DO LOCAL INSTITUTIONS AFFECT ALL FOREIGN INVESTORS IN THE SAME WAY? EVIDENCE FROM THE INTERWAR CHINESE TEXTILE INDUSTRY Peter Zeitz* Acknowledgements

Naomi Lamoreaux provided invaluable guidance throughout the project which is based on my doctoral dissertation completed at UCLA. I received particularly useful comments and suggestions from Daniel Ackerberg, Leah Boustan, Gregory Clark, Dora Costa, Xiaocai Feng, Richard Von Glahn, Edward Leamer, Aldo Musacchio, Jean-Laurent Rosenthal, Susan Wolcott, R. Bin Wong, and participants at the Shanghai Economic Forum (2008), the All-UC Graduate Student Workshop (2009), and the All-UC CEGE Conference (2011). This research was supported by a UCLA Economic History Travel Grant, National Science Foundation Doctoral Dissertation Improvement [grant number 0820584], and an Institute of International Education Fulbright fellowship (2008-2009).

*

Peter Zeitz, Assistant Professor, National University of Singapore Business School, Mochtar Riady Building #6-43, 15 Kent Ridge Drive, Singapore 119245. Email: [email protected]. Total Word Count: 12,345.

1

DO LOCAL INSTITUTIONS AFFECT ALL FOREIGN INVESTORS IN THE SAME WAY? EVIDENCE FROM THE INTERWAR CHINESE TEXTILE INDUSTRY

This paper analyzes the impact of employment institutions on Japanese-, British-, and Chinese-owned textile firms in China during the 1920s and 1930s. Despite Britain’s domestic position as a world productivity leader, Japanese firms enjoyed a 70 percent productivity advantage over both British and Chinese competitors. The divergent performance of Japanese and British investments in China is explained by differences in management practice. Japanese firms had domestic experience with employment institutions similar to China’s and applied labor management strategies that functioned well under these institutions. British firms lacked the institutional experience necessary to adapt management strategies to Chinese institutions.

INTRODUCTION

In the first half of the twentieth century, Japan rapidly absorbed industrial technologies from the West, while other Asian countries such as China and India lagged behind. In textiles, British managers and investors established plants in India, but productivity in these firms stagnated during the first half of the twentieth century, and many of these firms became unprofitable in the face of cheap imports from more efficient firms in Japan. A key question in understanding twentieth century development is why Japanese firms were able to transform productivity levels in their economy, while other 1

developing country firms, some of them run by experienced Western managers, tried and failed. To understand this puzzle, I investigate attempts of British and Japanese firms to transfer textile technology to China during the 1920s and 1930s. The analysis reveals some of the pitfalls which developed country firms face when they transfer technologies to countries with unfamiliar institutions, and explains why transfers may occur more readily across countries sharing similar backgrounds. The process of technology transfer is often seen in deceptively simple terms: firms in developing countries learning what firms in developed countries already know. In the popular knowledge-capital model of James Markusen, for example, multinationals deliver knowledge of ‘best-practices’ to foreign affiliates.1 Whether universal ‘best-practices’ exist, however, is questionable. Some knowledge is specific to characteristics—such as factor endowments, institutions, and culture—which vary across time and place. To illustrate this, I study transfers of managerial knowledge. Institutions and culture may be more important for management than for other types of technology. Barriers to the spread of managerial knowledge could help explain why low productivity persists in developing countries, even in foreign multinationals. If institutions and culture influence the effectiveness of management practices, one would expect firms in different locations to adopt different forms of organization. Nicholas Bloom et al. observe that institutional and cultural differences across countries influence the propensity of firms to centralize management.2 They also find that decentralized firms have higher productivity and that developing country firms tend to very centralized. 1

Markusen, Multinational Firms.

2

Bloom et al., “The Organization.”

2

Accordingly, Bloom et al. argue that developing country firms could improve their performance through decentralization. However, the development of effective managerial strategies may be more complex than this. Centralization does not break down cleanly by developmental level. For example, firms in Japan are well-known for innovative management practice, but are among the world’s most centralized. When firms set up affiliates in foreign countries, Bloom et al. find that they tend to be organized more like their parent firm than like local competitors. How foreign affiliates perform vis-à-vis local firms could depend on whether the practices they import meet local needs. In some cases, the transfer of inappropriate practices could put developed country multinationals at a disadvantage. I analyze transfers of organizational practices to China during the 1920s and 1930s. I focus on Japanese and British investments in Chinese textiles because their competition raises interesting questions for the theory of FDI. At this time, Japan was a relatively poor developing country and still relied on imported British textile technology. Productivity in Japan’s domestic textile sector lagged behind that in Britain. Nonetheless, Japanese corporations were able to dominate British competitors in China. Between 1918 and 1936, the Japanese share of the capital stock in the Chinese spinning industry grew from 21 to 42 percent, while the British share declined from 35 to 6 percent. Japanese success poses a puzzle: how can a difference in firm location allow a laggard multinational to leapfrog over a technological leader? I argue that institutional features of China favored the use of Japanese management techniques, disadvantaging firms which did not have access to this technology. A key feature of the Chinese environment was pervasive corruption. The Japanese management system was highly 3

centralized and monitoring intensive. By contrast, the British system was decentralized and relied on pecuniary incentives rather than intense monitoring. Decentralization worked well in Britain, but in China it facilitated corruption within lower levels of the managerial hierarchy. By contrast, centralized Japanese firms restricted decision-making authority of lower-level agents, curtailing their corrupt behaviors. The remainder of this paper analyzes productivity, institutions, and management practices in the Chinese textile industry. In the succeeding two sections, I use annual production statistics compiled by the Chinese Cotton Mill Owners Association and supplementary statistics describing product quality to compare total factor productivity levels in Japanese-, British-, and Chinese-owned textile mills. The results indicate that Japanese-owned firms were the most productive in the industry and British-owned firms were among the least. The next section examines the effects of China’s institutional context on the operation of British-, Chinese-, and Japanese-owned mills. I explain how local employment institutions caused problems at British- and Chinese-owned organizations and why Japanese-owned organizations were more successful at resolving these problems.3 The success of Japanese mill organizations inspired attempts at imitation among local competitors. In the penultimate section, I use bank investigations of organizational conditions and a case study of reforms at the largest Chinese-owned firm to show that adoption of Japanese-style practices was associated with productivity growth. Finally, I

3

My discussion of management and labor institutions in Chinese textiles draws from Sherman Cochran’s

work on foreign business in China and Tetsuya Kuwahara’s work on the management practices of Japaneseowned mills in China. See Cochran, “Encountering,” Kuwahara, “The Establishement,” and “The Local Competitiveness.”

4

conclude by reviewing the implications of context-specific knowledge assets for international flows of investment and ideas.

PRODUCTIVITY IN COTTON SPINNING

In this section, I outline the procedure and data I use to compare total factor productivity at Japanese-, Chinese-, and British-owned spinning mills. For a complete description of my data, procedures, and robustness checks, readers may consult the online data appendix which accompanies this article.4 Historical cotton spinning statistics typically report output in weight units, aggregated across yarns of different quality levels. Measurement of productivity in the cotton spinning requires adjustment for quality variation. Yarn fineness, or count, is the most important quality dimension. Pound for pound, high count yarn is much more costly to produce than low count yarn.5 Chinese sources record mill-level information on yarn output in weight units, but typically omit mill-level information on count. Information on count is available in aggregate statistics, which show the distribution of yarn counts spun in the Japanese-, Chinese-, and British-owned sectors. These data indicate that Japaneseowned mills spun significantly higher counts than other mills. Since value-added per pound

4

The data appendix is available at the author’s personal website.

5

Tim Leunig also adjusts for count in his comparison of spinning productivity in Britain and America, see “A

British Industrial Success.” He finds that count adjustment is necessary to obtain meaningful results.

5

increases with count, weight-based output measures understate the relative productivity of Japanese-owned mills.6 I adjust productivity for count differences using three types of statistical data on the Chinese spinning industry. The first type of data provides raw averages of labor and capital productivity in the Japanese-, Chinese-, and British-owned sectors. These sector-level averages reflect differences in both count and productivity. The second type of data provides the distribution of counts produced in the Japanese-, Chinese-, and British-owned sectors. The third type of data provides mill-level information on labor productivity, capital productivity, and count. I use these mill-level data to estimate elasticities of labor and capital productivity with respect to count. These elasticities allow me to correct capital and labor productivity figures for known differences in count. Finally, I use sector-level raw averages of capital and labor productivity, my elasticity estimates, and the sector-level count distributions to calculate quality-adjusted averages of capital and labor productivity for each sector. I confirm that my quality-adjusted averages are consistent with anecdotal reports of productivity at Japanese-, Chinese-, and British-owned mills spinning specific counts. Finally, I compute total factor productivity estimates for each sector as weighted averages of quality-adjusted capital and labor productivity.

6

Gregory Clark compares productivity at Japanese- and Chinese-owned textile mills in China and mills in

Japan circa 1929/1930, see Clark (1988), “Can Management.” Based on capital-labor ratios, he finds small productivity differences within China and large differences between China and Japan. Clark’s analysis has two major problems. Firstly, Clark greatly overestimates yarn counts produced in China and consequently his count adjustment procedure is seriously flawed. Secondly, Clark’s procedure ignores capital productivity differences. A more detailed description of problems in Clark’s work is in the online data appendix.

