農林業問題研究
Online ISSN : 2185-9973
Print ISSN : 0388-8525
ISSN-L : 0388-8525
大会講演
Rice as a Financialized Food in the Japanese Empire: Colonial Imports and Exchanges’ Futures Trade
Kiyotaka Maeda
著者情報
ジャーナル フリー HTML

2026 年 62 巻 1 号 p. 52-59

詳細
Abstract

Since 2024, rice prices have increased, raising public concerns regarding the affordability of staple foods and making food-related financial transactions a key issue. Japan has maintained strict oversight of its rice market since 1939. Before the post-1939 period, Japan actively and financially engaged in rice trading in thriving markets. These historical experiences offer important insights into current rice price policies. The various intrinsic factors that have historically influenced these markets must be recognized. This study focuses on colonial rice imports as a notable example. Major Japanese cities hosted rice exchanges that set index prices for the spot market. While these exchanges only traded domestic rice, major cities relied heavily on colonial rice supplies, and fluctuations in colonial rice prices would significantly affect domestic prices. Rice prices in Taiwan and Korea had different effects on domestic markets because their harvest seasons differed. Although the Japanese rice market generally served its function, it was sensitive to factors that affected the quantity and quality of rice distributed, leading to instability. Therefore, further historical studies are required to identify the factors that caused the instability of the rice market during this period.

1.  Introduction

Rice prices in Japan have remained high since 2024, making rice affordability a social concern. Under these circumstances, some believe that expanding market trade could benefit rice consumers (Nikkei, August 13, 2025). However, scholars and market participants have a limited understanding of the rice market because the literature has paid little attention to its role in rice pricing before the Second World War.

Japan has a long history of rice market trading. The Tokugawa Shogunate allowed rice futures trading at the Dōjima Kome Kaisho in the early 18th century, and rice was a major item listed on Japanese exchanges by the 1930s. Japan entered the Second Sino–Japanese War in 1937 and implemented economic regulations, eventually forcing the abolition of rice futures trading in 1939. After that year until 1995, the Japanese government operated the Shokuryō Kanri Seido (Food Control System). At the beginning of the 21st century, the government liberalized regulations for rice circulation, and in 2024, the Dojima Commodity Exchange resumed rice futures trading after an 85-year hiatus.

This background indicates the need to rely on historical examples when attempting to improve the current rice market. Therefore, this study examines the rice exchange system during the interwar period and presents insights into the instability of rice futures trading in the Japanese Empire by reviewing recent research.

2.  Financial transactions in exchanges

(1)  Rice futures trade and rice policies

Until the mid-1920s, rice was roughly equal in transaction value to stocks on Japanese exchanges. According to the Tokyo Stock Exchange (TSE, 1924–25), the average proportion of the former to the total transaction value was 49%, while that of the latter was 40% from 1923 to 1924. However, government regulations on rice prices led to sharp drops in the share of rice in 1925 and 1933.

In response to the Rice Riots of 1918, the government enacted the Beikoku Hō (Rice Law) in 1921 to balance rice supply and demand. In 1925, the law was amended to strengthen the government’s authority to directly control rice prices. Subsequently, the government’s actions stabilized prices, resulting in a decline in trading activity on rice exchanges. In 1933, the government replaced that law with the Beikoku Tōsei Hō (Rice Control Law), which further boosted its efforts to control rice prices by increasing its budget. The value of rice traded on exchanges declined even further, and consequently, the proportion of rice in the total transaction value on exchanges dropped to 6% of the 1934–38 average (TSE, 1935–39).

(2)  Futures contracts in rice exchanges

The two largest cities in Japan, Tokyo and Osaka, had major exchanges: the Tokyo Rice and Merchandise Exchange (TRME) and Osaka-Dojima Rice Exchange (ODRE). From 1921 to 1938, the average trading volume shares across the two exchanges were 22% and 30%, respectively (TSE, 1919–29, 1930a, 1930b, 1931–39). These exchanges served as Japan’s main rice-trading hubs and offered three futures contracts: deferred (three months), second-nearest (two months), and nearby (one month).

Japanese rice exchanges only traded domestic rice. Specifically, the TRME and ODRE handled rice from Saitama and Hyogo Prefectures, respectively. They allowed dealers to settle their balances without delivering rice; however, if dealers held nearby contracts at maturity, they were required to deliver or accept actual rice.

