to facilitate remittances between distant regions. We already saw in Chapter Two, the settlements by the Taiekigo, a Chinese merchant in Nagasaki, western Japan, heavily depended on Japanese banks. On the other hand, even after World War I, in China the bills issued by money exchangers in local cities could rarely be discounted in Shanghai. The popular circulation of BOJ notes appears to have been crucial. The role of the central bank after the establishment of the Bank of Japan in 1882 must not be neglected, but we should not reduce the differences between Japan and China simply to the degrees to which each had developed a modern banking system. The divergences reflected the historical backgrounds of each country. The issuance of the BOJ notes was, to some extent, secured with government

The world diversified and stratified  143

bonds and stocks in large firms, such as railroad companies (although a large portion of the capital of these firms also consisted of bonds). The popularity of bonds in modern Japan could not have emerged without the foundation in the domain system of the Tokugawa period as we have seen. As late as the 16th century, currency circuits prospered all along the Japanese archipelago. As late as the 18th century, proximate exchanges among locals kept their independence from distant trades. However, the combination of accumulating domain debt to merchants in central towns and institutional regulations discouraging local marketplaces happened to cement currency circuits in Japan before industrialization. Importantly, again, we should avoid a linear viewpoint under which monetary systems inevitably simply evolve from diversity to unification as the market becomes more advanced. As we saw in Chapter One, concurrent currencies in China did not represent backwardness in the degree of commercialization of peasants but resulted from the business activities of small land holders who were too independent to organize local markets. In terms of our four quadrants, in China, Quadrant IV had a higher density than the other three quadrants. Understanding the various characteristics of exchanges thus gives us a framework in which to understand a variety of market activities. The Chinese imperial government, which traditionally favoured a small government that rarely borrowed monies from merchants, had not plundered its subjects as much as other states, but was not suited to establishing a bridge between public finance and financiers’ business. In terms of our four quadrants, in China Quadrant II and Quadrant III were less dependent on each other. We need to combine institutional factors with the characteristics of exchanges in order to understand the diverse paths followed by China and Japan.

6 The third path: commercial oligarchies After the rapid contraction in the worldwide circulation of silver in the latter half of the 17th century, some local economies independently increased the local currency supply, or intensified Quadrant IV; others strengthened local coherence to create mutual credit systems or strengthened Quadrant III. In addition to the two directions for multiple currency supply and the direction for strengthening proximate credit, the third path increased its presence by responding locally to shortages of liquidity. Autonomous local currencies precluded independent commoners, particularly peasants, from possessing a high degree of freedom to access marketplaces; meanwhile mutual local credit was based on cohesive relationships between peasants, though they had less freedom to use the market. The third path represented a situation in which peasants depended on particular merchants or landowners who were able to engage in distant exchange. In this, peasants had difficulty making exchanges except with particular merchants. The exchanges were made in terms of money but means such as tokens distributed by shops were not available with other traders.

144  Global history of monetary delocalization

Those kinds of relationships in which less free peasants depended on commercial oligarchies had existed from the ancient period, but the global expansion of colonial economies after the 16th century took the system to an unprecedented size.31 In particular, the emergence of monoculture economies in the 19th century fostered this combination of commercial oligarchies and subsistence peasants. In Guatemala, which depended on coffee exports under the control of owners from Germany and the US, the work on fincas (‘plantations’) was performed chiefly by indigenous people. The growers provided basic sustenance for the workers, the food consisting chiefly of corn and beans (Young 1925). In 1898 Puerto Rico workers on plantations rioted in order to force the plantation owners to exchange the tokens of their issuance for official coins (Helleiner 2003, 171). The situation in the southern states of the US was similar. The case of a cotton plantation in 19th-century Louisiana shows a clear example (Lurvink 2014). Shops on the cotton plantations took cotton from sharecroppers in exchange for goods such as dry foods, cloth and the like. Unlike sharecroppers in the Netherlands where peasants could sell their products to other merchants, the peasants in the plantations had no other choice but to deal with the plantation shops. In order to reduce their debts to the plantation shops, sharecroppers paid in other products such as corn or in labour, as well as the usual cotton.The owners of shops often issued tokens, but the tokens were not easily redeemable for cash. Some tokens were only valid for particular goods, such as bread, meat or flour. Under these circumstances, though most transactions were made in terms of money, there were almost no currencies in circulation which might have enabled the peasants to have freer choices. Similar tokens issued by plantation owners emerged in many places across Latin America (Helleiner 2003, 71). In contrast with peasants, plantation owners or merchants running the plantation shops did their business through distant exchange using currencies which were available for settling long-distance trades. The system appears to combine Quadrant III for peasants with Quadrant I for plantation owners. However, unlike the case of mutual credit among proximate peasants, bookkeeping by monopolistic shops depended directly on the economic climate affecting distant exchanges and brought no autonomous device interfacing proximate exchanges with distant exchanges. Many local credit banks established in late 19th-century Japan originated from traditional rotary mutual credit organizations among local peasants. The Netherlands also organized farmer credit banks with sharecropping farmers. Plantation peasants such as those in Guatemala and Louisiana and cohesive villagers such as those in Japan and the Netherlands had a similarity in the point that they had little usage of currency, but this took different paths. Sharecroppers had no other choice but to depend on oligarchical shops. In contrast, some cohesive villagers could find endogenous money. Agricultural labours hired on fixed-term contracts by plantations seem to have been inclined to engage more widely in monetary transactions than sharecroppers, since the former were paid in cash monthly or annually, instead of being rewarded in kind as in the case of sharecroppers. However, the receipt of

The world diversified and stratified  145

rewards in the form of cash does not always mean that the labourers actually had the freedom to spend it. In the case of sugar plantations on the Island of Negro in the Philippines, commodities such as liquor, cloth and tobacco which labourers wanted were provided only by one particular agent. Labourers used to spend their wages on them soon after paydays. In addition to commodities, notoriously, local gambling places often waited to take their salaries. Thus, under an inside market structure, peasants as agricultural labourers had little possibility to choose transaction partners or negotiate the conditions for spending their cash. Money worked as a unit of account, but no negotiation existed. The Island of Negro appears to have had no circulation of tokens like the case of Louisiana, but the structure for exchange resembled it. Rapid expansion of sugar cultivation attracted sharecroppers and agricultural labourers to hacienda in the latter half of the 19th century. The economy depended heavily on sugar exports. Oligarchies, plantation owners and their agencies dominated the production and transportation of sugar (Echaúz 1978). Commercial oligarchies appeared with various forms of monetary systems. Trade monopolizers of foreign origin sometimes used native money as unit of account. In the case of the Royal Niger Company (1886–1899), controlled by Britain, all the company’s accounts in Africa were kept in cowries which were given a constant value of 1/3d per unit (2,000 cowries). All commodities possessed a relatively fixed value expressed in terms of cowries. For example, a local could bring 5/-d worth of palm oil to the company’s store and exchange it for 4 units worth of goods. Interestingly, real cowries, which could be obtained much more cheaply than 1/3d per head, never changed hands. The combination of the local economy’s dependence on foreign exports and a monopoly on foreign trade enabled the prices in terms of imaginary cowries to deviate from the prices in terms of real cowries (Vice 1983, 13). As we found in the comparison between China, Japan and England, a preference for one-time transactions (currency transactions) or for subsequent transactions (credit transactions) indicates whether people emphasized freedom or certainty. However, in both cases – local currency and mutual credit – there is no oligarchy monopolizing local exchanges such as the one we saw in this section. In this sense, the development of exchanges among proximate inhabitants depended on the degree of freedom of those carrying out the transactions. Wherever oligarchies dominated transactions, there was little space for people to create a currency or a mutual credit system. Thus, two different degrees of institutionalization, to what extent people could exchange independently and to what extent people preferred freedom or certainty, determined what kind of monetary system a society formed.

Notes 1 Minting of Spanish pesos was so centralized and consistent with a high fineness that the big silver dollars could work as a means of interregional settlement across the world. However, they could not meet flexibly the regional demands in Spanish America. That is why debased silver coins and inconvertible paper money became dominant according to region soon after the independence of Latin American countries. Irigoin 2009.

146  Global history of monetary delocalization 2 As we shall see later in Table 4.1, the wage for linen weavers in Valencia increased between 1551 and 1600 but rose more slowly than the price of linen. 3 The disparity of silver/gold exchange rates is a useful index for measuring the degree of demand for silver. However, it does not appropriately represent the facts to describe the silver inflow in exchange for gold outflow as significant. As repeated already, gold was not used as money in China after the Han dynasty and it is just one of the minor items that China exported. Flynn and Giráldez 1995, 207–8, 2002, 398. 4 We may assume that raw silk per pound in Manila was equivalent to 50 grams of pure silver in the beginning of the 17th century, when raw silk of the same quantity was priced around 130 grams of silver in Valencia. In 1640, the price of silk in Manila rose near 100 grams of silver, while that in Valencia fell under 100 grams of silver. Kimura 1989, 221–6. 5 According to Iwao 1953, Japan imported 404,000 kin of raw silk in 1634. It was equivalent to about one million kanme in silver. We may take it that Japanese rice production was around 20 million koku, which might have been equivalent to 10 million kanme in silver. 6 From 1501 to 1602 the dinar was equivalent to 0.1389 grams of pure silver. Hamilton 1934, 319. 7 Through the analysis of account books in Istanbul Pamuk found a big spike in prices expressed in grams of silver in the last quarter of the 16th century and the first quarter of the 17th century. Interestingly, food prices rose faster than manufactured goods and wages. Pamuk 2000, 118–27. 8 Bad harvests and hiking prices of grains still occurred in the late 17th century and the early 18th century in England (Appleby 1979). However, social disorder due to a dearth of grains occurred particularly during 1590 and 1648. Importantly, riots in grainproducing districts tried to prevent merchants from moving grains to the outside such as to London or foreign countries.The situation was quite similar to that in late 17th-century China. Outflows of grains mattered to locals. Walter and Wrightson 1976. 9 It is interesting that Sweden and England experienced price hikes in grains in terms of silver without new silver imports in the first half of the 16th century when the silver coinages underwent considerable debasement. It might have resulted from an increasing supply of currency whose values were small enough to fit the transactions in Quadrant IV. Hammarstrom 1957, 154. 10 Chengshi qingming jisi bu, possessed by the Institute of Economics, Chinese Academy of Social Sciences, no. si 39, b105. 11 A Swiss mercenary hired by the VOC visited China in the early 17th century to observe that local people cut a thin plate of silver and weighed a piece to exchange for goods. Giraud 1997, 185. See note 37 in Chapter Three. 12 It might not be just coincidence that European maritime traders transhipped both cowries and copper as profitable ballast items. Flynn and Giráldez 2002, 418. 13 The situation did not change through the 18th century. In 1787, the Royal Mint reported that only 8 percent of copper coins in circulation were official ones. Helleiner 2003, 24. 14 A survey of inventories in England show that the ratio of currency to credit by bond, bill obligatory and book was 1:1 in the period 1538–61, 1:4.5 in 1563–89, 1:12 in 1590–1620, 1:11 in 1621–39 and 1:6 in 1640–60. Inside bills of exchange were not counted. Kerridge 1988, 99. 15 Dutch war expenditure per capita in the late 17th century was nearly 2.5 times as high as England. Fritschy et al. 2012, 66. 16 A country bank is a merchant bank established outside London. Country bankers supplied all financial services including discounted bills on London, allowing overdrafts and issuing banknotes, in their locality. Country banking flourished in the late 18th century and early 19th century, especially after the suspension of conversion of Bank of England notes in 1797. By 1784, there were over 100 country banks, by 1796, 300 and by 1810, more than 600 Kerridge 1988, 77, 82.

The world diversified and stratified  147 17 Paper monies in the British North American colonies also worked as means of exchange for local transactions against precious metal species as device for interregional settlements. Their circulations were, however, far wider than a half day transportation. Grubb 2018. 18 According to one study, in 1830, weekly wages for a labourer were 12 shillings and for semi-skilled worker 20 shillings. Burnett 1969, 250–3. 19 It is safe to say that one monme was a popular denomination with hansatsu in spite of variety domain by domain. Wages also varied greatly. However, in some villages daily wages were regulated around 100 mon (or one monme and half) in the late 18th century. Nakagawa 2003, 296, 305. 20 It is hard to find consistent data on wages in rural China. Needless to say, in addition, there is a large degree of variation by region. For the moment, we take the cases of early 20th century Taiyuan according to the diary of Liu Dapeng, which we have already mentioned. Kuroda 1996. 21 In 1936 residents of the Nanzhaoji town, Yingshang County, Anhui, brought a lawsuit against local merchants to the Ministry of Finance, saying that when conversions of native notes were required, merchant A in the town handed to the bearer the native note issued by merchant B, merchant B did the same with the note issued by merchant C. They wrote that the native notes circulated across the area within a distance of 10–15 km from the town. Anhui qudi ge difang shanghao sifa zhibi. Academia historica, Taipei, no.018000027642A. 22 Chen XR. 2012, 223. 23 Dai and Chen 2011. 24 A typical case is the relationship between Kochi domain, who issued domain papers and financiers in Osaka who exchanged the paper monies through marketing sugar produced in the domain. Kobayashi 2015, 78–88. 25 A. Crump, The Key to the London Market, six edition, 1877, 49–50, cited in Yoji 1982, 256. 26 V. Stuckey, Ev., taken before the Secret Committee on Expediency of the Bank Resuming Cash Payments, 1819, 245–6, cited in Yoji 1982, 247. 27 Nakamura 1991, 133. On the Japanese cotton industry see Nakamura 1968, chapter 5 and Nakamura 1991, chapters 4, 5, and 6. 28 Daoguang shiyinian lixi zhang, Tongtaihao, Chinese National Library, Wenjinge, no. 49120-41. 29 According to a survey of a rural market, Madianji, in Shandong on 8 April 1934, around 20 persons sold manually-spun cotton thread and 40 persons sold manual woven cloth. The region is not far distant from Qingdao, where the cotton mill industry developed. Mizuno 1935, 35. 30 Interregional trade of manually-spun cotton threads existed but was far less significant than that in raw cotton. Shashi, a mid-Yangzi treaty port, recorded the export of manual cotton threads with a value of not more than 40,000 liang and raw cotton with a value of 1,333 thousand liang in 1899. Both were sold mainly in Sichuan, which cultivated little cotton. Cotton thread as well as raw cotton with seeds made in Wuji, Hebei, were sold in Guihuacheng, Shanxi in 1905. Kuroda 1994, 295. 31 In the 16th and 17th centuries, Poland and Lithuania expanded distant trade. Though information is limited, magnates possessing large scale landholdings appeared to monopolize the usage of cash, while serfs on their estates were paid in kind. Maczak 1976.

References Adelman, J. 1994. Frontier Development: Land, Labour, and Capital on the Wheatlands of Argentina and Canada, 1890–1914. Oxford: Clarendon. Anon. 1915. ‘Shanxi huobi zhuangkuang diaochabiao,’ in Qingdai minguo caizheng yusuan dang’an shiliao huibian, vol. 20 (2006). Beijing: Quanguo tushuguan wenxian suowei fuzhi zhongxin.

148  Global history of monetary delocalization Appleby, AB. 1979. ‘Grain prices and subsistence crises in England and France, 1590–1740,’ Journal of Economic History 39(4): 865–87. https://doi.org/10.1017/S002205070009865X Atwell, WS. 2005. ‘Another look at silver imports into China, 1635–1644,’ Journal of World History 16(4): 467–89. Barkan, ÖL. 1975. ‘The price revolution of the sixteenth century: A turning point in the economic history of the near east,’ International Journal of Middle East Studies 6: 3–28. https://doi.org/10.1017/S0020743800024302 Binney, JED. 1958. British Public Finance and Administration, 1774–92. Oxford: Clarendon. Boyer-Xambeu, MT., Deleplace, G. and Gillard, L. 1994. Private Money & Public Currencies: The 16th Century Challenge. Azodi, A. (trans.). Armonk, NY: ME. Sharpe. Brennig, JJ. 1983. ‘Silvers, in seventeenth-century Surat: Monetary circulation and the price revolution in Mughal India,’ in Richards, JF. (ed.) Precious Metals in the Late Medieval and Early Modern World, 477–98. Durham: Carolina Academic Press. Brewer, J. 1989. The Sinews of Power: War, Money, and the English State, 1688–1783. London: Unwin Hyman. Burnett, J. 1969. A History of the Cost of Living. London: Pelican. Chen, TS. 1936. ‘Shixing fabi hou de Julu nongcun,’ Dongfang zazhi 33(6): 113–14. Chen, XR. 2012. Minguo xiaoquyu liutong huobi yanjiu. Beijing: Zhongguo shehui kexue chubanshe. Dai, JB. and Chen, XR. 2011. ‘20 shiji 20, 30 niandai beifang nongcun qianpiao fanlan de kaocha,’ Zhonguo jingjishi lundan 1–11. Dickson, PGM. 1967. The Financial Revolution in England: A Study in the Development of Public Credit, 1688–1756. London: Macmillan. Echaúz, R. 1978. Sketches of the Island of Negros. Hart, DV. (trans.). Athens, OH: Ohio University Center for International Studies. Flynn, DO. and Giráldez, A. 1995. ‘Born with a “silver spoon”: The origin of world trade in 1571,’ Journal of World History 6(2): 201–21. Flynn, DO. and Giráldez, A. 2002. ‘Cycles of silver: Global economic unity through the mideighteenth century,’ Journal of World History 13(2): 391–427. Fontaine, L. 1996. History of Pedlars in Europe. Whittaker,V. (trans.). Cambridge: Polity Press. Frank, AG. 1998. ReOrient: Global Economy in the Asian. Berkeley: University of California Press. Fritschy, W., Hart, MT. and Horlings, E. 2012. ‘Long-term trends in the fiscal history of the Netherlands, 1515–1913,’ in Bartolome, YC. and O’Brien, PK. (eds.) The Rise of Fiscal States: A Global History 1500–1914, 39–66. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781139004237.003 Fujimoto, T. 1991. ‘Tokugawaki ni okeru shogaku tsuka: senka to hansatsu wo chushin ni,’ Shakai keizai shigaku 57(2): 179–205. Gelderblom, O. 2013. Cities of Commerce: The Institutional Foundations of International Trade in the Low Countries, 1250–1650. Princeton: Princeton University Press. Giraldez, A. 2012. ‘Cacao beans in colonial México: Small change in a global economy,’ in Munro, J. (ed.) Money in the Pre-industrial World. London: Pickering & Chatto. Giraud, Y. 1997. Voyages et aventures aux Grandes Indes (1617–1627). Pairs: Les Editions de Paris. Glamann, K. 1958. Dutch-Asiatic Trade 1620–1740. Copenhagen: Danish Science Press. Grubb, F. 2018. ‘Colonial Virginia’s paper money, 1755–1774: Value decomposition and performance,’ Financial History Review 25(2): 113–40. https://doi.org/10.1017/ S0968565018000057 Hamilton, EJ. 1934. American Treasure and the Price Revolution in Spain. Cambridge, MA: Harvard University Press.

The world diversified and stratified  149 Hammarstrom, I. 1957. ‘The price revolution of the sixteenth century: Some Swedish evidences,’ Scandinavian Economic History Review 5(2): 118–54. https://doi.org/10.1080/035 85522.1957.10411396 Helleiner, E. 2003. The Making of National Money: Territorial Currencies in Historical Perspective. Ithaca: Cornell University Press. Irigoin, A. 2009. ‘Gresham on horseback: The monetary roots of Spanish American political fragmentation in the nineteenth century,’ Economic History Review 62(3): 551–75. https:// doi.org/10.1111/j.1468-0289.2008.00430.x Iwao, S. 1953. ‘Kinsei nisshi boeki ni kansuru suryo teki kosatsu,’ Shigaku zasshi 62(11): 981–1020. Kagawa, T. 1996. Kinsei daimyo kinʾyushi no kenkyu. Tokyo:Yoshikawa Kobunkan. Kerridge, E. 1988. Trade and Banking in Early Modern England. Manchester: Manchester University Press. Keynes, JM. 1971. A Treatise on Money:The Applied Theory of Money. London: Macmillan. Kimura, M. 1989. Sakoku to siruba rodo. Tokyo: Saimaru shuppan. Kishimoto, M. 1997. Sindai Chugoku no bukka to keizai hendo. Tokyo: Kenbun shuppan. Kobayashi, N. 2015. Meiji ishinki no kahei keizai. Tokyo: Tokyo daigaku shuppankai. Kuroda, A. 1994. Chuka teikoku no kozo to sekai keizai. Nagoya: Nagoya daigaku shuppankai. Kuroda, A. 1996. ‘20 seiki shoki Taigen-ken ni miru chiiki keizai no genki,’ Toyoshi kenkyu 54(4): 103–36. Kuroda, A. 2005a. ‘The collapse of the Chinese imperial monetary system,’ in Sugihara, K. (ed.) Japan, China and the Growth of the Asian International Economy, 1850–1949, 103–26. Oxford: Oxford University Press. doi:10.1093/acprof:oso/9780198292715.003.0005 Kuroda, A. 2005b. ‘Copper-coins chosen and silver differentiated: Another aspect of “silver century” in East Asia,’ Acta Asiatica 88: 65–86. Kussmaul, A. 1990. A General View of the Rural Economy of England, 1538–1840. Cambridge: Cambridge University Press. Kuwahara, M. 1943. Echigo rokusaiichi no kenkyo. Tokyo: Nishigahara kankokai. Le Goff, TJA. 1973. ‘An eighteenth-century grain merchant: Ignace Advisse Desruisseaux,’ in Bosher, JF. (ed.) French Government and Society 1500–1850. Essays in Memory of Alfred Cobban, 92–122. London: Athlone Press of the University of London. Lin, ZX. 2002. Lin zexu quanji. Fuzhou: Haixia wenyi chubanshe. Lurvink, K. 2014. ‘Strapped for cash: Non-cash payments on Louisiana cotton plantations, 1865–1908’, Tijdschrift voor Sociale en Economische Geschiedenis 11(3): 123–51. doi: 10.18352/tseg.147 Maczak, A. 1976. ‘Money and society in Poland and Lithuania in the 16th and 17th centuries,’ Journal of European Economic History 5: 69–104. Mahapatra, PR. 1969–1970.‘Currency system in medieval Orissa,’ Quarterly Review of Historical Studies 9(2): 72–80. Mantetsu shanghai jimusho chosashitsu. 1940. Kososho Jojukuken noson jittai chosa hokokusho. Shanghai: Minami manshu tetsudo shanghai jimusho. Matsui,Y. 1935. Shincho keihi no kenkyu. Dalian: Minami manshu testudo. Mitra, DB. 1991. Monetary System in the Bengal Residency. Calcutta: Firma K.L. Mukhopadhyay & Sons. Mizuno, K. 1935. Santo no ichi noson ni okeru shakai keizai jijo. Dalian: Minami manshu tetsudo. Motomura, A. 1994. ‘The best and worst of currencies: Seigniorage and currency policy in Spain, 1597–1650,’ Journal of Economic History 54(1): 104–27. https://doi.org/10.1017/ S0022050700014017 Muldrew, C. 1998. The Economy of Obligation. London: Macmillan.