6

Data on raw productivity levels come from annual statistics collected by the Chinese Cotton Mill Owners Association (CCMOA) between 1924 and 1936. These data provide the most extensive source of information on the Chinese cotton textile industry. The data record the number of workers, spindles, and looms at each mill, and the number of bales of yarn and bolts of cloth produced, together with the weight of cotton consumed. The CCMOA data suffer from a number of minor flaws. The data on yarn production sometimes exclude yarn woven into cloth. Since cotton consumption statistics do not have this problem, I use cotton consumption data to measure yarn output. Another problem is that the CCMOA data record the total number of workers at each plant, but do not indicate how workers were divided between spinning and weaving in vertically integrated mills. To deal with this problem, I use a monthly plant-level survey conducted in 1947, the 1947 Textile Survey, which provides explicit data on the number of spinning workers and weaving workers at 38 integrated mills located in Shanghai, Jiangsu, Zhejiang, and Anhui.7 Based on the 1947 Textile Survey data, I estimate that, on average, operation of one loom required 40.5 times as much labor as operation of one spindle. Based on this proportion, I infer the percentage of workers in the spinning and weaving department at each mill from spindle to loom ratios reported in the CCMOA data. Count differences greatly affect productivity comparisons across sectors and require thorough analysis. Aggregated statistics from tax records indicate the distribution of yarn

7

Sixth Area Mechanized Textiles Industrial Committee, National Textile Factory Survey.

7

counts produced in each sector for 1929, 1932, 1933, 1934, and 1935.8 For other years, I need to make assumptions to deal with the absence of explicit count data. My choice of assumptions may have a significant effect on productivity estimates for years prior to 1929, but is not likely to affect productivity estimates for years from 1929 and onwards. For 1924, I assume that all three sectors produced the same distribution of counts as the Chineseowned sector in 1929.9 For 1936, I assume that the distributions of counts produced in each sector were identical to the distributions produced in 1935. For 1925-1928, 1930, and 1931, I fill in the count distributions using linear interpolation. The next step is to estimate elasticities that specify how capital and labor requirements varied as a function of count. I use two basic estimating equations. Equation 1 specifies log labor productivity as a function of log count, plant fixed effects, and time dummies. Equation 2 specifies log capital productivity as a function of the same variables. Rather than utilize a production function, I estimate the two separate equations separately because available data do not allow for simultaneous observation of capital and labor productivity. (1) (2)

8

Sector-level data are reported in Gang Zhao and Zongyi Chen, Chinese Cotton Textile History, and

Zhongping Yan, A Draft History. Yan verified these data against mill-level records that have since been lost, see Wang, The Last Hurrah. 9

Evidence supporting this assumption comes from The Situation of the Jiangsu Textile Industry, a mill level

survey conducted in 1919/1920 by the Jiangsu Employment Office.

8

In the two equations, i indexes firms, j indexes product types, and t indexes time;

is

labor productivity at mill i in product type j at time t, measured in either pounds of yarn per worker-day or pounds of yarn per unit of wage expenditure;

is capital productivity

measured in pounds per spindle-day at mill i in product type j at time t;

is the count of

product type j at mill i at time t;

are time

dummies; and are

and

and

are plant fixed effects;

and

are error terms. The coefficients of interest in Equations 1 and 2

, the elasticity of labor productivity with respect to count, and

, the elasticity of

capital productivity with respect to count. These elasticities indicate how much additional capital and labor is required to achieve an increase in count while holding the weight of output fixed. I estimate Equations 1 and 2 using three Chinese datasets. The first dataset, the Wang and Wang data, is a cross-section of production statistics from Chinese- and Britishowned mills operating between 1932 and 1933.10 The Wang and Wang data contain 90 observations of output per spindle-hour by product type from 37 plants. The second data source, the Shenxin data, is a panel of unit wage costs by product type for four Chineseowned plants controlled by the Shenxin Company and operating between 1934 and 1936.11 The Shenxin data contain 66 observations of unit wage costs from 4 plants. The third data source, the Qingdao Yearbook data, is a panel of monthly production statistics from eight

10

These data were collected by Wang Zijian and Wang Zhenzhong, see Wang and Wang, A Report.

11

These data come from the Shanghai Municipal Archive collection of Shenxin Company records.

9

government-owned mills operating between 1946 and 1948 and located in Qingdao.12 The Qingdao mills had been owned by the Japanese prior to 1945 and production conditions in these mills were likely similar to those in Japanese-owned mills during the 1930s.13 The Qingdao Yearbook data report monthly observations of unit wage costs by product type, yearly observations of output per spindle by product type, and monthly plant-level observations of yarn output per worker and average yarn count. Estimates of the elasticities of labor and capital productivity with respect to count are reported in Table 1. The Wang and Wang data and the Qingdao Yearbook data yield identical estimates of the capital productivity elasticity, -1.15. This value is similar to that implied by capital productivity-count conversion tables used by mills in Japan.14 The Shenxin wage data and the Qingdao Yearbook wage data yield almost identical estimates of the elasticity of labor productivity, -0.66 and -0.65, respectively. The Qingdao Yearbook output per worker data yields a slightly larger estimate, -0.77. I use elasticity estimates of 1.15 for capital productivity and -0.65 for labor productivity because they yield the most conservative estimates of Japanese-owned mills’ productivity advantage. [Insert Table 1]

12

These data are published in China Textile Construction Company, Three Years of Qingdao Spinning and in

Qingdao Textile Factory Statistical Office, 1947 Annual Statistical Report. The Shanghai Library holds copies of these publications. 13

Zhihuan Jin describes the Guomindang’s management of these mills after World War II, see Jin, Research.

The Guomindang retained Japanese engineers in management, utilized machinery acquired prior to the war, and operated under a scheme of labor organization similar to that in place prior to the war. 14

See tables in Keizo Seki, The Cotton Industry.

10

I combine the raw capital and labor productivity levels from the CCMOA data, the known count distributions, and the elasticity estimates to compute quality-adjusted productivity levels for each sector. I convert raw capital and labor productivity levels into capital and labor productivity levels in terms of 20-count equivalents. These are predictions of what capital and labor productivity levels would be if the sector produced only 20 count yarn. Computation of quality-adjusted productivity levels is shown in Equations 3 and 4, where

indexes the sector under comparison,

are labor and capital productivity in sector h in terms of 20 count equivalents,

and and

are

unadjusted levels of labor and capital productivity in sector h, εl,c and εk,c are the elasticities of labor and capital productivity with respect to count, and

is the fraction of

yarn count c produced in sector h, where c is one of seven possible yarn counts: 9.5, 12, 15.6, 20, 32, 42, or 60.15 [∑ [∑

( )

∑

]

( )

∑

]

The raw levels of labor and capital productivity,

and

(4)

, are not comparable across

sectors because each sector produces a different set of products, { product set, {

(3)

}. Changes in this

}, will alter raw labor and capital productivity levels, even though the real

productivity level in the sector remains fixed. I use observations of { the elasticities εl,c and εk,c to convert

and

} and estimates of

into levels of capital and labor productivity

15

For each sector, the count data indicate aggregate output by weight within seven count ranges. I impute an

approximate average count for each of the seven count ranges.

11

in terms of a single product, 20 count yarn. The resulting values,

and

, are invariant

to the product set produced, allowing comparisons across sectors spinning different counts. Evaluation of Equations 3 and 4 generates capital and labor productivity series for the Chinese-, Japanese-, and British-owned sectors in terms of 20 count equivalents. Since spinning productivity figures are typically reported in terms of pounds per hour, I convert them into an hourly framework using a 320 day work-year, 24 hour operation in Chineseand British-owned mills, 22 hour operation in Japanese-owned mills, and the conversion factor of 420 pounds per bale. As Table 2 shows, my estimates of the productivity gap between Japanese- and Chinese-owned mills are corroborated by small samples of data reported in Zhao and Chen and Peter Duus. Table 2 also reports productivity statistics for Qingdao mills in 1947 based on data in the Qingdao Yearbooks. The Qingdao Yearbooks provide full information on the counts produced the division of workers across spinning and weaving, operating hours, and the number of days worked. Assuming that productivity did not change significantly during World War II, the Qingdao Yearbooks figures for 1947 should give estimates of capital and labor productivity in Japanese-owned mills similar to those in mid-1930s, and this is what the data show.16 [Insert Table 2] Using these adjusted labor and capital productivity figures, it is simple to produce a TFP index comparing the performance of the Japanese-, Chinese-, and British-owned sectors in spinning. Arno Pearse reports estimates of Chinese spindle costs, labor costs, 16

Studies of the post-war Chinese cotton industry suggest that productivity at formerly Japanese-owned mills

during the late 1940s remained similar to productivity at Japanese-owned mills during the 1930s. See Jin, Research, and Wang, The Last Hurrah.

