But usually only temporarily. Unlike those in England or the Soviet Union, Chinese peasants were not dispossessed of their property; though the state continued to own all agricultural land, thirty-year leases gave families effective control. Workers could and did move back and forth between farms and factories, knowing that they had something to return to in their home villages.22
In most cases they had to return home, like it or not, because of the Chinese hukou system of residency permits, instituted in the 1950s. Chinese citizens need a permit to live in particular areas and most social benefits, including health care and public schooling, are linked to the specific hukou they possess. Migrant workers received temporary residency permits arranged by their employers, which expired when they left their jobs. Obtaining a permanent shift of residency permit to a city was all but impossible. For the first generation of migrant workers, factory jobs (and urban construction work and service jobs) were necessarily interludes, usually lasting a few years, often between the time of finishing or dropping out of school and beginning a family, much as had been the case for New England mill workers.23
Migrant factory workers had a different and inferior social status than workers in state-owned enterprises. Until reforms that began in the late 1980s, state-owned and collective employers in China provided a broad range of benefits, including permanent job tenure, training, housing, lifetime medical care, pensions, and other welfare provisions, even subsidized haircuts. Generally, the intensity of work was light and managerial discipline minimal.24 This was not the case in the privately owned factories that blossomed as the state-owned enterprises began to shrink. Job turnover in the special economic zones was astoundingly high. Many companies provided dormitory housing to migrant workers for free or a fee but otherwise took no responsibility for their welfare. Whatever benefits workers were eligible to receive—including educational opportunities for their children and pensions—came from their home area, where they were registered under the hukou system. Private employers were legally required to contribute to social benefit funds for their workers, but like minimum wage and overtime regulations the requirement was frequently ignored. The intensity of work in private-sector factories was high and discipline harsh.25
In effect, China developed two quite different systems of factory production, one state or collectively owned, the other privately owned, with different informing ideologies, laws, customs, standards of living, and workforces. Even the terminology for workers differed. Employees at state enterprises were gongren (“workers”), holding, at least in theory, the highest social status in China during its communist heyday. Rural migrant workers, by contrast, were often called by newly coined terms dagongmei or dagongzai (“laboring girl”) or (“laboring boy”) with the connotation of hired hand, a low-status appellation.26
The migrant labor system provided employers with a vast workforce, expandable and shrinkable at will. The pool of rural young men and women was so large that it took nearly two generations before labor shortages began. And it was a pool of cheap labor. Most factories paid migrant workers the legal minimum wage (which in China is set by local governments) or less, as enforcement generally was minimal. Recruiting out of a rural labor market, where living standards and wages were far below urban norms, the coastal export factories did not have to match wage levels for local workers or what state-owned enterprises paid, able to attract workers because the low wages they offered were substantial by village standards. Furthermore, because the factories did not pay for most social welfare benefits for their workers, they were effectively being subsidized by the rural governments that did, allowing their labor costs to be below the cost of social reproduction in the areas they were located. Like Stalinist industrialization, Chinese industrialization has depended on squeezing wealth out of the countryside.27
Lodging workers in company dormitories was both a necessity and an advantage for big export factories. Migrant workers, because of housing shortages in the factory boomtowns and their lack of permanent resident status, often had difficulty finding lodging. To attract workers, factories provided it themselves, just as the Lowell mills and the Soviet industrial giants had done. Doing so allowed them to pay workers less than they would have to if those workers had to obtain housing on the open market.
In the early years of private factory growth, most of the migrant workers were young women, so lodging them in company dorms also had an element of providing a chaste environment. One large electronic firm required as a condition of employment that all young, unmarried, single women live in dormitories within the factory complex. Even after men began to be hired for production jobs, dormitories generally remained sexually segregated.
The dormitory system gave companies extraordinary control over their workers. As in the Lowell-style mills, many Chinese factories had (and have) detailed rules for behavior, imposing fines not only for being late to work, poor-quality work, or talking on the job but also for littering or leaving dormitory rooms untidy. Foxconn forbids workers of the opposite sex from visiting one another in their rooms, bans drinking and gambling, and imposes a curfew.
Having workers in company housing allows factories to mobilize large numbers of workers rapidly when rush jobs come in and makes it easier to have large numbers of young women working night shifts. Extremely long working hours—sometimes twelve hours at a time or more, a common practice, especially during busy seasons—are easier to demand if workers live right at the factory.28
In the mid-1990s, there were an estimated 50 million to 70 million migrant Chinese workers. In 2008, 120 million. By 2014, more than 270 million, nearly double the number of employed civilian workers of any kind in the United States, an oceanic movement of population from farms to factories and back.
Hometown networks play an important role in the movement, as migrant workers tell sisters and brothers and neighbors about the opportunities and city life and help them find jobs. Provincial and local governments have facilitated the flow. Interior provinces helped recruit workers for factory labor elsewhere, prizing the remittances they sent back home. Some local governments set up offices in Shenzhen to connect workers from their region to foreign-owned factories. Without active state support, the whole system would not have been possible.