The trading volume among these contracts showed gaps. Deferred contracts were the most significant trades, accounting for approximately 70% of rice futures trading in the TRME (TCC, 1901–02; TCG, 1904–09, 1911a, 1911b, 1913–30; TRME, 1915–16, 1929–39). Accordingly, previous studies have focused on the prices of deferred contracts. In this investigation, rice imports from Japanese colonies require attention.

3.  The empire’s rice supply system

(1)  Colonial rice imports

Increased living standards and urban growth led to higher rice consumption. Japan began importing rice from Southeast Asian countries, causing its self-sufficiency ratio to drop below 100% in the 1890s. During this period, Japan also acquired Taiwan and Korea as colonies in 1895 and 1910, respectively. Japan encouraged these colonies to boost rice production, leading to an increase in colonial rice imports beginning in the 1910s. By the early 1930s, most foreign rice imports had largely ceased (MAC, Food Control Bureau, 1944; MAF, Economy of Agriculture and Forestry Bureau, 1955). Japan created a rice supply system that depended partly on colonial rice, especially in major cities that relied heavily on it.

Panels (A) and (B) of Fig. 1 show the monthly proportion of rice supplies in Tokyo and Osaka, respectively, from January 1917 to June 1939. Panel (A) indicates that Tokyo boosted its Korean rice imports after the Great Kanto Earthquake in 19231. Furthermore, Tokyo began actively importing rice from Taiwan in the early 1930s. By the 1930s, domestic rice accounted for less than 30% of Tokyo’s total rice supply, and the city relied heavily on rice from Taiwan and Korea.

Panel (B) reveals that even in the late 1910s, domestic rice comprised less than half of the rice in Osaka. The city served as an arrival point for sea routes from Korea as well as a hub for Korean commodities (Maeda and Kahm, 2026). From the mid-1920s onward, Osaka relied heavily on Korean rice.

This situation resulted from agricultural policies implemented by colonial governments. The following subsection focuses on rice policies in Taiwan.

(2)  Hōrai rice production in Taiwan

The Taiwan Sōtokufu (Governor-General of Taiwan) required drastic policies to boost Taiwanese rice production. Originally, Taiwan harvested an indica variety suitable for tropical climates, whereas Japan, which has a temperate climate, cultivated a japonica variety. The two varieties differ in length and texture, leading Japanese consumers to reject Taiwanese standard rice. In the mid-1920s, the colonial government addressed this by developing Hōrai rice, a japonica variety adapted to Taiwan’s tropical climate. Hōrai rice production rose from the mid-1920s, significantly contributing to Taiwanese rice exports by the late 1930s (GGT, Industrial Bureau, 1934, 1941).

Based on the historical facts presented in this section, colonial rice imports should be incorporated into this analysis of rice futures pricing during the interwar period.

4.  Rice price linkage within the empire

(1)  Historical decomposition

Maeda and Tsai (2025) utilize a vector error correction (VEC) model, which is categorized as a time-series model. The VEC model enables the calculation of historical decomposition (HD), as suggested by Burbidge and Harrison (1985). The following equation is used to calculate the HD:

  
y t + j = i = 0 j 1 ψ i ε t + j 1 + [ X t + j β + i = j ψ i ε t + j 1 ]

where yt+j is a multivariate stochastic process, ε is its multivariate noise process, X is the deterministic part of yt+j, and i is the number of periods. The first term on the right-hand side represents the proportion of yt+j triggered by the shock. The following term on the same side indicates the prediction of the price series originating from the information available at time t, which corresponds to the event data.

An HD highlights the market’s temporal changes by examining the cumulative impact of subsequent shocks (Kilan and Lütkepohl, 2017). Maeda and Tsai (2025) apply an HD to daily rice price data in Tokyo and Taipei from 1935 to 1939, when Tokyo relied on rice from Taiwan.

(2)  Daily rice price data

Extant literature on the Japanese rice market before the Second World War that uses time-series models is available. Ito et al. (2017, 2018) analyze market efficiency following Fama (1970). Although they quantitatively and empirically examine time-varying changes in the rice futures market, they only identify annual shifts because they use the monthly weighted average prices of rice in Tokyo and Osaka. The monthly average prices represent one data point per month. Therefore, an analysis based on monthly data fails to capture monthly fluctuations in the market.

Maeda and Tsai (2025) address this by collecting daily rice price data from the TRME (1935–1939), archived at Ritsumeikan University and Osaka Metropolitan University, and Taiwan Rice Market Association (1935–1939), held at Academia Sinica in Taiwan. They are the first to utilize daily rice prices for both the metropole and colonies of the Japanese Empire.