12

interest rates, and depreciation allowances that suggest equal spindle and labor cost shares in 20 count spinning.17 Accordingly, I calculate total factor productivity as the exponent of the average of quality-adjusted log spindle and log labor productivity. To put Chinese spinning productivity levels in international perspective, I report spinning productivity levels for each sector as a percentage of productivity levels in Japan circa 1930-1934.18 The results are shown in Figure 1. During the 1930s, average productivity levels in the Japanese-, Chinese-, and British-owned sectors were 0.99, 0.59, and 0.46, respectively. Clearly, the Japanese-owned sector enjoyed a substantial productivity advantage within China. In spinning, Japanese-owned plants in China were just as productive as those in Japan. [Insert Figure 1 here] To understand China’s international competitiveness, it is useful to estimate costs of production and compare them with those of Japan, a leading textile exporter. I formulate a production cost index for 1930-1936 by adjusting the TFP index for wage differences.19

17

Pearse, The Cotton Industry.

18

To calculate total factor productivity levels in Japan, I use the Japanese spinning productivity series reported

in Susan Wolcott and Gregory Clark, “Why Nations Fail.” Mills in Japan, Japanese-owned mills in China, and British- and Chinese-owned mills in China operated for 16, 22, and 24 hours a day, respectively, see Pearse, The Cotton Industry. Shorter operating hours imposed higher capital costs on Japanese mills. Thus, in my TFP calculation, I adjust the Japanese and Japanese-owned capital productivity series downwards by factors of 16/24 and 22/24, respectively. 19

I use plant-level average wages reported in the Wang and Wang data to measure Chinese-owned wages in

1932/1933, Wang and Wang, A Report. Panel data report average wages at the Shenxin Company, the largest Chinese-owned firm, for 1930-1936, see Shanghai Academy of Social Sciences, Historical Materials. I use

13

The results indicate that Japanese-owned firms in China were exceptionally competitive internationally. Normalizing unit costs in Japan at 1, I estimate production costs for the Japanese-, Chinese-, and British-owned sectors as 0.79, 1.21, and 1.58, respectively. Evidence corroborating high productivity and low production costs among Japanese-owned mills in China comes from a 1939 Manchurian Railway Company survey of Japanese mills in Qingdao.20 The Railway Company survey reports that, circa 1940, average output per work hour at Japanese-owned spinning mills in Qingdao was 90 percent of the average output per work hour for spinning mills in Japan. Simultaneously, wages at Qingdao mills were only 49 percent of the Japanese average. Thus, labor costs in Qingdao spinning were only 55 percent of those in Japan. These cost levels are consistent with the Japanese-owned sector’s export performance. While Chinese-owned firms focused on supplying low count yarns to domestic consumers, Japanese-owned firms in China exported increasing quantities of yarn to Japan, India, and Southeast Asia. In fact, starting in 1928 and continuing until at least 1932, China exported twice as much yarn as Japan annually. Japanese-owned firms were responsible for the vast majority of these exports.21 PRODUCTIVITY IN WEAVING the Shenxin series to estimate wage changes over time within the Chinese-owned sector. To impute wages for the Japanese- and British-owned sectors, I use estimates of relative wages across sectors reported in Duus, “Japanese Cotton Mills.” For firms in Japan, I use a spinning wage series reported in Mark Ramseyer, “Credibly Committing”. I convert wage data into US dollars using annual exchange rates reported online at Laurence Officer, measuringworth.com. 20

Youshiyuki Yoshida and Hiroshi Wada, The Investigation.

21

See Pearse, The Cotton Industry and Moser, The Cotton Textile Industry.

14

In this section, I outline the procedure and data I use to compare total factor productivity at Japanese-, Chinese-, and British-owned weaving mills. The primary dimension of differentiation in weaving is the fineness of the cloth woven. Fine cloth is woven from high count yarn, whereas coarse cloth is woven from low count yarn. However, unlike in spinning, raw sector-level measures of weaving productivity are not greatly affected by differences in cloth quality. Firstly, there were no marked differences in sectorlevel weaving specialization. In all three sectors, mills wove primarily coarse cloth. Secondly, differences in capital and labor requirements between coarse and fine weaving are much smaller than those in spinning. 22 Accordingly, I use unadjusted units reported in the CCMOA data to measure weaving output. The availability of a meaningful mill-level measure of output allows weaving productivity to be assessed in a simple Cobb-Douglas framework. I estimate capital and labor shares using a Cobb-Douglas production function and compute annual sector-level total factor productivity levels from the CCMOA data based on these shares. To estimate the production function, I combine the 1947 Textile Survey data with the Qingdao Yearbook data. The combined dataset is a monthly panel of 46 plants covering the period from January 1947 to June 1947. Twenty of these plants were Japanese-owned during the 1930s and transferred to national ownership in 1945. The remaining 26 plants were under private Chinese ownership. The data contain plant-level observations of spinning

22

In the data appendix, I show data on weaving specialization patterns in each ownership sector and estimates

of the difference in capital and labor requirements in coarse and fine weaving. The results imply that any adjustments for differences in weaving specialization would have negligible effects on sector-level productivity estimates.

15

employment; weaving employment; bolts of cloth produced; weight of cotton consumed; operating shifts per month; and numbers of spindles, automatic looms, and plain looms. Unlike in spinning, weaving machinery varied across sectors and this could have affected productivity. During the 1930s, Japanese-owned firms installed Toyota automatic looms, while Chinese- and British-owned firms continued to use cheaper plain looms.23 Among firms in the combined 1947 sample, the share of automatic looms was 88 percent on average among formerly Japanese-owned firms, but only 28 percent among private Chinese-owned firms. To assess whether this difference affected productivity, I include the share of automatic looms in each plant’s loom stock as a control variable in production functions. The regression specification is shown in Equation 5, where yit is log bolts of cloth output per weaving shift in mill i at time t; lit is log weaving employment; kit is log looms; sit is the automatic loom share; Zit is a vector of additional controls for mill-level productivity differences; γt is a set of month dummies; and εit is an error term. (5) Results from specifications of Equation 5, reported in Table 3, suggest that weaving production exhibited constant returns to scale with approximately equal factor shares. Specifications 1 and 2 are OLS estimations of Equation 5, which are identical except that Specification 1 omits the automatic loom share. The sample size is smaller in Specification 2 because information on loom type is missing for around one-third of the plants. In Specification 2, the estimated coefficient on the automatic looms share is positive and significant, suggesting that the use of automatics increased productivity. However, since 23

See Pearse, The Cotton Industry and Wang, The Last Hurrah.

16

formerly Japanese-owned mills used automatics more intensively, this result may be due to spurious correlation. Specification 3 includes a dummy variable to control for productivity differences associated with past Japanese ownership. The ownership dummy is positive and significant, and indicates that circa 1947 average TFP in formerly Japanese-owned mills was 31 log percentage points higher than that in Chinese-owned firms. The coefficient on the automatic looms share is very near zero, suggesting that the use of automatic looms did not significantly affect productivity. In Specification 4, I use within-firm variation to try to identify the productivity effects associated with automatic looms. I control for firm-wide productivity levels using firm-specific time dummies and identify regression coefficients using within-firm variation in multi-plant firms. The assumption here is that most productivity variation occurs across firms, and therefore that control for ownership reduces bias in coefficient estimates. In Specification 4, the coefficient on the automatic looms share is positive, but very small in magnitude and not statistically significant. Based on the results of Specifications 3 and 4, I conclude that any productivity advantage enjoyed by Japanese-owned firms was not due to their more widespread use of automatic looms. A final concern, addressed in Specifications 5 and 6, is the potential for unobserved productivity differences to bias estimates of the capital and labor coefficients. Since productivity change in textiles during this period was labor-saving, productivity heterogeneity could bias estimates of the labor coefficient downwards and the capital coefficient upwards. In Specification 5, I include log labor productivity in each plant’s spinning section as a proxy for productivity in its weaving section. Comparing Specification 5 to Specification 1, inclusion of the control increases the labor coefficient and decreases the capital coefficient. In Specification 6, I apply the control function 17

approach due to James Levinsohn and Amil Pertin which proxies for unobserved productivity differences using data on material consumption.24 Since direct measures of yarn inputs entering the weaving process are unavailable, I use log cotton consumption and log spinning labor as proxy variables. The results are extremely similar to those in Specification 5, and imply constant returns to scale, a capital elasticity of around 0.45 and a labor elasticity of around 0.55. [Insert Table 3 here] I use the CCMOA statistics to estimate sector-level total factor productivity, computed as residuals of a sector-level constant-returns-to-scale Cobb-Douglas production function, with a labor share of 0.55 and a loom share of 0.45. The results shown in Figure 2, which plots aggregate TFP estimates for Japanese-, British-, and Chinese-owned plants, indicate that Japanese-owned firms enjoyed a large productivity average in weaving. From 1930-1936, Japanese-owned plants’ TFP in weaving exceeded that of Chinese-owned plants by 79 percent on average, and that of British-owned plants by 72 percent. Productivity among British-owned plants collapses from 1934 to 1936, reflecting reductions in operating hours at British-owned firms. In the mid-1930s, some Britishowned mills cut back operations to one shift per day to reduce operating losses.25 In computing productivity measures, I exclude mills that shutdown completely, but cannot observe the timing and extent of reductions in operating hours. Accordingly, hours reductions appear as drastic productivity declines. 24

See Levinsohn and Petrin, “Estimating.” I apply the Stata algorithm published in Levinsohn et al.,

“Production Function Estimation.” 25

See Leonard Wu, “The Crisis.”

