Shopping in urban China takes place in a transformed landscape. In stark contrast to the shortages of the planned economy years, a surfeit of consumer goods now competes for the attention of shoppers who pulse through market spaces. Massive glass-and-chrome structures house the new shopping complexes and modern department stores of China’s largest cities, replacing the dour shopping spaces that characterized pre- and early-reform periods. A similar “upgrading” of the service interactions conducted within store walls has paralleled these improvements. More goods and more stores have generated more competition, and newly-acquired mantras like “The customer is never wrong,” unheard of in the days when store clerks could act as surly gatekeepers to scarce merchandise, reflect the dramatic rise in stature of that once-neglected entity in China, the consumer.

China’s bustling stores and marketplaces are not just barometers of economic growth, however. They are also spaces where social change and new inequalities are negotiated on a daily basis. This article explores how these everyday experiences of inequality are organized and understood in urban China by focusing upon one setting in which people from differing social classes may encounter one another: the sales counter. By extending Bourdieu’s notion of social distinction into a workplace centered on the sale of consumer goods and services, I demonstrate that service work can centrally involve the production and consumption of social difference, as workers respond to customer claims to class position and social status. In such cases, service work becomes distinction work.

However, the recognition and performance of social difference in service interactions is not inevitable. In China, an increasingly stratified social context and the introduction of a competitive market, where the customer is a voluntary participant in the interaction, have been central to the emergence of distinction work in urban retail settings. Within department stores, a close examination of distinction work reveals the concrete social practices through which social difference is produced in the course of everyday life. This study focuses upon the micro-dynamics involved in producing, and consuming, social distinctions in the form of service.

A comparison of two different department stores in China throws the elements of distinction work into sharp relief. Drawing upon ethnographic field research conducted in the northeastern Chinese city of Harbin, I compare service interactions at a state-owned department store, which I call the “Harbin No. X” department store, with those performed at a privately owned department store, the luxury “Sunshine” department store. At Harbin No. X, service was delivered in a working-class environment relatively free (for urbanites) from class judgments and evaluations. Because service work at the state-owned department store was originally organized under the conditions of a centralized, planned economy and involved interactions between relative social equals, service work at the store was not organized to produce class distinctions. It was not, in this sense, distinction work. By contrast, at Sunshine, service work and customer–worker interactions centrally involved the recognition of customers’ class positions, primarily signaled through salesclerks’ deferential behavior and closely enforced by store management. The contrast between the two settings highlights the emergence of a new, class-stratified culture of entitlements in contemporary urban China and reveals the practical forms that inequality takes in the realms of work, consumption, and daily social interaction.

Service work as “distinction work”

Many scholars portray class as practice (e.g., Bourdieu 1998), arguing that social interactions are a key site where class operates in everyday life (Bettie 2003; Lamont and Fournier 1992). West and Fenstermaker (1995) characterize class as a “situated accomplishment” realized through concrete social interactions that naturalize, normalize, and legitimate structures of inequality. Interactions serve as mortar and bricks for relations of inequality, structures that “are not automatic but must be constantly reproduced in practice” (Bettie 2003, p. 55). Interactions also become a site for the cultivation and expression of cultural and social sensibilities that make certain groups feel more entitled to certain social goods. The sense of entitlement that people carry with them into social interactions in different institutional settings becomes a practical expression of social hierarchy and social location (Lareau 2003; Williams 2006).

Consumption has long been identified as a realm for performances of class (Bourdieu 1984; Holt 1997; Veblen 1899[1934]) and shopping as an avenue through which consumers, consciously or not, enact their class (or other) identities (Halter 2000; Zukin 2004). In China, recent studies of elites and relatively privileged middle-classes suggest that consumption and lifestyle have become important markers of class-affiliation and identity (Tomba 2004, 2005). Among retail settings, department stores have long been associated with notions of middle- and upper-class respectability and have acted as key sites for the production of class distinction through consumption practices in the West (Benson 1986; Miller 1981). In North America, American department stores were pioneers in cultivating customer expectations of deferential service that were tied to “new standards of gentility” sought by the middle and upper classes (Benson 1986, p. 134). In China, the establishment of Western-style department stores in the early 1900s was tied to industrialization and the emergence of new urban middle- and upper-classes (MacPherson 1998). With the rise of the Chinese Communist Party to power in 1949, however, stores were gradually purged of their bourgeois associations. But economic reforms initiated in 1978 have brought the state’s control over retailing to a gradual end. Aging state-owned department stores have been joined by sleek new private and even foreign-invested department stores that target China’s new economic elites with luxury goods and claims of solicitous service.

Recent ethnographic studies of service interactions in US and Chinese settings (Otis 2007; Sherman 2006; Williams 2006) illustrate how social interactions provide the stage on which performances of social class and recognition of class entitlements are enacted. Part of what service workers in high-end service settings do, Rachel Sherman (2005, 2006) demonstrates, is to recognize high-class consumers as entitled to luxury service, a marker of class privilege.