The urban employment of rural workers has turned the Spring Festival week around Chinese New Year into an epic of logistics, emotion, and labor recruitment. Each year, millions upon millions of migrant workers return home for the holiday, to be reunited with parents, children, and village friends, in what has become the world’s largest, regular human migration. In 2009, the Chinese railway system expected to carry about 188 million passengers during the holiday period. Huge crowds fill stations and spill over into neighboring streets. Ticket systems crash under the weight of demand. Trains and buses are crammed and overcrammed with people and baggage (though the recent expansion of the Chinese railroad system has somewhat eased the chaos). When the holiday ends, not everyone goes back. Each year, millions of migrant workers decide to stay home, forcing factories and other employers to scramble to find replacements.29
Why So Big?
Migrant labor made possible the rapid expansion of export-oriented manufacturing in China—and also Vietnam—but it does not explain the creation of factories larger than any ever seen before.30 For the most part, their size is not a result of technical requirements of production. Look at a photograph of a large sneaker factory in, say, Vietnam and what you will most likely see will be rows of workers sitting at individual workstations assembling precut pieces. (Sneakers and casual footwear are made by gluing and stitching together pieces of rubber, synthetic fabrics, synthetic leather, and sometimes actual leather.) Masses of workers may be under the same roof, but for the most part their labor is individual or in small groups, doing work identical to other individuals or groups nearby, without interacting with them.31 In this respect, these plants are less like River Rouge or Magnitogorsk and more like the early English textile mills,
where weavers or spinners stood side by side doing individual tasks.
Even when products require more complex assembly, there is often no clear relationship between the number of workers needed to make a particular product and the size of a factory. In the EUPA factory, the Taiwanese-owned small appliance plant featured in Manufacturing Landscapes and one of Edward Burtynsky’s best-known photographs, assembly workers are housed in a vast, modern, single-story shed. But each assembly line within it is short and relatively simple. Thirty lines made electric grills, but each had an average of only twenty-eight workers, not the hundreds found on integrated assembly lines in automobile or tractor plants. Rows of assembly workers face each other across a slowly moving belt. For the most part they use simple hand tools, without mechanical pacing of production, taking pieces on and off the belt rather than working on moving components, as in an auto plant.
Figure 7.1 Workers making Reebok shoes in a factory in Ho Chi Minh City, Vietnam, 1997.
Electronics firms are notoriously secretive, so it is difficult to get a full sense of their manufacturing processes. But one account of an Apple production area within the Foxconn Longhua complex described assembly lines ranging from dozens to more than a hundred workers each, larger than the lines in shoe or small appliance factories but still very modest in size compared to the overall size of the factory, with its several hundred thousand workers.
Vertical integration adds to plant size. Some footwear plants make the synthetic materials that go into sneakers and shoes, mold and cut pieces, and embroider logos. EUPA manufactures most of the parts used in the goods it produces. Foxconn makes some of the components that go into the devices it assembles, though most of the high-end elements come from elsewhere.
Still, even adding in parts manufacturing, technological requirements do not explain giant plant size. Rather it is like Alfred Marshall’s comment about cotton spinning and weaving, that “a large factory is only several parallel smaller factories under one roof.” At Foxconn City, that was almost literally the case, with separate buildings used to assemble similar products for different companies.
Beyond some point, economies of scale in production diminish or disappear. In his classic study Scale and Scope Alfred D. Chandler, Jr., after noting that at one point close to a quarter of the world’s production of kerosene came from just three Standard Oil refineries, wrote: “Imagine the diseconomies of scale that would result from placing close to one-fourth of the world’s production of shoes, textiles or lumber into three factories or mills! In those instances the administrative coordination of the operation of miles and miles of machines and the huge concentration of labor needed to operate those machines would make neither economic nor social sense.” Yet something close to that has happened in the production of electronic devices and some types of footwear. In the case of Apple, production concentration has gone beyond what Chandler imagined as absurd; every iPad is assembled in a single factory and most iPhone models in just one or two.32
Why are the factories so large? The answer seems to lie in economies of scale and competitive advantages, not for manufacturers, but for the retailers that sell the products they make. This reflects a fundamental shift in relations between the two parties. Until fairly recently, the design, manufacture, and marketing of consumer products generally occurred within the confines of one company. But since the 1970s they have been delinked. And, as sociologist Richard P. Appelbaum has argued, in contemporary global supply chains it is retailers and branders (designers and marketers that depend on others for manufacturing) who have the most power to establish the arrangements and terms of production, not factory owners. Factory giantism serves their interests.33
Early in the history of factory production, some of the most successful manufacturers established their dominance by selling their products under brand names and controlling distribution networks. In the United States, the Lowell mills pioneered this approach, which was adopted by such iconic companies as the McCormick Harvesting Machine Company. The Singer Manufacturing Company extended the model to a global scale, as its salesmen and distribution agents sold sewing machines across Europe and the Americas, largely produced in just two factories. The big automobile manufacturers used the model as well, selling cars that they branded—Fords and Chevys, Chryslers and Cadillacs—through independently owned dealerships that they effectively controlled. General Electric, IBM, and RCA likewise sold or leased their products under their own names and exerted considerable influence, if not total control, over distribution networks.