150  Global history of monetary delocalization Muldrew, C. 2012. ‘ “Th’ancient distaff ” and “whirling spindle”: Measuring the contribution of spinning to household earnings and the national economy in England, 1550–1770,’ Economic History Review 65(2): 498–526. https://doi.org/10.1111/j.1468-0289.2010.00588.x Nakamura, S. 1968. Meiji Ishin no kiso kozo: Nihon shihon shugi keisei no kiten.Tokyo: Miraisha. Nakamura, S. 1991. Nihon shoki shihon shugi shiron. Kyoto: Mineruva. Nakagawa, S. 2003. Osaka ryogaesho no kin’yu to shakai. Osaka: Seibundo. Pamuk, S. 2000. A Monetary History of the Ottoman Empire. Cambridge: Cambridge University Press. Prakash, O. 2004. Bullions for Goods: European and Indian Merchants in the Indian Ocean Trade 1500–1800. New Delhi: Manohar. Pressnell, LS. 1956. Country Banking in the Industrial Revolution. London: Clarendon. Sabean, DW. 1990. Property, Production, and Family in Neckarhausen, 1700–1870. Cambridge: Cambridge University Press. Shanghai yinhang diaochabu. 1932. ‘Shinianrai Shanghai xianjin liutong zhi guancha,’ Yinhang zhoubao 16(43): 22–5. Shi, ZH. 2008. Qingdai hubu yinku shouzhi he kucun tongji. Fuzhou: Fujian renmin. Spooner, FC. 1972. The International Economy and Monetary Movements in France, 1493–1725. Cambridge, MA: Harvard University Press. Supple, BE. 1957. ‘Currency and commerce in the early seventeenth century,’ Economic History Review, second series 10(2): 239–55. doi:10.2307/2590860 Tajiri, I. 1892. ‘Oini chiho ginko no hattatsu wo nozmu,’ Tokyo keizai zasshi 609: 158–66. Tawada, M. 2007. Kinsei shinshu no kokumotsu ryutsu to chiiki kozo. Tokyo:Yamakawa. Toa Dobunkai. 1920. Shina shobetsu zenshi, vol. 17. Tokyo: Toa Dobunkai. Tsurumi, M. 1991. Nihon shinyokiko no kakuritsu: Nihon ginko to kin’yu shijo.Tokyo:Yuhikaku. Usher, AP. 1943. The Early History of Deposit Banking in Mediterranean Europe. Cambridge, MA: Harvard University Press. Valenze, D. 2006. The Social Life of Money in the English Past. Cambridge: Cambridge University Press. van der Wee, H. 1977. ‘Monetary, credit and banking systems,’ in Rich, EE. and Wilson, CH. (eds.) Cambridge Economic History of Europe, vol. 5, 290–392. Cambridge: Cambridge University Press. https://doi.org/10.1017/CHOL9780521087100.006 Vice, D. 1983. The Coinage of British West Africa and St. Helena 1684–1958. Birmingham: Peter Ireland Format. Wakita, H. (ed.) 2002. Iwami ginzan. Kyoto: Shibunkaku. Wallerstein, I. 1974. Modern World-System: Capitalist Agriculture and the Origins of the European World-Economy in the Sixteenth Century. New York: Academic Press. Walter, J. and Wrightson, K. 1976. ‘Dearth and the social order’ Past & Present 71: 22–42. Wang, XN. and Liu, JM. 2001. Zhongguo Shanxi minjian piaotie. Beijing: Zhonghua shuju. Yasukuni, R. 2016. Nihon kinsei kaheishi no kenkyu. Kyoto: Shibunkaku. Yoji, S. 1982. Igirisu shin’yo kaheishi kenkyu. Fukuoka: Kyushu daigaku shuppankai. Young, JP. 1925. Central American Currency and Finance. Princeton: Princeton University Press. Zhang, GH. 1989. Wanqing qianzhuang he piaohao yanjiu. Beijing: Zhonghua shuju.

5 Nationalized money Backstage of the international gold standard regime, c.1900

1 The age of international silver dollars: an alternative to the territorial currency system The viewpoint of the three different paths local monies took after the global silver march c.1600, as argued in the previous chapter, suggests that differences between the modern economic situation society by society has depended on conditions in the past, particularly on how a particular society responded to the general monetary contraction in the early modern period. After first accelerating the divergences in monetary conditions in different places, the three patterns ended up forming the modern world economy together. However, in the late 19th century at the latest, nationalized monies began to spread throughout the world, and the difference between the three paths subsequently became obscured. This apparent convergence emerged with the proliferation of the gold standard formula. Contrary to the common understanding that territorial state currencies became dominant with the advent of the international gold standard, as argued by scholars such as Helleiner, this chapter will reveal that alternative monetary systems survived beyond state control in many places around the world, and the incorporation of multiple monetary devices operating at the ground level into single national systems was not yet complete in the period before World War II. Table 5.1 indicates the long-term movement of precious metal production across the world. Until the first half of the 19th century silver production increased faster than that of gold and was accompanied by lower silver-gold exchange rates. In contrast, however, from 1851 to 1874, gold production increased more rapidly than silver but was not accompanied by a depreciation of gold against silver. This period, known for its ‘gold rushes,’ was followed by another period in which major countries shifted their monetary systems to the gold standard, under which silver became a material used only for subsidiary coinages. This new trend thus clearly shows that the institutional framework of the gold standard gave gold a higher value in spite of an increasing supply which might otherwise have caused gold to depreciate. Table 5.2 shows us what changes the adoption of the gold standard system in 1879 brought to the circulation of currencies in the US. Comparing the

152  Global history of monetary delocalization Table 5.1 Estimates of annual production of gold and silver in the world accessible from Europe periods

gold (kg) annual product (a)

silver (kg) annual product (b)

exchange rate in Europe (c)



1493–1600 1601–1700 1701–1800 1801–1850 1851–1879

6,990 9,123 19,001 23,698 187,973

211,400 372,340 570,350 654,480 1,312,300

11.5 14 15 15.7 15.85

80,385 127,722 285,015 372,058.6 2,979,372

0.380251 0.343025 0.499719 0.56848 2.270344

Source: Conférence monétaire internationale avril-mai 1881, procès-verbaux. Paris: Imprimerie Nationale, 1881, 59.

Table 5.2  Monies in circulation in 1879 and 1905, US (dollar) 1905/1/1 Gold coin (including bullion in Treasury) Gold certificates Standard silver dollars Silver certificates Subsidiary silver Treasury note of 1890 United States notes National bank notes Total

649,548,528 466,739,689 80,039,395 468,017,227 102,891,327 10,940,054 342,287,627 449,157,278 2,569,621,125

1879/1/1 96,262,850 21,189,280 5,790,721 413,360 67,982,601 310,288,553 314,339,356 816,266,721

Source: Johnson 1906, 390.

breakdown in 1906 with that in 1879 in Table 5.2, spanning the period that the gold standard system became established, state-issued paper monies had become dominant in value. Between 1879 and 1905, the issuance of gold and silver certificates increased by 45 times, meanwhile subsidiary coins increased by less than 50 percent and national banknotes (actually issued by private banks under state license) by 40 percent. The rapid growth of monetary transactions in value was made possible by the issuance of certificates convertible with gold. It seems natural enough that the silver subsidiary currency occupying only 4 percent of the total money in circulation in value has been overlooked; scholars have tended to pay more attention to precious metal monies than to monies of non-precious materials. If we measure the total amount of transactions that currencies mediate, small denomination currencies such as subsidiary silver always play a role in only a small proportion of all exchanges. Meanwhile, large denomination currencies such as gold coins invariably mediate the vast majority of all transactions. However, even in 1905, a significant proportion of transactions in terms of frequency must still have been conducted with subsidiary coins. In 1879, when

Nationalized money  153

silver subsidiary coins occupied 8 percent of the value of the total currency, it might not be wrong to assume that the majority of transactions by frequency was made with small currencies. As Zelizer impressively discovered, in the same period when this estimate was made, when most payments depended on small subsidiary coinages, ordinary people distinguished them for purchasing different items through physically earmarking them for different purposes, as if they were different currencies (Zelizer 1994). However, the shift to the gold standard framework was not the only global monetary transformation to occur in the second half of the 19th century and early 20th century. In some regions certain silver dollars circulated internationally and their importance in transactions grew. The difference between these two paths was more than an issue of the choice of metallic material, gold or silver. Under the gold standard system, gold worked fully as a unit of account but did not function as a means of exchange to the extent that silver did under other systems. Instead, paper monies and subsidiary coinages (including silver coins) circulated to mediate transactions settled physically by currency. Those currencies thus had to be convertible with the gold standard coin in fixed ratios. Meanwhile, in the international silver dollar system, currencies ranging from small denominations to large ones were not exchanged at a fixed exchange ratio. Even among silver coinages, one silver dollar was exchanged for different quantities of silver cents depending on the demand/supply for each currency. Formally, the gold standard’s set of convertible currencies appears to provide a degree of certainty to people engaged in exchanges. By contrast, the international silver dollar system seems to have suffered from large transaction costs. However, from the viewpoint of flexibility, the international silver dollar system could adjust to situations in which monetary demand/supply varied in different tiers of the market: local, state-wide and interstate. Meanwhile, the gold standard system seems to have worked well for economic activities that were organized at the state-wide level. In short, the rivalry between the gold standard system and the international silver dollar system reflected a competition between two ways of organizing the global economy: in one, the international circulation of gold connected national economies in which convertible currencies mediated domestic transactions; in the other, international silver dollars circulated freely both domestically and abroad, and local economies adjusted independently to monetary demand/supply. Despite the popular perception today that the gold standard prevailed across the world in the second half of the 19th century, the circulation of international silver dollars actually increased during the same period. Many countries minted silver dollars of similar weight and fineness. In the late 19th century, the US, Britain, France and Japan issued large quantities of silver dollars, mainly for international circulation or usage outside their home countries: US Trade dollars 36 million, 1873–1887; British India 151 million, 1895–1903; French Indochina 68 million, 1885–1903; Japanese Trade dollars 165 million, 1871– 1897. The champions among them were the Mexican dollar in the East and the Maria Theresa dollar (minted in Austria) in the West: Mexico 674 million,

154  Global history of monetary delocalization

1874–1903; Austria 176 million, 1751–1903 (Andrew 1904, 356). The former dominated international trade in East Asia until World War I, while the latter played an important role in Africa and the Middle East throughout the 19th century and survived in the Red Sea region even until World War II. The prevalence of international silver dollars challenges the popular idea that money must be guaranteed by a state in order for people to accept it. Mexico had no political influence in East Asia, nor did Austria in Africa and the Middle East. Issuance of the Maria Theresa dollar actually increased even after the Austrian government demonetized its own territory in 1860. Territorial currencies, meaning that a set of currencies circulate homogeneously within a state by expelling foreign coins and domestic forgeries, emerged in some countries in the early 19th century. However, the popularity of foreign silver dollars did not decrease but rather increased in the latter half of the 19th century. We will see that in both the period in which the gold standard was ascendant, and in the period when it was already out of date, the Maria Theresa dollar continued to dominate transactions in Africa and the Middle East. The important concepts we confirmed in previous chapters, such as multi-layered markets, currency circuits, the stagnancy of currency and, above all, complementarity among monies, will all help us to understand these phenomena.

2 The riddle of the Maria Theresa dollar The Maria Theresa dollar (or Maria Theresa Thaler) issued in Vienna remained in circulation for over one and a half centuries in Africa and the Middle East, even though Austrian sovereignty never reached these regions. This riddle has yet to be fully explained. Particularly in the Red Sea region the coin bearing the image of the Empress Maria Theresa and the inscription 1780, the last year of her reign, was still popular during World War II.Though Ethiopian emperors, the Italian colonial government and the British Army each in turn attempted to substitute their own currencies for the Maria Theresa dollar, all such efforts eventually failed. Before World War II, the ‘mystery’ of the Maria Theresa dollar’s popularity attracted the attention of a number of social scientists, including Max Weber and J. M. Keynes, as well as numismatic students and other contemporaries (Weber 1924, 217; Keynes 1971, 12; Stride 1956). The most popular explanation was that the coin itself served as a guarantee of its silver content. In other words, although the effigy of the empress and the year on the coin seemed to matter to people in northern Africa and the Middle East, they must also have trusted its contents. Its physical appearance was a promise of its material worth (Robertson 1948, 48–9). In history, those who had territorial ambitions in Ethiopia or the Red Sea region shared this interpretation and tried to substitute their own silver coins for the Maria Theresa dollar. First, in the 1890s, Italy issued coins whose fineness was no lower than that of the Maria Theresa dollar. However, the Italians soon found that their coins were not accepted for circulation. The next

Nationalized money  155

challenger was a local ruler. Between 1897 and 1915, with the assistance of the Paris mint, the Ethiopian emperor Menelik attempted to introduce new coins just as Italy had. The popularity of the emperor himself could not, however, popularize his coins (Pankhurst 1968, 479–82). Though they were in no way inferior in content to the Maria Theresa dollar, people valued them 25 percent less. Eventually the emperor was thought to have accumulated between five and seven million Maria Theresa dollars by the time of his death (Pope 2001, 261, 266). These and many other failures to replace the Maria Theresa dollar suggest that people did not accept it solely due to its intrinsic value.1 To some extent, even the local rates of the Maria Theresa dollar were influenced by the international price of silver. As we will see later, the long-term trend of the dollar’s rate in terms of the pound sterling had a positive relationship with the fluctuation of silver prices in London. However, more important is the fact that, except for one period influenced by war and extraordinary speculation, local prices of the Maria Theresa dollar usually exceeded its intrinsic value in terms of the current silver price. The overvaluation of the Maria Theresa dollar gave the Austrian government an incentive to keep issuing it for one and a half centuries.2 The Red Sea region continued to attract the dollar at high enough rates to easily cover the cost of transportation from Vienna. The flow was overwhelmingly unidirectional: it appeared almost as though large quantities of the dollar disappeared into the desert.3 Thus, the metallic value of this particular coin does not persuasively explain the strong demand it enjoyed in these regions. At the same time, its popularity was clearly not enforced politically, as both native governments and colonial powers tried in vain to replace it. Unlike the pound sterling issued by Britain before World War II, or the US dollar after that, the Austrian dollar was by no means an international standard currency supported by a global hegemony. All in all, neither the coin’s intrinsic value nor any extrinsic pressure can provide a convincing explanation for what caused the Maria Theresa dollar to be accepted. The mystery remains unsolved and has encouraged the revival of the simplistic modernist view that primitive economies prefer a reliable currency of external origin to an internal creation of less certain value (Ida 1953, 14). Some have suggested that the use of the empress-bearing coin in ornaments and dowries indicates cultural reasons for its popularity.4 However, students as well as contemporary observers have failed to take into account the very points that we have argued in the preceding chapters. That is, the Maria Theresa dollar was not the sole currency in circulation but functioned in the region as one of several coexisting monies, which included the Indian silver rupee, the Italian lire (a paper currency), salt bars (amole) and forms of cloth. Not only contemporary rulers but also researchers have assumed the dollar to be substitutive for other monies (Pankhurst 1962, 213–47). Consequently, they have not attempted to determine if other monies could have done what the Maria Theresa dollar could do and vice versa. In other words, the complementary relationship between monies has rarely been considered in

156  Global history of monetary delocalization

this context.Yet herein lies a clue to the question posed at the start. The dollar alone could not mediate between buyers and sellers: it could work well only in association with other monies. The key to solving the mystery is the nature of the complementary relationship among the monies that circulated alongside it. It is not clear when and how the Maria Theresa dollar began to circulate in the Middle East and northern Africa. As early as 1762, a Danish geographer found the dollar during a journey to the Arabian Peninsula.5 Though this has not been conclusively proven, it has been said that the dollar’s circulation in the Middle East commenced with the purchase of cotton from Egypt and coffee from Moca (Mukha).6 In 1780, Empress Maria Theresa passed away. The dollar bearing her effigy and the year 1780, however, continued to be cast after her death. During the 19th century, the geographical sphere of the Maria Theresa dollar’s dominance gradually shrank. A map in the 1898 Geschichte des Maria-Theresien Thalers shows that at this time the dollar circulated in Ethiopia, Eritria, the Arabian Peninsula and northern Nigeria but had disappeared from Egypt and the coastal area of the Black Sea (Peez and Raudnitz 1898, 86). A prohibition against its use by the Ottoman Empire may have prompted this shrinking in the Middle East, while in West Africa the British colonial policy of substituting its own silver coins for the Maria Theresa dollar began to take effect after 1900 (McPhee 1971, 233–7). Despite the spatial retreat of the Maria Theresa dollar in the 19th century, the quantity cast by the Vienna mint continued to increase until after World War I (Figure 5.1).7 Increased exports of local products from the Middle East and northern Africa may have stimulated the demand for the Maria Theresa dollar. In Ethiopia, the development of coffee exports, unlike the previous century’s slave trade, is thought to have increased demand for the dollar.8 The coffee price in Addis Ababa was 2.25–2.50 dollars per farasulla (16.83 kilos) in 1899, rising to 5–6.50 dollars in 1913 and 10 dollars in 1926. Ethiopian coffee exports via Djibouti were only 619 metric tons in 1915–1919, but rapidly increased to 8,991 tons in 1920–1924, 34,337 tons in 1925–1929 and 42,115 tons in 1930–1934 (McClellan 1980, 74–7).9 It is clear that rising prices stimulated the rapid expansion of coffee exports from Ethiopia. At the beginning of the 20th century, Ethiopia, one of the few independent African states, was in the process of establishing Western-style public finances, including a national bank. The government paid its officers and soldiers in Maria Theresa dollars. According to MacGillivray’s report, the sum of 1,651 thousand Maria Theresa dollars was collected as a part of the revenue for the Emperor in 1904.10 As seen later, this accounted for more than half of the entire revenue for the year. Thus, the administration relied heavily on the popularity of the Maria Theresa dollar. It was not surprising that the Bank of Abyssinia issued its bank notes denominated in the Austrian dollar (Pankhurst 1965, 396). Considering that even the founder of Addis Ababa, Emperor Menelik, failed to establish his own legal tender, it can be said that the high popularity of the dollar resisted all attempts to substitute it. MacGillivray stated that the Maria

Nationalized money  157 16000 14000 12000 10000 thousand

8000 6000 4000 2000 0 1867





Figure 5.1 Annual issuance of MTD from Vienna Source: Hans 1961, 18.