Sherman suggests that a key way in which this happens is through what she dubs the “deference imperative” Sherman (2006, p. 45). Erving Goffman (1967) argued that expressions of asymmetrical deference – where one side defers to the other, and deference is not reciprocated – are “status rituals” that recognize unequal social positions. Asymmetric deference can be seen as a way to convert economic and cultural capital (class position) into “a capital of recognition” (Bourdieu 1998, p. 102) though which dominant social groups both portray and receive recognition of their way of life as worthy of esteem (Bourdieu 1990, p. 135). Service interactions that involve acts of deference, then, become practical enactments of relative social locations, a “doing” of social difference.

For Sherman, a central aspect of luxury service work is that “recognition” of guests involves the subordination of workers’ own selves. As a result, Sherman’s research focuses upon issues of worker consent and normalization of inequality in contexts in which workers willingly allow the extraction of their emotional and physical labor for guests. Eileen Otis’s (2007) study of luxury hotels in China reveals similar patterns of subordination and worker strategies to reinterpret, or invert, the guest–worker hierarchy that produce worker consent.

But whereas Sherman focuses on the inter-subjective elements of luxury service and the work of recognition as they relate to the worker’s “self,” here I concentrate on the performances of social distinctions and of social distance. In particular, I seek to pinpoint the production of difference and hierarchy by contrasting deferential, “entitlement” producing service interactions with more mutual, relatively deference-free ones. Rather than focus upon worker compliance and consent – which in my research sites was secured by more despotic means than Sherman or even Otis identify (Hanser 2007) – the goal here is to examine the practical and symbolic content of service interactions themselves.

I label interactive service work that is organized to produce class distinctions distinction work, and I demonstrate that demands for deference can be acts of power that both enact and constitute relations of inequality, not simply reflecting symbolic capital and relative class positions but also reproducing structures of inequality (Rollins 1985). While the production of class difference is also intricately linked to other axes of social difference in China, especially gender and generation (Hanser 2005, 2006), this article focuses upon customer–salesclerk interactions and the micro-politics of class, where clerk expressions of deference serve as a key means for recognizing and marking class privilege and social entitlements in urban China.

Research methods and field sites: Retail in China

Data for this study were gathered in China between 2001–2002, when I conducted participant observation in two large department stores in the northeastern city of Harbin. In each site I spent about two-and-a-half months working 7-h days, 6 days a week. I also spent lengths of time observing in a number of other stores, markets, and service work settings elsewhere in the city. I supplemented ethnographic work with interviews of workers, store managers, merchandise suppliers, and other industry experts and conducted archival research on institutional changes to China’s retail sector since the introduction of economic reforms in 1979.

Harbin represents an example of China’s large, urban settings in the throes of economic and social change. A city of roughly 3 million and the capital of Heilongjiang Province in China’s far northeast, Harbin over the past 20 or so years has witnessed innumerable changes. A center for largely Russian and Eastern European émigrés in China in the early 1900s (Wolff 1999), Harbin has since become an unequivocally “Chinese” city (Carter 2002) and a center of heavy industrial production and state planning in the early Mao era. With the introduction of economic reforms by the Chinese party-state in 1979, Harbin like other Chinese cities witnessed the rise of private business, the decline of state industry, and an increasingly visible gap between the material circumstances of the city’s richest and poorest residents. As part of China’s northern “rust belt,” reform-era policies have resulted in high regional levels of unemployment (Lee 2000, 2007). In this context, Harbin’s increasingly stratified retail sector serves as a barometer of these broader changes.

Changes to retailing and the service sector more generally have been enormous. A sector characterized by scarcity and dominated by state enterprises has become dramatically competitive and dominated by private business. For example, between 1980 and 1995, state-owned retailers saw their share of total retail sales drop from over 50% to less than 30% (Wang and Jones 2001); their sales officially stood at just 13% of the total in 2003.Footnote 1 In Harbin, between 1993 and 1995, the city center’s four large state-run department stores were joined by at least six competitors (representing state, private, and joint-venture investments), and another three department stores had opened by 1997. As in other Chinese cities, department stores saw profit margins shrink from 25–30% to a scant 3–5% (Wang and Jones 2001, p. 32).

Many new department stores target China’s newly rich by offering not only luxury merchandise but also luxury service; these stores often sit side-by-side with older, state-run operations where dramatically different patterns of service interaction unfold. My first site was one such aging retailer, a state-owned department store I call the Harbin No. “X” Department StoreFootnote 2 and one of Harbin’s oldest and largest department stores, employing almost 3,000 people. Despite numerous organizational reforms, this store retained many aspects of the traditional work unit, providing secure employment and numerous benefits to its workers. From the time it was nationalized in the late 1940s, the store symbolized the bounty of state socialism and enjoyed a privileged position in the state’s centrally controlled distribution system. Working-class salesclerks dealt with mostly blue-collar shoppers. Access to Harbin No. X was gained though social connections to a manager, who communicated my research intentions to upper levels of store management. Here I sold down coats in the women’s department.