The manufacturer-dominated system of branded products stayed in place in Europe and the United States through the 1970s. Goods producers like Volkswagen, GM, Siemens, Sony, Ford, Whirlpool, Levi Strauss, and Clarks shoes (which first garnered wide attention when its products won awards at the 1851 Crystal Palace exhibition) persisted as household names. The companies, their products, and the factories that produced them remained tightly bound to one another in reality and image.34
The severe global recession of the 1970s and a series of subsequent developments unraveled the ties. With profit rates declining as a result of increased international competition, rising energy and labor costs, tight credit, and inflation, many American corporations, under pressure from corporate raiders, sought to reduce costs and shed less profitable operations. To become leaner and more flexible and show a rapid drop in spending, they began outsourcing to other firms functions they had traditionally performed themselves. They tended to start with support services, such as data processing and communications. But over time, companies began outsourcing core functions, too, including manufacturing.35
Take sneakers. From their introduction in the nineteenth century through the 1960s, sneakers generally were designed and made by the same companies, mostly large, stodgy rubber firms like United States Rubber Company (Keds) and BF Goodrich (PF Flyers). But then dominance shifted to companies like Adidas, Puma, Reebok, and Nike that were built around athletic footwear and clothing rather than rubber and focused on technological innovation, fashionable design, and marketing. While into the 1980s most of the industry leaders, including Nike, did at least some of their own manufacturing, increasingly they contracted out production, until they became essentially just branders.36
In the electronics and computer industries as well, leading corporations began contracting out some of their manufacturing. Sun and Cisco, two Silicon Valley success stories, worked with specialized contract manufacturers, like Solectron and Flextronics (before the rise of Foxconn, the largest such firm), to manufacture advanced products, sold under their brand names. Some companies, including IBM, Texas Instruments, and Ericsson (a large Swedish telecommunications manufacturer), sold off individual factories or even whole manufacturing divisions to smaller firms, with which they then contracted to do their manufacturing. Over time, contract manufacturers became increasingly sophisticated in their design and logistics capacities, partnering with their clients in integrated, multifirm production systems, stitched together by electronic data communication.37
During the same years, a revolution in selling took place as well. It had two facets, the rise of new, giant, low-price retailers and the burgeoning of global brand companies that did little or no manufacturing themselves.
In the United States, the new mass retailers had their origins in the 1960s, when a series of discount store chains, including Wal-Mart and Target, were founded. But it was not until the 1980s that they really took off. Wal-Mart, using a combination of low-wage labor, low prices, advanced technology, and highly efficient logistics, grew into the largest retailer in the world. In 2007 it had 4,000 stores in the United States and 2,800 elsewhere. Though no company came even close to Wal-Mart in size, other retailers based in Europe and the United States, like Carrefour, Tesco, and Home Depot, ballooned through expansion and acquisitions.
With their massive purchasing power, giant retailers won an edge over their suppliers, whether well-known companies like Levi Strauss or obscure
firms that made products sold under the retailers’ house labels. New communications and logistics technology, including bar codes, computer tracking systems, and the internet, allowed retailers to monitor, communicate with, and direct suppliers on an almost instantaneous basis. Faced with the possibility of the loss of massive orders, companies that made goods for megaretailers were at their mercy and often restructured their operations to meet their needs and desires.38
A parallel process developed in the growth of branded product companies like Apple, Disney, and Nike. Such firms achieved massive global sales by concentrating on product design and, above all else, marketing, making their products symbols of hipness, worldliness, modernity, and fun. Some of the big brands at one point or another did some of their own manufacturing, but typically they eventually outsourced most or all of the production of the goods they sold. Koichi Nishimura, the CEO of Solectron, in 1998 said of his customers that “The more sophisticated companies work on wealth creation and demand creation. And they let somebody else do everything in between.” Apple initially manufactured its own products, some in factories near its Silicon Valley headquarters. But in the mid-1990s it began selling and shutting down plants, contracting out almost all of its physical production. In 2016 Apple made only one major product, a high-end desktop computer, in the United States. Similarly, in the 1990s Adidas, which had made most of its footwear in factories in Germany, began getting out of the manufacturing business, closing down all of its plants except for one small operation it used as a technology center.39
Behemoth Page 32