2024 年 41 巻 p. 37-52
The purpose of this article is to introduce research on the histories of currency and finance in Japan during the final years of shogunate (i.e., the Bakumatsu years) and the dawn of the Meiji period that followed (1858–1871). The article focuses on three areas: (1) the structure of ledgers, (2) currency units of calculation and values, and (3) the realities of merchant family management. It introduces those texts that should be read when doing research on three processes: (1) that of how the “local” set of bookkeeping rules referred to as the washiki chōai method that had become entrenched in early modern Japan switched over to the Western-style double-entry bookkeeping method in the early Meiji period; (2) that of how the diversity of the issuers of currency and units of calculation in the early modern period came see a consolidation of issuers and units of calculation in the early Meiji period; and (3) that of how the major financiers such as those who would issue loans to feudal lords advanced into the banking business or fell into run in the early Meiji period. At the same time, it also explains the historical significance of joint venture called Yamahiroya-Chōbei (山廣屋長平) having been established before corporations based on the joint-stock company form had emerged in Japan.
The purpose of this article is to introduce in English the research that has been conducted in Japan on the history of Japanese currency and finance in Bakumatsu and early Meiji periods (1858–1871) that relies particularly on sometimes difficult-to-read handwritten texts. My goal is not to provide an exhaustive introduction; rather, I intend to focus in particular on research based on analyses of ledgers.
Ledgers are an important primary historical resource for analyzing individual business entities. Information such as how much profit was earned in each and in what form were assets held has a bearing on how to evaluate management overall. Accordingly, while analyses of business histories using ledgers have been undertaken for many years in Japan, analyzing ledgers from the Bakumatsu-Meiji Restoration years ultimately calls for specific knowledge and skills.
First, as a prerequisite, it is necessary to decipher the writing used at the time. In order to read the cursive kuzushiji script that was generally used in Japan during the early modern period, it will likely be necessary to refer repeatedly to a kuzushi dictionary such as that published by Kodama (1993). Recent years have seen the development of artificial intelligence-based apps that can read kuzushi, as well as internet-based platforms where large groups of people work to decipher them;1 however, ultimately the author’s own judgement and responsibility are what come into play.
Based on whatever techniques for reading kuzushi we use, let us turn now to part of one ledger. Figure 1 shows the introductory “current accounts ledger” (tōzachō)2 section of the Kōnoike Zen’emon family papers (Kōnoike Zen’emon-ke monjo). Figure 2 is a direct English translation of that section and parts that follow. According to the table of contents, the creator was an entity known as “Yamahiroya-Chōbei,” and the document itself was written in 1870 (Meiji 3).
Source: Yamahiroya-Chōbei, “Tōzachō,” Meiji 3.2.22 (March 23, 1870), Kōnoike Zen’emon-ke monjo.
The original text is probably illegible to most readers today at first sight. While one might be able to decipher the characters in modern language, that does not mean one will understand the content.
What, then, was the purpose of a “current accounts ledger”? For the average Japanese family, keeping a single household accounts book may be sufficient today. However, just 150 years ago even the most prominent merchant families in Japan kept multiple ledgers. Without understanding how those ledgers are related to one another, we cannot understand the meaning of the numbers and names written in those ledgers. We should be able to infer the outlines of that relationship by investigating what sort of bookkeeping system was established in Japan during the period in question.
Figure 1 records ryō as units of calculation, and kensatsu as a type of currency. So, just how large an amount was 25,000 ryō? It may easily be surmised that kensatsu refers to a “note,” but what sorts of paper notes circulated at the time? When analyzing ledgers of the time, we need to do so based on the units of calculation that were then in use and the sorts of currency that were then in circulation.
Ultimately, just who (or what) was Yamahiroya-Chōbei? Kōnoikeya-Yamanaka Zen’emon (also known as “Kō-Zen” or “Yamanaka”) was a particularly wealthy merchant from Osaka, who built his fortune mainly from providing loans to daimyō. The Thirteenth National Bank that Kō-Zen later founded is also known as part of the family tree for the present-day MUFG Bank. Kō-Zen used both the trade name “Kōnoike” and the surname “Yamanaka;” “Yamanaka” can be seen in Figure 2. The name “Yamahiroya-Chōbei” was formed by taking the “Yama” (山) from “Yamanaka” (山中), “Hiro” (廣) from Kajimaya-Hirooka Kyūemon（加島屋廣岡久右衛門, i.e., Ka-Kyū), and “Chō”3 (長) from Kajimaya-Osada Sakubei (加島屋長田作兵衛, Ka-Saku); “Yamahiroya-Chōbei” thus does not represent the name of an actual person, but rather is a sort of code (trade name) that represents a business entity. What sort of operations did they conduct using this joint name? It is possible to clarify the realities of merchant family management by identifying who prepared the historical records in question and analyzing the ledger.