18

[Insert Figure 2 here]

EXPLAINING THE CROSS-SECTOR PRODUCTIVITY GAP

Why were Japanese-owned mills so much more productive than their competitors in spinning and weaving? Commonly cited reasons for productivity differences, such as political institutions, differences in laborers’ educational attainment and experience, and capital quality, cannot explain the performance gap because the mills operated in the same cities, using similar labor sources and machines. Workers in Japanese- and Chinese-owned mills had similar tenure, so that differences in worker experience do not explain the productivity gap either.26 Finally, the legal and political environment could not explain performance differences between British and Japanese firms; most of these firms were located in Shanghai’s International Settlement, where they benefited from Western legal protections.27 Moreover, if the political environment had any effect, Chinese boycotts of Japanese products during the 1920s and 1930s and Britain’s dominant role in Shanghai’s governance should have disadvantaged Japanese-owned firms. A difference which might explain poor performance in the Chinese- and Britishowned sectors was their use of a form of subcontracting known as the foreperson system. Under this system, managers ceded control over labor recruitment, training, and

26

See data on work tenure in Chinese- and Japanese-owned mills presented in Fong, Cotton Industry and

Pearse, The Cotton Industry. 27

For a description of Shanghai’s institutions, see Debin Ma, “Shanghai-based Industrialization.”

19

organization to forepersons.28 The structure of this system was quite similar to that used throughout British manufacturing. Among textile firms in Britain, forepersons were given the authority to hire their own assistants, set their wages, and make production decisions.29 Even though British firms used a similar management system both in China and at home, consequences of the system were different in the two locales. In Britain, subcontracting remained the predominant mode of manufacturing organization until well after World War II.30 William Lazonick has argued that subcontracting in textiles encouraged forepersons to cooperate with management to raise productivity.31 Under subcontracting, British textile firms were able to produce at or near the global productivity frontier.32 On the other hand, a similar system of labor arrangements in China led to frequent conflict between forepersons and management and prevented managers from making efficient staffing decisions.33 The inconsistency of the results obtained under the British management system reflected differences in the British and Chinese institutional environments. In early twentieth century textiles, productivity improvement was linked with the intensity of labor. More productive workers tended more machines simultaneously, meaning that efficient 28

For a description of the foreperson system based on interviews with retired Shanghai workers, see Emily

Honig, Sisters. 29

For a description of textile labor organization in Britain, see John Jewkes and E. M. Gray, Wages.

30

See Steven Tolliday and Johnathan Zeitlin, “Employers.”

31

Lazonick, Production Relations.

32

See Leunig’s comparison of spinning productivity in the Britain and the US, Leunig, A British Industrial

Success. 33

Cochran, Encountering Chinese Networks.

20

firms had lower staffing levels. In Britain, industry-wide collective bargaining set stable piece-rates for forepersons, guaranteeing that they could earn rents if they reduced staffing levels while maintaining output. In China, labor recruitment institutions established perverse incentives that encouraged forepersons to hire unnecessary workers. In the Chinese labor market, bribery or ‘squeeze’ had a central role in job search. Workers obtained positions by offering forepersons sums equivalent to two weeks’ pay, and maintained employment through regular gifts. Since adding new workers increased forepersons’ income, they preferred to hire more labor than tasks demanded.34 Overstaffing led to considerable idleness within mills; supernumerary workers found time to loiter, sleep, and even absented themselves from the mill entirely.35 Clearly, to the extent that forepersons could bring idle workers into the mills, they obstructed productivity change. The ability of forepersons to force firms to hire more workers than necessary reflects the influence of the Green Gang, a powerful criminal organization. Most mill labor brokers, forepersons, and guards belonged to the Gang, and membership helped them to negotiate with mill managers over hiring practices.36 In textiles, the Gang sold labor, property protection, and strike prevention services, but was simultaneously active in

34

A government-commissioned study of failing mills in Tianjin noted that foreperson-mediated employment

led to the use of three times as many workers as were actually necessary; see Gail Hershatter, The Workers. 35

Under the ‘let-go-policy’, forepersons afforded workers leave in exchange for bribes. Lee, B.Y. “Real

Causes Behind Japanese Mill Strikes.” The China Weekly Review (February 1925) 36

Estimates of Green Gang membership vary widely. For a discussion of these estimates, see Smith, Like

Cattle. A reasonable estimate might be twenty percent of the labor force.

21

vandalism and strike instigation.37 Acting both as an informal union for supervisory workers and as a labor broker for unskilled workers, the Gang protested changes which disadvantaged its constituents. These disputes were often related to staffing levels and personnel changes.38 Progress in the textile industry depended upon reforming hiring practices to curtail bribery. However, foreperson opposition made these reforms difficult. When mills dismissed forepersons, the forepersons drew on Gang connections to mobilize worker resistance. Since workers invested in their relationships with forepersons through bribery and risked blacklisting if they abandoned the relationship, managers could not easily dismiss forepersons while retaining the workers under them.39 Similarly, management attempts to curtail forepersons’ right to select their own assistants brought on violent

37

Descriptions of the Green Gang’s involvement in racketeering in the textile industry are contained in Honig,

The Contract Labor System, Honig, Sisters, and Frazier, Mobilizing. 38

Barker and Barker, The Textile Industries, noted that “in China the principal trouble [in management-labor

relations] is that a mill is practically forced to employ more people than is necessary...” 39

The Gang used blacklists to discipline workers who obtained jobs improperly; see Honig, The Contract

Labor System and Honig, Sisters. Contested foreperson dismissals in Chinese-owned firms regularly caused large-scale walkouts of shop floor workers; see Lee, B.Y. “Real Causes Behind Japanese Mill Strikes.” The China Weekly Review (February 1925).

22

strikes.40 Given the high costs of opposition, managers usually ceded to forepersons’ demands.41 Japanese-owned mills operated under a management structure which absorbed many of forepersons’ functions into specialized bureaucratic agencies. Personnel offices controlled hiring and the assignment of employees to tasks. Instructors trained workers in classrooms and workshops on specially designated equipment. Production decisions came under the control of trained technicians. Forepersons were essential to fewer operations, used in smaller numbers, and were paid less.42 Importantly, loss of authority over hiring, firing, and compensation decisions prevented forepersons from extracting bribes from the workforce.43 Japanese firms’ development of centralized management structures reflected a response to labor management problems in Japan. Firms in Japan engaged foremen (oyakata), who controlled labor gangs, and mediated the supply, training, and management

40

An example is the 1919 strike at the Rihua mill described in Jiangsu Employment Office Third Department,

The Situation. 41

According to Lee, “…[forepersons] generally do not do much work nor know much and usually do not stay

in the mills all the time. Most Chinese managers are afraid of them and would not dare to do or say anything directly against their will...” Lee, B.Y. “Real Causes Behind Japanese Mill Strikes.” The China Weekly Review (February 1925) and cited in Cochran, Encountering. 42

According to 1920 survey of industry conditions, Japanese-owned mills used one-tenth as many supervisory

personnel per production worker as Chinese-owned mills. Jiangsu Employment Office Third Department, The Situation. 43

Lee reports that ‘squeeze’ was limited in Japanese-owned mills, and that this angered forepersons. Lee, B.Y.

“Real Causes Behind Japanese Mill Strikes.” The China Weekly Review (February 1925).

23

of labor. Japanese foremen mobilized patronage relationships with workers and informal networks to influence managerial decisions.44 Japanese corporations responded by replacing subcontractors with corporate hierarchies that emphasized centralization, promotion-based incentives, and formal training.45 In Japan, problems with foremen mainly occurred in mining and heavy industry, but organizational responses had a broader influence on all forms of large-scale manufacturing. Recruitment problems were another important impetus to centralization in the Japanese textile industry. In the early twentieth century, Japanese mills relied on networks of rural recruiters to supply their labor needs. Recruiters often exploited deception to arrange placements, and as a result many workers quit shortly after their arrival. Since recruitment fees were a significant component of total labor costs, turnover rates directly affected profitability. To reduce turnover, Japanese textile firms established internal recruitment offices that enhanced company control over employment contracts and worker selection. In China, Japanese-owned firms encountered profound foreperson resistance in establishing a managerial system that marginalized forepersons’ authority. Labor disputes occurred much more frequently at Japanese-owned mills than British- or Chinese-owned mills. Mark Frazier argues that frequent strikes reflected a response to Japanese attempts to

44

See Solomon Levine, “Labor Markets.” Kazuo Nimura provides a superb case study of conflict between

patronage networks and management in Japanese copper mining; see Nimura, The Aisho Riot. 45

For discussions of Japanese labor management during the first half of the twentieth century, see Sanford

Jacoby, “The Origins” and Chiaki Moriguchi, “Implicit Contracts.”