Theresa dollar cost 2 fr. 22 c. in Europe, and 2 fr. 35 c. at Aden and cost about 3 c. to transport it from Aden to Addis Ababa.11 The price gap was far larger than the transportation cost from the coast to Addis Ababa (about 1 percent of the price), and we therefore know how highly the Maria Theresa dollar appreciated there.12 The large profits to be had from minting the Maria Theresa dollar must have been what encouraged other European mints to start issuing the coin.The Italian government in particular was keen to issue the dollar. Unlike other rivals, Italy had inherited the dies of the Maria Theresa dollar when Austria ceded Milan and Venice (which both had mints) in the 19th century. In addition, territorial ambitions in East Africa were a strong reason for the Italian government’s interest. Vienna ceded the right to issue the Maria Theresa dollar to Rome in 1935, the very year Italy under Mussolini began its invasion of Ethiopia. In the course of the war, the Maria Theresa dollar became a crucial topic, since, as mentioned earlier, it had appeared to serve as almost legal tender in Ethiopia. That is why the acquisition of the right to issue the key currency carried a great advantage. Thus the mint in Rome began to issue large quantities of the Maria Theresa dollar for their military operations in rural regions of Ethiopia, meanwhile, in order to control Ethiopian economy, an embargo on the dollar’s shipment to Ethiopia was declared.13 However, the embargo by the Italian government caused exporters of local products in the Red Sea Region to suffer from shortages of the Maria Theresa dollar. At that time, the intrinsic value of the coin and the cost of transportation from Europe to Aden was as high as 1 shilling 5 pence for each Maria Theresa

158  Global history of monetary delocalization

dollar. However, the exchange rate in Aden, which was under British sovereignty, recorded 2 shillings 1 pence or more per dollar.14 In 1936, London also commenced minting the Maria Theresa dollar. Paris and Brussels then joined the race. Claims from Rome that any minting outside Italy was illegal had no effect. The clearest argument against the Italian assertion was supplied by the French government. They insisted that, once the original issuer, Austria, had abandoned its right of issuance, the dollar was no longer a currency belonging to a specific country but a medal that happened to be exchanged. According to this logic, it was quite legal for the French mint to issue the very same ‘medal,’ responding to requests from French nationals.15 As long as the gap between the cost of collecting the Maria Theresa dollar in the Red Sea region and that of importing it from Europe was maintained, European mints could earn profits from issuing the dollar. Rome minted 19,446,729 dollars between 1935 and 1939, Paris 4,512,750 dollars, 1935–45 and Brussels 9,845,000 dollars, 1937–38 (Hans 1961, 19) and London 4,700,000 dollars, 1936–38.16 In spite of the Italian embargo, huge amounts of dollars entered Ethiopia. This was the result of Italy’s failure to popularize the lire currency.17 Even in the regions occupied by Italian troops, the dollar remained necessary for trade. As long as such strong demand continued, it was impossible to prevent the dollar from entering Ethiopia illegally. In the autumn of 1938, the exchange rate fell to 1 shilling 5 pence per dollar, which rendered shipments from European mints almost unprofitable.18 Shortly after this, Italy completed its annexation of Ethiopia. Following the outbreak of World War II, the battle between Britain and Italy in northeast Africa again made the Maria Theresa dollar a key issue. War made it necessary to collect materials urgently in a limited time for military use. In particular, it was necessary to secure food for dispatched troops in order for them to continue fighting. In purchasing food, British as well as Italian troops could not help but depend on the Maria Theresa dollar, since Arabian and other merchants that dominated markets in the Red Sea region preferred payment in this currency for the commodities they supplied.19 The Bombay mint became the major supplier of the Maria Theresa dollar at this stage. Bombay had advantages over London in minting. Not only was it closer to Aden, but Bombay also had an ample stock of silver, since the colonial government in India had begun to substitute paper currency for the silver rupee. German submarine attacks made transportation from London extremely difficult. Thus, military urgency made it necessary to establish a new mint in a location closer to the East African front. The Bombay mint struck the ‘Austrian’ dollar from December 1940 to July 1941. Its total issue amounted to 18,864,537 dollars.20 This quantity was far greater than that of the Royal Mint, which had minted 4.7 million dollars from 1936 to 1938. Between December 1940 and August 1941, 9.5 million dollars were shipped from Bombay to Khartoum, 3 million to Nairobi, 0.5 million to Aden and 2 million to Berbera. Totally 15 million dollars were shipped, while 3.8 million dollars ultimately remained unshipped at Bombay. Three months

Nationalized money  159

later, after the last ship carrying the dollars left Bombay.21 Britain finally won the battle against Italy in northeast Africa and gained a safe route through the Red Sea. In February 1943, 0.7 million dollars in Khartoum, 0.9 million in Nairobi, 4.7 million in Aden, 3.2 million in Harar and 2.5 million in Asmara were estimated to be stocked.22 The Maria Theresa dollar had taken root so deeply in the Red Sea region that Britain had to bring an imitation of it from Bombay in order to win victory on the East African front. In the following section, we shall see what made this silver dollar so indispensable to business in regions far from its place of issue.

3 The reality of the Maria Theresa dollar’s circulation The most detailed information on the circulation of the Maria Theresa dollar was conveyed by a correspondent whom C.S. Collier, former managing director of the National Bank of Abyssinia, dispatched to Aden.23 This report was made in July 1937, after the Italian embargo seriously affected local trade. The exchange rate of the Maria Theresa dollar at Aden in August 1937 was 2 shillings or more, which far surpassed the total cost of minting in Europe and transport to Aden. The correspondent was sent to enquire into the reason for the appreciation:24 I have made careful enquiries and I find that the final destination of Maria Theresa dollars shipped to Aden is Arabia (Hodeidah, Makalla, Shehr and Jeddah).   The dollars that we have so far sold in Aden to the merchants known as Shroffs are not used by them, but they re-sell them mostly to big Arab firms and to firms like – A. Besse, F. Livierato, B. Tivari etc., and to small local traders. The Arab firms and the other firms mentioned are buying dollars solely for the purpose of sending them to places like Hodeidah, Makalla, Shehr and Jeddah to buy Hides, Skins, Coffee, Incense, Honey and other export produce. Local dealers who buy dollars are generally Banians and Jews who sit in the open market and sell dollars to Arabs and Somalis coming from the interior. The daily retail sale in the market done by those local dealers amounts to: from October to April (full season) Four to Eight thousand dollars, and from May to September (slack season) Two to Four thousand dollars per day. There are also Banian and Jew speculators who buy dollars in the market from the Shroffs and re-sell them when the price offers a little benefit to local Arab and other firms, and to local retail dealers. (Aden, 30 June 1937) This report identifies three points regarding the circulation of the Maria Theresa dollar. First, the dollars imported to Aden were transported to other districts on the Arabian Peninsula. In fact, the destinations of the transferred dollars

160  Global history of monetary delocalization

were not only cities on the peninsula but also those in East Africa across the Red Sea. Second, as expected, the dollar was used for the purchase of export products such as hides. Third, demand for the dollar fluctuated seasonally.25 Daily sales in the busy season were twice as high as in the slack one. The last point vividly shows that the dollar was bought and sold for other monies such as the Indian rupee or the pound sterling. The need to exchange currencies suggests that one money could do what another could not. The Austrian dollar appeared to be especially preferable in collecting export goods like hides and was taken to places producing them. What this report fails to mention is that the Maria Theresa dollar was too expensive for ordinary people to use in daily transactions. The lack of subsidiary coins was often mentioned as a defect of this popular Austrian dollar. The shortage of smaller denomination currency in this region was frequently noted by contemporary observers. In 1918, an American report stated that only about one million dollars’ worth of small coins were in circulation in Ethiopia. At that time, the number of Maria Theresa dollars in circulation was estimated at five million, and that of dollars buried or stored between 25 and 50 million (Pope 2001, 267). However, no doubt bars of salt, iron, rifle cartridges, glass beads and other items accounted for the majority of ‘small monies’ used in daily transactions (Pankhurst 1968, 486).26 Table 5.3 shows that salt and cloth accounted for 20–30 percent of the Ethiopian emperor’s revenue, but local markets might have been far more dependent on them.The point is not that commodities worked as currencies, but that local preferences dictated their use. In the early 20th century, the number of pieces of salt, amole, exchangeable for a dollar at Addis Ababa was 4, while it was 10 at Goré. Regional preferences were clearer in beads. In Dizu, small transparent

Table 5.3  Revenue and expenditure of the Ethiopian emperor Revenue (dollar) Gold Dollars Salts Ivory Cloths Total Expenditure (dollar) Gold Dollars Salts Ivory Cloths Total Source: PRO, FO 401/8, 50.

1902 327,560 2,069,122 854,427 175,100 216,720 3,642,929

1903 384,840 903,887 904,988 98,770 128,485 2,420,970

1904 501,000 1,650,763 666,085 201,280 112,300 3,131,428




17,520 920,941 647,864 166,600 174,630 1,927,515

93,080 645,977 686,679 170 119,970 1,545,876

95,040 933,450 421,274 175,950 48,785 1,674,499

Nationalized money  161

sea-blue beads the size of peas exchanged at the rate of 1,200 to the dollar, and no other type was acceptable. In Chako and Gurafarda, on the other hand, the preference was for opaque violet-blue hexagonal beads or small opaque white beads (Pankhurst 1968, 460–6). More important is that the exchange rate between the Maria Theresa dollar and local small monies fluctuated from day to day.27 Incidents like famine and military actions made the exchange rate of salt appreciate against the dollar. In 1905, MacGillivray wrote on Ethiopia, ‘Nearly every tenth man is a money changer evidently doing good business.’28 Wherever small money is always in use, a small money exchanger can make profits. Smaller denomination currencies used locally for daily transactions were not always in a traditional or primitive form such as the salt bar. The circulation of 10-lire notes provides a clear example. Rome had originally planned to replace the Maria Theresa dollar with the lire currencies after its annexation of Ethiopia. However, the replacement completely failed as, in spite of the embargo, the Maria Theresa dollar continued to circulate (Pankhurst 1970b). However, the 10-lire note was an exception to this. Unlike other Italian currencies, the 10-lire note became so popular in local markets that the British army could not replace it with East African currencies and had to recognize its use in the region it occupied.29 The success of the 10-lire note, valued at a sixth to a quarter of the Maria Theresa dollar, suggests that local markets had a serious need for such a denomination.30 Although Britain set a fixed rate between the Maria Theresa dollar and the 10-lire note, the active rate fluctuated constantly and varied in different regions. For example, in February 1942, the rate was 60 lire per dollar in Massawa, a port by the Red Sea in Eritrea, while it was 48 lire in Asmara, an inland city. The distance between the two towns is as much as 50 kilometres. The sudden outflow of the Maria Theresa dollar from Massawa caused it to appreciate. The leakage of the dollar was a result of payments for imports from Yemen, as the owners of the Dhow ships in Massawa engaging in coastal trade transferred a large quantity of Maria Theresa dollars.31 This incident suggests that the dollar worked as a measure for payments in distant trade, while the 10-lire note was used for relatively local transactions. Another episode supports this assumption. In the period that the Latin Monetary union operated in Europe, the Italian government once issued 5-franc coins for Eritrea. However, Rome found that the francs were useless because they needed to purchase the Maria Theresa dollar for payments outside Eritrea. As a result, they had to recall two-thirds of the 5-franc coins and abandon them.32 This failure makes it clearer that the demand for Maria Theresa dollars was closely related to payments beyond local markets. From the viewpoint of international business conducted in pounds sterling or US dollars, the Maria Theresa dollar worked as a ‘local’ currency. However, for local traders, the foreign silver coin played a significant role as an interregional currency.The report from Aden at the beginning of this section proves that this assumption is persuasive. It is important to note that such a division of function among monies

162  Global history of monetary delocalization

was not organized through any form of regulation but appeared spontaneously through local trade. Given such a weakly regulated system, it would be a great challenge to establish any monetary principle with fixed rates among concurrent currencies. In East Africa, during World War II, the British government intended to allow the currencies to circulate only at fixed rates. For example, in the autumn of 1941, they allowed notes and coins with face values of less than 10 lire issued by the former Italian government to be in circulation at the following rates: 492 lire = 1 Egyptian pound, 480 lire = 20 East African shillings, 36 lire = 1 rupee, 45 lire = 1 Maria Theresa dollar.33 In order to respond to the demand for small denomination currencies, these less-than-10-lire currencies were accepted, but fixing exchange rates was thought to be indispensable to control the markets. However, setting fixed rates was one thing; maintaining them was another. In the case of the rates for the Maria Theresa dollar, its active quotation in markets was always far from the official rate. In Aden, May 1942, the Bank of Aden exchanged 140 rupees for 100 dollars (MTD) according to the official regulation, but 100 dollars were traded for 175 rupees in the black market. The truth was that, in purchasing vegetables for the British navy from Yemen, payment had to be made in Maria Theresa dollars, or the black market rate had to be adopted if paid in rupees.34 Britain had struggled to keep the ratio of 1 dollar (MTD) = 1 shilling 10.5 pence, which was not far from its intrinsic value. However, as late as the end of 1942, active rates in major markets in this region were quoted at more than 3 shillings. As a result, the occupation government raised its official rate in the food trade to 3 shillings in order to avoid a serious loss in exchange.35

4 Currencies working as complementary buffers: currency circuits survived Thus, we can find three layers of monetary circulation. Above the layer of the dollar’s circulation, the pound sterling or its convertible rupee flowed in international and interregional circuits. At the boundaries between the two layers, depending on fluctuating exchange rates, native exchangers such as shroffs and traders engaging in interregional trade competed for profits through speculation. At the same time, below the dollar’s flow, a variety of smaller monies, such as the 10-lira note, copper coins, rifle cartridges, cloth, salt bars and beads, were in circulation with strong regional preferences. At the border between the lower two layers, there were many small but prosperous businesses interchanging the currencies. Whether in the international market or local markets, the Maria Theresa dollar and other currencies were exchangeable, but without fixed rates. The rates fluctuated from day to day and differed by region. In the triple-layer monetary system, the Maria Theresa dollar might well have worked as a buffer between international and local currencies. Few statistics regarding this region remain for us to test this hypothesis, but fortunately a series of monthly rates of the

Nationalized money  163

Maria Theresa dollar in Addis Ababa is recorded in reports from the British consul. Table 5.4 shows monthly movements of the silver price in London and the exchange rate of the Maria Theresa dollar to the pound sterling in Addis Ababa from January 1926 to June 1927, from November 1928 to December 1929 and from October 1930 to June 1935. The co-efficient of correlation between the two trends during the first period is −0.95, the second period −0.96 and the

Table 5.4 Monthly movements of the Maria Theresa dollar rate in Addis Ababa and silver price in London 1926 January–1929 December

1926 Jan

1926 Jul

1927 Jan

1927 Jun 1928 Nov 1929 Jan

1929 Jul

MTD rate in Addis Ababa, dollar per £

Silver Price in London, pence per ounce

8.43 8.56 8.87 9.06 9.1 9.05 9.35 9.42 9.33 10.05 9.95 10 10.02 9.88 10.11 10.15 10.05 10.01 10.27 10.34 10.36 10.38 10.45 10.92 11.64 11.58 11.68 11.79 11.85 11.89 12.25 12.72

31.31 30.8 30.3 29.68 30.13 30.25 29.85 28.77 27.89 25.27 25.2 24.73 25.86 26.85 25.66 26.14 26.05 26.2 26.71 26.35 26.25 25.91 26 25.74 25.11 24.31 24.3 24.29 23.69 23.04 22.69 22.26 (Continued)

Table 5.4 (Continued) 1930 October–1934 June

1930 Oct 1931 Jan

1931 Jul

1932 Jan

1932 Jul

1933 Jan

1933 Jul

1934 Jan

1934 Jun

MTD rate in Addis Ababa, dollar per £

silver price in London, pence per ounce

15.82 16.82 19.5 19.88 20.18 20.21 20.22 20.53 21.73 21.74 22.16 20.26 18.32 16.77 16.04 16.14 16.15 16.95 17.75 17.75 18.07 17.87 16.95 16.49 16.6 16.37 16.41 16.33 16.27 16.14 16.26 16.19 16.24 16.32 16.35 16.38 16.44 16.4 16.41 16.26 15.65 15.07 15.22 15.35 15.48

16.56 16.63 15.2 13.79 12.42 13.52 13.21 12.86 12.71 13.18 12.82 14.1 17.15 19.39 20.02 19.6 19.57 18.34 16.92 16.87 16.84 16.94 18 18 17.81 18.1 17.11 16.88 16.89 17.59 18.44 19.05 19.08 18.32 17.88 18.28 18.22 18.41 18.67 19.38 20.07 20.28 19.74 19.28 19.98

Sources: MTD rate 1926–1929, Bank of England Archives OV133/1, 1930–1934, PRO, FO 401/28, ‘Addis Ababa Intelligence Report for the Quarterly,’ passim; Silver, Treasury Department, Bureau of the Mint, Annual report of the Director of the Mint for the fiscal year ended June 30, (Washington) passim.

Nationalized money  165

last period −0.92, so the dollar’s price in Addis Ababa appears to have followed silver’s in the international market. During the late 1920s, increasing coffee exports must have caused the demand for the Maria Theresa dollar to synchronize with international silver price movements. However, in the shorter term, we can notice that the correlation between the silver price and the dollar’s exchange rate depended on the conditions of the export trade from Ethiopia. The co-efficient was −0.94 during 1931 but decreased to −0.79 during 1932 and became almost zero (−0.02), during 1933. Behind these movements in exchange rates lay rapid changes in international demand for coffee produced in Ethiopia. An export tax instituted in Brazil in 1931 created unfavourable conditions for Brazilian coffee and allowed Ethiopian coffee to become successful in the international market. After the return of Brazilian coffee the next year, however, Ethiopia’s coffee lost its competitiveness and unsold coffee was stockpiled in the territory in 1933.36 Unlike during 1931 and 1932, the exchange rate of the Maria Theresa dollar in Addis Ababa moved little during 1933. This indicates that the stagnation of coffee exports did not increase demand for foreign exchange in the middle of the year as before and consequently decoupled the dollar’s domestic rate from international intrinsic quotations. It is probably safe to conclude that favourable export trade brought the dollar value in the local market into synchronization with the silver price in the international circuit, while the disruption of that trade kept them apart. It is true that arbitrage between the international silver price and the local price of the Maria Theresa dollar influenced the movement of the Austrian coins. However, this did not mean that local demand for it could not exist independently from the business of silver dealers.The Maria Theresa dollar appeared to serve as a device that switched local markets on or off the international market according to whether conditions were favourable or unfavourable for export. Far from the simple but popular view that a currency is accepted due to its intrinsic valuation, the Austrian dollar in Ethiopia worked as an interface between two widely separated levels of monetary circulation: the pound sterling and various local currencies such as salt bars. Could anyone have intentionally introduced such a flexible buffer? The process by which the 10-lire note gained unexpected popularity under British rule suggests that it would be unlikely. The Italian colonial government valued one Maria Theresa dollar at five lire after their occupation of Addis Ababa in 1936, but in the markets it continued to appreciate. Consequently the rate of the dollar against the lire was frequently revised upwards, as would happen later under British rule. With the fall of the Italian colony in 1941, the lire currencies seriously depreciated and almost disappeared. However, this depreciation stopped at between 40 and 60 lire to a Maria Theresa dollar, and, unlike other lire currencies, the 10-lire note survived and remained in circulation under British rule.37 The reason was simply that a currency valued at a quarter or a sixth of the dollar was required in local transactions; despite its depreciation to one tenth of the initial official rate, the 10-lire note happened to fill the gap between the Maria Theresa dollar and currencies of far lower denominations. Nobody had designed such a

166  Global history of monetary delocalization

complementary relationship among currencies. It was produced by the markets themselves. While the demonetization of the Maria Theresa dollar occurred in such regions as West Africa and Egypt, it retained a significant role in the Red Sea region. What were the differences between the regions that demonetized the dollar and those that kept it? Increasing exports of peasant products and large gaps in prices between international and local markets, as mentioned in the previous section, must have been common phenomena throughout the Middle East and East Africa. The elastic supply of local currency was also recognized in Egypt as a crucial problem in purchasing raw cotton from peasants (Arminion 1911, 392). Nevertheless, the Maria Theresa dollar disappeared from Egypt. A report from a British attaché in Cairo in 1887 stated that the dollar had become useless, and that they wanted to dispose of their stock of the dollar.38 During World War II, a considerable number of dollars poured into Egyptian territory, but they were thought to have been melted down for use as silver.39 Why did such developments not occur in the region that included Ethiopia and the southern Arabian Peninsula? Gathering miscellaneous information from local reports, we find that the flow of Maria Theresa dollars formed a round circuit connecting certain city markets (Kuroda 2007, 104). First, Maria Theresa dollars landed at Aden and were transported by Arabian or Indian merchants to regions producing export commodities. The dollars for purchasing coffee were sent to Gore, in western Ethiopia, where the route of the dollar divided into two streams. Some dollars were collected through levying taxes to be sent to Addis Ababa, where they were distributed to soldiers and officials as salary. Some of them must have been transported for trade in grains or cotton textiles (mostly Japanese-made) via Djibouti, before finally returning to Aden. Other dollars from Gore were carried north by Sudanese merchants and exchanged for necessary goods such as salt and cloth (probably British-made).Thereafter, they passed through Gambaila and Khartoum to Port Sudan by the Red Sea, where the demand for the Maria Theresa dollar was low, and then they were finally sent to Aden, where high demand was expected (Figure 5.2). Thus, we find that by crossing territorial borders, Maria Theresa dollars made their way in grand circuits. Dollars brought to Aden from Europe entered and journeyed around the grand circuits before returning to Aden. As mentioned earlier, there were many circuits of the dollar connecting distant cites, for which Aden functioned as a hub. The current of the dollar moved along the circuits unidirectionally rather than in both directions. Although Aden geographically worked as an important crossing point for some circuits, it did not work as a centre that distributed the dollars to towns only to promptly recall them. Herein lies one of the reasons why no one authority could succeed in replacing the Maria Theresa dollar.The grand circuits of the dollar crossed several authorities’ territories. No single government could easily substitute a new currency of its own issue for the Maria Theresa dollar, and consequently the dollar survived in the region crossing the

Nationalized money  167 Port Sudan

Gambeila Aden

Gore Djibouti Addis Ababa Figure 5.2  Circuits of the Maria Theresa dollar in East Africa

Red Sea and the Gulf of Aden longer than in other regions such as Egypt. The length and unidirectional nature of this circuit might also have discouraged any merchant group from trying to monopolize business with their own money. Thus, besides the vertical movements providing an interface between peasant households and the international market that we confirmed previously, the dollar also worked well to link urban cities through long journeys along grand circuits. Maria Theresa dollars leaving Aden continued to travel, rather than return by the same route, and some of them finally ended up back in Aden after a series of one-way trips. Figure 5.3 indicates that the streams of the Maria Theresa dollars diminished as they followed the circuits due to unaccountable losses. Thus, horizontal movements between cities as well as the vertical ones between urban and rural markets were also unidirectional rather than bilateral. Here we find a clear example of a complementary relationship among currencies in which any single money could not represent actual liquidity but an assortment of monies could. We should stress that the Maria Theresa dollar and the functions it performed in Africa and the Middle East was not an isolated case. Circulation of foreign silver dollars was a well-known global phenomenon in the 19th century and early 20th century. Cases in China are helpful for us to consider the roles played by foreign silver dollars in comparative terms. In Jiujiang, an inland port in the mid-Yangzi region, the Mexican silver dollar and the Japanese silver dollar were in circulation with other currencies in the early 20th century. The Mexican dollar made a round trip mainly between Shanghai and Jingdezhen located