Work based on such a procedure is required for analyzing Japanese ledgers from the early modern era through the first part of the Meiji period (roughly 17th through 19th centuries CE). Accordingly, in the work that follows, I will introduce previous research that should be read with regard particularly to (1) the structure of ledgers, (2) currency units of calculation and values, and (3) the realities of merchant family management, while also offering some discussion regarding Figures 1 and 2 as an actual example of putting this into practice.
According to Tsumura (2019), no instruction manuals on techniques for doing bookkeeping were published during the Edo period (1603–1867); ledgers were prepared using styles that varied from merchant family to family. Even so, there were certain procedures that were loosely shared; these are referred to today as washiki-chōai, or Japanese-style account balancing. The oldest surviving Japanese ledger is the San’yōchō—literally, “notebook for calculations”—of the Ise family. Dating to 1638, this nearly four century-old ledger used a single-entry bookkeeping system for calculating assets; it shows total amounts of assets and liabilities.
Throughout the two-plus century-long rule of the Tokugawas (1600–1867), though not entirely without interchanges with Western society bookkeeping in Japan basically developed domestically on indigenous lines. The method that became established was to record each and every transaction every day as preliminary step in a bookkeeping journal (shiwakechō), and then posting them to the ledger. The types of bookkeeping journals were referred to by such names as a daifuku-chō or oboe-chō, depending on the family. The incoming and outgoing of cash and drafts would be temporarily consolidated in a single ledger. Multiple ledgers would later be created corresponding to the nature of the receipt or disbursement; for example, cash disbursed for a new loan would be posted to a loans ledger. Accordingly, it is said to be a double record based on “double entry calculations” or a “multiple-ledger system.”
Structural analyses of this washiki-chōai method have been conducted with a focus mainly on the merchant families of Osaka, which in the early modern period was a hub of finance and distribution. The ledgers of Kō-Zen in particular have drawn attention, and a string of studies have been produced.
Yasuoka (1960) and Yasuoka ([1970] 1998) both use the San’yō-chō to elaborate on the extent to which capital was accumulated. In the case of wealthy merchants like Kō-Zen, while they would create many types of business ledger at the same time, the one labeled San’yō-chō would be the final accounting ledger.
With the San’yō-chō, when performing calculations, they would first calculate the assets, liabilities, and capital, and then subtract the liabilities and capital from the assets to determine the ari-gin (here, meaning net assets at the end of the fiscal year). Profits and expenses would then be reckoned in, and finally the ari-gin—which combines start-of-fiscal-year capital and current-year net income or loss—would be reconfirmed. As to itemization of assets, in some cases the amount of accrued interest has been listed alongside to the amount of the loan. There is also a section that corresponds to assets and liabilities (uchioikata), and here borrowed capital would be itemized along with the equity capital of the Kōnoike family. Thus, by carefully persuing the San’yō-chō, it is possible to reconstruct to some degree the asset calculations and the profit and loss calculations for each fiscal year. Sakudo (1966) discusses the relationship between the San’yō-chō and other ledgers.
Shinbo (1971) notes that the account-keeping methods of major merchant management in the Edo period generally adopted a multiple ledger system and a transaction double-entry system. He notes that while there is little doubt that they used double-entry accounting, there were differences in the account-keeping methods used by the respective merchant’s management. The San’yō-chō that Kō-Zen introduced in 1670 provided a single system for complex and mixed account books. It made it possible to prepare financial statements at the end of the fiscal year. Shinbo points out that while the concept of “capital” to all intents and purposes was established in San’yō-chō, it was inadequate compared to later modern accounting. Namely, the fixed assets that should be a component of capital were not recorded; capital and liabilities were not explicitly distinguished; and the details regarding equity capital (“Kōnoike family disbursals”) and borrowed capital (“disbursals outside the family”) were not clear. Kō-Zen’s San’yō-chō has a limitation in that it cannot strictly show the total amount of assets since it does not include fixed assets such as land and houses.
While Kō-Zen was a wealthy merchant in the so-called “three cities” (Edo, Kyoto, and Osaka), bookkeeping spread to provincial cities and farm villages. Ogura (1967) is a pioneering study that compared Kō-Zen’s ledgers with the account-keeping methods of the Nakai family in Gōshū (today’s Shiga Prefecture). In the case of Kō-Zen, he established systematic account-keeping methods quite some years after the company was founded. This was largely due to improvements and development that had occurred internal to management. However the Nakai family would come to have systematic account-keeping methods within a short period of time after they were founded in the mid-18th century. More than being a distinctive characteristic of individual families, this means rather that washiki-chōai methods had become popularized in early modern Japanese society.