24

control hiring and shop floor practices.46 In these labor disputes, Japanese managers were not uniformly successful. From 1925 to 1927, a series of particularly violent and disruptive strikes occurred at Japanese-owned mills. The Green Gang appears to have played a central role in fomenting these disputes. Strike negotiations led to some management changes that benefited Green Gang interests. Compensation for forepersons at Japanese-owned mills was increased dramatically, while that for ordinary shop floor workers barely changed. At the same time, Japanese-owned mills made accommodations with Gang interests by agreeing to source most of their workers from gang-controlled recruitment networks.47 However, within mills, Japanese-owned firms continued to maintain much stricter control over staffing levels, job allocation, and work practices than Chinese- and British-owned competitors.48 This outcome can be interpreted as an efficient solution to a dispute over the allocation of the rents associated with the Japanese management system. By offering concessions to the Green Gang in the form of higher foreperson wages and greater reliance on the Gang’s recruitment networks, Japanese-owned mills were able to retain a system of strict shop floor discipline that boosted productivity. Contrasts in the way Japanese and British investors in China approached labor management were not unique to textiles. Evidence from the coal industry suggests that 46

See Frazier, “Mobilizing a Movement.” An illustrative example is the 1919 strike at the Japanese-owned

Rihua mill described in Jiangsu Employment Office Third Department, The Situation and Frazier, “Mobilizing a Movement.” Rihua purchased this mill from a British firm and attempted to reform its managerial system, firing child workers under incumbent forepersons and replacing them with workers selected by management. This precipitated a strike in which forepersons fought to retain control over hiring. 47

See Frazier, “Mobilizing a Movement,” and Frazier, The Making.

48

Cochran, Encountering

25

problems with subcontracting were common in China generally, and that Japanese and British managers’ responses to them followed a consistent pattern. The two largest coal mining firms in early twentieth century China, Kailuan (British-owned) and Fushun (Japanese-owned), faced tremendous difficulty in managing tens of thousands of laborers, and relied on decentralized contract systems to meet their labor needs. Agency problems made these contract systems woefully inefficient. Contractors refused to maintain mine works and used bribery and violence to usurp managerial authority.49 The Japanese were able to transfer experience gained from the reform of contract mining systems in Japan to Fushun. British managers enviously noted that unlike at Kailuan, where the contract system continued to discourage innovation, at Fushun “labor management improved so much that the mines [were] able to enjoy all the benefit generated from either the innovation or improvement on coal faces.”50 Following reforms, labor productivity at Fushun doubled, greatly surpassing performance levels at all other coal mines in China.51

49

Dixin Xu, “The Wage System,” compiles records describing organizational problems at Kailuan. Nisaburo

Murakushi, The Transfer describes labor management improvements at Fushun. Tim Wright describes subcontracting in the Chinese coal mining’ industry in Wright, “A Method.” 50

Quoted in Xu, “The Wage System.”

51

The comparison considers six year averages of labor productivity at the two largest mines in China, Fushun

and Kailuan, using data series in Murakushi, The Transfer, and Xu, “The Wage System.” Preceding the onset of reforms in 1927, average labor productivity levels at Fushun [1920-1926] and Kailun [1920-1926] were 115 and 153 tons of coal per man-year respectively. After reforms, labor productivity levels at Fushun [19301936] and Kailun [1930-1936] were 243 and 149 tons of coal per man-year respectively.

26

ORGANIZATIONAL REFORM UNDER OPPOSITION

Though the productivity gap between Japanese- and Chinese-owned mills was large on average, some Chinese-owned mills made significant advances in productivity through organizational reforms. 52 In other cases, attempts to modify mill organization led to strikes and riots, compelling would-be-reformers to roll back changes. Successful reform generally involved significant investments, including the development of in-house training for Chinese engineers and technicians, changes in management structure that allowed technicians to directly control production, and the design of employment systems that benefited workers and reduced foreperson control. Many mills, however, adopted a more piecemeal approach, attempting only to intensify monitoring without reorganizing power structures. In these mills, forepersons used strikes and violence to derail the reform agenda. This section considers two waves of reforms, contrasting a less successful period of piecemeal change in the mid-1920s, with a more successful phase that followed in the late 1920s and early 1930s. In both periods, reforms were not evenly distributed throughout the textile sector; the largest two Chinese-owned firms, Shenxin and Yongan, took the lead, while many others, especially British-owned firms, did not participate. Despite these limitations, by the early 1930s reformers controlled a growing share of Chinese-owned

52

Plant-level investigations of management practices conducted in 1932 indicate significant developments in

management practices during the 1920s and 1930s. For a description of management practices circa 1920, see Jiangsu Employment Office Third Department, The Situation. The 1932 survey comes from the Shanghai Municipal Archive collection of Jincheng Bank records.

27

capacity and their activities were generating productivity growth within the Chinese-owned sector. The Shenxin Company, the largest Chinese-owned textile firm, provides an excellent case study of the reform process. Sherman Cochran has studied Shenxin in detail and I draw heavily from his analysis.53 I also expand on Cochran’s qualitative study by using production statistics from the CCMOA data to show that reforms at Shenxin plants conducted during the late 1920s and early 1930s significantly improved productivity. Though one company does not comprise an industry, Shenxin was at the vanguard of the Chinese-owned sector, and changes within the firm were closely watched. The first period of reform relied heavily on labor markets as a source of new talent. In 1924 and 1925, Shenxin assigned shop floor positions to technicians recruited from Japanese-owned firms. Reform began with an experiment: managers assigned older, poorly performing American spindles to the new technicians, while forepersons retained control of newer British machines.54 By improving production methods, the technicians quickly exceeded the performance levels attained by the forepersons, convincing Shenxin’s plant manager of the need for full-scale reorganization. Technicians took control of the shop floor, directly monitoring foreperson and worker activities. Managers also established maintenance and experimental departments to address core governance deficiencies in machine upkeep and optimization. Although these policies did produce an immediate

53

See Cochran’s case study of the Shenxin company in Cochran, Encountering.

54

Zhongmin Zhang describes the experiment in Zhang, Scientific Management.

28

performance improvement, the benefits were only transient.55 Limited in number, technicians depended upon cooperation from subordinates to implement changes, but because technicians increased workloads and beat workers frequently, their presence was greatly resented. After several months, Shenxin’s forepersons organized a workplace riot, expelling technicians from the mill. Following negotiations, managers dismissed the technicians and reinstituted the prior governance system.56 Shenxin’s second period of reforms focused on educating and training technicians internally. As was characteristic of Chinese-owned firms, initial efforts to develop talent were centered on members of the company director’s extended family, many of whom were sent to textile schools abroad. Simultaneously, Shenxin developed its own educational capacity through the founding of a company professional school, the Shenxin Managerial Training Institute. Students attending the Institute’s year-long courses engaged in classroom learning under foreign-educated instructors and concurrently practiced their knowledge in the Institute’s simulated factory. In developing a central educational institution to support multi-plant operations, Shenxin followed an approach that resembled the international training routines in place in Japanese-owned mills. Centralized training at Shenxin provided a framework for the introduction of managerial innovations across

55

During the temporary reorganization, each worker attended 25 percent more machinery. Shanghai Academy

of Social Sciences, Historical Materials. 56

For records of events at Shenxin, see Shanghai Academy of Social Sciences, Historical Materials.

Foreperson opposition was common at other mills as well. See Frazier, “Mobilizing.”

29

multiple plants, including accounting techniques which allowed inter-plant comparisons of unit costs.57 In-house training provided the company with the capacity to reform itself. By 1933, 81 graduates had matriculated from the Institute, providing Shenxin with the nucleus of talent in middle management necessary to mount a successful reform of its Wuxi plant. This time, Shenxin’s management co-opted workers through the provision of a range of benefits, including improved housing, dining, recreation, and religious facilities, and greater voice in company governance. The company’s Wuxi campus, renamed ‘The Community of Self-Governing Workers’ because of the use of worker-elected judges to resolve labor disputes, created an autonomous social system which encouraged workers to identify their interests with those of management. Once the new system of managementlabor relations had taken hold, forepersons were dismissed from the mill and replaced with technical staff. The new staff rationalized work procedures, laying off supernumerary workers and nearly doubling the machine-tending demands for those who remained. After success at Wuxi, Shenxin began to introduce reforms at other mills. At the firm’s Hankou mill, similar reforms were coupled with the introduction of literacy requirements for new workers, who entered an extensive classroom-based training program upon employment.

57

In 1927, a similar accounting system was introduced at Yongan, the second largest Chinese-owned firm.

See Shanghai City Textile Industrial Bureau, Yongan Spinning.