168  Global history of monetary delocalization

Figure 5.3 Triple layer of currencies interfaced by the Maria Theresa dollar

inland via Jiujiang. Meanwhile, the Japanese dollar circulated along a larger circuit; it flowed from Fujian’s coastal regions, including Amoy, into mountainous southern Jiangxi, and into Shanghai via Jiujiang before returning to Amoy. Along these currency circuits, porcelains from Jindezhen were brought to Shanghai, native cotton cloth produced in Ji’an, southern Jiangxi, was sold to Fujian, and kerosene oil was brought to southern Jiangxi from Shanghai via Jiujiang. The large circuit of the Japanese silver dollar in China (Figure 5.4) is clearly similar to that of the Maria Theresa dollar in East Africa (Figure 5.2) Like the Maria Theresa dollar, the Japanese dollar in China was available neither for daily transactions by local traders, who used copper coins for this purpose, nor for circulation for international transactions, which had become dependent on the gold standard system. In addition, the exchange rates among

Nationalized money  169



Jian Fuzhou

Amoy Figure 5.4  Circuit of Japanese silver dollar in Southern China

concurrent currencies incessantly fluctuated. Thus, we can assume a universal trend in which a foreign silver coin worked vertically as an interface in multiple market transactions while charting a journey along a grand circuit connecting local markets horizontally. In this situation, it was neither a local currency nor an international one. The aim of this chapter has not been to find what led the Maria Theresa dollar to go into circulation in the Middle East more than two centuries ago, but instead to determine why it remained indispensable in market activities in the Red Sea region even into the early 20th century. I have argued that the dollar was popular not only because of its silver content but also because it played a significant role in complementary relationships among monies in two ways. First, vertically, the Maria Theresa dollar functioned as a buffer between the international monetary circuit represented by the pound sterling and local markets mediated by various smaller currencies ranging from the 10-lire note to salt bars. Second, horizontally, the dollar helped form unidirectional circuits connecting local cities. Based on these functions of the dollar, the export of local products was able to develop in this region.That is why no currency could successfully substitute for this coin that bore a particular effigy and year. The material substance of the Maria Theresa dollar, namely, silver, helped it bridge markets vertically and horizontally. Ten-lire notes or beads could hardly

170  Global history of monetary delocalization

play the same role as the silver coins. In this sense, the role of a currency in relation to other monies, to some extent, depends on its own material substance. A precious metal for interregional trade certainly differs from a cheap material for daily transactions. This dual materiality and functionality of the Maria Theresa dollar might mislead some to describe the Maria Theresa dollar simply as a silver commodity in circulation. The fact that we cannot detach its materiality from its functioning does not mean that we can ignore the actual mechanisms adjusting demand and supply. As I have proved, the Maria Theresa dollar was not always priced by its intrinsic value during actual transactions, therefore we cannot take it as a mere silver commodity. The fluctuating demand for the Maria Theresa dollar in the market was the decisive factor determining its exchange rate. More importantly, the multiplicity of markets allowed the monetary demand to differentiate and made it difficult to synchronize demands from different currencies. Furthermore, since the layers of multiple markets were rather fuzzy, the demand and supply of one specific currency could not be entirely independent from another. Therefore political attempts to fix exchange rates in multiple markets usually failed. In multiple markets, the total supply of money cannot be measured without considering the complementary relationship among multiple currencies. This unsubstitutability among monies often made the quantity of a currency not neutral in determining prices, such as in one example of the sudden shortage of Maria Theresa dollars that led to a hike in commodity prices in 1939.40 We shall consider this in the conclusion. The case of the Maria Theresa dollar clearly demonstrates the effectiveness of the new viewpoints this volume provides. The complementary relationship among monies means that no money works independently, but a combination of monies can do what a single money cannot. The apparent imperfections of each currency resulted from the division of labour among monies, not from their materiality. However, the Maria Theresa dollar was obviously not made to pair with salt bars or 10-lire notes. Did anyone ever intentionally design a variety of monies to work together? It is important to note that it was neither a government nor a merchant group that organized this assortment of monies, but the market itself. The process by which the 10-lire note took root (after the Italian surrender!) provides a vivid example of a local market establishing a money. Self-organization in each circuit made the entire market multiple-layered. This multiplicity gave a segmented appearance to the system of markets, and the absence of a fixed rate among monies appeared to be less institutional.41 However, this did not result in irrational backwardness but rather provided sufficient flexibility to stabilize transactions. The Maria Theresa dollar and its complementary partners do not fit a pattern of linear evolution, such as from imperfect to perfect money or from a segmented market to an integrated one. The complementary role of the Maria Theresa dollar resulted from a self-organizing process implemented by markets themselves to smooth dealings, although its rationality might be beyond our contemporary common sense, in which we presume that one single currency works in one single-layer market.

Nationalized money  171

5  Paper monies nationalized peasant economies As we saw in some detail in the previous section, the Maria Theresa dollar provided complementary interfaces between global and local trade circuits rather than supplying the demand for money of all kinds. Mexican dollars in East Asia as well as the Maria Theresa dollar in Africa still continued to circulate widely in the early 20th century, just 100 years ago. The striking fact that until the end of the 19th century currencies of foreign origin had so easily performed significant roles reveals that a homogenous money inside a nation is just a convenient myth created only very recently. It is a quite modern invention for a single currency to circulate exclusively in a country across all layers: individual households at the bottom; various types of firm at the intermediary level; and the government at the top. However, the homogeneousness of currency, which had played only a minor role on a global scale until the late 19th century, began to dominate state by state, coinciding with the proliferation of the gold standard system across almost the entire world. From the viewpoint of monetary usage by ordinary people which this book has emphasized, it was not through the circulation of gold coins but rather the popular circulation of paper monies and subsidiary coinages defined in a fixed proportion to a gold standard unit that the gold standard system really transformed the manner of transactions at the bottom layer of a society. In previous chapters we have already seen how independent local exchanges adjusted monetary demand and supply across the world and throughout human history. Particularly, the seasonality of transactions and the preference for small denomination currencies often caused devices for local exchanges to deviate from those for distant exchanges. In addition, we have already confirmed that paper monies emerged differently and worked in varying ways country by country in the early modern age. Local paper monies, such as English country banknotes, Japanese domain hansatsu and Chinese local qianpiao, could exist for the purpose of meeting fluctuating local monetary demand. Could the supply of national paper money and subsidiary coins under the gold standard system flexibly meet demand at the level of end users? From the end of the 19th century to the early 20th century, except for China and Indochina, where silver-based systems were retained, a monetary system which maintained convertibility among all currencies in terms of gold had spread across Asia, regardless of whether they were politically independent like Japan or colonial like India. Needless to say, the standardization of gold units did not mean that gold coins actually circulated in the market. Currencies more frequently used hand to hand were coins of silver, copper and nickel, and paper monies of various denominations.Whether the new system based on the gold-term standard worked or not at the level of end users must be judged by whether non-standard coins worked as subsidiary to standard ones and by whether paper monies circulated according to their face value in terms of the gold standard. We have already seen that, unlike the modern common-sense assumption of the homogenized circulation of currencies, it was not easy for any issuers to

172  Global history of monetary delocalization

make small coins subsidiary to a large one and to keep the actual value of paper monies at par with their face value. Especially in the case of an economy which depended on peasant households, the size of their transactions were far smaller than the value of a gold coin, thus it must have been a major challenge to bridge this gap.The seasonality of monetary demand in peasant-dominated economies made the task more difficult, since no authority could have easily supplied small denomination currencies flexibly enough to meet demand subject to substantial seasonal fluctuations. We have already confirmed that modern statistics show the circulation of small denominations to have had a more seasonal bias in Chapter Two. Nowadays such a disproportion among different denomination currencies has little effect on the entire economy, since cash settlements now account for only a small part of all transactions. However, the story would inevitably be quite different if we analyze a period such as the early 20th century when economic activity depended on cash to a higher degree and with greater seasonality than now. China and Indochina, which had not adopted any gold standard monetary system, both faced the difficulty of supplying small currencies flexibly enough to meet demand subject to large seasonal fluctuations. The two decades before World War I saw favourable terms of trade for certain agricultural products, which encouraged peasant-dependent countries to export them for industrial resources and food for wage-earning labourers in industrialized countries.42 An example is organic oil, demand for which rapidly increased during this period from the chemical industry in Europe. As Table 5.5 shows, compared with the price of wheat, the prices of palm oil and linseed rose to a greater extent in London. As the movements of prices in Calcutta and Lagos synchronized with those in London, demand for peasant products in industrialized countries absorbed such products from peasant-dependent countries during this period, although we should remember the fluctuating trend over the longer term as already shown in Table 1.4. Asian countries other than India also increased their exports of raw cotton, soya, sesame, rice, etc., which were mostly produced by small peasants. The rapidly increasing export of soya beans from Manchuria show us a typical case in which a transformation of currency supply could meet the increasing demand for currencies with which merchants acquired soya beans from peasants (Kaminishi 2013). This global connection of peasants to international trade offered an opportunity to narrow the gaps between trade made in terms of gold and transactions with peasants using smaller monies through new intermediary devices. We should note that common global conditions at the turn of the 20th century did not erase the difference in the association of the Four Quadrants country by country. A comparison between Canada and Argentina in this period shows that differences in exchange and monetary conditions brought serious divergence in political economies. Both countries shared two fundamental factors in the late 19th century: becoming major wheat exporters to Europe and depending on immigrants for agricultural labour. On the pampas in Argentina, landowners preferred inviting tenants for sharecropping to hiring workers, since the former were more inclined to

Nationalized money  173 Table 5.5  Prices of agricultural products, 1889–1923

1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923

London palm oil £/ton

linseed s/qr

wheat s/qr

25 27 26 24 28 24.5 23 22 22 23 25 27.5 26 27.5 28 27.5 27 30.5 33 27.5 29 35 34.5 33 35.5 38 35 44.5 46 45 69 70 37 35 36

42 43 42 39 42 38 37 33 33 36 40 54 53 50 39 33 39 43 44 45 49 66 70 60 45.5 49 57 80 112 131 139 157 72 75.5 78

29.8 31.9 37 30.3 26.3 22.8 23.1 26.2 30.2 34 25.7 26.9 26.8 28.1 26.8 28.3 29.7 28.3 30.6 32 36.9 31.7 31.7 34.8 31.8 35 53.9 58.4 75.7 72.7 72.8 80.6 72.8 47.8 42.2

Calcutta linseed s/cwt

12 12.2 11.9 8.8 7.1 8.9 10 10 10.7 11.3 15.5 16.4 14.6 10.1 10.3 9.3 10.4 9.4 11.1 31.1 27

Lagos palm oil £/ton

24 21.8 24 22.6 18.3 17.3 21.9 21.3 43.9 42.3 17.7 21.7 23.6

Sources: Sauerbbeck (1904, 1911, 1924); Statistical Abstract Relating to British India, 1911–12/1920–21, 218; Helleiner GK. 1966, TABLE 2-B-2.

avoid risk under the turbulent oscillation of world prices in agricultural products. Mostly, tenants made short-term contracts which caused them to specialize in cultivating wheat from which producers expected a short-term return. Banks mainly lent to commercial sectors in major cities, or else to landowners. Tenants must have borrowed from informal lenders such as dealers of wheat.43

174  Global history of monetary delocalization

On the prairies in Canada, homestead farmers became the main producers of wheat. Generally modest family-based farms dominated. Between 1901 and 1914, the number of branches of chartered banks in the prairie regions increased from 71 to 870. By 1914, almost every prairie town hosted a bank, most of them branches of the established banks (Adelman 1994, 206). In Canada, through bank branch networks, Quadrant II extended to Quadrant III to provide farmers with seeds and machines. In Argentina, Quadrants II and III were separated. Both needed the largest labour supply at harvest time. On the pampas, a huge number of short-term labourers met the demand, while on the prairie, mutual cooperation among local farmers played the most significant role. The rapidly increasing monetary demand for paying peasants in Asia was met by supplies of national (or provincial) paper monies and multi-denomination coins which had not been popular before. As shown in Table 5.6, from the beginning of the 20th century to the 1920s, both the issuance of governmental paper money in Thailand and the Indochina Banknote had increased, as both countries expanded their exports of rice. In the same way, the issuance of banknotes by the Imperial Bank of Persia increased in tandem with the growing export of cotton from Iran. Meanwhile, in the northern region, the Russian ruble began to circulate. Russia was the best customer of Iranian cotton (Issawi 1971, 389; Jones 1986, 126–7, 354–7). These regional differences prove that increasing demand for currency in collecting cotton caused new currencies to be accepted, the device varying depending on the situation. The issuance of silver certificates, a paper money, continued to increase after its introduction into the Philippines in 1903 as well. Their total circulation increased from 9,057,127 peso in 1904, to 41,528,608 peso in 1909 and to 52,575,118 peso in 1914 (Luthringer 1934, 36). The success of paper monies astonished some contemporary American observers in the Philippines, since

Table 5.6  Paper monies in Thailand and Indochina

1903 1908 1913 1918 1923 1928 1933

Thailand governmental paper money million baht

Indochina banknote issuance million franc

note reserve million franc

3.5 14.8 26.1 59.7 91.7 135.3 114.3

40.2 57 86.5 174.4 831.1 1841.1 956.4

31 65.7 71.2 60.6 324.6 708 822.5

Sources: Central Service of Statistics, Siam 1940, 322–3; Gonjo 1993, 372–5.

Nationalized money  175

they had thought that local people would accept only metallic monies. They mistook the popular circulation of metallic coins as an absence of fiduciary currencies or incorrectly assumed that local people accepted coins simply because of their metallic content. However, as Kemmerer aptly described, ‘the Philippines were not upon the silver standard’ but ‘upon a fiduciary coin standard.’ There, silver coins with a wide variety of different silver content circulated at the same value, while the Mexican silver peso appreciated to the extent that its value at the market was 20 percent higher than its bullion value (Kemmerer 1916, 252–3). However, without the concept of the currency circuit, even Kemmerer failed to pay further attention to who actually shared the fiduciary currency and could not perceive that one fiduciary currency was not available to those who traded with a different one. At the turn of the 20th century, China also seemed to foreign observers to be remote from a situation of popular circulation of paper monies. Banknotes issued by foreign banks and domestic Western-style banks had begun to circulate in China from the late 19th century. However, until World War I they were only available within trade ports or particular business circuits closely connected with foreign trade. And yet, as we have already seen, in the 12th through the 14th centuries, official paper monies had once played a significant role across China. In addition, as we often saw in previous chapters, from the 19th century to the early 20th century, a variety of privately issued native notes became popular across China. Thus, depending on the situation, paper money could certainly work in China. Following the concept of the Four Quadrants of Exchange, we can interpret the popular circulation of native notes to mean that paper monies in various forms mediated exchange in Quadrant IV, but they did not work in the sphere of Quadrant I, in other words, paper monies circulated locally had no stable connection with any devices for interregional settlement. In this sense, the emergence of provincial paper monies and their partial success in several provinces, such as Hubei, during the first decade of the 20th century are significant evidence that paper monies issued officially could circulate interregionally beyond the boundary of a county (Kuroda 2005, 111–2).We have already seen the circulation of provincial paper monies in 1930 Yingkou in Chapter Two. Importantly, across Asia in the early 20th century, traditional currencies which had previously mediated transactions between peasants disappeared, while national paper monies and new types of coins increased in circulation. Copper coins in Indochina and cowries in Thailand ceased to work. In Iran, a shortfall of currency caused nickel coins to replace copper ones. Representing this trend, the circulation of copper cash with a square hole in the centre which had mediated exchanges in China for two millennia finally came to an end, and machine-minted copper coins with no square hole substituted for them at the beginning of the 20th century (Kuroda 2005, 108–9). Thus, whether a currency was metallic or fiduciary was a superficial criterion in understanding actual monetary usages in East Asia. The monetary systems actually worked within various currency circuits, a combination of a

176  Global history of monetary delocalization

particular currency and a trade circuit, as we have established. The same coin might circulate at a premium in one circuit, while the same item could be traded only as metal in another circuit. As we saw, the dominance of spot trade and strong seasonality in business made it difficult for any arbitrage to bridge the gaps between circuits. However, it is clear that unprecedented changes happened simultaneously across Asia from the turn of the 20th century. National paper monies began to intrude into circuits near to peasant households, while many traditional local currencies of small denomination, such as copper cash and cowries, disappeared and mechanically made coins of higher denominations took their place. Importantly, these movements appeared also in regions other than Asia. Currency circuits were also ubiquitous in Africa until the early 20th century. Table 5.7 shows the case of British East and Central Africa and the Uganda Protectorates. As we saw, the circulation of the Maria Theresa dollar, cowries (Hogendorn and Johnson 1986) and glass beads (Pallaver 2009) occurred within the circuit. However, the same transformation as happened in Asia could be found in Africa too. In the case of British West Africa, in which peasant products, such as palm oil, increased exports in the early 20th century, the substitution of new currencies convertible with sterling for traditional currencies such as cowries continued to be promoted by the colonial government. The change advanced slowly, but the sharp increase of exports to Britain during World War I caused a serious shortage of currency and consequently bronze subsidiary units of one pence, half pence, and one tenth pence began to prevail (McPhee 1971, 235–8). Table 5.5 shows the rapid rise of the palm oil price towards the end of the 1910s.

Table 5.7  Currency of British East and Central Africa and the Uganda Protectorates, 1896 Province


Kingdom of Uganda

Central Western

Busoga Bukedi Unyoro Toro Ankole

Nile Source: PRO, T1/9082B.

Dodinga Bari Shuli

Media of Exchange Rupee accepted Government Stations: Pice Outlying District: Cowries One rupee = 800–1,000 cowries Cowries, Beads Ivory trade: Rupees Government stations: Pice Outlying: Cowries Government stations: Pice Outlying: Cowries Government stations: Pice Outlying: Cowries Government stations: Rupee Outlying: Beads

Nationalized money  177

Keeping these similarities in mind, we should also note that, broadly speaking, compared with the case of the African colonies, even the colonial regions in Asia were under less restriction from home countries with regard to supplying currencies to local economies.44

6 How did banknotes reach down to the ground? The case of inter-war China As touched on in the previous section, in China banknotes were accepted only within major cities like Shanghai before World War I. Through the war period, the Chinese economy experienced major transformations including import-substitutive industrialization such as the rapid growth of cotton mills in Shanghai. These changes accompanied monetary transformations including the proliferation of the Chinese silver dollar – the Yuan Shikai dollar – which subsequently encouraged banknotes convertible with the dollar to enlarge their spheres of circulation. As we saw, for more than one millennium China had used silver mainly by weight as money, not by count. Importantly, yinliang or the silver tael as a unit of account was commonly found in all cities across China, though there was significant local variety. However, a survey of the usage of inland notes and currencies in cities across China in 1924 reveals that many cities had already lost the circulation of silver ingots and their own local liang had disappeared. For example,Yidu, which had used the qingping, changed its dominant usage to the Shanghai liang. Suzhou followed the same way (Jin 1925, 295). The proper silver liang in Suzhou, buhuiwen, was used only for transferring amounts between traders by account (Jin 1925, 59). Recalling that, as we saw in Chapter Two, a grocery store in 19th-century Ningjin recorded actual entries of silver ingots. Thus, until the middle of the 1920s, the usage of silver ingots in terms of a local liang had already disappeared from most cities across China. Even in the case of cities keeping their own unit in liang, such as Yingkou, the liang worked only as an imaginary money for transfer, again as we saw in Chapter Two. Most reports on the suspension of silver ingot usage attributed the change to silver shortages in each region. However, it is worth noting that even before the world depression and the silver purchase policy by the US government in the first half of the 1930s, which was to drive silver out of rural areas in China, the drain of silver had already happened in local cities. Before the nationwide decree abolishing silver liang in 1933, the silver liang system had already ended its historical role except in some major treaty ports such as Shanghai which retained the system. In contrast with the disappearance of silver used by weight, Chinese silver dollars rapidly became popular across China after the middle of the 1910s. In 1924, 960 million silver coins were estimated to be in circulation in China, of which 750 million were Yuan Shikai dollars and only 30 million were foreign dollars, with the remainder being accounted for by other Chinese dollars. The Mexican dollar which had dominated business conducted in terms of silver dollars lost its premium position soon after trade with Europe was depressed

178  Global history of monetary delocalization

by the outbreak of World War I. The change had already begun before the suspension of minting silver coins with the introduction of the gold standard in Mexico. The prevalence of the Chinese silver dollar also was enhanced by a drop in copper currency usage over a vast area centred at Hankou, where we saw a strong preference for copper coins in Chapter One and a subsequent switch to the silver dollar by the middle of the 1920s. In addition to the replacement of foreign dollars and the disappearance of copper-preferring regions, the prevalence of the Yuan Shikai dollar was interrelated and tied together with this through the circulation of banknotes by major banks. Increasing demand for raw cotton due to the development of the Shanghai cotton mills caused the silver dollar that was in circulation in the Lower Yangzi area to penetrate the cotton-growing regions of the central Yangzi area. The Yuan Shikai dollar gained popularity because of its role as a currency for payments to cotton-growing peasants, and bank notes which were convertible with the Yuan Shikai dollar steadily expanded their spheres of circulation. In the mid-1920s, notes issued by major banks like the Bank of China were broadly accepted for transactions involving raw cotton in cotton trading cities such as Zhengzhou (Kuroda 2005, 118). From the viewpoint of the two silver centuries (c.1300 and c.1600) which were major epochs in global monetary history, the final remnants of the first silver century finally finished the history of silver by weight in 1933, as the silver dollar representing the second silver century in the end replaced it. However, the latter also followed the first in finishing its historical role just two years later, as we shall see later. In tandem with the transformations in currency circulation described earlier, in the Middle and Lower Yangzi region an institution of networking banks and financiers emerged. Through the 1920s, it became a widespread practice for small banks to hold reserves of banknotes from major banks such as the Bank of China and the Bank of Communications, which were called lingyong. We must take note that this was not the general form of banknotes that financiers received from major banks. In addition to ordinary banknotes, major banks made a particular kind of banknote with a secret code for a bank or exchanger who had a contract for receiving a certain amount of banknotes. The banknote was called anjiquan, ‘secret-code-note.’ Every note had an extra code printed or stamped on it.This code was typically a particular Chinese character, sometimes numbers and sometimes an image. One was printed by the issuer, a major bank, another was stamped by the users, a financier or a non-major bank. The allocation of the code to a financier who made a contract for receiving a certain number of banknotes was kept secret from third parties. In principle, the bearer of the banknote could ask the bank issuing the note to convert it into coins. In a case where the bank exchanged silver coins for banknotes with a secret code, the bank would claim the cost of conversion from the financier who held the contract for using the coded note. The secret code notes would become invalid when a financier using them went bankrupt. For example, in 1932, when the Tongtai exchanger in Shanghai

Nationalized money  179

holding a contract for using 400,000 yuan of Central Bank notes closed their business, the Central Bank instructed its branches to make the note with the code ‘Qian 4’ invalid. Many native financiers in Shanghai made the same sort of contract with the Central Bank. Each native bank received a different banknote with a particular secret code. In the case of the Tongtai, ‘Qian 4’ was the allocated secret code which was printed on the Central Bank note.45 Did secret code notes circulate beyond cities? Let us look at a case in a rural region. In 1933, in Gaoyou, a rural town in the Lower Yangzi, ‘banknotes consisted 70 per cent of currencies and silver dollars 30 per cent’ among the circulating currencies. A native exchanger circulated banknotes from the Bank of China and Zhongnan with secret codes.46 In this case only along the commercial network of a native exchanger could the banknotes reach the local marketplace. Thus, the secret code note was theoretically convertible, but its circulation could not extend too far beyond the circuit of customers for each financier. Financiers receiving a major bank’s notes must have deposited at the bank the same amount of cash as the number of banknotes. If a native exchanger wished to make a contract for receiving banknotes of 400,000 yuan, they would have to bring 400,000 yuan of coins or ingots to the safe of the major bank issuing the notes. In return the bank would pay interest to the financier. In the case of the Central Bank in 1935 (before 4 November) it had sent checks equivalent to 7 percent of the principal annually to the financiers who circulated its banknotes.47 The rate of cash deposits by financiers at the major banks issuing the banknotes varied according to the bank and specific situation. Sometimes 70 percent of the amount of the banknotes received was deposited in cash or bullion; the rest was done in securities and other forms (Dong 2009, 168–9). Thus, through the 1920s among financiers in the Middle and Lower Yangzi regions, the practice of keeping major banks’ notes became widespread. However, most of the secret code notes had denominations of either five yuan or ten yuan. The denominations suggested that they were not expected to circulate among small traders. As far as the circulation of banknotes in denominations of five yuan and ten yuan was concerned, the most important region in economic terms in China already had to prepare for accepting the switch to the issuance monopolized by the four major banks in 1935, which we will see later. The banknotes with secret codes, anjiquan, issued before 4 November 1935 were allowed to remain after that date, but the contracts between the issuing banks and local financiers for the notes had to discontinue after their expiry.48 According to the analytical scheme of the Four Quadrants, the banks located in Quadrant II and native financiers in Quadrant III connected with each other through the circulation of high-denomination banknotes registered with secret codes. The way the banknotes circulated seems to have been midway between a named relationship and an anonymous one. Banknotes with a particular code issued by a bank circulated mainly among the customers of a particular local financier.