Nishimukai (1997) is a must-read article for analyzing the ledgers of merchant families in early modern Japan. It talks about the best ways for creating a catalogue when examining merchant family documents, and also provides a careful explanation about the structure of ledger organization in the early modern period. After dividing the organization based on its inner parts of the “living organization,” meaning the house, and the “management organization,” meaning the shop (tana), the first important step in elucidating the organizational structure of business ledgers is to find from among the many business ledgers the general business ledger that represents the whole of the house’s management or assets. In the case of the Hashimoto family of Onomichi (present-day Onomichi City, Hiroshima Prefecture) that Nishimukai analyzes, they had a honke sō kanjō chō, which corresponds to a general business ledger for assets; based on this, he ascertains the profit-and-loss accounts (profits and expenses) and the asset accounts for each fiscal year.
They do not combine the requirements of modern bookkeeping such as double-entry sorting of loans, left-right contra accounts, and principle of balance averaging; strictly speaking, this cannot be deemed double-entry bookkeeping.4 Furthermore those items like land and houses that today we call real estate is frequently not included in the general business ledgers of other merchant families and not just Kō-Zen’s San’yō-chō. Despite these limitations, some merchant families created ledgers that give a rough idea of their cash flow and stock of movable property for each year. We can see that a relatively sophisticated bookkeeping system had taken shape in 18th century Japan.
2. Meiji Restoration PeriodIn Japan, a new government was formed on January 3, 1868 (Keiō 3.12.9), and after a civil war (the Boshin War) the old shogunate was toppled. This Meiji Restoration government is known for implementing a series of reforms to modernize the country and for adopting technologies and systems from Western countries. One of these was the introduction of Western-style double-entry bookkeeping. For this, the achievements of two hired foreigners (oyatoi-gaikokujin) have been notably recognized.
The first person was a Portuguese man named Vicente E. Braga. Nishikawa (1974) takes up his achievements in detail. Braga had been an accountant at the former Hong Kong branch of the British Royal Mint. He arrived in Osaka in 1870 at the request of the Japanese government and took charge of doing bookkeeping and handling computations at the Ministry of Finance’s Imperial Mint (Zōhei-ryō). The Imperial Mint in Osaka began operations on November 27, 1870. The Imperial Mint produced the new tender that would be the first coins of the Meiji government and that would be the first currency based on units of yen (en). After working for nearly five years at the Imperial Mint, from March 1875 to July 1878 Braga was attached to the Ministry of Finance itself. After resigning from the Ministry, he became an instructor of bookkeeping at Mitsui. Braga’s method for organizing ledgers began with preparing vouchers in the pattern shown here.
Voucher→Waste Journal→Journal→General Ledger→Daily Balance→Balance Sheet/Profit and Loss Account
In short, unlike those of the early modern period, this format called for preparing a voucher for each transaction. The journal (nikki)—used to tabulate the cash receipts and disbursements for every day, based on those vouchers—forms a left-right debits-and-credits account, with the outgoing, i.e., debits (出), written on the left side, and the incoming (納), i.e., credits, written on the right. Japanese government agency bookkeeping has been based on double-entry bookkeeping since 1879. This is said to have been due to Braga’s guidance (Japan Mint 1976, 95), and the contributions he made upon the introduction of modern bookkeeping were immense.
The second person was a Scotsman named Alexander A. Shand. Shand was formerly a manager of the Yokohama subbranch of the Chartered Mercantile Bank of India, London and China, and was hired as a secretary attached to the Ministry of Finance in 1872. Shand was 28 years old when he went to work for the Ministry. He was dismissed from his position in 1877, and later went on to become a manager at Parr’s Bank Limited in London.
Hotori (2024)—a landmark work that provides an overview of the modern Japanese financial industry based on banking inspections carried out by the Ministry of Finance—gives credit to Shand’s success in spreading bookkeeping firsthand by accompanying Ministry of Finance inspections. National banks modeled on the national banks of the United States were established in 1873. At this point in time, the Bank of Japan had yet to be established as a central bank. The Ministry of Finance was supervising the soundness of management at the national banks, and also instructing those banks in modern knowledge such Western-style double-entry bookkeeping.
In addition to direct guidance to the national banks, Shand also published various manuals. Ebihara and Umehara (1873) is a manual written in Japanese based on a Shand lecture. It explains vouchers (i.e., the source records), three main types of ledger, 34 auxiliary types of ledger, and computation tables. As with Braga bookkeeping, the original record of the transaction was not that which was written in some sort of ledger like a bookkeeping journal but rather a voucher.