30

Achievements were more limited in Shanghai where stronger opposition from the Green Gang made it more difficult to challenge the foreperson system.58 Based on reform dates for individual Shenxin plants provided in Cochran (2000), I test whether reform improved spinning and weaving productivity as measured using the CCMOA data. For spinning, I estimate plant-level TFP as the residual of a constant returns-to-scale Cobb-Douglas production function with equal factor shares. I regress spinning TFP on year dummies, plant-level fixed effects, and a reform dummy variable which is equal to one in the years following reform and equal to zero in the year of reform and all preceding years. This productivity measure is not count-adjusted, and OLS estimates of reform coefficient are biased downwards because Shenxin spun higher yarn counts than the Chinese-owned average. I try to control for this problem using plant fixed effects, but the coefficient is still likely biased downwards since average counts at Shenxin increased over time.59 Measurement of weaving productivity is less problematic because changes in cloth type have less dramatic effects on output. For weaving, I compute plantlevel TFP as the residual of a Cobb-Douglas production function, using 0.45 and 0.55 as the capital and labor shares. Estimates, reported in Table 4, indicate that reform dramatically improved company performance. Focusing on the fixed effects estimates reported in Specifications 1 and 3, reform increased spinning TFP by an average of 27 log percentage points, and weaving TFP by 78 log percentage points. Since Shenxin had

58

The personnel department in one of Shenxin’s Shanghai plants continued to be run by a gangster, see

Frazier, The Making. Reforms in Shanghai were likely limited by the Green Gang’s greater influence there, see Cochran, Encountering. 59

Wu reports that Shenxin began producing higher yarn counts around 1934. See Wu, “The Crisis.”

31

acquired 20 percent of Chinese-owned capacity in both spinning and weaving by 1936, success of the company’s reforms contributed greatly to aggregate productivity growth in the Chinese-owned sector. [Insert Table 4 here] Records of the Jincheng Bank, a Shenxin creditor, indicate that banks could play an active role in encouraging organizational change. Changes at Shenxin were closely observed by Jincheng and other banks; indeed, Jincheng participated in a consortium which took a direct role in managing some of Shenxin’s underperforming Shanghai mills. Jincheng placed observers in many of its debtors’ mills, relying on periodic reports from these agents to assess creditworthiness. Under the depressed market conditions of the early 1930s, continued access to credit became conditional on the acceptance of operational changes recommended by Jincheng staff.60 Research conducted by the Jincheng Bank allows another test of the hypothesis that organizational change influenced productivity. In 1932, Jincheng observers conducted an investigation of 32 Chinese- and British-owned mills in Shanghai and Jiangsu. At each mill, investigators recorded the levels of capital productivity, labor productivity, product quality, and unit costs attained in spinning 20 count yarn. Unusually, these performance measures were combined with assessments of potential explanatory factors, including machinery value, managerial quality, welfare institutions, maintenance procedures, and working conditions. Machinery value in particular is important because it provides an excellent proxy for machinery quality which is an alternative explanation for productivity differences.

60

Shanghai Municipal Archive collection of Jincheng Bank records.

32

To facilitate econometric analysis, I convert descriptive information in the Jincheng dataset into numerical codes. I use sentence-long descriptions of managerial quality, welfare institutions, and working conditions to code a dummy variable measuring each mill’s organizational quality. For 21 out of the 32 miIls, I assign an organizational quality dummy equal to one; these mills had positive assessments containing words like ‘orderly’, ‘scientific’, or ‘among the best’. I assign a zero value to mills with negative assessments, described using phrases like ‘lacking order’ or ‘employing the foreperson system, and thus impossible to organize.’ For product quality, here referring to dimensions such as the yarn’s appearance and strength rather than count, I reduce simple verbal descriptions such as ‘top grade’ and ‘middle grade’ to a 4-tier numerical scale (0, 1, 2, 3). The treatment of machinery quality is also important. The bank only provides an assessment of the combined value of each mill’s spindles and looms. To generate mill-level estimates of spindle quality, I regressed the assessed value of production equipment on the total number of spindles and looms. Based on this regression, I estimate the percentage of capital value attributable to spinning equipment at each mill, and use this to estimate the average value per spindle. As one might expect, spindles owned by recently established mills had a higher estimated value than those of older mills. Table 5 presents the results of a series of regressions of performance measures on spindle value and the organizational quality dummy. The regressions indicate that organizational quality had a significant impact on the performance of Chinese- and Britishowned mills, regardless of the performance measure used. Improvements in organization appear to have had a more dramatic effect on labor productivity (0.29 log points) than capital productivity (0.09 log points). This suggests that effective organization primarily 33

aided mills by allowing them to reduce staffing levels. Well-organized mills also spun significantly higher-quality yarn, suggesting that focusing only on quantity and count may miss some performance-relevant features. Although the differences are too small for statistical significance, more expensive capital equipment does appear to have been associated with slightly higher levels of capital productivity. More expensive machines were also more heavily staffed, however, meaning that effects on total factor productivity are ambiguous. [Insert Table 5 here] The results from analysis of the Jincheng investigation suggest that variations in organizational practice influenced productivity levels within the Chinese- and Britishowned sectors. British-owned mills were among the worst managed mills included in the survey; out of four mills whose managerial conditions were described in particularly harsh terms, two were British. Though no Japanese-owned mills were included in Jincheng’s investigation, surveys of general industry conditions conducted in 1920, 1929, and 1930 indicate that Japanese-owned mills were the best managed in China.61 The finding that management, and not machinery differences, were behind within-group productivity differences among Chinese-owned mills, strongly suggests that the same was true of aggregate differences between Chinese- and Japanese-owned mills. Even after nationalization following World War II, formerly Japanese-owned mills continued to be more productive than their competitors. In 1951, a Shenxin company investigation of these

61

Surveys comparing Japanese-, Chinese-, and British-owned management include Jiangsu Third

Employment Office, The Situation, Pearse, The Cotton Industry, and Moser, The Cotton Textile Industry.

34

factories concluded that advanced training programs and management techniques, and not machinery differences, were the key factors behind their superior performance.62

CONCLUSION

The Chinese experience with the absorption of textile technology reveals several points of interest to students of technology transfer, foreign investment, and development. The organization of textile production in China was heavily influenced by the local institutional environment. Difficulty in negotiating local institutions caused British-owned textile firms to perform worse than Chinese-owned firms, even though they, like all firms in the industry, were operating with primarily British machines. For firms with the appropriate knowledge of how to organize Chinese labor, good results could be obtained. Japanese-owned firms, equipped with locally-appropriate organizational techniques, were able to achieve productivity levels which approached those in Japan. The experience of Japanese- and British-owned firms in China shows that context affects the success of technology transfer. A managerial system may be best practice in one context, but perform poorly in a different institutional and cultural environment. The implication is that knowledge may only be useful within a certain scope, defined by input characteristics. Transfers of knowledge to regions with dissimilar inputs may fail to improve productivity.

62

To support this conclusion, the survey notes that outdated equipment in formerly Japanese-owned mills

performed better than new equipment in Chinese-owned mills. See Shenxin Company, A Visit.

35

One interesting implication of the theory is that foreign direct investment follows a matching process; investors match differentiated firm-level knowledge stocks to differentiated input sources, and obtain a productivity level which reflects goodness of fit. Based on this theory, one would expect countries with similar input characteristics, for instance those sharing common cultures, to have unusually strong investment relationships. Between WWI and WWII, this appears to have been true of Japan and China. For example, between 1914 and 1923, Japanese-owned firms grew from 19 percent to 62 percent of the total number of foreign firms registered in Shanghai, China’s premier industrial center.63

63

Bruce Reynolds, The Impact.

36

REFERENCES Barker, Aldred F. and Kenneth C. Barker. “The Textile Industries of China: Their Present Position and Future Possibilities, A Report Presented to Chiao-Tung University.” Publisher Not Listed, Held in Shanghai Library, 1934. Bloom, Nicholas, Raffaella Sadun, and John Van Reenan. “The Organization of Firms across Countries.” Working Paper 15129, NBER, 2009. Bruhn, Miriam, Dean Karlan, and Antoinette Schoar. “What Capital is Missing in Developing Countries?” American Economic Review 100, no. 2 (2010): 629-33 Chin, Rockwood Q.P. “Cotton Mills, Japan’s Economic Spearhead in China.” Far Eastern Review 6, no. 23 (1937): 261-67. Chin, Rockwood Q.P. “Japanese-Owned Cotton Mills in China: A Study in International Competiton.” Ph.D. Diss., Yale University, 1937. China Textile Construction Company Qingdao Branch.《青纺三年》(Three Years of Qingdao Spinning.) China Textile Construction Company Qingdao Branch, 1949. Clark, Gregory. "Can Management Develop the World?: Reply to Wilkins." The Journal of Economic History 48, no.1 (1988): 143-48. Cochran, Sherman. Encountering Chinese Networks: Western, Japanese, and Chinese Corporations in China, 1880–1937. Berkeley: University of California Press, 2000. Duus, Peter. “Japanese Cotton Mills in China, 1895-1937.” in The Japanese Informal Empire in China, 1895-1937, Peter Duus, Ramon H. Myers, and Mark R. Peattie, eds., Princeton: Princeton University Press, 1989. Field, Fredrick V. “China’s Foreign Trade,” Far Eastern Review 4, no. 5 (1935): 33-40. 37

Fong, H.D. Cotton Industry and Trade in China. Tianjin, China: Nankai Institute of Economics, 1932. Frazier, Mark W. “Mobilizing a Movement: Cotton Mill Foreman in the Shanghai Strikes of 1925.” Republican China 20, no. 1 (1994): 1-45. ______. The Making of the Chinese Industrial Workplace, Cambridge, UK: Cambridge University Press, 2002. Fujino, Shozaburo, Shiro Fujino, and Akira Ono. Estimates of Long-Term Economic Statistics of Japan Since 1868: Textiles. Tokyo: Toyo Keizai Shinposha, 1979. Hershatter, Gail. The Workers of Tianjin, 1900-1949. Stanford: Stanford University Press, 1986. Honig, Emily. “The Contract Labor System and Women Workers: Pre-Liberation Cotton Mills of Shanghai.” Modern China 9, no. 4 (1983): 421-54. ______. Sisters and Strangers: Women in the Shanghai Cotton Mills, 1919-1949. Stanford: Stanford University Press, 1986. Izumi, Takeo. Transformation and Development of Technology in the Japanese Cotton Industry. Tokyo: United Nations University, 1980. Jacoby, Sanford. “The Origins of Internal Labor Markets in Japan.” Industrial Relations 18, no. 2 (1979): 184-96. Jewkes, John and E. M. Gray. Wages and Labour in the Lancashire Cotton Spinning Industry. Manchester: Manchester University Press, 1935. Jiangsu Employment Office Third Department.《江苏省纺织业状况》 (The Situation of the Jiangsu Textile Industry.) Shanghai: Commercial Press, 1920.