(3)  Empirical results

Maeda and Tsai (2025) calculate the monthly average ratio of the HD value (RHDV) from January 1936 to December 1938, indicating the relative contribution of a cause variable to the fluctuations in an outcome variable.

Panels (A)–(C) of Fig. 2 illustrate the three main pricing elements for Tokyo’s futures and spot markets and Taipei’s spot market, respectively: price changes in Tokyo’s futures, Tokyo’s spot, and Taipei’s spot markets. As Panel (A) shows, the RHDV of Tokyo’s futures market mainly includes it and Taipei’s spot market. Thus, Taipei’s spot market continuously influenced Tokyo’s futures market.

Fig. 2 

RHDV of Tokyo’s Futures and Taipei’s Spot Markets, Jan. 1936–Dec. 1938

Panel (B) shows that Taipei’s spot market had both direct and indirect effects on Tokyo’s spot market. Although its influence was approximately half that of Tokyo’s futures market, Taipei’s spot market significantly affected Tokyo’s futures market, which was a key contributor to price fluctuations in Tokyo’s spot market. Consequently, pricing in Taipei’s spot market was crucial in driving these price changes in Tokyo.

Panel (C) indicates that Taipei’s spot market accounted for 49% of its RHDV, whereas Tokyo’s futures market contributed 40% on average. Taipei’s spot market was not merely a subordinate market to Japan. The RHDV of Taipei’s spot market varied seasonally, increasing from May to July and from October to November. The average values were 55% and 54%, respectively, during these periods and 46% in the other months. These changes stemmed from the seasonality of rice harvests in Taiwan.

In Taiwan, the first and second harvest seasons are from May to July and October to December, respectively (GGT, Industrial Bureau, 1935). As most of Japan reaped rice only once per year, from September to October, the first harvest season in Taiwan was approximately three months before the Japanese harvest season. Thus, Taiwan exported its products before the Japanese harvest season because rice prices in Japan had reached their seasonal peak. Taipei’s spot market generated Taiwanese rice prices, referencing the harvest conditions in Taiwan, and influenced price movements in Tokyo.

Recent research suggests that the seasonality of colonial rice harvests affected Japanese markets not only in major cities but also in small rural towns that did not rely on colonial rice.

5.  Colonial influences on the Japanese market

(1)  Methodology and data

Maeda (2025) discusses how Korean rice imports and trade activities in Osaka affected rice prices in Kumamoto. Kumamoto was a rural city that primarily consumed domestic rice. Surrounded by rice-producing areas, it did not need to import colonial rice. The city hosted the Kumamoto Rice Exchange (KRE), which was founded in 1894. It was a typical small rural exchange, ranking 10th out of 17 rice exchanges in 1930–39 based on the average annual volume of rice futures traded (TSE, 1930a, 1930b, 1931–39).

Maeda (2025) presents two analyses that employ the VEC model to calculate HD. The first investigation analyzes a series of monthly weighted average rice prices for deferred contracts across three rice exchanges (TRME, ODRE, and KRE) from 1900 to 1939. Maeda (2025) references statistics published by the central and local governments to generate price data series2. This investigation involves two steps: calculating the absolute HD ratios (AHDR) for each exchange, which indicate the relative contribution of a cause variable to an outcome variable, and conducting multiple regression analyses to examine the KRE’s pricing using its AHDR.

  
ln K R E = β 1 ln T V + β 2 ln I M P + β 3 ln O D R E

where KRE is the AHDR to and from the KRE, TV is the volume of futures trade in the KRE, IMP is the import volume of Korean rice in Osaka, ODRE is the AHDR from the ODRE to the KRE.

The second analysis examines three series of monthly weighted-average rice prices from 1919 to 1939 in Osaka: the futures prices of deferred contracts for the ODRE and spot prices of domestic and Korean rice3. It also includes two steps: calculating the AHDR for each price and performing a multiple regression analysis to assess how fluctuations in Korean rice prices influenced the ODRE’s prices using the ODRE’s AHDR.

  
ln A K P = β 1 ln T V O R + β 2 ln I M P + β 3 ln A O D R E

where AKP is the AHDR from Korean spot prices to futures prices in the ODRE, TVOR is the volume of deferred contract trade in the ODRE, IMP is the import volume of Korean rice in Osaka, AODRE is the AHDR to and from the ODRE.

Maeda (2025) uses the six-month moving averages of each variable in both equations to eliminate the effects of minute changes in the variables on the results.