Even after Shand was gone, the Ministry of Finance continued to press forward with establishing double-entry bookkeeping through the publication of various practical guides.5
Thus, in the early 1870s—a point in time when what we today refer to as the washiki-chōai method style of bookkeeping had already been established in Japan since early modern times—Western-style methods of double-entry bookkeeping were coming in. The modern bookkeeping system first introduced in government offices gradually spread to the private sector thanks to the private engagement of hired foreigners (private sector employment) and the publishing of Ministry of Finance inspections, manuals, and instructional books.
In analyzing businesses, we should always bear in mind what a given ledger tells us about a business in connection with other ledgers. The work needs to be based on the historical background when it comes to the bookkeeping systems upon which these ledgers are based.
Next, let us consider the currency that people actually used at the time.
In Japan today, coins that the independent administrative agency Japan Mint produced and the government issued and banknotes the National Printing Bureau produced and the Bank of Japan issued are in circulation. Both use “yen” as their unit of calculation. Since coins are a small-denomination currency of 500 yen or less, it is for all intents and purposes possible to pursue monetary adjustment through the issuance and collection of Bank of Japan notes, i.e., banknotes of large denominations of 1,000 yen or more. In Japan since the modern period, the Bank of Japan as a central bank has mainly had a coordinating function for finance.
The Bank of Japan was established in 1882, and the first Bank of Japan note was issued in 1885. Accordingly, in Japan prior to this the monetary adjustment functions that a central bank would perform were not available. Instead, as the shogunate and the domains before it had done in the Edo period, the new Meiji Restoration government issued its own currency and implemented its own monetary policies.
Takizawa and Nishiwaki (1999) is a high-quality handbook that deals with currency minted from the early modern period through the first years of the modern period. Kobayashi (2016) surveyed existing studies, as well as those studies that can be viewed online. The Bank of Japan Currency Museum (2017) also serves as a digital guidebook.
Based on these materials, we know that during the early modern period gold coins such as the Keichō koban and Man’en nibukin, silver coins such as the chōgin and mameita-gin, and coins (senka) made of non-precious metals like the Tenpō tsūhō were in circulation. Focusing on units of calculation, these could be classified into three types: gold (kin)-denominated units (such as the ryō, bu, and shu); silver (gin)-denominated units (including monme, and fun, and non-precious-metal (zeni)-denominated units (mon). In short, the currency system comprised gold, silver, and non-precious-metal (strictly speaking, various materials including copper, iron, and zinc) coins as materials for the currency itself, and was based on denominations that used gold, silver, and non-precious-metals as units of calculation. With few exceptions, the shogunate basically did not issue paper money; it issued only minted coins. It would then issue new currency at irregular intervals with differing grades and weights. In short, they would recoin the currency.
The gold-based units of calculation were based on a quaternary numeral system, while those for silver and non-precious-metals were based on a decimal numeral system. Normally, a single ryōban would be referred to as a koban, a single koban would be equivalent to four ichibukin, and one ichibukin would be equivalent to four isshukin (1 ryō = 4 bu = 16 shu). In gold denomination, 1 kan was equivalent to 1,000 monme and in turn 10,000 bu, while in silver denomination 1 kan was equivalent to 1,000 mon. In principle, the units of calculation for gold coins are gold-denominated, and those for silver coins and non-precious-metal coins are, respectively, silver-denominated and non-precious-metal-denominated. However, with the issue of the Nanryō nishu-gin counter currency in 1772, many gold-denominated silver coins also came into circulation.
Aside from currency minted by the “central” government, a wide variety of other authorities also issued their own currencies. These included hansatsu (notes issued by domains), hatamoto-satsu (notes issued by direct vassals of the shogun), kōzan-satsu (notes issued by mine owners), jisha-satsu (notes issued by temples or court nobles), shukuba-satsu (notes issued by inn at one of the stations along one of the major highways), chōsonsatsu (notes issued by a town or village), and—used particularly in western Japan—ginme-tegata. Most of these notes was guaranteed by the issuing entity to be convertible to either gold, silver, or non-precious-metal coins. They were assigned units of calculation that were denominated in gold, silver, or non-precious-metals. While the situation regarding notes issued nationwide has yet to be fully ascertained, at least 234 daimyō issued their own domainal notes, and 1,231 types of private note other than domainal notes are in private hands or have been confirmed. Of these, in Harima Province (a region in the southern part of present-day Hyōgo Prefecture) where they have the largest number of notes issued anywhere in Japan, the issuance of 24 hansatsu, 10 hatamoto-satsu, 10 kōzansatsu, 115 jissha-satsu, 11 shukuba-satsu, and 139 chōsonsatsu can be confirmed. Naturally, shogunate-issued coins were also in circulation in Harima Province, which suggests the diversity of currency in the early modern period.