38

Jin, Zhihuan.《中国纺织公司研究 1945-1950》(Research on the China Textile Construction Company, 1945-1950.) Shanghai: Fudan University Press, 2006. Koll, Elizabeth. From Cotton Mill to Business Empire: The Emergence of Regional Enterprises in Modern China. Cambridge, MA: Harvard University Press, 2003. Kraus, Richard A. Cotton and Cotton Goods in China, 1918-1936. New York: Garland, 1980. Kuwahara, Tetsuya. “The Establishment of Oligopoly in the Japanese Cotton-Spinning Industry and the Business Strategies of Latecomers: The Case of Naigaiwata and Co., Ltd.” in Japanese Yearbook on Business History:1986. Nakagawa Keiichiro and Morikawa Hidemasa, eds. Tokyo: Japanese Business History Institute, 1986. ______. “The Local Competitiveness and Management of Japanese Cotton Spinning Mills in China in the Inter-war Years.” in International Technology Transfer: Europe, Japan, and the USA, 1700-1914. David Jeremy, ed. Brookfield, VT: E. Elgar, 1991. ______. “The Development of Factory Management in Japan during the Early Stages of Industrialization: the Kanegafuchi Cotton-Spinning Company before the First World War.” in The Fibre that Changed the World: the Cotton Industry in International Perspective,1600-1990s. Douglas Farnie and David Jeremy, eds. New York: Oxford University Press, 2004. Lazonick, William H. “Production Relations, Labor Productivity, and Choice of Technique: British and U.S. Cotton Spinning.” The Journal of Economic History 41, no. 3 (1981): 491-516.

39

Lee, B.Y. “Real Causes Behind Japanese Mill Strikes.” The China Weekly Review, February 28, 1925. Leunig, Timothy. “A British Industrial Success: Productivity in the Lancashire and New England Cotton Spinning Industries a Century Ago.” Economic History Review 56, no. 1 (2003): 90-117. Levine, Solomon. “Labor Markets and Collective Bargaining in Japan.” In The State and Economic Enterprise in Postwar Japan. W.W. Lockwood, ed. Princeton: Princeton University Press, 1965. Levinsohn, James, and Amil Petrin. “Estimating Production Functions Using Inputs to Control for Unobservables.” The Review of Economic Studies 70, no. 2 (2003): 31741. Levinsohn, James, Amil Petrin, Brian P. Poi. “Production Function Estimation in Stata Using Inputs to Control for Unobservables.” The Stata Journal 4, no. 2 (2004): 11323. Ma, Debin. “Shanghai-Based Industrialization in the Early 20th Century: a Quantitative and Institutional Analysis.” Working Paper 18, LSE Global Economic History Network, 2006. Markusen, James R., Multinational Firms and the Theory of International Trade. Cambridge: MIT Press, 2004. Moriguchi, Chiaki. “Implicit Contracts, The Great Depression, and Institutional Change: A Comparative Analysis of the U.S. and Japanese Employment Relations, 19201940.” The Journal of Economic History 63, no. 3 (2003): 625-65.

40

Moser, Charles K. The Cotton Textile Industry of Far Eastern Countries. Boston: Pepperell, 1930. Murakushi, Nisaburo. The Transfer of Coal-mining Technology from Japan to Manchuria and Manpower Problems: Focusing on the Development of the Fushun Coal Mines. Tokyo: United Nations University, 1981. Nimura, Kazuo. The Ashio Riot of 1907: A Social History of Mining in Japan. translated by Terry Boardmen and Andrew Gordon. Durham, NC: Duke University Press, 1997. Odell, Ralph. Cotton Goods in China. Washington, DC, US Department of Commerce, 1916. Officer, Laurence H., “Exchange Rates Between the United States Dollar and Forty-one Currencies.” www.measuringworth.com (accessed September 21, 2008). Pearse, Arno S. The Cotton Industry of Japan and China. Manchester: Taylor, Garnett, Evans, & co., 1929. Qingdao Textile Factory Statistical Office. 《三十六年度统计年报： 中国纺织建设公青岛分公司》(1947 Annual Statistical Report: China Textile Construction Company Qingdao Branch.) Publisher Not Listed, No Date. Ramseyer, Mark J. “Credibly Committing to Efficiency Wages: Cotton Spinning Cartels in Imperial Japan.” John M. Ohlin Law & Economics Working Paper 13, University of Chicago, 1993. Reynolds, Bruce L. “The Impact of Trade and Foreign Investment on Industrialization: Chinese Textiles, 1875-1931.” PhD. Diss., University of Michigan, 1975. Shanghai Academy of Social Sciences Economics Research Institute. 《荣家企业史》 41

(Historical Materials on the Rong Family Enterprises.) Shanghai: Shanghai People’s Publishing House, 1962. Shanghai City Textile Industrial Bureau. 《永安纺织印染公司》(Yongan Spinning, Weaving, Printing, and Dyeing Company.) Beijing: Zhonghua Publishing Bureau, 1964. Shanghai City Textile Industry Association. 《中国棉纺统计史料》(Historical Statistical Materials on the Chinese Cotton Spinning Industry.) Shanghai: Shanghai City Textile Industry Association, 1950. Shanghai Municipal Archives. Chinese Cotton Mill Owners Association 1933 Annual Report, Shanghai, China, 1933. Shanghai Municipal Archives. Jincheng Bank Records, Shanghai, China, 1932-1934. Shanghai Municipal Archives. Shenxin Company Records, Shanghai, China, 1932-1952 Shanghai Municipal Archives. Sixth Area Mechanized Textiles Industrial Committee. 《国纺织工厂调查表》 (National Textile Factory Survey), Shanghai, China, 1947. Shenxin Company. 《青岛天津北京参观》(A Visit to Qingdao, Tianjin and Beijing.) Shanghai: Shenxin Company Publications, 1951. Seki, Keizo. The Cotton Industry of Japan. Tokyo: Japan Society for the Promotion of Science, 1956. Smith, Stephen A. Like Cattle and Horses: Nationalism and Labor in Shanghai, 1895 1927. Durham: Duke University Press, 2002.

42

Sugiyama, Shinya. “Marketing and Competition in China, 1895-1932: The Taikoo Sugar Refinery.” in Commercial Networks in Modern Asia. Shinya Sugiyama and Linda Grove, eds. Richmond: Curzon Press, 2001. Tolliday, Steven and Johnathan Zeitlin. “Employers and Industrial Relations Between Theory and History.” in The Power to Manage? Employers and Industrial Relations in Comparative Historical Perspective. Steven Tolliday and Jonathan Zeitlin, eds. New York: Routledge, 1991. Wang, Ju.《近代上海棉纺业的最后辉煌 1945-1949》(The Last Hurrah of the Modern Shanghai Cotton Spinning Industry, 1945-1949.) Shanghai: Shanghai Academy of Social Sciences, 2004. Wang, Zijian, and Zhenzhong Wang.《七省华商纱厂调查报告》(A Report on the Investigation of Chinese Spinning Mills in Seven Provinces.) Shanghai: Commercial Press, 1935. Wolcott, Susan, and Gregory Clark. "Why Nations Fail: Managerial Decisions and Performance in Indian Cotton Textiles, 1890-1938." The Journal of Economic History 59, no. 2 (1999): 397-423. Wright, Tim. “’A Method of Evading Management’—Contract Labor in Chinese Coal Mines before 1937.” Comparative Studies in Society and History 23, no. 4 (1981): 656-78. Wu, Leonard. “The Crisis in the Chinese Cotton Industry.” Far Eastern Survey 4, no. 1 (1935): 1-4. Xu, Dixin. “The Wage System and Wage Levels in Old Kailuan Mines.” Chinese Economic Studies 23, no. 4 (1990): 9-125. 43

Yan, Zhongping.《中国棉纺织史稿》(A Draft History of the Chinese Cotton Textile Industry.) Beijing: People’s Press, 1965. Yoshida, Youshiyuki and Hiroshi Wada. 《青島紡績勞働調查》(The Investigation of Qingdao Textile Labor.) Dalian: Manchurian Railway Company Investigation Bureau, 1940. Zhang Zhongmin.《20 世纪 30 年代上海企业的科学管理》(Scientific Management in Shanghai Enterprises during the 1930s.) Shanghai Economic Research 6 (2003): 72-79. Zhao, Gang and Zhongyi Chen. 《中国棉纺织史》(Chinese Cotton Textile History.) Beijing: Agricultural Press, 1997.

