Economic Development and Labour Supply in Underdeveloped Regions: An Analysis of the Labour Supply of Domestic Servants in Northern Akita Prefecture, Japan, 1910–1924

Abstract

In industrialising Japan, labour supplies depended on regional differences in economic development. Large parts of Japan were underdeveloped regions characterised by delayed industrialisation and relatively low agricultural productivity. However, little is known about how economic conditions affected labour supplies in such regions. To answer this question, I investigated the employment of domestic servants in northern Akita Prefecture, a typical underdeveloped region, from the 1910s to the early 1920s. Until the early 1910s, wealthy families in this region recruited daughters from peasant families as domestic servants at a low fixed wage. This indicates that the supply of labour was unlimited, as defined by W. Arthur Lewis. From the late 1910s to the early 1920s, however, this region achieved remarkable agricultural growth. As a result, peasant families could obtain almost the same amount of income by having daughters work on farms as by sending them elsewhere to work as domestic servants. Employers of domestic servants therefore offered higher wages to recruit workers. This implies that in the underdeveloped regions, what had been an unlimited supply of labour was transformed into a limited supply due to agricultural growth.

This chapter is a translation of an article that originally appeared in Shakai Keizai Shigaku 75(4) (November 2009), pp. 27–49.

Appendices

Appendix 1 Estimate of an Annual Net Outflow of Labour

The following is an explanation of how to estimate an annual net outflow of labour from an area. Official statistics divide temporary residents into two types defined as follows (Umemura et al. 1983; Saito 1998). The first are temporary residents who had migrated from an area to another; the second are temporary residents who had migrated to an area from another. The former denotes those who had registered an area as their place of permanent residence but who resided outside the area at the end of each year. The latter indicates those who resided inside an area at the end of each year but who had registered other areas as their places of permanent residence. I subtracted the number of temporary residents who had migrated to an area at the end of each year from the number who had migrated from the area at the same time. The annual increase (or decrease) in this difference is defined as the annual net outflow of labour from the area.

According to official data on temporary residents, the annual net outflow of women in northern Akita Prefecture was, on average, 762 between the ends of 1909 and 1915 (Akita-ken 1912–1925, 1909 and 1915 versions). At the end of 1913, the cohort aged 15–19 accounted for 11.5% of all female temporary residents who had migrated from this region (Akita-ken Kōbunshokan, Akita Kenchō Monjo, no. 12271). By multiplying 762 by 11.5%, I estimated the annual net outflow of women aged 15–19 at 88. However, this is an underestimate. When people left their homes to stay in other places as temporary residents, they had to register the places of their residence. The official data on temporary residents were based on this registration, but a considerable portion of temporary residents neglected the registration.

To assess the degree of underestimation, I calculated the exact difference between the number of female temporary residents who had migrated from and the number who had migrated to northern Akita Prefecture at the end of 1920. The 1920 census lists the exact number of those who stayed in this region at the time the census information was taken, namely on 1 October 1920 (Naikaku Tōkeikyoku 1927, pp. 2–5). Moreover, Akita Prefecture’s statistical yearbooks specify the number of those who had registered this region as the location of their homes at the end of 1920 (Akita-ken 1912–1925, 1920 version). Of these two numbers, the latter was larger than the former because the outflow of labour exceeded the inflow of labour. Thus, by subtracting the former from the latter, I obtained the exact difference between the numbers of temporary residents who had migrated from and to northern Akita Prefecture. This difference was 1.93 times greater than the counterpart obtained using only the official data on temporary residents. By multiplying 88 by 1.93, I revised the annual net outflow of women aged 15–19 to 170. In the same manner, I estimated the annual net outflow of women aged 15–19, on average, at 728 between the ends of 1915 and 1920.