The economy of early modern Japan, with its complex and diverse currency system, is taken up by Iwahashi (2019), with a focus on small-denomination currency. Iwahashi defines small-denomination currency as, “Currency with face values of amounts that are indispensable to commonfolk going about their daily lives as consumers.” He defines small currency as gold coins and counter currency worth up to one bu (two shu or less), as well as silver-denominated notes and silver monme notes worth under five monme or less as small-denomination currency.
The system in the early Edo period was such that the role of small-denomination currency was played mostly by non-gold-coin currencies (silver coins, non-precious-metal coins, etc.). However, as the demand for small-denomination currencies in the marketplace grew alongside the increase in the volume of commodity transactions, it would seem that even small-denomination gold coins were issued in large volumes to respond to this. The proportion that small-denomination currency accounted for among all of the currency issued by the shogunate grew, and even silver coins were exchanged for counter currency suitable for small-denomination transactions.
In parallel with this, the issuance of hansatsu was also on the rise, and this played a role in conserving specie. Most hansatsu were small-denomination notes, and were mainly used for economic activities within the domain in a role supplementary to specie. In Japan during the early modern period—a time when the shogunate had exclusive control over the minting of specie but was unable to fully develop a nationwide monetary policy—in local areas it is possible to evaluate how hansatsu supported economic development by providing a flexible supply of money. The increased amounts in circulation of small-denomination shogunate-minted coins and of hansatsu alone indicates that currency had gone into daily use.
Further, aside from actual currency, rice was also a means of settlement serving as a medium for value. Domainal lords in the early modern period managed their finances centered on an annual tribute (nengu) system based on rice collections. The annual tribute—a tax of sorts—would be levied in rice, and vassals would frequently be paid their stipend in rice. For this reason, domainal lords would handle their finances by dividing them into income and expenditures based on currency-based units of calculation and income and expenditures based on rice-based units of calculation. There is a vast amount of research into the financial history of the shogunate—meaning here the Tokugawa family, who were the largest domainal lords; Ōguchi (2020) is one such example from more recent years. The shogunate kept track of expenditures and income in terms of ryō, the unit of calculation for currency, and koku, the unit of calculation for rice. Understandably, koku appears repeatedly in the ledgers.
In sum, the currency and settlement systems used in early modern Japan were complex. For example, when reading the original documents, one must remember that when the character 分 appears in a ledger, one must determine whether it refers to the gold-denominated bu or the silver-denominated fun. Furthermore, even notes of the time that had one ryō as their face value could not necessarily be exchanged for the equivalent of one ryō in specie. One had to distinguish between one ryō of koban of shogunate-minted coin and a one ryō note of gold-denominated hansatsu. Furthermore, even for the same minted currencies, the market price of gold plus silver and gold plus non-precious-metal fluctuated daily. We need to look at the amounts in the ledgers based on the foregoing actual realities.
2. Meiji Restoration PeriodDuring the Meiji Restoration years (1868–1871), in terms of currency policy the government decided to issue the Dajōkan-satsu (Council of State note) while abolishing silver-denominated currency. Sawada ([1934] 1966) is a classic study in this regard.
In 1868, the Meiji government—even with a civil war (the Bōshin War) under way—decided to issue these Dajōkan-satsu. The Dajōkan-satsu were gold-denominated notes, and for that reason were also referred to as kinsatsu (“gold notes”). Of the total amount of notes issued, ten ryō and five ryō notes were the highest-value notes, accounting for more than half of the value of everything issued.
The commonly accepted theory up to that time was that they had been issued to supplement inadequacies in military spending, e.g., for “financial reasons,” but Sawada notes that most Dajōkan-satsu were used not for military expenditures but rather for promoting industry. The drafters had a policy plan to issue Dajōkan-satsu throughout the country and provide loans to manufacturers and the like.
Furthermore, the “Administrative Proclamation No. 381” issued in 1868 has been referred to as the “Order to abolish silver-denominated currency.” This ordinance banned not only silver-denominated (ginme)silver coin, such as chōgin and mameita-gin, but also silver-denominated lending and borrowing, and ordered all prefectures to reconvert transactions to gold or non-precious-metals according to the market price on the day they were conducted. The point here is that the order to the abolish silver-denominated currency produced a distrust in silver-denominated drafts, and that this produced a run at the money-exchange shops that were the main actors for remitting drafts.