44

TABLE 1 ELASTICITY OF CAPITAL AND LABOR REQUIREMENTS WITH RESPECT TO COUNT Dependent Variable Log output Log inverse wage costs

per worker-

Log output per spindle-hour

month Dataset

QY

SH

QY

WW

QY

-0.65**

-0.66**

-0.77**

-1.15**

-1.15**

(0.06)

(0.15)

(0.13)

(0.05)

(0.06)

8 to 60

4 to 60

10 to 42

8 to 80

Fixed effects

N/A

Yes

Yes

Yes

Yes

Time dummies

Yes

Yes

Yes

N/A

Yes

4

8

37

8

Robust

Plant-level

Plant-level

Plant-level

Plant-level

OLS

clusters

clusters

clusters

clusters

0.92

0.78

0.27

0.90

0.96

28

66

232

90

141

Log count

Range of counts

Averaged Number of mills

across 8 mills

Standard errors R2 Observations

Standard errors in parentheses, clustered errors are bootstrapped; * 5% significance ** 1% significance Table 1 presents estimates of the elasticity of labor and capital productivity with respect to count based on longitudinal production statistics in the Qingdao Yearbook, Shenxin, and Wang and Wang data. The three datasets are identified in the table as QY, SH, and WW, respectively.

45

TABLE 2 CAPITAL AND LABOR PRODUCTIVITY IN CHINESE-, JAPANESE-, AND BRITISH-OWNED MILLS IN CHINA, AND IN MILLS IN JAPAN Pounds per worker-hour Location

Pounds per spindle-hour Japand

China

Year

Chinese

British

Japanese

1.02

0.97

1.37

Japand

China Chinese

British

Japanese

1.88

0.027

0.025

0.033

0.042

1924-1925

0.75

1924-1925a

0.78

1927-1928

0.64

1.16

0.94

1.82

2.56

0.033

0.025

0.040

0.042

1929

0.59

1.19

0.97

2.02

2.56

0.032

0.026

0.047

0.042

1929b

0.55-0.7

0.025-0.033

0.037-0.046

0.045

0.038

0.046

0.045

1929a 1930

0.53

1.14

1.28

2.16

3.54

0.030

0.027

0.045

0.048

1931

0.44

1.15

1.24

2.64

3.54

0.028

0.028

0.045

0.048

1932

0.44

1.20

1.24

2.72

3.54

0.032

0.027

0.043

0.048

1932a

0.034

1933

0.50

1.35

1.10

2.69

3.54

0.030

0.024

0.041

0.048

1934

0.51

1.36

0.76

2.69

3.54

0.029

0.017

0.043

0.048

1934a

0.032

1935

0.50

1.54

0.79

3.07

3.99

0.030

0.015

0.046

0.045

1936

0.49

1.51

1.02

3.07

3.99

0.031

0.023

0.048

0.045

1936a

0.46

1947c

2.73

0.045

Table 2 compares estimates of labor and capital productivity based on CCMOA data to estimates reported by other sources. The figures refer to either 20 counts or 20 count equivalents. Years for CCMOA estimates are bolded. Data from other sources indicated below. a

Zhao and Chen, Chinese Cotton Textile History

b

Duus (1989), citing Takamura (1982)

c

Qingdao Yearbook Data

d

Wolcott and Clark, “Why Nations Fail.”

46

TABLE 3 COBB-DOUGLAS PRODUCTION FUNCTION FOR WEAVING Dependent variable: Log cloth output (measured in bolts per shift) OLS Independent variables

LP

(1)

(2)

(3)

(4)

(5)

(6)

0.53**

0.53**

0.50**

0.46**

0.44**

0.46*

(0.13)

(0.14)

(0.13)

(0.12)

(0.14)

(0.22)

0.53**

0.51**

0.46**

0.56**

0.56**

0.55**

(0.15)

(0.16)

(0.12)

(0.12)

(0.11)

(0.16)

0.19*

-0.01

0.04

(0.10)

(0.10)

(0.10)

Log looms (βk)

Log employment (βl)

Auto looms share

Prior Japanese

0.31**

ownership

(0.10)

Log spinning labor

0.46**

productivity

(0.12) 0.06

0.04

-0.04

0.02

-0.01

0.01

(0.06)

(0.09)

(0.09)

(0.07)

(0.07)

(0.23)

Yes

Yes

Yes

Test of CRS (βk+βl-1) FirmTime dummies

Yes specific

Total observations

257

173

173

116

257

R2

0.78

0.82

0.84

0.95

0.81

257

Standard errors are adjusted for plant-level clustering and are reported in parenthesis. * 5% significance ** 1% significance Table 3 reports the estimations of Cobb-Douglass production functions for weaving. The results suggest constant returns-to-scale, capital and labor shares of around 0.45 and 0.55, and that the use of automatic looms did not significantly affect productivity. Specification 3 indicates that formerly Japanese-owned firms continued to enjoy a productivity advantage under government management

47

TABLE 4 PRODUCTIVITY EFFECTS OF REFORMS AT THE SHENXIN COMPANY Dependent variable Log spinning TFP Independent variable

Log weaving TFP

(1)

(2)

(3)

(4)

0.27*

0.05

0.78**

0.55**

(0.14)

(0.09)

(0.22)

(0.15)

Year dummies

Yes

Yes

Yes

Yes

Mill fixed effects

Yes

No

Yes

No

Number of observations

440

440

182

182

R2

0.01

0.01

0.11

0.32

Reform dummy

Standard errors clustered at the plant-level are reported in parentheses. * 5% significance ** 1% significance Table 4 presents a series of OLS regressions sh that organizational reforms at Shenxin improved productivity. The regression sample is restricted to Chinese-owned mills operating between 1929 and 1936. Production statistics are taken from the CCMOA data and organizational reforms are dated according to Cochran, Encountering. In both spinning and weaving, I measure plant-level log total factor productivity as the residual of a Cobb-Douglas production function, where the capital and labor shares in spinning are 0.5 and 0.5, and in weaving 0.45 and 0.55. The coefficient on the reform dummy in Specification 2 is biased downwards because Shenxin spun higher counts than the Chinese-owned average. Specification 1 partially corrects for this problem using fixed effects, but estimates may still be biased downwards because the counts spun at Shenxin increased over time.

48

TABLE 5 PRODUCTIVITY EFFECTS OF ORGANIZATIONAL QUALITY Dependent variable Independent variable

Log output

Log unit

Product

per spindle

cost

quality

Log output per worker

0.09**

0.27**

0.07**

-0.08*

0.8*

(0.03)

(0.09)

(0.03)

(0.03)

(0.35)

0.07

-0.07

0.04

0.06

0.2

(0.04)

(0.10)

(0.03)

(0.05)

(0.70)

19.0

19..

19.0

1900

1900

2.

2.

2.

.2

20

Organizational quality dummy

Log capital value Per spindle

0.97** Log spindles per worker (0.06) R2 Number of observations

Robust standard errors are reported in parentheses. * 5% significance ** 1% significance Table 5 presents a series of OLS regressions demonstrating that well-organized mills had significantly higher output per worker, output per spindle, and product quality, and significantly lower unit costs. The data is taken from the Jincheng Bank Records and the statistics refer to 20 count yarn production.

49

FIGURE 1 PRODUCTIVITY LEVELS IN SPINNING TFP Index 1.2

1

0.8

0.6

0.4

0.2

0 1924

Japanese-Owned

1926

1928

Chinese-Owned

1930

1932

British-Owned

1934

1936

Year Figure 1 displays total factor productivity indices for the Japanese-, Chinese-, and British-owned sectors. Productivity levels are reported as a percentage of average spinning productivity in Japan circa 1930-1934. With the exception of 1932, the aggregated indices include all mills for which output data is available in CCMOA statistics. In 1932, the Shanghai Incident forced Japanese-owned mills in Shanghai to repatriate Japanese employees and close temporarily, and thus the 1932 Japanese series excludes mills in Shanghai.

50

FIGURE 2 PRODUCTIVITY LEVELS IN WEAVING TFP Index 3

2.5

2

1.5

1

0.5

0 1928

Japanese-Owned

1929

1930

1931

Chinese-Owned

1932

British-Owned

1933

1934

1935

1936

Year Figure 2 displays indices of sector-level total factor productivity in weaving. I normalize the productivity indices by setting Chinese-owned productivity in 1928 equal to one. With the exception of 1932, the aggregated indices include all mills for which output data is available in CCMOA statistics. In 1932, the Shanghai Incident forced Japanese-owned mills in Shanghai to repatriate Japanese employees and close temporarily, and thus the 1932 Japanese series excludes mills in Shanghai.

51