Appendix 2 Estimate of the Living-Cost Index in Northern Akita Prefecture

The following is an explanation of how to estimate the living-cost index in northern Akita Prefecture. Ohkawa et al. (1967) have already estimated the national index for living costs and I used the same method for my estimations. Based on the case of a typical household, I classified its living expenses by category and, when necessary, divided each category into items. Then, using the Laspeyres formula, I fixed a benchmark year and weighted each category or item. Finally, I multiplied the price index of each category or item by its weight and aggregated the resulting figures to calculate an index of the cost of living. To examine how price changes affected the household budget of a lower-class farming family, I used a 1934 survey of the household budget of a peasant family cultivating tenanted lands in Kita-Akita District (Akita Eirinkyoku 1935, pp. 12–17). I classified the living expenses of this family by category or item and weighted each category or item. Here, living expenses are defined as regular household expenses. Thus, I excluded irregular expenses, such as those for ceremonial occasions, from the household expenses of the family. I also converted the consumption of self-produced foods into its monetary value using price data available in the source. Then, I set 1910 as the benchmark year and calculated the living-cost index using price data from Akita City (Akita-ken 1912–1925, 1910–1924 versions).⁷

The categories, items, and weights (in percentage) of the model household’s living expenses are listed at the end of this appendix. The classification of categories is based on Ohkawa et al. (1967). The food expenses category, weighted at 79.02, is divided into items. For each item, or each category other than food expenses, I used the price of a specified good or service. In some cases, however, the survey includes a number of goods or services under the heading of a single category or item. In such cases, I chose as the representative good or service the one for which the expenses were largest and then used its price data. Price data are unavailable for a number of goods listed in the survey. For these goods, I used the prices of their substitutes. The final list specifies as “representative” or “substitute” those categories or items for which I adopted a representative good or service or a substitute good.

• A. Food expenses 79.02

  Items: rice (brown rice: substitute) 56.37, barely 3.32, soybean 2.52, soy source 1.18, salt 1.31, sugar (imported white sugar: substitute) 1.61, dishes other than staple food (dried bonito: representative and substitute) 7.15, beverages (saké: representative) 5.56

• B. Clothes expenses (cotton cloth: representative) 4.56

• C. Fuel and lighting expenses (kerosene: representative and substitute) 3.74⁸

• D. Housing expenses (carpenter’s daily wage: representative) 10.07

• E. Miscellaneous expenses (writing papers: representative and substitute) 2.63.

Primary Sources

• Kōbunshokan, Akita-ken (Akita Prefectural Archive), Akita City, Akita Kenchō Monjo (Akita Prefectural Office Documents), Taishō ninen jūnigatsusue genzai naikaku kunrei ni junkyoseru shirabe, jinkō tōkei zairyō, toshibetsu, Akita-ken tōkei shunin (Investigation on the population by city and district at the end of 1913, based on the order by the Cabinet, Chief of the Statistical Section of Akita Prefecture), no. 12271.

• Hosei University, Ohara Shakai Mondai Kenkyūsho (Ohara Institute for Social Research), Machida City, Kyōchōkai Shiryō (The Kyōchōkai Papers), Taishō jūnen gogatsu jūyokka yori dō jūrokunichi ni itaru chōsa, Kanegafuchi Bōseki Kabushiki Kaisha Tokyo honten kōjō rōdōjijō chōsa hōkoku (Report of the investigation on the working conditions of the Tokyo head office factory of the Kanegafuchi Cotton Spinning Co., Ltd. from 14 to 16 in May 1921), unfiled.

• Shiryōkan, Kokubungaku Kenkyū (hereafter KKS) (National Institute of Japanese Literature), Tachikawa City, Nakada-ke Monjo (The Nakada Family Papers), no. 42, Meiji sanjūkyūnen chiryō kinkoku shūnyū (Collection of rents in 1906).

• KKS, Tachikawa City, Nakada-ke Monjo, Zatsuyōchō/Zakki (Miscellaneous records)

  no. 52 (June 1923 to 30 September 1925)

  no. 54a (January 1911 to 1912)

  no. 54b (1913 to 1916)

  no. 54c (January 1917 to February 1920)

  no. 54d (March 1920 to May 1923)

  no. 55a (From 1909)

  no. 55b (1919)

• Ōdate Shiyakusho Hinai Sōgō Shisho (Municipal Office of Ōdate City, Hinai Branch), Ōdate City, Hinai chōshi hensan shiryō (Collection for compiling history of Hinai Town), Shotoku chōsabo, Higashidate-mura yakuba (Survey on household incomes, Higashidate Village Office), unfiled.