Thus, while monetary reforms may have taken place during the Restoration period, the monetary system continued to be gold and zeni-denominated. Yamamoto (1994) argues that, the Man’en nibukin that were issued in the final years of the shogunate in large amounts following the Man’en-era (ca. 1860–1861) monetary system reforms became the price standard, and that this Man’en nibukin system continued until around 1874. It is around 1874 that Yamamoto believes the yen takes root among the Japanese people in Japan. The modern currency system was established with the work done to adjust existing notes completed in 1885 by Matsukata Masayoshi, who was then the minister of finance. Yamamoto’s achievement is significant in that he explained the smooth connection from ryō to yen by situating as being within the same currency system from the end of the shogunate through the early years of the Meiji period.
Kobayashi (2015) continues some of the findings of Sawada ([1934] 1966) and Yamamoto (1994). However, he argues that the Dajōkan-satsu—the first notes in Japanese history that the central government guaranteed to circulate nationwide—had become the settlement currency intermediating bilateral transactions, and that they revised the impact of the abolition of silver-denominated currency downward.
Thus, by around 1874 or 1875, yen-denominated currency issued by the government was in wide circulation in Japan. For some time thereafter, one yen notes and one yen coins (gold or silver) did not have the same face value. However, when the Bank of Japan issued convertible notes in 1885 and the Meiji government guaranteed the convertibility of government notes in 1886, one-yen bills became equivalent to one-yen silver coins. Furthermore, while Japan currently does not have a standardized currency system, people recognize that 1,000 yen in coins has the same value as 1,000 yen in notes. Given such a currency system, it is sufficient to show the units of calculation as “yen” in the ledger.
That said, the mix of currencies that were used in each region differed in Japan during the late shogunate and early Meiji periods. It is precisely for this reason that the units of calculation such as ryō or monme are clearly indicated for each transaction in the ledger. Furthermore, even if the units of calculation are the same, the value may differ depending on whether it is minted unit of currency made of gold, a note issued by a domain, or paper currency issued by the Meiji government. For that reason, one may also come across here and there place where the type of currency is entered.
Accordingly, when analyzing ledgers, one does so—within Japan as much as anywhere—on the premises of the period, the region, as well as how the units of calculation and values of currency fluctuated.
Finally, when it comes to anaylzing the management of merchant families using ledgers, I would like to take up research on the Kō-Zen, Ka-Kyū, and Ka-Saku families of Osaka. All were major financial merchants working as money changers (money exchange shops).
First, Kō-Zen. For them, starting with Miyamoto (1959) would be best. This work presents the basic facts: Shinroku, who is regarded as the founder, was born in 1570. He started a sake-brewing business in the village of Kōnoike (present-day Kōnoike, Itami City, Hyōgo Prefecture). Shinroku’s eight son Masanari began doing business as a money changer and became the first-generation Zen’emon. Finally, they began loaning money to daimyō from around the mid-17th century and had relationships with their various families. Although the account of what happened after the Meiji Restoration is thin, the work does include a brief chronology up to 1954, which is useful.
Subsequent research on Kō-Zen increased in the detail of the evidence, culminating in Mori (1970) and Yasuoka ([1970] 1998). These works made it clear that the earnings rate on loans to daimyō had fallen off by the end of the Edo period owing to the unilateral suspension of interest and principal payments by daimyō (o-kotowari) and backward-looking management by Kō-Zen. This can be understood not just as a character of Kō-Zen, but also of management in regard to lending to daimyō in general.
Next, the Ka-Kyū is a merchant family about which research has suddenly been growing in recent years. This was due to the impact of two collections of historical records having been made available in recent years: the papers of Daidō Life Insurance and the papers of the family of Hirooka Kyūemon. According to Hirookake Kenkyūkai (2017), Ka-Kyū (the Hirooka family) traces its origins to Hirooka Tomimasa (1603–1680), who founded Kajimiya Kyūemon in Osaka. At the time of its founding, the family was in the rice trading business, but later they specialized in lending to daimyō. They founded Kajima Bank after the Meiji Restoration, and furthermore went on to become the founders of Daidō Life Insurance company. However, historical records related to management from their days as money changers through their days as a life insurance company were stored in scattered locations. A series of investigations into these historical records have been undertaken, and arrangements are currently being made to make them available to the public.6
Several research findings have come out of this. The most important of these is Takatsuki (2020). This work includes numerous new factual discoveries, such as how Ka-Kyū moved from rice trading to handling loans to daimyō, the realities of doing lending based on the long-term relationships that Ka-Kyū developed through daimyō loans, and how they managed to avoid crises at the time of the Meiji Restoration.
Unfortunately, research on merchant families cannot get away from being histories of the winners. This is because there are few business ledgers that remain from families that went bankrupt. However, the fact that some historical records from Ka-Kyū remain extant is valuable for analyzing the history of bankrupt companies. Yuki (2020) clarifies how debts were settled and how membership rights to shareholders were dealt with when Kajima Bank went bankrupt.