180  Global history of monetary delocalization

As we have seen throughout this section, in China, banknotes finally percolated down to circulate at the ground level in some districts during the interwar period, but their circulation was still limited. The majority of transactions in local markets were mediated by another paper money issued informally by local merchants. That was native notes, as we have already seen. The situation in Tongshan County in the Lower Yangzi region around 1930 shows us a more comprehensive situation of the locality of money relating to the demand for small denomination currencies and to the circulation of native notes (Jin 1931, 23–4). Few banknotes worked there. As far as currencies of one yuan were concerned, across Tongshan county the Yuan Shikai silver dollars issued by the central government, Mexican dollars, Bank of China notes, Bank of Communication notes and others circulated in the same way as in other counties. However, a great variety of additional smaller denomination currencies of different kinds also circulated within the county. Particularly, native notes issued by local merchants, whose denominations ranged from 50, 100, 200, 300, 500 and 1,000 wen, accelerated the multiplicity of currency circulation. There were two kinds of native notes: one was issued by exchangers and creditworthy shops in the county town; another was issued by shops in rural areas. The former could circulate across the county, while the latter circulated as widely as across 40–50 square kilometres. The smallest denomination coins, 10-wen copper coins, always suffered from shortages, thus they appreciated against higher denomination copper coins (20, 50, 100 and 200 wen) and native notes by around 10 percent. Meanwhile, the 20 cent silver coin was undervalued against the 20 cent note by around five percent. The disparity reflected the local commercial custom of using 20 cent notes more frequently than 20 cent coins. The usage of small denomination currencies depended on local customs prevailing in a space of a few hours’ walking distance. Thus, local currency circuits still independently supported proximate exchanges.

7 Seasonality and temporality in monetary demand still mattered: towards 1929 Until the outbreak of World War I, the Bank of England had astonishingly stuck to its strict domestic policy of not issuing banknotes with a face value of less than five pounds.This policy caused ordinary people to suffer from shortages of currency. By contrast, the Presidential Bank in India had issued mainly smaller denomination notes. In addition, these notes were accepted legal tender only within one district, which means they must have been discounted in conversion outside the region (Keynes 1913, 28–9). Large local paper money supplies in relatively small denominations made it possible for peasants to accept them with less difficulty. This combination of small denominations and locality was the key to paper monies reaching peasant households. As already mentioned, in the case of early 20th-century China, provincial paper monies played an important role in supplying currencies to collect peasant products to the ports

Nationalized money  181

(Kuroda 2005, 111–3), and banknotes began to circulate among local merchants in some regions after World War I. The importance of localities resulted from the strong seasonality in monetary demand in the peasant economy. In the case of India, during summer, the slack season, notes held by the Presidency Banks and the Treasury increased, whereas in winter, the busy season, they decreased. For example, their holdings of paper money decreased from 13.59 million rupees in August 1906 to 9.11 million in January 1907 and from 17.47 million in August in 1907 to 8.62 million in January 1908 (Keynes 1913, 54). This large gap between the busy season and the slack season in Indian finance attracted keen interest from Keynes. He remarked that large and regular seasonality afforded the ‘differentiation between the Indian Market and those which we are familiar with in Europe’ (Keynes 1913, 242–3). In order to secure large quantities of peasant products, significant amounts of ready cash must have been transported to rural markets in every harvest season.The discount rate also showed clear seasonality: high interest rates in winter, low interest rates in summer. Monthly discount rate of the Presidency Bank of Bengal rose from 5 percent in August 1906 to 9 percent in January 1907 and from 3 percent in August in 1907 to 9 percent in January 1908 (Keynes 1913, 240–1). The seasonal contrast clearly relieved the movement of cash from city to villages. The same situation also appeared in China as we saw in Chapter Two. Thus, sufficient currencies were required to be held locally. On this point we can see a dilemma not easily solved. Statistics from the late 20th century US, which we touched on in Chapter Two, support the observation that far earlier redemption of currency in small denominations over large denominations proves a higher frequency of using the former than the latter, and that currency in small denominations responded more readily to seasonal increases in monetary demand than currency in large denominations. Thus, one might easily imagine that seasonal tensions of monetary demand and supply after the harvest should accompany the problem of maintaining different denominations in proportionate supply. Potentially flexible monetary supplies to peasants could have a negative relationship with the principle of keeping convertibility between standard money and paper monies or subsidiary coins. Convertibility among monies should mean no transaction costs when changing one money into another. However, it does not mean that a flexible supply could be guaranteed for both monies, as we argued in Chapter One. Extending convertible monies to small peasants in Asia heralded a new stage in the world economy. However, it also meant that monetary tensions which had been solved locally had escaped the control of the local authorities. According to this interpretation, it is no coincidence that most global financial crises from the late 19th century to the early 20th century were ignited in autumn, the harvest season (Miron 1986). The financial crisis of 1907, the largest global panic before World War I, is one example. In August 1906, a speculative plunge in the New York Stock Exchange brought about serious tension in bank reserves. In order to rescue the banks, the US Treasury used official money

182  Global history of monetary delocalization

to make a contract to purchase $36 million in gold from Europe in September. This move in turn placed a serious burden on the London financial markets, where, in October, the interest rate rose to the highest it had been since 1899. Then it was the movement of gold to Egypt which had had a good harvest of cotton that spurred a shortfall of liquidity (Noyes 1909, 357). Financial instability appeared to calm for a while, but then a credit panic went around Hamburg, Alexandria, Tokyo and Santiago. Finally, in the autumn of 1907, the cash demand for collecting foreign harvests hampered London in settling a large number of financial drafts, and consequently the entire credit system crashed. This panic caused significant damage to actual production. For example, steel production in 6 major countries decreased by 23 percent (League of Nations 1931, 275). Most trade statistics recorded large falls in 1907–1908.49 From autumn 1907 to winter 1908, a large quantity of gold was recorded as having returned to the Presidency Banks in India. It showed that, while the monetary situation in India was deeply influenced by that of London, a significant amount of the gold which could be potentially assembled was held by individuals. As long as multiple monies could work, financial panic at the global level did not always affect the local economy, since fluctuating rates between monies could absorb the impact. Meanwhile, even if an unprecedented harvest occurred across a wide area, urgent demand for currencies could be met by local devices. Thus, modern convertible systems brought transparency to transactions but stripped away the buffers which had mitigated unpredictable impacts. After the end of World War I, the export of Asian peasant products resumed its increase. However, the patterns of finance were different from what they had been before World War I. Before World War I, very high reserve rates in issuing banknotes were common. In India the note reserve was required to fully cover the total amount of banknote issuance. In 1913, the total amount of note issuance was £46 million, while £11 million in silver rupees, £19.5 million of gold in India, £6 million of gold in London and £9.5 of securities were reserved (Keynes 1913, 34). Full reserves were required in Iran, too, but in actual practice they were no more than two thirds (Jones 1986, 80). In the case of the Hubei Provincial Bank, which successfully issued the largest amount of provincial paper money in early 20th century China, the proportion of silver reserves against total issuance was 44 percent in December 1908 and 26 percent in July 1910, but the total reserves, including securities, always covered the total issuance of paper money (Xie 1988, 256). Thus, until World War I, the reserve rate against issuance in Asian countries was not always low, even if they were lower than those of industrialized countries. However, such a high proportion of reserves disappeared after World War I. In the case of Indochina Bank notes, until World War I the proportion was never less than 70 percent. However, in 1918, the issuance increased to nearly three times reserves. On 31 December 1918, according to an order by the French President, the maximum of note issuance was allowed to be eight times reserves (Nawata 1932, 124). After that, the amount of issuance continued to soar until just before the 1930s.

Nationalized money  183

Under colonial (or semi colonial) conditions, some paper monies mainly worked to transmit international economic conditions. In the case of Syrian paper money in the 1920s, the price movements of flour followed the fluctuation of the exchange rate between Syrian piasters and US dollars (Said 1935, 78). The coefficient between the two movements was high (0.88). Local merchants were obliged to set prices in terms of paper piasters by governmental regulation. This is in contrast with the cases we saw in which local currencies of endogenous origin could function as a buffer to absorb external fluctuations and subsequently provide local economies with a haven in which locals could exchange in relatively stable conditions. Besides issuing banknotes, banks also expanded their supply of credit enormously. For example, banks’ lending to mutual credit associations in Cochin China expanded from 99,642 piasters in 1913 to 15,108,778 piasters in 1930. The mutual credit associations actually worked to finance landowners to grow rice (Robequain 1944, 170–2). Similarly, agricultural credit associations in Burma also reached their peak in the 1920s (Cheng 1968, 274). This ‘credit inflation’ as it was called by Robequain at the level of peasant households accelerated across the world. Over-lending to monoculture economies, such as occurred in Brazil when it suffered from an oversupply of coffee, had been thought to be one cause of the world depression beginning in 1929 (Kindleberger 1986, chapter 4). However, through the 1920s, borrowing in the local peasant economy actually increased. As long as the prices of their products could be maintained, over-lending could be concealed. Once prices began to fall, non-performance would potentially aggregate across regions.Towards the end of the 1920s, the prices of major peasant products plunged. The conditions for the worldwide credit crisis had been already been set.

8 The paper money standard in China in 1935: unification at the top and variety on the ground As we saw in the previous sections, after the turn of the 20th century, Asian countries whose economy fundamentally depended on peasant households rapidly became accustomed to using paper monies. Centralization and nationalization of monetary issuance appeared to be an irreversible transformation spreading across the world, including peasant-dominated Asia, as banks backed by the state monopolized note issuance in most countries and colonies. The world depression and the collapse of the international gold standard accelerated the tendency towards the dominance of national paper monies issued by central banks which no longer had the backing of any material content. Was this the end of the history through which, so far, we have found that the heterogeneity of money derived from a variety of exchanges? After the proliferation of managed currency systems supported by central banks across the world, the complementarity among monies appeared to survive only in the relationship between an international currency guaranteed by world hegemony, the US dollar and a number of national currencies which homogenously and

184  Global history of monetary delocalization

exclusively circulated within national territories. Did this exogenous transformation by a state successfully tailor the endogenous heterogeneity of exchanges into a consistent form? In this section, we will find the answer to this to be in the negative in 1930s China, where, under an apparently unified surface, in reality a multiplicity of different monies still existed. According to the global trend after 1929 described earlier, the emergence of the paper money standard in China on 4 November 1935 seemed simply to follow the universal tide. In particular, compared to the similar transformations of the monetary system in Mexico in the same period, it is apparent that the artificial rise in the international price of silver because of US governmental policy in 1934–35 caused silver to flow out from both China and Mexico, where the monetary systems were dependent on silver more than anywhere else in the world and encouraged both governments to demonetize silver through uniform paper money (Helleiner 2003, 156). However, the acceptance of unified paper money in exchange for silver was possible only by depending on the historical transformations which China had experienced through and after World War I. Through this process, monetary units in silver by weight which the first silver century had brought about finally ended. It was no coincidence that the circulation of silver dollars which the second silver century had established also reached a final stage in East Asia during the same period as the Maria Theresa dollar in the Red Sea region was ending its dominance, as we saw earlier. The Nationalist Government in China declared that national banknotes issued by four major banks should substitute for all existing currency and that all monetary metals, including silver dollars which were in dominant circulation, should be exchanged for the national banknotes and be in the possession of the central government. Even if we overlook the global trends in the 1930s, a dematerialization of currency might seem to be evolutionary, and a unification of monetary devices into one currency also seemed to promise a reduction of transaction costs. From these viewpoints, the transition from the dependence on metallic currencies to the unified paper money standard in 1935 appeared to demonstrate that China was finally on the track of a reasonable monetary system following a ‘chaotic period.’ Thus, in a teleological way, historians have tended to describe the monetary system in China as if it had been designed to advance towards a dematerialized and unified currency under the Nanjing Nationalist Government. Table 5.8 shows what proportion the unified paper money occupied among the entire currency in use province by province at the end of 1935. These estimates were made by the post office. Considering that only less than two months had passed since the declaration of the monetary unification by national paper money, and in spite of significant provincial biases such as low rates of conversion to the new currency in Shanxi and Guangdong, some people may be struck by the efficiency of substituting new paper money so quickly and profoundly. However, beneath the surface of the apparently ongoing unification, at the bottom layer of exchanges, fragmental localization of currency issuance appeared

Nationalized money  185

all across China. The situation appears to have been quite contradictory, since the introduction of a paper money standard system at the national level paradoxically spurred the increasing issuance of native notes at the county level after 1935.The banknotes issued by the four central banks mostly dominated nationwide circulation, while native notes with no official guarantee proliferated in local markets. In Chapter One and Chapter Four, we have already seen the popular circulation of native notes with no official guarantee in some counties. A clue to understanding this diversification is that the majority of unified banknotes were of five yuan and ten yuan denominations, while most of the native notes covered denominations of less than one yuan. Why did such a denomination gap between national and county levels appear in China in 1935? We can find in Table 5.8 that the Middle and Lower Yangzi region (Jiangsu, Shanghai, Anhui, Jiangxi and Hubei) successfully established a high level of national paper money circulation. The greater success in this region compared to the rest of China resulted from the transformation which the region had undergone over two decades, as we showed in Section Six earlier. Looking at the processes occurring after the declaration of the currency reform on 4 November 1935, it appears that the actual emphasis of the Nationalist Government in China was on demonetizing silver rather than on currency unification. The central government permitted local people to continue using provincial currencies already in circulation, though no further issuance of such Table 5.8 The proportion of the unified paper money (fabi) among the entire currency in use at the end of 1935 (percent) Jiangsu Shanghai Zhejiang Anhui Jiangxi Hubei Hunan Dongchuan Xichuan Shandong Hebei Beiping Henan Shanxi Shaanxi Gansu Fujian Guangdong Guangxi Yunnan Guizhou

80 80 74 82 88 84 59 85 93 84 77 64 73 16 63 58 92 23 2 20 65

Source: Zhonghua minguo youzheng huidui chujin ju 1937, 69.

186  Global history of monetary delocalization

currencies was allowed. Meanwhile, financiers had to make a report of how much cash in banknotes and in silver coins and wares of precious metals that they possessed. For example, as shown in Table 5.9, native financiers in Wuxi in the Lower Yangzi made reports to the authorities on the number of banknotes and silver coins they kept on 3 November 1935. In contrast with the amounts reported by modern-style banks in Wuxi, the amounts of coins which the native financiers in the same city possessed appear to have been very small. And we should note that the amounts of coins kept were exactly the same as the amounts of silver coins the native financiers reported as converted into the legal money, fabi, eight days later, on 11 November. Thus, we should be cautious about assuming that the amounts of coins reported by financiers did not reflect the amount they actually possessed. Even if financiers might have reported lesser amounts of silver coins than they really possessed, it is likely that native financiers held large quantities of banknotes as assets. We can confirm the same tendency in other local cities such as Shashi in the Middle Yangzi in 1935.50 The presence of so many banknotes in the safes of native financiers suggests that banknotes in the 1930s worked quite differently than they did in the 1910s, when traditional financiers rarely used banknotes in their business. These were the circumstances in which unified paper money, fabi, could be exchanged for silver so quickly after the decree on 4 November 1935. In Section Six, we saw what happened to native financiers’ business during this period. As far as the Middle and Lower Yangzi region is concerned, the popularity of the secret code note system shows that native financiers, operating locally, had already established a close connection with major issuing banks, operating interregionally. However, the denominations of secret code notes were all of five or ten yuan. We should keep in mind that the breakdown by denomination shows that the financiers in Shashi possessed a larger number of small denomination banknotes. Contemporary foreign observers had serious concerns about whether a legal tender with no metal backing would be acceptable in China. What the Table 5.9 Amounts of banknotes and silver dollars possessed by native exchangers in Wuxi 1935 November (yuan)

Fuyaun Zengda Fuchangsheng Shenyu Fuyu Defeng Yuanfeng Yonghengfeng


Silver dollars

12,563 3,205 12,189 3,753 17,467 2,999 1,014 2,553

1,000 200 100 200 212 110 162 100

Source: Shanghai municipal archives. Q55-2-73.

Nationalized money  187

Nationalist Government did most successfully was eventually to keep the exchange rate of the yuan against the US dollar and pound sterling within narrow margins.51 The issuance biased towards higher denominations, five yuan and ten yuan, might have made it easier for the issuer to control the balance between issuance and collection.The control over the quantity in circulation was necessary to keep a parity between the legal tender and foreign currency. The stability of the yuan against major foreign currencies would encourage traders engaging in foreign-trade businesses to accept the new paper money. On the other hand, accompanying the decree of currency unification on 4 November 1935, the fee for domestic transfers was unified. Prior to the reform, fees for domestic transfer differed according to the specific combination of the distance between two cities and fluctuated also according to the economic situation. The fees were a source of profit for financiers. However, after the decree, the rate of the fee was unified to one percent regardless of distance. As far as we can confirm through the business reports of major banks such as the Bank of Communications, the new flat fee seems to have been adopted. In this sense, the unification of currency brought the unification of domestic transfer fees at a reduced rate and subsequently the reforms reduced the cost of interregional settlements.52 Thus, as far as interregional settlements, domestic and foreign, were concerned, the monetary reformation in 1935 appeared to have been successful. However, the denominations of five and ten yuan, though not extremely high, were far higher than the denominations suitable for settling business transactions between ordinary people. As we saw, success in such a unification in Quadrant I is not always followed by success in Quadrant IV. At least, before the currency unification decree in November 1935, local banks or branches of major banks in rural region supplied one- yuan notes and subsidiary notes whose face values were less than one yuan. Table 5.10 shows the case of Shashi. In May 1933, responding to financial tension, Zhejiang Regional Bank issued banknotes of 3 million yuan in total; 500,000 yuan in 10-yuan notes, 750,000 yuan in 5 yuan notes, 1,000,000 yuan in 1 yuan notes, 500,000 yuan in 2 jiao (two-tenths of 1 yuan) notes and 250,000 yuan in 1 jiao (one-tenth yuan) notes. It is clear that means of payment in small denominations were required to resolve liquidity shortages.53 However, the monetary unification in 1935 blocked the ways that local banks previously had flexibly supplied what local markets needed. Thus, as soon as the unification policy of the fabi was introduced, a shortage of small denomination currencies emerged. Responding to a survey on the situation of the market after the unification reform, many branches of the Bank of Communications reported to the headquarters in Shanghai how seriously local markets suffered from the shortfall of one-yuan notes. In reality, even the denomination of one yuan was not small enough to meet end users’ demand. Responding to the shortage, traders in Zhangjiakou created ‘native goods notes’ of 1 yuan, 2 jiao and 1 jiao to meet the demands for small change.54 These

188  Global history of monetary delocalization Table 5.10 Amounts of banknotes and silver dollars possessed by native exchangers and banks in Shashi 1935 November (yuan) Silver dollar Bank notes 10 yuan note 5 yuan note 1 yuan note Subsidiary notes Qianxin 1,201 Qainyu 1,587 Jixiang 0 Jiuyu 0 Jin’an 1,564 Bank of China 277,251 Bank of 64,407 Communication Peasant Bank of 100,100 China

248 0 2,560 260 657 150 2,315 430 6,739 180 2,710,228 61,000 160,019 21,000

0 330 210 320 3,600 62,000 134,210

240 1,020 180 361 1,531 146,159 4,201

8 350 1,165 1,204 1,428 1,869 608

146,540 36,900




Source: Shanghai municipal archives. Q55-2-728.