(2)  Empirical results

Table 1 presents the results of both multiple regression analyses. The first regression analysis indicates that the price-formation function of the KRE had proportionate relationships with the volume of Korean rice imports in Osaka and the ODRE’s role in setting the index prices for the KRE. These results indicate that the KRE enhanced its price-formation function in response to an increase in the ODRE’s effects. In contrast, the price-formation function of the KRE had an inverse relationship with its futures trading volume. This inverse relationship resulted from seasonal fluctuations in futures trade volume. The futures trade volume increased before the harvest season, when rice prices peaked. Furthermore, because Japan and Korea harvested rice during the same season, Osaka actively traded imported Korean rice after the Japanese harvest. Therefore, during Japan’s harvest season, the futures trading volume of the rice exchange had an inverse relationship with the volume of Korean rice imports.

Table 1. 

Results of Multiple Regression Analyses on the Pricing of the KRE and ODRE

Explained Variable ln K R E ln A K P
β1 Coefficient −0.256*** −0.232***
t-value −10.11 −3.572
β2 Coefficient 0.245*** 0.233***
t-value 8.418 2.703
β3 Coefficient 1.000*** 1.404***
t-value 24.01 21.73
R ¯ 2 0.626 0.690
N 456 233

Note: *** indicates significance at the 1% level. These equations have no multicollinearity because the variance inflation factors of all explanatory variables are less than five. The results of the analyses of variance indicate that both equations are significant at the 1% level.

The second analysis indicates that the function of the Korean rice spot trade in Osaka in forming index prices for the ODRE had proportional relationships with the volume of Korean rice imports in Osaka and the ODRE’s price-formation function. Maeda (2025) further explores the relationship between the explained variable and β3 and finds that the futures trade of domestic rice and spot trade of Korean rice drove the futures prices in mutually opposite directions. Therefore, the futures market improved its price-formation function to adjust prices in response to the influence of Korean rice spot prices. In contrast, the function of the Korean rice spot trade in Osaka in forming index prices for the ODRE had an inverse relationship with the volume of deferred contracts traded on the ODRE. This relationship stemmed from seasonal fluctuations in futures trade volumes, as in the case of the KRE.

These results demonstrate that the KRE set its prices independently when the Korean rice trade expanded in Osaka. Unlike Taiwan, Japan and Korea harvested rice during the same season. Thus, after the Japanese and Korean harvests, the Korean rice trade in the Osaka spot market thrived. The Korean rice trade increased its influence on the ODRE, reducing the trading volume after harvest. However, the sluggish Korean rice trade in Kumamoto made the ODRE’s prices inappropriate as an index for the KRE during the harvest season. Consequently, during the same season, the KRE temporarily improved its price formation function to offset the impact of the ODRE. Despite differences in the geographical distribution of colonial rice, it still affected price formation, even in rural areas where it was not imported.

6.  Conclusions

Rice futures trading was a major financial transaction in Japan until the late 1930s. Japan’s major cities had rice futures markets that created index prices for the rice market. They required colonial rice to meet the growing demand driven by improvements in the standard of living and urbanization. The markets within the empire were closely linked, as were rice prices in remote locations.

As Maeda and Tsai (2025) note, the strong linkages between rice markets within the Japanese Empire triggered political conflicts during the interwar period. Japanese farmers criticized the decline in rice prices caused by colonial rice imports. Several members of the Rikken Seiyūkai (Constitutional Party), one of the largest political parties in Japan until 1940, owned agricultural land. Landowners and farmers needed to increase the prices of agricultural products. Accordingly, the Constitutional Party required the government to prevent a decline in rice prices from the 1920s onward. In response to these demands, the government strengthened its rice price regulations, including the Rice Law and Rice Control Law, as noted in Section 2. Eventually, it attempted to curb colonial rice imports. However, the colonial governments fiercely opposed the central government’s plan. This history suggests that the expansion of colonial rice imports triggered political conflicts among domestic farmers, the central government, and the colonial governments.

While examining historical events before the 1930s can offer valuable insights into current rice price policies, the various intrinsic factors that have affected the market in the past are crucial to consider. Although the rice market generally served its functions, they have changed over time. They were sensitive to factors affecting the quantity and quality of rice distributed in the market. Colonial rice imports provide a prime example. Further studies are required to identify the factors that caused instability in the historical rice market.

Funding

The author acknowledges the Japan Society for the Promotion of Science for its financial assistance provided through the Grant-in-Aid for Scientific Research (C) (no. 24K04991).

Notes
1  The Great Kanto Earthquake prevented the Tokyo Chamber of Commerce and Tokyo City Government from collecting and compiling data on rice imports from June 1923 to January 1924.

References
 
© 2026 地域農林経済学会
feedback
Top