Finally, there is Ka-Saku. These, too, are historical records about a family who went through a business failure; they are in the collection of the National Institute of Japanese Literature in Tokyo.7 According to Shiryōkan (1968), Sakubei I began living in Osaka in 1626, and from around the time of Sakubei III the family took the trade name of “Kajimaya.” At first, their business was that of rice brokers, but later like Kō-Zen and Ka-Kyū they went into providing loans to daimyō. Ka-Saku was creditor to around 60 domains in the years from the middle of the early modern period onward.
Ka-Saku (the Osada family) went bankrupt around 1874, for reasons that Senda (1980) lays out. The high proportion of domain debts among their assets, and regardless of that the high proportion of abandoned debts when going through the process of adopting or rejecting feudal debts had a negative impact on the Osada following the abolition of domains and the establishment of prefectures. The failure to find a way to new businesses to replace lending to daimyō led to the decline of even wealthy merchants like the Osada family.
Why, then, did these three wealthy merchant families undertake joint ventures around the time of the Meiji Restoration? In the next section, I will look into the forms by which merchant families undertook businesses together.
2. Yamahiroya and Joint Ventures in the Early Meiji PeriodIn Japan, the National Bank—established under the National Bank Act of 1872—is said to have been the country’s first joint-stock company (Shinbo 1969; Takamura 1996). Thus, while the joint stock company—a modern corporate form in the strictest sense—did not exist in the earlier period, this is not to say that there had been cases of multiple business entities operating joint ventures.
Shinbo (1969) brings up kanyū-gashi (lit., “participatory lending,” but essentially joint financing) practiced by numerous merchant families from early modern times onward as a domestic condition for the rapid spread of joint-stock companies upon their introduction to Japan at the start of the Meiji period. He then went on to reevaluate the exchange companies and trade companies established under the leadership of the Meiji government in 1869.
In response to Shinbo’s discussion, Miyamoto (1969) examined Hōraisha (which was a bank of sorts) that emerged at the point when the shift from exchange companies to national banks was underway.
In 1873, Osada Sakubei, Yamanaka Zen’emon, and others jointly signed a written request asking that Gotō Shōjirō’s eldest son Taketarō serve as the representative (shusai, i.e., president) of Hōraisha. According to the agreement of the following year (Hōraisha jōyaku), Hōraisha was established as a financial institution with its head office in Tokyo that was engaged mainly in deposit and lending services. The Yamanaka and other merchant families paid a total of 3.4 million yen, and Uesugi Narinori and Hachisuka Mochiaki paid 1.05 million yen.
Miyamoto recognizes Hōraisha as being a “joint venture” and praises the progressiveness of its organizational form. However, he also recognizes its limitations in that the fact that this business went bankrupt in 1876 shows the pessimism of Osaka merchants in regard to making investments, as well as in the fact that it was a company based on the outmoded nature of its business.
Hiroyama (1993) discusses the Jinrikigumi—which was jointly established early in the Meiji Period by the Yamanaka, Hirooka, and Osada families—along with Yamahiroya-Chōbei, Hōraisha, and others. His assessment is that they were not effective. He sees the approximately 15-year-long period of stagnation from the final years of the shogunate into the years around the Meiji Restoration as a major limiting factor for the corporate activities of Kōnoike. With the revision of the National Bank Act in 1876, Yamanaka established the Thirteenth National Bank on his own; Hiroyama’s interest is in the management of this national bank. Using San’yō-chō, he lays out the asset trends for the period from 1876 to 1897.
Thus, although there is considerable research, when it comes to historical records for Kō-Zen from the early years of the Meiji period only extremely fragmentary materials remain (Yasuoka [1970] 1998, 156). When it comes to the current level of research, we know that there is still much we do not know about Yamahiroya.
Based on this sorting through of the research history, we can now finally attempt to interpret Figures 1 and 2.
The period around 1870 is when Western-style double-entry bookkeeping was first introduced to central government institutions by hired foreigners; it had yet to circulate among the wealthy merchants. Figure 1 shows bookkeeping based on the traditional washiki-chōai method. The current accounts ledger corresponds to a bookkeeping journal rather than an accounting ledger, and can be thought of as the primary ledger into which transactions are recorded as they occur.
Yamahiroya, the creator of this document, is the trade name of the joint venture that the Yamanaka family (Kō-Zen) of early modern wealthy Osaka merchants launched together with the Hirooka and Osada families. From Figure 2, we see that the Yamanaka invested 8,400 ryō and the Hirooka and the Osada invested 8,300 ryō each, launching the endeavor with a total of 25,000 ryō. The three families were almost equals. On the 22nd day of the 2nd month, this entire amount was lent to Yamanaka Zengorō,8 who put up 9,900 sacks of rice from Chikuzen (present-day Fukuoka Prefecture) as collateral (hiki’ate).