Table 5.11  Exchange rates of 10 wen copper coin per 1 yuan at the end of 1935 (wen) Jiangsu Shanghai Zhejiang Anhui Jiangxi Hubei Shandong Hebei Beiping Shanxi Shaanxi Gansu Fujian Guangdong Guangxi Yunnan Guizhou

3,300 3,250 3,350 3,250 3,600 3,000 4,800 5,050 4,050 4,750 3,800 6,000 3,360 2,900 3,580 4,900 3,300

Source: Zhonghua minguo youzheng huidui chujin ju 1937, 70.

pretended to be coupons, not currencies. In Yantai, Northern China, postal and fiscal stamps were reported to be in use for small transactions.55 The shortfall of small currencies inevitably affected the circulation of copper coins. Table 5.11 shows the exchange rate on average in the year 1935 in 17 provinces. While four major jurisdictions (Jiangsu, Shanghai, Zhejiang and Anhui) in the Lower Yangzi region had lower rates from 3250 wen to 3350 wen, the northern three provinces (Shandong, Hebei and Shanxi) had higher rates from 4750 wen to 5050 wen.

Nationalized money  189

According to the currency unification decree in November 1935, existing copper coins had to be collected to be melted down. Table 5.12 shows a large gap between urban and rural areas nine months after the declaration of the reform. Nanjing, the national capital, had a small amount of copper coins and Ningbo, a financial city, also possessed very low numbers. The large numbers in Nanchang, Jiangxi province, might simply be attributed to high transportation costs. However, the cases of Nantong and Qingjiangbu, which were cotton and grain producing districts and not so far from Shanghai and Nanjing, suggest there was strong demand for copper coins there. It was not strange that copper coins began to appreciate against the official rate. In Yantai, though local traders set the exchange rate for copper coins at 3,800 wen per yuan, the actual rate was between 3,100 and 3,200 at the end of December 1935. In Wuhan in the middle Yangzi, the exchange rate of copper coins to fabi became 2,800, despite the official rate of 3,000, in December 1936. Copper coins usually appreciated at the close of the year, thus such a seasonal pattern accelerated the shortage. It is interesting that the unification policy paradoxically made the localities stand out more, even in the regions where the ties between local financiers and interregional banks strengthened as mentioned earlier. The exchange rates of small denominations against the legal tender differed between Jinhua and Yuyao, both of which are located in the vicinity of Hangzhou.56 Several days after the currency reform on 4 November 1935, in Jinhua one yuan was quoted at 3,400 wen in copper coins, while in Yuyao it was 4,000 wen. The distance between Jinhua and Yuyao is 100 km in a straight line. The gap appears to have been due to more than transportation costs. Thus, the unification of currency exogenously imposed by the central government, ironically, reignited local currency circuits across China which autonomously adjusted the demand and supply of money. We already saw the importance of native notes in Chapter Two and Chapter Four. An index of native notes’ popularity can be found in the collections of numismatists. Shi Changyou, for example, collected more than 1,700 notes from the Republican Table 5.12  Quantity of copper coins possessed by bank branches 1936 August (in pieces)

Hangzhou Wenzhou Nanchang Nantong Changshu Ningbo Qingjiangbu Nanjing Taicang

Bank of China

Bank of Communications

2,310,000 24,870,000 59,250,000 13,075,200 9,063,000 510,000 2,535,000

1,590,000 9,720,000 7,650,000 6,120,000 900,000

Source: Shanghai municipal archives. Q55-2-643.

7,599,486 128,550 2,885,700

190  Global history of monetary delocalization

period.The number of new native notes from northern provinces in his collection increased from 63 notes from the 1910s to 393 notes from the 1920s and to 753 notes from the 1930s (Dai and Chen 2011). It seems safe to conclude that, in spite of no institutional support, native notes increased in popularity through the early 20th century. We revisit Julu County, which we saw in Chapter Four. An article in a contemporary journal wrote about the situation in the county in 1936. No unified official money worked there yet. Provincial banknotes still circulated, but they consisted of only 50 percent of circulating currencies. The rest were composed of native notes, though it was difficult to tell forged notes from genuine ones (Chen 1936). It indicates that even these native notes could not sufficiently meet the demand for money. The apparent monetary unification through official paper money on the surface in fact had induced local paper monies to flourish at the ground level.

Notes 1 Tschoegl noticed that the difficulty of replacing network externalities was a more persuasive explanation for the durability of the Maria Theresa dollar than the stability of its silver content, but he did not make clear what kind of network the dollar circulated in. Tschoegl 2001, 454. 2 After World War II, Vienna resumed minting and has still continued to issue them. Between 1998 and 2001 nearly 60,000 dollars were struck to respond to demand from coin collectors. These later issuances are beyond the scope of this book. Semple 2005, 40. 3 A portion of the Maria Theresa dollars, together with other species, was exported to India and other places due to trade deficits in the cities of the Arabian Peninsula. Pope 2001, 271. 4 We can see ample examples of ornamental usages in Semple 2005, especially, 92, 109. 5 PRO, MINT/2253/19. 6 The continuous debasement of Ottoman coinages in the same period might have given an advantage to the Maria Theresa Dollar for circulation in the Middle East. Pope 2001, 268. 7 The export of silver from Mexico also increased. She exported 28 million Mexican dollars in 1882–83, 55 million dollars in 1892–93 and 77 million dollars in 1902–03. Andrew 1904, 355. 8 The slave trade also stimulated the imports of manillas, brass currencies, to West Africa. Most of them were made in Birmingham and Nante. Guyer 2004, 72. 9 On farasulla see Pankhurst 1970a, 65. 10 PRO, FO401/8, p. 50. D. MacGillivray was an agent despatched by the National Bank of Egypt during the establishment of the Bank of Abyssinia. For an account of this bank, see Schaefer 1992, 361–89. 11 PRO, FO401/8, p. 51. 12 Sending one mule load of seven bags of Maria Theresa dollars (3500 coins) from Harar to Addis Ababa cost 35 dollars on average. Pope 2001, 263. 13 During the initial stage of the invasion, Italians distributed the dollar even into the countryside of Ethiopia. Schaefer 1996, 93. 14 BLIO, J1836/8/1. Sir S. Barton to Mr. Eden. 15 PRO Mint 1182/2/1937. 16 BLIO F2563/1941. 17 Contraband imports of the Maria Theresa dollar and exporting lire was immensely profitable. Pankhurst 1970b, 106–11.

Nationalized money  191 18 PRO, MINT F817/1937. At that time it cost 1 shilling 6 pence to manufacture one dollar in London. 19 ‘Arabian’ is, in fact, too broad a description. For example, Hadhrami was one of the traders dominating the slave and coffee trades in Aden and other cities in this region. Ewald and Clarence-Smith 1997. 20 BLIO, F6048/1942. 21 BLIO, F2563/1941. 22 BLIO, F580/1943. 23 As for Collier, see Schaefer 1992, 379, 387. 24 BLIO, F3573/1937, ‘Report from Aden on Maria Theresa dollars.’ 25 Monthly movements of the exchange rate of Maria Theresa dollars against the rupee in Muscat show a tendency to reach its highest point in October and its lowest in April, which is similar to the situation in Aden. Pope 2001, 259. 26 Gras rifle cartridges exchanged at the rate of ten to one Maria Theresa dollar. To promote the circulation of the new Menelik coin, the government tried in vain to prohibit its use. Interestingly, the four quarter-dollar Menelik coins were valued higher than one full dollar in Menelik coin. Pope 2001, 263. 27 For the seasonal fluctuation of exchange rates between salt bars and the dollar, see Pankhurst 1962, 228. 28 PRO, FO401/8, p. 51. 29 Manchester Guardian 6 Sep. 1941, ‘Currency and Banking in Ex-Italian Colonies.’ 30 The British authorities warned their troops in Ethiopia against accepting Italian lire but allowed accepting Italian coins as small change because there was in fact no subsidiary coinage for the M.T. dollar with the condition of a fixed rate. ‘Currency in Ethiopia and Occupied Italian Colonies’ p. 9. PRO, FO371/27555, 15 February 1941. 31 BLIO, F4895/1942. 32 BLIO, F360/1933. 33 Op cit. Manchester Guardian, ‘Currency and Banking.’ 34 BLIO, CMD/22318 cipher, from G.O.C. in C. East Africa to the War Office, 7 May 1942. 35 BLIO, OET/09646 cipher, from G.O.C. in C. East Africa to the War Office, 14 June 1943. 36 PRO, FO 401/28, Further Correspondence respecting Abyssinia, ‘Intelligence Report for the Quarterly ending March 31, 1932’ and ‘ending March 31, 1933.’ 37 PRO, WO 32/9414 Overseas: Abyssinia Currencies and rate of exchange 1941–49. 38 PRO, Mint23/22 Cairo Treasury Chest: disposal of Maria Theresa dollars, 1885. 39 BLIO, OET/08201 cipher, from C.in C. Middle East to the War Office, 5 June 1942. 40 According to a letter from the Aden Administration in May 1939, the prices of vegetables, eggs, and meats also rose. They came from inland and were paid for in Maria Theresa dollars. The short supply of the dollar was said to have raised the prices of food. BLIO, F1863/1939. 41 For example, see Webb 1982. 42 For example, the cloth-purchasing power of cocoa in Ghana was kept high until 1916. Austin 2005, 294–5. 43 Argentina, an export economy and a debtor to London, kept domestic convertibility of notes for gold when her exports were in a favourable condition, but, unlike other export countries like Australia and Canada, export-producing oligarchies abandoned convertibility when the conditions became unfavourable (Ford 1962).Through the 20th century, Argentina had experienced the concurrence of different provincial paper monies. Théret 2018. 44 The case of Liberia shows that, in Africa, even a politically independent country suffered from a severe limitation of monetary supply under the international gold standard regime (Gardner 2014). As the case of cocoa farming in Ghana showed, production of exported commodities heavily relied on credit supplies originating from European capitals, regardless of whether it was intermediated by African brokers or by European ones (Austin 2005, 278–303). That is why it was difficult for export economies in Africa to establish a monetary supply independent from European countries.

192  Global history of monetary delocalization 45 Shanghai municipal archives, S178-2-19. 46 ‘Ershi’er nian Gaoyou zhi shangye ji jinrong’ Jiaohang tongxin 1–4. 47 Shanghai municipal archives, S178-2-19. 48 ‘Gehangzhuang zhaojiu lingyong zhongjiao anjiquan,’ Yinhang zhoubao 20(16). 49 Interestingly, the exchange rate of the Maria Theresa dollar against the rupee fell in 1907–08. The dollar was used to settle the trade deficit with India. Pope 2001, 259. 50 Shanghai municipal archives, Q55-2-728. 51 The international relationship between China and the Powers in the period of the currency reform in 1935 was made clear by Shiroyama (2008). She suggests that the importance of the Chinese government’s silver sales to the US provided a significant reserve to support its currency reform. Shiroyama 2008, 192. 52 However, in practice there were difficulties in the case of remittances between cities in which none among the Bank of China, the Bank of Communications and the Central Bank had any branch. According to an archive of the Shanghai Commercial and Saving Bank, at the end of 1935, an extra charge of 3 to 5 yuan was required in transferring 1,000 yuan to those cities. Shanghai Municipal Archives, Q275-1-812, 1, 81. 53 ‘Zhejiang difang yinhang faxing duihuanquan’ Jiaohang tongxin 2–9. 54 Shanghai municipal archives, Q55-2-731. 55 Shanghai municipal archives, Q55-2-726. 56 Shanghai municipal archives, Q55-2-728.

References Adelman, J. 1994. Frontier Development: Land, Labour, and Capital on the Wheatlands of Argentina and Canada, 1890–1914. Oxford: Clarendon. Andrew, AP. 1904. ‘The end of the Mexican dollar,’ Quarterly Journal of Economics 18(3): 321–56. Arminion, P. 1911. La situaion economique et financiere de l’Egypte, le Souden Egyptien. Paris: No publisher. Austin, G. 2005. Labour, Land and Capital in Ghana: From Slavery to Free Labour in Asante, 1807–1956. Rochester: University of Rochester Press. Central Service of Statistics, Siam. 1940. Statistical Yearbook Siam, 1939. Bangkok: Central Service of Statistics. Chen, TS. 1936. ‘Shixing fabi hou de Julu nongcun,’ Dongfang zazhi 33(6): 113–14. Cheng, SH. 1968. The Rice Industry of Burma 1852–1940. Kuala Lumpur: University of Malaya Press. Dai, JB. and Chen, XR. 2011. ‘20 shiji 20, 30 niandai beifang nongcun qianpiao fanlan de kaocha,’ Zhonguo jingjishi lundan 1–11. Dong, X. 2009. Zhongguo yinhang Shanghai fenhang yanjiu. Shanghai: Shanghai renmin chubanshe. Ewald, J. and Clarence-Smith, WG. 1997. ‘The economic role of the Hadhrami diaspora in the Red Sea and Gulf of Aden, 1820s to 1930s,’ in Freitag, U. and Clarence-Smith, WG. (eds.) Hadhrami Traders, Scholars and Statesmen in the Indian Ocean, 1750s–1960s, 281–96. Leiden: Brill. Ford, AG. 1962. The Gold Standard 1880–1914; Britain and Argentina. Oxford: Oxford University Press. Gardner, L. 2014. ‘The rise and fall of sterling in Liberia, 1847–1943,’ Economic History Review 67(4): 1089–112. doi:10.1111/1468-0289.12042 Gonjo,Y. 1993. Banque coloniale ou banque d’affaires: la Banque de l’Indochine sous la III République. Paris: Imprimerie nationale.

Nationalized money  193 Guyer, JI. 2004. Marginal Gains: Monetary Transactions in Atlantic Africa. Chicago: University of Chicago Press. Hans, J. 1961. Maria-Theresien-Taler: zwei jahrhunderte, 1751–1951, epilog 1951–1960. Leiden: Klagenfurt. Helleiner, E. 2003. The Making of National Money: Territorial Currencies in Historical Perspective. Ithaca: Cornell University Press. Helleiner, GK. 1966. Peasant, Agriculture, Government, and Economic Growth in Nigeria. Homewood, IL: RD. Irwin. Hogendorn, J. and Johnson, M. 1986. The Shell Money of the Slave Trade. Cambridge: Cambridge University Press. Ida, G. 1953. Colonial Monetary Conditions. London: HM. Stationery Office. Issawi, C. 1971. The Economic History of Iran, 1800–1914. Chicago: University of Chicago Press. Jin, BP. 1925. Guonei shangye huidui yaolan. Shanghai: Shanghai shangye xuchu yinhang. Jin, WJ. 1931. Tongshan nongcun jingji diaocha. Zhenjiang: Jiangsu nongmin yinhang. Johnson, JF. 1906. Money and Currency: In Relation to Industry, Prices, and the Rate of Interest. Boston: Ginn. Jones, G. 1986. Banking and Empire 1889–1952: The Imperial Bank of Persia. Cambridge: Cambridge University Press. Kaminishi, M. 2013.‘The seasonal demand for multiple monies in Manchuria: Re-examining Zhang Zuolin’s government’s economic policy during the 1920s,’ Financial History Review 20(3): 335–59. https://doi.org/10.1017/S0968565013000140 Kemmerer, EW. 1916. Modern Currency Reform: A History and Discussion of Recent Currency Reforms in India, Port Rico, Philippine Islands, Straits Settlements and Mexico. New York: Macmillan. Keynes, JM. 1913. Indian Currency and Finance. London: MacMillan. Keynes, JM. 1971. A Treatise on Money:The Pure Theory of Money. London: Macmillan. Kindleberger, CP. 1986. The World in Depression, 1929–1939, revised and enlarged edition. Berkeley: University of California Press. Kuroda, A. 2005. ‘The collapse of the Chinese imperial monetary system,’ in Sugihara, K. (ed.) Japan, China and the Growth of the Asian International Economy, 1850–1949, 103–26. Oxford: Oxford University Press. doi:10.1093/acprof:oso/9780198292715.003.0005 Kuroda, A. 2007. ‘The Maria Theresa Dollar in the early twentieth-century Red Sea Region: A complementary interface between multiple markets,’ Financial History Review 14(1): 89–110. https://doi.org/10.1017/S0968565007000376 League of Nations. 1931. The Course and Phases of the World Economic Depression. Geneva: League of Nations. Luthringer, GF. 1934. The Gold-Exchange Standard in the Philippines. Princeton: Princeton University Press. McClellan, CW. 1980. ‘Land, labour and coffee: The south’s role in Ethiopian self-reliance, 1889–1935,’ African Economic History 9: 74–7. doi:10.2307/3601388 McPhee, A. 1971. The Economic Revolution in British West Africa, 2nd ed. London: Routledge. Miron, JA. 1986. ‘Financial Panics, the seasonality of nominal interest rate and the founding of the Fed,’ American Economic Review 76(1): 125–40. Nawata, M. 1932. Futsuryo Indoshina kaheishi.Yokohama:Yokohama Specie Bank. Noyes, AD. 1909. Forty Years of American Finance. New York: G. P. Putnam’s Sons. Pallaver, K. 2009. ‘ “A recognized currency in beads”. Glass beads as money in 19th-century East Africa:The central caravan road,’ in Eagleton, C., Fuller, H. and Perkins, J. (eds.) Money in Africa. London: The British Museum.

194  Global history of monetary delocalization Pankhurst, R. 1962. ‘ “Primitive money” in Ethiopia,’ Journal de la Société des Africanistes 32(2): 213–48. Pankhurst, R. 1965. ‘The history of currency and banking in Ethiopia and the Horn of Africa from the Middle Ages to 1935,’ Ethiopia Observer 9: 358–408. Pankhurst, R. 1968. Economic History of Ethiopia, 1800–1935. Addis Ababa, Ethiopia: Haile Selassie I University Press. Pankhurst, R. 1970a. ‘A preliminary history of Ethiopian measures, weighs and values (Part 3),’ Journal of Ethiopian History 8(1): 31–54. Pankhurst, R. 1970b. ‘The perpetuation of the Maria Theresa dollar and currency problems in Italian-occupied Ethiopia 1936–1941,’ Journal of Ethiopian History 8–2: 89–117. Peez, C. and Raudnitz, J. 1898. Geschichte des Maria-Theresien Thalers.Vienna: G. Graeser. Pope, G. 2001.‘The Maria Theresia Dollar in the late nineteenth and early twentieth century,’ in Bachinger, K. and Stiefel, D. (eds.) Auf heller und cent. beiträge zur finanz- und währungsgeschichte, 253–78.Vienna: Ueberreuther. Robequain, C. 1944. The Economic Development of French Indo-China. Ward, IA. (trans.). London: Scranton. Robertson, DH. 1948. Money, 4th ed. Cambridge: Cambridge University Press. Said, H. 1935. Monetary and Banking System of Syria. Beirut, Syria: American Press. Sauerbbeck, A. 1904. ‘Prices of commodities in 1903,’ Journal of the Royal Statistical Society 67(1): 85–99. Sauerbbeck, A. 1911. ‘Prices of commodities in 1910,’ Journal of the Royal Statistical Society 73(3): 315–29. doi:10.2307/2339852 Sauerbbeck, A. 1924. ‘Wholesale prices of commodities in 1923,’ Journal of the Royal Statistical Society 87(2): 256–77. doi:10.2307/2341227 Schaefer, C. 1992. ‘The politics of banking: The Bank of Abyssinia, 1905–1931,’ International Journal of African Historical Studies 25(2): 361–89. doi:10.2307/219391 Schaefer, C. 1996. ‘Serendipitous resistance in fascist-occupied Ethiopia, 1936–1941,’ Northeast African Studies, new series 3(1): 87–115. Semple, C. 2005. A Silver Legend:The Story of the Maria Thresa Thaler. Orrel: Barzan. Shiroyama, T. 2008. China During the Great Depression: Market, State, and the World Economy, 1929–1937. Cambridge, MA: Harvard University Asia Center. Stride, HG. 1956. ‘The Maria Theresa Thaler,’ Numismatic Chronicle, sixth series 16: 339–43. Théret, B. 2018. ‘Monetary federalism as a concept and its empirical underpinnings in Argentina’s monetary history,’ in Gomez, GM. (ed.) Monetary Plurality in Local, Regional and Global Economies, 84–113. London: Routledge. Tschoegl, AE. 2001. ‘Maria Theresa’s Thaler: A case of international money,’ Eastern Economic Journal 27(4): 443–62. Webb, JLA. Jr. 1982. ‘Towards the comparative study of money: A reconsideration of West African currencies and neoclassical monetary concepts,’ International Journal of African Historical Studies 15(3): 455–66. doi:10.2307/218147 Weber, M. 1924. Wirtschaftsgeschichte. München und Leipzig: Duncker & Humblot. Xie, HS. 1988. ‘Qingmo gesheng guanyinqianhao yanjiu (1894–1911),’ Zhongguo shehui kexueyuan jinji yanjiusuo jikan 11: 199–274. Zelizer,V. 1994. The Social Meaning of Money. New York: Basic Books. Zhonghua minguo youzheng huidui chujin ju. 1937. Zhonghua minguo ershisi niandu youzheng huidui chujin ju shiye nianbao. Nanjing: Zhonghua minguo youzheng huidui chujin ju.