At this point in time, notes declared as “yen” had yet to be issued; the units of calculation used since early modern times were ryō. Kensatsu are thought to refer to exchange company notes9 issued by Osaka Exchange Company. The Osaka Exchange Company was in essence led by the Yamanaka, Hirooka, and Osada families; they likely withdrew loans in exchange company notes and then lent them on to other parties.
Furthermore, looking at other sections of this current accounts ledger, we see that Yamahiroya was trying to expand into the business of commodity-backed finance for the general public during the period around the Meiji Restoration. Without a doubt, daimyō power was in danger of coming apart after the civil war that began in 1868. Yamahiroya’s was not a continuation of the lending to daimyō of early modern times, but rather was oriented toward commodity-backed finance for the general public. Even viewed from a business organization perspective, it may be said that this was a joint venture that preceded Hōraisha.
However, there are few empirical studies of Yamahiroya, and the few studies that have touched on it have not given the Yamahiroya business much recognition. If we are to view Yamahiroya—where they shared the capital and personnel shared, and which was operated in the same location and under the same trade name—as a joint venture, then how should we assess this as an organizational form in a time before the joint-stock company? There is likely room for reconsideration of them as a part of Kō-Zen (i.e., Yamanaka family) business history.
Since the dawn of the 21st century, the digitalization of research has made striking advances even in Japan. Digitalization has proceeded in three directions: (1) books, (2) academic papers, and (3) historical records. In Japan, the best example of a digitalization effort on books is the National Diet Library Digital Collections, viewable at <https://dl.ndl.go.jp/>. With regard to open-access databases for academic papers, the movement to make works available over the internet in Japan has spread to include academic journals in the humanities and social sciences. Digitalized articles can be accessed through such platforms as J-STAGE <https://www.jstage.jst.go.jp/> and the institutional repositories of various universities. As to historical records, work is being done independently by various archival institutions such as the National Archives <https://www.digital.archives.go.jp/> to digitalize and make available such records.
In that connection, much of the research taken up in this article can also be viewed over the internet. This makes it available to researchers anywhere in the world, and I hope they will make the most of the opportunity.
This work was supported by KAKENHI grant no. 21K01601 from the Japan Society for the Promotion of Science.
1 Examples of such software include the “Komonjo Camera” smartphone application provided by Toppan Holdings for deciphering antique documents using artificial intelligence-assisted optical character recognition <https://camera.fuminoha.jp/>, and “Minna de honkoku,” a reprint platform based on public participation for historical texts and materials.
2 Yamahiroya-Chōbei, “Tōzachō,” 1870, Kōnoike Zen’emon-ke Monjo [Kōnoike Zen’emon Family Papers], 5–6, Graduate School / School of Economics, Osaka University.
3 The Japanese character that corresponds to “Osa” in the surname “Osada” can also be read as “Chō.”
4 However, there is also a view (Sakudo 1966) that partially accepts the principle of average balancing.
5 Ministry of Finance, Japan, ed., Ginkō boki reidai [Bank bookkeeping example], 1879 (Tokyo: Ministry of Finance, Japan); Ministry of Finance, Japan, ed., Ginkō zasshi [Bank bookkeeping magazine], 1877–1889 (Tokyo: Ministry of Finance, Japan).
6 At this writing, the plan is to make these available to the public in autumn 2024 as the “Kajimaya Hirooka Family Archive <http://ccss.kobe-u.ac.jp/research/database/bunsho/archive/index.html>.”
7 The remaining part of the collection, which had been in foreclosure in the warehouse of the Osaka branch of Mitsui Bank under seal, was distributed to the Mitsui family council (Mitsui dōzoku-kai), and the remainder of the collection was housed in the Mitsui Archives. The National Institute of Japanese Literature (preceded by the Ministry of Education Repository of Historical Documents) apparently received the same historical records from the former Mitsui Archives.
8 Yamanaka Zengorō was from an influential branch family of Kōnoikeya Zen’emon and was also an exchange merchant during the early modern period. Later, in 1871, together with Yamanaka Eizaburō, who was also from a Kōnoike branch family, he established Ōsaka Sōsan Kaisha. However, faced with the crisis of the family falling into ruin in 1872, Zengorō seems to have suddenly decided the following year to enter the priesthood and become a recluse (Nakagawa 1989, 53–54).
9 Starting in 1869, they issued five type of gold-denominated notes (kinken) and four types of silver-denominated notes (ginken).