Conclusion Money as social circuit

1 A global history of monetary delocalization History tells us that, to the extent that people exchange proximately, whether by anonymous currency or by named debt, money works endogenously. No exogenous institution in anywhere in the world substantially intervened in autonomous local transactions until the late 13th century. Then, for the first time, a common unit of account emerged beyond the boundaries of a single civilization.1 However, even in upper-level markets, no material yet circulated universally. Only in the sphere of Quadrant I did a commensurable form of monetary unit appear to work. Local monies remained independent almost as they had been before. Three centuries later, a common unit materialized through the global silver march. Silver, including large dollar coins, made Quadrant I affluent to an unprecedented extent and affected the other three Quadrants. Responding to the contraction in silver circulation, local exchanges transformed in three ways: by creating local currency, organizing local credit and consolidating exchanges centred on oligarchies. In the course of competition between these three paths, some states happened to organize Quadrant III along the networks for Quadrant II and eliminated the autonomous Quadrant IV.2 Then a nationwide space in which a set of currencies was proportionately denominated in a single unit of account covering all levels of transactions appeared. However, most of the human business activity was conducted in local currencies until the 19th century. Though the precise time and place are still a matter of debate, it appears that the transferability of bills of exchange in Europe launched in the late 14th century, and it became common after the early 17th century (Trivellato 2019, 32). It is probably not wrong to assume that the rapid shrinkage in large currency for distant trade (Quadrant I) caused credit devices for distant trade (Quadrant II) to be established after the first silver century and proliferate after the second silver century. Through the formation of the international gold standard regime, which absorbed local currency circuits particularly in the peasant economies most commonly seen in Asia and Africa, convertible monies seemed destined to

196  Conclusion

pervade the world in the early 20th century (Kuroda 2008b). However, the currencies under the gold standard system were too convertible to accommodate monetary demands diversified horizontally as well as vertically. After a significant fall in international trade through the 1930s and the proliferation of managed currency systems, in which a central bank monopolized the issuance of currency, the complementarity among monies appeared to vanish, with the exception of a division of labour between the standard currency for international settlements, US dollars and a number of currencies homogeneously mediating transactions within national borders. Thus, local monies created endogenously are now marginalized or made illegal through the exogenous institutions of states. However, the minor role of such local monies at the present time does not mean that the variety of exchanges shown in the four quadrants disappeared. Nor does the current popularity of national money justify prohibiting locals from creating money.3

2  Modern ‘common sense’ uncommon in history The global history of money viewed from the ground reveals that modern common sense about money is not always applicable to the past. For example, in the preceding chapters, we have seen ample historical cases disproving the assertion in academic discourse that ‘bad money drives out good money.’ At the ground level, old superior coins often worked in tandem with new inferior coins. The substitutive idea has blinded us to the complementary relationship among monies that was once found universally throughout global history (Kuroda 2008a). Emancipation from the substitutive framework draws our attention to the invalidity of other common sense assumptions relating to money. When the circulation of a currency suddenly decreases in any particular circumstances, what happens to prices? According to the quantity theory of money, a common sense assumption of our age, prices must fall according to the decrease of monetary supply. However, in history, there are many cases showing exactly the opposite, contrary to the quantity theory of money, where a shortage of currency caused a spike in prices. Take one example from a story we presented earlier. In May 1939, Aden suffered from a shortage of the Maria Theresa dollar. According to a letter from the Aden Administration, prices of vegetables, eggs and meats rose.These products came from inland and were paid for in Maria Theresa dollars. The short supply of the dollar was said to have raised the prices of food.4 To cite another example, in Chapter Three we saw that the government in Beijing ordered a prohibition on shroffing copper coins, since the shroffing caused prices to hike. The authorities judged that it was not an increase in inferior coins but the removal of some coins and selecting particular coins by merchants that triggered the spike in prices. Other sources in history also show that it was not the shortage of currency in general but the shortage of particular small currencies that caused local price rises. For example, observing the transition of the circulating medium from copper coin to official

Conclusion  197

paper money in the late 13th-century Lower Yangzi, a bureaucrat wrote that shortages of small denomination notes, xiaochao, caused prices to rise more than twice as high as before (Cheng 1983–86, 117). A price in reality is not an inverse function of the monetary supply as the quantity theory of money indicates. It must be, primarily, an index showing the degree of difficulty to obtain a particular good or service. All three cases mentioned earlier show that a decrease in a currency made it difficult for locals to purchase necessities like food. Following the concept of the four quadrants of exchange, we realize that the disappearance of a means of exchange made anonymous/proximate exchanges hard to sustain.5 This causality in history leads us to notice another consequence that is against the common understanding. Many thinkers, such as Hicks, believe that markets create money (Hicks 1969, 63), but in some situations, this relationship of causality can be reversed: a monetary supply can create markets.There were cases when increasing supplies of currency stimulated the emergence of marketplaces. As we saw in Chapter Three, the emergence of new marketplaces in Holland and Japan followed after an increasing supply of currencies during the period of the Mongol regime, even though in a later period the rulers of domains in Japan appeared to believe that a stable currency supply was a precondition for successfully establishing a new marketplace (Abe 1965, 52–3). A marketplace provides people with opportunities for anonymous exchange. Even if a seller and a buyer are acquainted, in markets they can conduct one-time transactions. The principle of one-time transactions – i.e., a transaction settled without any expectation of a subsequent transaction – makes cash indispensable in the marketplace as we have seen (Kuroda 2018). If currency is indispensable for a marketplace, then providing currency can generate a new marketplace. We can generalize this consequence in a broader sense. A transformation of the means of exchange can change the ways of exchange. History reveals that this certainly happened. Take the case of a mountainous district in late 18th-century Colombia. Antioquia produced gold dust, but there was no small-denomination currency there. Gold dust worked as units of account, but people had difficulty in actually exchanging what s/he possessed for what s/he wanted. Day labourers were paid with cloth which was imported from coastal regions. An episode in which a governor lost a bet and paid up in cloth shows how cloth had appreciated sharply in value. Probably cloth worked as a mean of exchange to some extent but very inadequately. In this situation there was no store selling general goods anywhere in Medellin, whose population was 16,000. The introduction of silver coins changed this situation in the 1780s under the Bourbons. A new governor encouraged the merchants who imported necessities including cloth in exchange for gold dust to use silver coins instead. In the 1790s, 26 general stores were licensed in Medellin.The circulation of silver coins among locals in Medellin created a domestic market for themselves. Numerous local dealers replaced the small number of distant traders who directly exchanged cloth for gold dust. Thus, currencies transformed the exchanges in the region from an oligarchic model to an anonymous one (Twinam 1982).6

198  Conclusion

Here is a clue for us to understand why local currency circuits persist so stubbornly in history. In this case, a silver coin was introduced along the chain of exchanges: gold dust for silver coin and silver coin for cloth. Once this chain was established, as long as locals sold gold dust and purchased cloth, it was difficult for other currencies to substitute for the silver coin. The concept of the chain of exchanges reveals us the flaw in another common framework of understanding.

3 The misuse of the concept of arbitrage: no equilibrium amidst streams As we have demonstrated, a multiplicity of currencies can take the form of several different, coexisting currency circuits. Under such conditions, two currencies often had different exchange rates depending on their circuits. Such gaps could be greater than any losses incurred through transportation costs. As we saw in Chapter One, peasants in the mid-Yangzi region preferred 1,000 copper coins to one silver dollar, even though the latter could be exchanged for 1,300 copper coins. It seems natural to wonder why arbitrage did not result in a convergence of the two exchange rates. However, such ‘common sense’ actually depends on the false assumption that holders of copper coins and holders of silver dollars would only consider the exchange rate between copper coins and silver dollars. In practice, copper coins might have been useful for purchasing raw cotton in rural areas, while silver dollars might have been used to import kerosene oil through a treaty port. If the trade in raw cotton could yield greater profits than the exchange of copper coins and silver, there would be no incentive for holders of copper coins to exchange them for silver dollars. By the same logic, if the business of kerosene oil was expected to yield large profits, holders of silver dollars would not have exchanged them for copper coins. Copper coins might have been dominant in the rice trade as well, while silver dollars might have been essential for importing machine-made cotton yarn. Thus holders of copper coins or silver dollars not only considered the exchange rate between copper coins and silver dollars but also calculated the profitability in purchasing goods which they could secure only with copper coins or silver dollars respectively. The point is that the complementarity of monies reflects diversified chains of exchanges that could not be integrated through a single exchange rate. The chain of exchanges for raw cotton and rice through copper coins and the chain of exchanges for kerosene oil and cotton yarn through silver dollars could be connected, as a merchant might engage in both the export of raw cotton and the import of cotton yarn. However, the two chains could not be easily integrated. The former might have been dominated by proximate exchanges and therefore inclined to spot transactions, while the latter might have extended interregionally and been more open to deferred payment. Spatially, the former chains formed a cluster of transactions which were made in a single marketplace, while the latter chains extended to weave a long-distance network which

Conclusion  199

linked cities. Thus, one cannot assume that the relative value of two monies along one chain of exchanges could converge with that along another chain of exchanges into a third, intermediate value. There have also been chains of exchanges even in modern times. Let us consider some cases of the network type from after the world depression. American monetary advisors to foreign countries after the 1930s insisted on them abolishing the circulation of foreign currencies, including US dollars. The expulsion of money of foreign origin and the establishment of a central bank with a monopoly on note issuance seems consistent with the political movements that pursued independence from colonialism. However, traders often resisted the demonetization of foreign currencies with which they had made many transactions. For example, those engaged in trade with the US related to industries in the Philippines and Puerto Rico opposed the bans against the circulation of US dollars.The prohibition of US dollars required traders to change their chains of exchanges and use new national currencies. Chains of exchanges gave monies a certain inertia that led traders to continue using them. These chains could extend across administrative boundaries along streams of international currencies. This is exactly why a new government with a central bank monopolizing note issuance wanted to disconnect the chains: to keep control over domestic monetary supplies. Thus, for example, after independence the Indian government substituted a new, external rupee for the original rupee that had dominated business in the Gulf region. The external rupee, not available in India, could not attain acceptability in the Gulf region, and states in the region subsequently established their own national monies (Helleiner 2003, 167, 209–10). A currency mediating one trade circuit and another currency mediating another are exchangeable but not fungible. Money is a social circuit connecting people who want to exchange.

4  Escaping the teleology of monetary history Without understanding the concept of the complementarity of monies, the idea that money has evolved steadily towards dematerialization seems very persuasive. First, commodity currencies such as grains or cloth were replaced by metal currencies, metals were then replaced by paper monies, and, finally, notes evolved into electronic devices. This teleological thought is obviously closely related to experience of the history of Western Europe, in which paper monies developed only in a very recent period. However, global comparisons clearly show many cases in which new monetary usages tended towards materialization rather than dematerialization. The disappearance of official paper monies and the proliferation of silver by weight in China after the 15th century is a typical case of this, as we saw in Chapter Three. Until recently, no paper money could cover all transactions.7 Actually, it only worked well in combination with other devices, even when authorities prohibited the use of other currencies, as in China in the period under Yuan rule and again after 1935, which we demonstrated in Chapter Three and Chapter Five.

200  Conclusion

As we found in our comparison of local notes in England, China and Japan in Chapter Four, the relationship between paper monies in Quadrant IV and devices in the other three quadrants differed by country. Here we should note that the idea that money evolves towards dematerialization is dominated by a substitutive viewpoint that overlooks the ways in which a division of labour in monetary devices can cover a variety of exchanges. A global history of money also reveals that the idea of mono-functional money is an illusion. Bulky and fragmentally valued cowries, shell monies, might have appeared to work only as a means of exchange. However, the attributes of cowries did not prevent them from serving as devices for accumulating assets. And in fact at times people and authorities in certain regions stocked them as assets. The framework of mono-functional money also neglects the division of labour among monetary devices. Cowries and copper coins worked mainly as means of exchange in a complementary relationship with other devices, such as silk or silver, that served to accumulate wealth. The myth of mono-functional money also exposes the flaws of another teleology. Without knowing the historical facts, it seems superficially persuasive that, as the market economy advanced, imperfect monies, or mono-functional monies, of the pre-modern era evolved into a perfect money that fulfiled all monetary functions in the modern era. In fact, each currency could potentially fulfil all monetary functions, but clusters of traders endogenously associated certain devices in complementary ways to more effectively cover a range of transactions. After the 20th century, it became common for a nation state with a central bank to organize monetary devices exogenously into a convertible set. However, in this case money did not evolve by itself; rather, an exogenous institution replaced endogenous ones to structure exchanges. The myth of an evolution towards perfect money is closely related to the idea that the three monetary functions (means of exchange, unit of account and store of wealth) should form a trinity. However, this notion obscures the fact that two of those functions, means of exchange and unit of account, have contradictory features. No currency comprised of a material substance can be supplied infinitely, but an infinite amount of money can be made through a unit of account. A currency without a denomination cannot work practically, but the denomination does not matter in transactions made only through a unit of account (or transactions without substance, such as bookkeeping). That is why, in history, a means of exchange very often deviated from the unit of account, as we saw in many cases throughout this volume. The concept of perfect money obscures the opposing features of these two functions of money and has prevented thinkers from noticing the complementarity of monies. However, is it true that nobody ever really noticed it?

5 Alternative ideas about money by contemporaries In the middle of the 17th century, Huang Zongxi, a noted Chinese Confucian scholar, insisted on abolishing silver (and gold) and using only copper cash. He

Conclusion  201

thought that the use of silver widened the gap between the rich and the poor and made it easy for corrupt officials to collect bribes (Huang 1985, 38). His arguments were so notional and ethical that they sound impractical to modern readers, but, at the time, the growing use of silver was generating many social problems, such as the grain riots that we saw in Chapter Four. Another Confucian in a neighbouring country submitted even more detailed proposals on the same topic. In 1727, Ogyu Sorai, a prominent Japanese Confucian, prepared a book in which he discussed contemporary political issues, which he was allowed to present to the Shogun, Tokugawa Yoshimune. Sorai recognized that, unlike gold and silver, bulky copper cash did not return to the cities but stayed in the countryside. In principle the issuance of copper cash was the monopoly of the central government, but Sorai argued that, instead, the domains should be allowed to mint coins locally to make up shortages in provincial cities. This proposal was not derived from abstract philosophy, but was based on his own experience living in the countryside before moving to the capital city.8 Though he cannot be called an economist in the modern sense, he realized that one currency circulated differently from another, and that some currencies were more stagnant than others. Contemporary thinkers identified the different features of currencies in their time and proposed resolutions to various problems that arose due to those features. They did not, however, develop ideas about how to combine the features of different currencies to form a complementary system. Similarly, after the establishment of the political economy, social scientists did not construct a framework to explain the division of labour among monies.9 In describing the circulation of banknotes in early 19th-century England, Tooke hinted at a relationship between lower denomination notes and local monetary demand (Tooke 1884, 127–8). Keynes demonstrated a similar division of labour between smaller denomination notes used locally and larger ones used for interregional settlement in early 20th-century India (Keynes 1913, 28–9). However, neither of them developed structural explanations for their observations. Keynes theorized a separation between liquidity for speculation and liquidity for transactions, but he did not develop the idea that the latter might be further divisible depending on space and denomination. In developing the quantity theory of money, Fisher originally separated actual money and deposits subject to transfer by check, since he thought, correctly, that they had different velocities (Fisher 1911, 48). However, he did not relate the different velocities of currencies to institutional factors. In the modern era, the effect of seasonality on the entire economy became negligible, and cash transactions came to make up only a marginal portion of all transactions. People became unaware of the high wastage rates of currency and ignorant of the choice between freedom and certainty in transactions. As a result, ideas about money began to deviate from earlier realities, and theories became confined to models in which currencies worked as liquids that were homogeneous and lacking in viscosity.

202  Conclusion

6  Institutions for flexibility as well as for certainty Different kinds of exchanges worked in association with a variety of rules that traders observed, implicitly or explicitly. Subsequent transactions are stabilized through codes of practice directly encouraging the consequences of exchanges. Typically the codes contained punishments for violators. Meanwhile, onetime transactions were maintained not through any codes directly designed to enhance anonymous exchanges but through more general institutions which indirectly discouraged an oligarchy from monopolizing exchanges. Under the Chinese imperial regime, a range of regulations, such as the equal division of inherited properties among sons, open bidding in adopting local measurers who weighed goods in transactions to earn a commission, the ban on appointing locals as governor, and others, incidentally, also indirectly guaranteed an open market system in which nobody could establish a monopoly over transactions in the marketplace. Each individual regulation was not specifically directed towards commercial activities, but the whole range of indirect regulation created a foundation in society which served to maintain open access to markets. Historical trace of ten periodic markets in Qingyuan County reveals that the rural market system was so deeply embedded in Chinese society that the cycle of the opening day in every ten day period underwent no change throughout the Kangxi period (1662–1722), the Republican period (1912–49) and even in 1986, soon after the revival of the ‘free market,’ ziyoushichang, in rural areas (Cui 1990, 61). The introduction of railroad services in the early 20th century and the prohibition of commercial activities during the Cultural Revolution did not eradicate the system through which ordinary peasants make local exchanges. We can find that the preference for one-time transactions among peasants remained so strong precisely to sustain opportunities to exchange anonymously in person. The Qing dynasty, the Nationalist Government and the reformist regime after Mao had no specific policies to promote rural markets. However, without explicit support, the establishment of wider institutions worked indirectly to encourage peasants to exchange through marketplaces. The government maintained their centralizing power by preventing oligarchies from monopolizing local businesses, meanwhile locals could decentralize commercial activities at the ground level. Ubiquitous petty exchangers, which we have seen through this book, represent the structure through which ordinary peasants engaged in one-time transactions at marketplaces that worked independently from distant exchanges. From a conventional viewpoint, the proliferation of small-scale moneychangers seems to be a case in which short-sighted traders kept heterogeneous currencies in order to secure small profits through the exchange of money, and thereby sacrificed trade growth and imposed large transaction costs on the entire economy. However, we now know that the use of small currencies in marketplaces provides people a greater degree of freedom in exchanges.We also know that small currencies were useful for mediating anonymous/proximate exchanges, but that these devices also tend to stagnate and were difficult to supply on demand. From the viewpoint of the end users who used currency

Conclusion  203

in exchanges for daily necessities, petty moneychangers thus facilitated free trade and prevented professional dealers from monopolizing currency supplies to gain an oligarchic dominance over marketplaces. In addition to institutions that integrated transactions and brought certainty to exchanges, there were other institutions that secured flexibility and thereby domesticated the rampant instability brought to exchanges by outside forces. The conventional belief that a standardization of transactions raises effectiveness in business often overlooks the risk of losing elasticity against fluctuating circumstances. Again, the coexistence of plural monies was functional rather than incidental, in that exchanges were not homogenous. To avoid artificially standardizing the heterogeneity among exchanges sometimes brings stability to entire transactions. Thus, one money could fulfil a function that another could not, and vice versa. In such a situation, the relationship between one money and another is complementary, not substitutive. Adam Smith remarkably depicted money as being something like a highroad (Smith 1896, 191). It creates nothing by itself, but nothing can circulate without it. However, he failed to pay attention to the plurality of money that was still dominant across the world in his period. Not even Western Europe was an exception. Cesare Beccaria, an Italian economist, wrote of 22 different gold coins and 29 silver coins, as well as black money and copper coins, that circulated in Milan before the French Revolution (Einaudi 1953, 241–3) – a situation similar to that of late 18th-century Dacca, East India (Mitra 1991, 54). Extending Smith’s metaphor, we can say that roads also occur side by side in different complementary forms and can practically work only in a combination of highways, streets, lanes and paths. Some connections may allow direct operation, while others may require transfer. How to organize monies determines what a society is like.

Notes 1 Pamuk rightly suggested the possibility that silvers from Central Asia caused the rise of silver usage in the eastern Mediterranean in the late 13th century (Pamuk 2000, 25–6), though in fact these silvers originated from China. 2 Referring to Weber and Schumpeter, Ingham stressed that the integration with the debt finance of the state strengthened the acceptability and transferability of private banknotes. Ingham 2015, 15. 3 The IMF Articles of Agreements prohibits multiple currency practices. Blanc 2016, 244. 4 BLIO, F1863/1939. 5 In the opposite direction along this understanding, a larger supply of a currency sufficient to prevail among end users could generate a premium with the currency against other devices as a means of exchange mediating local transactions. Grubb argued that the case of Colonial Virginia’s paper money showed more ‘moneyness’ value against tobacco, etc. Unlike the quantity theory of money, the more ubiquitous the paper monies became, the more of a premium in local transactions they won. Grubb 2018. 6 Land reform in Bolivia shows an example of the relationship between an oligarchy and the biased usages of currency. Before the reform, landlords made large-sum monetary transactions to sell products in their farms to city and mines, while tenants had few

204  Conclusion opportunities to use currencies. After the reform, which gave peasants the rights to manage land, purchases of cloths and furniture by peasants substantially increased. Clark 1968. 7 In 1820, in France, the denomination of the smallest banknote issued by the Banque de France, 500 francs, was equivalent to more than six months of a worker’s wages. Baubeau 2016, 5. 8 Ogyu 1987, 133–46. Guillaume Carré brought this important essay to my attention. 9 In analyzing paper money, Adam Smith distinguished the circulation among dealers and the circulation between dealers and consumers. He noticed that large denomination notes suited the transactions of the former, while the transactions of the latter needed small denomination ones. However, he failed to formulate this in a spatial context. Smith 1970, 421–3.

References Abe, 1965. Kakeiri nominshi. Tokyo: Shibundo. Baubeau, P. 2016. ‘Is cash an endangered species?’ Politique Internationale: 1–6. Blanc, J. 2016.‘Unpacking monetary complementarity and competition: A conceptual framework,’ Cambridge Journal of Economics 41(1): 239–57. https://doi.org/10.1093/cje/bew024 Cheng, JF. 1983–86. Xuelou ji in Wenyange siku quanshu, vol. 1202. Taipei: Shangwu yinshu. Clark, RJ. 1968. ‘Land reform and peasant market participation on the north highlands of Bolivia,’ Land Economics 44(2): 153–72. doi:10.2307/3159311 Cui, XL. 1990. ‘Jiating, shichang, shequ: Wuxi Qingyuan nongcun shehui jingji bianqian de bijiao yanjiu (1929–1949),’ Zhonguo jingji shi yanjiu (1): 42–65. Einaudi, L. 1953. ‘The theory of imaginary money from Charlemagne to the French Revolution,’Tagliacozzo, G. (trans.) in Lane, FC. and Riemersma, JC. (eds.) Enterprise and Secular Change. London: George Allen & Unwin. Fisher, I. 1911. The Purchasing Power of Money: Its Determination and Relation to Credit Interest and Crises. New York: Macmillan. Grubb, F. 2018. ‘Colonial Virginia’s paper money, 1755–1774: Value decomposition and performance,’ Financial History Review 25(2): 113–40. https://doi.org/10.1017/ S0968565018000057 Helleiner, E. 2003. The Making of National Money: Territorial Currencies in Historical Perspective. Ithaca: Cornell University Press. Hicks, JR. 1969. A Theory of Economic History. Oxford: Clarendon Press. Huang, ZX. 1985. Huang Zongxi quanji, vol. 1. Hangzhou: Zhejian guji. Ingham, G. 2015. ‘ “Great divergence”: Max Weber and China’s missing links,’ Max Weber Studies 15(2): 1–32. Keynes, JM. 1913. Indian Currency and Finance. London: MacMillan. Kuroda, A. 2008a. ‘What is the complementarity among monies? An introductory note,’ Financial History Review 15(1): 7–15. https://doi.org/10.1017/S0968565008000024 Kuroda, A. 2008b. ‘Concurrent but non-integrable currency circuits: Complementary relationship among monies in modern China and other regions,’ Financial History Review 15(1): 17–36. https://doi.org/10.1017/S0968565008000036 Kuroda, A. 2018. ‘Strategic peasant and autonomous local market: Revisiting the rural economy in modern China,’ International Journal of Asian Studies 15(2): 195–227. https://doi. org/10.1017/S1479591418000049 Mitra, DB. 1991. Monetary System in the Bengal Residency. Calcutta: Firma K.L. Mukhopadhyay & Sons. Ogyu, S. 1987. Seidan. Tokyo: Iwanami.

Conclusion  205 Pamuk, S. 2000. A Monetary History of the Ottoman Empire. Cambridge: Cambridge University Press. Smith, A. 1896. Lectures on Justice, Police, Revenue and Arms. Oxford: Clarendon. Smith, A. 1970. The Wealth of Nations. London: Penguin. Tooke,T. 1884. A History of Price, and of the State of the Circulation, from 1839 to 1847. London: Longman, Brown, Green and Longmans. Trivellato, F. 2019. The Promise and Peril of Credit: What a Forgotten Legend About Jews and Finance Tell Us About the Making of European Commercial Society. Princeton: Princeton University Press. Twinam, A. 1982. Miners, Merchants, and Farmers in Colonial Colombia. Austin, TX:  University of Texas Press.


Model 1: the reason making streams of currency unidirectional

Figure A.1  Unidirectional streams of currency with frictions

ρu Vu Au = ρl Vl Al A = f(s, p, t) v: velocity of currency (in annual average) s: degree of spatial dispersion of traders p: population independently engaging in trade t: temporary bias in demand for money such as seasonality If the fluid is incompressible, the density (ρ) must be constant. The difference between Al and Au becomes larger, as the rural market (a lowerlevel market) covers more space (s), has more traders (p) and its monetary

Appendix  207

demand has more temporal bias (t). Assume that the size of both the purchase and the sale from urban (upper-level) to rural (lower-level) is the same in terms of a monetary unit. For example, the urban side purchases a bottle of wine of 10 cents 1000 times, while it sells a cloth of 10 dollars ten times. From the urban side viewpoint 100 pieces of one dollar coin appear to be paid and returned. However, in the sphere of the rural market the dollar should be converted into another currency, such as brass penny whose face value is less than one dollar, as if a streamline would be divided into multiple sub-lines. Then inelastic supply of small money disables all the money to convert as if the vortexes (of dollar coins) would appear in the left corners. In return to urban market small monies should be converted to a full dollar. However, some small coins dispersed in rural market fail to assemble as if the vortexes (of penny coins) in the right corners would emerge. The vortexes accounted for the unaccounted loss of coins out of visible circulation.Thus, the same monetary unit has different contents between urban market and rural market, and as long as the conversions are not perfect, additional supplies of currency should be necessary to keep the size of trade. Since the usefulness of coins differs between the two markets, the exchange rate can be different according to the market and change depending on situation, such as harvest season or slack season. The fluid dynamics assumes that when the fluid is compressible, the density (ρ) must be inconstant.1 If we are allowed to take it, the streamline of the currency in the sphere of the rural market can have a different density from that in the urban one. Keeping the parity between dollar coin and penny coin, a covertible currency system adjusts monetary supply to keep the density constant, while adjusting the exchange rate between dollar coin and penny coin, the incovertible one accommodates the densities to supplement the inelastic supply.

Note 1 A part of Figure A.1 was originally taken from Ishiwata 2000, 101 before Kuroda modifies it.

Reference Ishiwata, R. 2000. Ryutai rikigaku nyumon. Tokyo: Morikita shuppan.


Note: Page numbers in italic indicate a figure and page numbers in bold indicate a table on the corresponding page. Abu-Lughod, J. L. 67 Addis Ababa 156 – 7, 163 – 4, 167; dollar’s price in 165; occupation of 165 agriculture/agricultural 20; labours 144 – 5; land, rent charge against 98; productions, seasonality of 2; products, prices of 173; societies 12, 49 – 52; unfavourable conditions for 22 alternative monetary system 151 annual sales and purchases per households 23, 24 annuities 130 – 1; capital value of 128 – 9; contracts for 96; purchasers of 98; rates of 97 anonymity 3, 5, 40, 41 arbitrage 120, 165, 176, 198 – 9 Argentina 124, 172 – 3 Aristotle 1, 3 asper 83, 93 – 4 assets: forms of 38; proportion of 136 Austrian dollar 156, 158, 160, 165 bad money 11 – 14, 196 bank accounts 136 bank branch networks 174 banking systems 10, 16, 46 – 8, 130, 142; development of 47 – 8 banknotes 177 – 80, 183, 185, 186 – 7, 188; high-denomination 179; presence of 185 Bank of Communications 187 Bank of England 13, 131, 133 – 6 Bank of Japan (BOJ) 141 – 3 baoyin system 77; establishment of 79 Battuta, I. 90 Bombay 158 – 9 Britain 145, 153, 161 – 2 bulk transactions 68 – 9

bureaucrats 48, 102, 137, 141, 197 Burma, agricultural credit associations in 183 Chagatai silver coins 93 chain of exchanges 198 – 9 chaotic period 184 China/Chinese: banknotes 177; bronze bell in 87; cotton cloth 139 – 40, 140; native notes in 133; operational coins in 106; raw silk 121; silk from 72 – 3, 89, 123; silver dollar 178; transformations in monetary usage 126 Cipolla, C. 12 clustering coefficient 40 coins: excavations of 46 – 7, 69; gold 134; investigation of 45 – 6; operational 15; standard 103, 106; subsidiary 152 – 3; zinc 31; see also copper coins; silver coins Collier, C.S. 159 colonialism, independence from 199 commercialization 91 commercial oligarchies 143 – 5 commercial taxation 77 commodities 27, 67 – 8; currencies 68; by distance 24; exchange for 24 – 5; to Longxianghao 54; prices 119 – 20; value of 23 common sense 196 – 8 community in credit transactions 36 complementary monies 61 consecutive state bankruptcies 117 contemporary foreign observers 185 contemporary thinkers 201 conversion 32 convertibility 134 – 5, 181; of domain notes 135

Index  209 convertible monies 195 – 6 copper: cash, quantities of 49; supplies of 127; see also copper coins copper century 119 – 28 copper coinages 102; variety of 116 copper coins 69, 72, 74, 85, 125, 189; abundance of 88; circulation of 68; demonetization of 86; distribution of 47; exchange rate of 188 – 9, 198; form of 107; hoards of 107; in Indochina 175; in Japan 105; overabundance of 86; proliferation of 47; quantity of 189; in Thailand 175; transporting 116 – 17 cotton: cloth 139 – 40, 140; cultivation 33, 33, 140; industry 139; plantations 144 credit: to cash transactions 36; and community 36; devices 195; economy, formation of 128 – 32; inflation 183; instruments 118; oriented societies 95 – 100, 119; preconditions 3; system 107; transactions 3 crop farming 139 currencies 138 – 43, 201; acceptability 46; anonymous 195; of British East and Central Africa 176; circuits 62, 143, 176; circulation of 6, 46, 85, 171 – 2, 178, 196; and credit 3; demand and supply for 3, 13; denomination of 9; dependency on 38; distribution of 50, 51; in East Asia 118; economy 119 – 28; endogenous 52 – 3; end users of 11; exchange of 55; exogenous 45 – 7; and finances 140; functional differentiations of 59; honour substitutes for 35 – 9; issuance, fragmental localization of 184 – 5; locally differentiated 58 – 62; local usages of 125; multiplicity of 198; outstanding 48; popular form of 71 – 2; propensity of 48; quantity of 2; shortages of 39, 133, 135, 176, 196; significant proportion of 47; of smaller divisions 70; societies 119; stability 3; stagnant nature of 46; super-fragmental 32; suppliers of 45; supplies 75, 116; systems, proliferation of 196; transformation of 172; triple layer of 168; types of 12 – 13; unidirectional streams of 47 – 9, 206, 206 – 7; unification of 75; unintentional stagnation of 45; unit of 77; use of 35; wastage rate of 12, 47; working as complementary buffers 162 – 70 currency: unification 186 – 7, 189 custom economy 8

daily transactions 25, 25 day labour markets 21 Decennial Report 27 deferred payments 25 – 6 dematerialization 58 – 62, 199 demonetization, of copper coins 87 – 8 denomination 46, 84, 133 – 5, 179, 186; copper coins 180; of currencies 9, 31 – 2, 53, 126, 137; currency 6; imperfect substitution of 12; of money 31; problem 2 – 3, 12; substitution of 12 Desan, C. 13 distant trades 5 East Asia, actual monetary usages in 175 – 6 economic transformations 138 Edkins, J. 62 England, country bank notes in 135 Ethiopia 158, 165; emperor, revenue and expenditure of 160 Eurasia 116; age of commerce 88 – 95; commensurability across 84 – 5; silver century 73 – 6 Europe, credit-oriented systems in 95 – 100 excavations, functional differences of 71 exchange: proportion of 2; of services 8 – 9 expenditure 55, 56 false teleology 140 familiarity, degree of 5 family-based farms 174 feudal systems 141 finances 182; currencies and 140 financial instability 182 Fisher, I. 201 food distribution patterns 52 – 3 foreign currencies 199 foreign silver dollars 167 foreign trade 104 four quadrants of exchanges 6, 6, 9, 108, 175, 197 Frank, A. 116 funded debt 97, 97 – 8, 100 German silver mines 117 globalization 9 global monetary delocalization 67 global silver flows 10 global silver march 195 global synchronization 101 gold: annual production of 152, 152; coins 118 – 19, 134; dust 197; imports of 124; rushes 151; standard unit, fixed proportion to 171

210 Index gold standard system: adoption of 151 – 2; currencies under 196; framework 153 good money 11 – 14, 196 governmental bonds 132 grains 29; distant trade in 124; merchants 130; paid for rent 130; prices of 124 – 5; riots, problem of 125; supply 124 Great Depression 45 Gresham’s Law 11 Guyer, J. I. 13 – 14 Hayek, F. A. 13 Helleiner, E. 13, 151 Hicks, J. 45, 197 high-denomination banknotes 179 hired labour 33 Hubei Provincial Bank 182 hyperinflation: in Africa 51 – 2; in Asia 51 – 2 illegal trade networks 106 imaginary monies 61 – 2 Imperial Bank of Persia 174 import-substitutive industrialization 177 incomes, investigations of 22 Indian silver rupee 155 individual transaction 2 Indochina Bank notes 182 industrialization 11, 19, 33, 90, 138 – 43 Ingham, G. 13, 30 institutions for flexibility 202 – 3 interest rate of credit 129 international currencies 199 international gold standard system 9 – 10, 151; formation of 195 international market 162, 165 international silver dollar: prevalence of 154; system 153 international silver price 165 international transactions, circulation for 168 interregional currency 6, 124, 126, 161; supplies of 126 – 7 interregional payments 118 interregional settlement system 117, 141 interregional trade 120, 123, 130; development of 129 interregional transactions 27 – 8 Italian lire 155 Japan/Japanese 121; bronze bell in 87; commercial centre of 133; commercialized region in 34; cotton cloth 139 – 40, 140; credit banks 144; domain notes in 133 – 5; financial system 141 – 2; rice prices in 101; silver dollar

167, 169, 169; subsequent transactions in 37 Java 50 – 1, 51 Keynes, J. M. 201 Kiyotaki, N. 13 Korea, bronze bell in 87 labour: exchange of 34; shortages of 8 – 9, 21 land tax 76 liquidity 10, 182; shortages of 127 – 8, 143 litigation 35 local currency 40, 41, 138 localities, importance of 181 localization, of money 102 local monies 196 London, silver price in 163, 163 – 4 long-distance trade 91 – 2 lower denomination notes 201 macacas 119 MacGillivray’s report 156 – 7 machine-made yarns 140 managed currency systems 183 manillas 30 Maria Theresa dollar 154 – 62; in Addis Ababa 165; at Aden 159, 166; circulation 159 – 62, 169; complementary partners 170; cost of collecting 158; demand for 161; demonetization of 166; dual materiality and functionality of 170; in East Africa 167; exchange rate of 161, 163, 163 – 4; geographical sphere of 156; local rates of 155; material substance of 169 – 70; in Middle East and northern Africa 156; overvaluation of 155; own issue for 166 – 7; popularity of 154; replacing 166; right to issue 157; shortages of 157 – 8; spatial retreat of 156; supplier of 158; triple-layer monetary system 162; triple layer of currencies 168 market-based economy 8 marketplaces 23, 32, 35, 138 – 43; in China 20; density of 138; disappearance of 38 – 9; distribution of 138; emergence of 90 – 1, 197; exchanges in 30 – 1; in Holland 197; in Japan 197; one-time transactions at 202; products in 20; rural 23; in rural areas 3 markets: in Holland 90, 90 – 1; in Japan 90, 90 – 1 material silver 57 Mauss, M. 7 means of exchange 1, 3

Index  211 means of settlement 6 metallic currencies 68 – 9, 175 Mexican silver dollar 167 minting periods, distribution of 70 modern banking systems 53 monetary delocalization: Chinese silver 78 – 84; commensurability across Eurasia 84 – 5; copper coins 85 – 8; credit-oriented systems in Europe 95 – 100; currency circuits 100 – 7; Eurasian age of commerce 88 – 95; Eurasian silver century 73 – 6; global history of 195 – 6; independent monetary system 67 – 73; Mongolian taxation system 76 – 8; second silver century 107 – 8 monetary demand: existence of 1; seasonality of 172; seasonal shifts in 141 – 2 monetary economies 116 monetary history, teleology of 199 – 200 monetary issuance 183 monetary supplies, monopoly of 10 monetary system 68, 73 monetary transactions 2; growth of 152 monetary usage, viewpoint of 171 money/monies 1 – 5, 11; alternative ideas about 200 – 1; bad and good 11 – 14; chain of exchanges 198; in circulation 152; complementarity of 198; complementary relationship 155 – 6, 196; denomination of 31; exchange and institutional setting 7 – 9; exchange and multiplicity of 3, 7; exchange ratio among 61; four quadrants of exchange 5 – 7; global history of 196, 200; multiple strata of 86; multiplicity of 12 – 14; nature of 1, 3 – 4; quantity theory of 196; substitutive relationship between 11; understandings of 14; unification of 14; and usages in Yingkou 61 Mongolian taxation system 76 – 8 monmesen 134 monoculture economies 183 mono-functional money 200 moral economy theory 19 MTD, annual issuance of 157 multilateral clearance system 57 – 8, 60 – 1 multilateral counter-trade 58 multiple exchange zones 23 mutual credit 40, 40, 145 national currencies 183 – 4 nationalized money 151; international silver dollars 151 – 4; Maria Theresa dollar 154 – 62 national paper monies: dominance of 183; monetary unification by 184 native financiers 185

native notes: in China 138; convertibility of 134; distribution of 137; popular circulation of 137; popularity of 138 network theory 39 Nevskiy, A. 77 – 8 new/newly minted coins 103, 104 notes: conversion of 134; majority of 136 one-time transactions 21, 25 – 7, 32 – 3; preference for 41 operational coins 103, 106 order economy 8 palm oil, price of 30 paper currency 73, 80 paper monies 94, 102, 132 – 8, 180, 184; changes in 80; of denominations 171; in Indochina 174; introduction of 185; issuance of 81, 82; in Japan and England 136; nationalized peasant economies 171 – 7; popular circulation of 175; popularity of 94; standard 134, 183 – 90; taxation in terms of 80; in Thailand 174 paper piasters 183 payers, discouragement of 105 peasants: exchange with peasants 19 – 23; freedom or certainty 39 – 41; honour substitutes for currency 35 – 39; local commercialization 39 – 41; marketplace for one-time transactions 23 – 7; natives preferred local currencies 27 – 30; products, sales of 20, 20; subsequent transactions 32 – 5; ubiquitous fractional currencies and petty exchangers 30 – 2 peddlers 130 ‘people’s money’ 128 perfect money, concept of 200 physical substance, absence of 6 plural monies 203 Popkin, S. 19 postal relay system 78 prices: hike 196; movement of 26; revolution 119 – 20 private notes, issuance of 134 private remittances 141 proliferation of drafts 38 prominent annoyances 62 provincial banknotes 190 public bonds 132 public creditors 131 public debts 130 – 1 public finances 141 public money, transfers of 99

212 Index Qingyuan County 30 rate of interest 97 – 8, 98 raw cotton 35 raw silk, price movements of 121 Redish, A. 12 redistributive system 8 requisition tax 78 reserves, proportion of 182 revenue 55, 55 rice: cultivation 33, 33; money balance of 51; price, annual movement of 29, 29 Rome 157 – 8 Royal Niger Company 145 rural commercialization 35 rural markets 8, 23 – 4, 31, 36, 39 Scott, J. 19 seasonality 11, 136; degrees of 37, 141; in monetary demand 180 – 3 seclusion policy 132 secret code notes 178; denominations of 185 shroffing 103, 104 – 5, 196 silk, monetary role of 68, 71, 72, 73, 77, 83, 89, 93, 120 – 1, 123, 124, 141 – 2 silver 59 – 60, 67, 116 – 17, 195; annual production of 152; availability of 76; bars 84 – 5; certificates, issuance of 174 – 5; circulation of 80, 98, 119, 129, 195; coinage in London 74; conceptual function of 93; contraction of 29; copper money against 29; currency 84 – 5; demand, diversity in 118; disappearance of 177; drain of 137; in exchange 132; exchange of 198; imports of 124; inflow 126; ingots 57; international price of 184; introduction of 73; mining 72; minting of 75; price in London 155, 163 – 4; production of 72, 117, 119 – 20; quantities of 132; rupees, minting of 82; scarcity of 74; shortages of 75 – 6; supplier of 121; supply 85; transactions in terms of 125; uncoined 78 – 9, 94; usage of 83; use for 80 silver-based systems 171 silver coins 75 – 6, 83, 116 – 19, 197; circulation of 197; fiscal system and issuances of 78; form of 80; minting of 79; supply of 92; suspension of 62, 178 silver dollars 185, 186 – 7, 188; exchange of 198; global stream of 118; importance of 117 – 18; settlements in 117 silver famine 67 silver-gold exchange rates 151

silverization 77, 85; in taxation 82 Sino-Spanish trade 120 small currencies 127; disinclination of 12; unilateral movement of 49 small denomination notes 2 – 3, 69, 128, 161; shortages of 196 – 7 small-size transactions 60 Smith, A. 203 society 19 – 20; with clustering coefficient 40, 40; with preference for anonymity 40, 41 Sorai, O. 201 Spain 117 spot trade 176 standard coins 103, 106 stratified markets 88 – 95; in agricultural societies 49 – 52; business in 54 – 8 subsidiary coins 152 – 3 super-fragmental currencies 32 synchronization 106 Syrian paper money 183 tamga 93 taxation 107; forms of 76 – 7; silverization in 82; system 117; in terms of paper money 80; unit of 78 taxes, collection of 101 taxpayers 77 tax payments 73, 134 – 5 tax revenues 96 teleology of monetary history 199 – 200 temporality in monetary demand 180 – 3 tenants 173 ten-lire note 169 – 70 textile industries 139 Tongtaihao/Huangxueguang transaction 56 Tongtaihao/Longxiang transaction 56 traditional currencies 176 transactions 2 – 3, 34, 56 – 7, 59 – 60, 102, 138, 202; bulk 68 – 9; in China 137; by currency 3; of daily goods 94; decentralized features in 27; levels of 195; proportion of 70, 116; significant proportion of 152 – 3; standardization of 203; using IOUs 129 transformation 139 transition 79 – 80; from payment 80 transportation 23 – 4 triple-layer monetary system 162 Ueda domain 37 uncoined silver 78 – 9, 94 unfixed monetary institutions 14

Index  213 unification 184 – 5; of money 14; policy 187, 189 unified currency system 12 unified monetary system 32 unified paper money 185, 185 universal currency 60 Valencia, prices in 121 – 3, 121 – 3 Vienna 157 Wallerstein, I. 116 Wenqiaoyidui excavation in China 71

West Asia silver 67 wheat price 91 – 2, 92 Wright, R. 13 Yamawaki family, advance and purchase between 34, 34 Yuan Shikai dollar 178, 180 Zelizer,V. 13 zinc coins 31 Zongxi, H. 200