The second setting was a high-end, privately owned department store – the “Sunshine” Department Store – a glistening structure located in Harbin’s downtown where both the rich and style-conscious liked to shop. This luxury department store first opened in the early 1990s, employed a staff of over 1,000 and offered six floors of expensive merchandise to Harbin shoppers. The store was run by a private mainland Chinese business group and was generally acknowledged by shoppers and retail industry specialists alike as the city’s most exclusive, and probably most successful, department store. Entry into Sunshine was arranged through a cashmere sweater company that supplied merchandise to both stores. Here I worked as a uniformed salesclerk in a cashmere sweater boutique. Although I never really “blended in” in either of these two sites, in each setting my presence achieved a kind of normalcy as I became another fixture of the environment, and my foreignness meant that my research was never covert, that I could take field notes openly and raise all sorts of issues and questions with my informants.

Although the two department stores ostensibly operated on the same organizational model with regards to merchandising and commission-based pay schemes, the structure of work was different in key ways (Hanser 2007). Sunshine, for example, operated with a flexible, despotic work regime in contrast to the job security and worker autonomy found at Harbin No. X. In contrast to Harbin No. X’s stable workforce made up of largely middle-aged women, Sunshine’s more deferential and “distinctive” workforce of young women under 30 experienced high levels of turnover. These factors suggest that the two organizations and their managers were differently invested in the production of distinction and social difference on the sales floor – an issue I return to in the conclusion – but they tell us nothing about the actual content of those interactions and the texture of inequality in urban China today.

“Serving the people” in a state-owned department store

In the fall of 2001, I was assigned work at the down coat counter in the women’s department of the Harbin No. X Department Store. This state-owned department store had retained many aspects of a state-socialist workplace and an urban Chinese work unit. Particularly notable, however, was the store’s proletarian quality of service that extracted little deferential behavior from salesclerks. Managers, focused more on the bureaucratic hierarchy of the state than a stratifying consumer market, intervened little in service interactions and did not structure sales work around the production and marking of social distinctions. As a result, workers and customers, drawn from similar socioeconomic strata, could view each other as roughly social equals, and store workers even established a hierarchy of knowledge and expertise from which they actively disciplined – and were rarely disciplined by – customers. Salesclerk deference was not an expected part of the service interaction and did not serve as a means for expressing the class distinctions that would be such an important part of the service interaction at the luxury department store.

A socialist (service) organization

As both a physical and social space, the state-owned department store was characterized by a distinctly proletarian, even revolution-era, feel. The store’s material form was vast, covering an entire block in the heart of the city, a massive ring of eight towering floors of cement. On one inner wall a colossal mural of a hard-hatted worker reminded store employees to be vigilant against fire, and when the store opened its doors every morning, customers and workers alike were broadcast a highly-operatic version of the store song, steeped in revolutionary fervor (and sung in CCTV military chorus style), entitled “Soar, Harbin No. X!”

Great ambitions to expand, like the surging ocean!

The spirit to engage struggle, like the lofty mountain!

Genuine smiles win the customers’ love!

Exquisite merchandise is exchanged for a golden reputation!

Glorious Harbin No. X! Glorious Harbin No. X!....Go create a splendid new era!

The song goes on to tell us that civilized business practices are the source of store unity, the workers’ warmth and charisma moving to see. Although composed for the opening of the store’s new building in the early 1990s, the song was infused with the language of the Chinese revolution and extolled the contributions of China’s workers to a glorious future. The store itself was a prize of the revolution, originally a foreign-owned and managed store that was nationalized even before the formal founding of the People’s Republic in 1949.

As an organization, the Harbin No. X of 2001 retained important institutional orientations rooted in China’s pre-reform, planned economy era, despite numerous organizational changes and reforms. Most significantly, nominal ownership changes had left the enterprise under the thumb of the municipal commerce bureau. Prior to retail sector reforms, Harbin No. X’s position in the state hierarchy determined its access to goods and other resources, and its position in the retail “market” was more or less bestowed by government agencies. In the late 1990s, store ownership was converted to stock, yet the government had maintained a formal 33% stake in the enterprise and controlled roughly another 33% informally (the remaining shares were owned by staff and management), giving municipal commerce officials de facto control over any major organizational decision. The store owed loans exceeding 500 million yuan (over US$60 million) to the city government (Ji 2002), placing the store’s financial fate in the hands of city officials. And Harbin No. X’s general manager was not an internal selection but a municipal commercial authority appointee who was expected to approve every managerial decision of any importance. As a result, the managers below him, many of whom had worked their way up the store hierarchy from salesclerk positions, expressed little sense of control over the major business affairs of the store. As one manager commented, “We basically have to respond to the demands of those above us,” meaning city officials.Footnote 3

Because economic and enterprise reforms had not substantially altered the store’s orientation to the bureaucratic hierarchy of the state, managers did not cultivate either the physical space of the store or the workforce itself to convey a class-specific market position. Instead, the store was an open, permeable, and even democratic space, in the sense that the world “outside” would frequently spill in and was subject to little, if any, filtering: people visited the store not only to shop, but also for a bit of exercise, to socialize, or even to engage in a little social activism. A senior citizen’s group called “Setting Sun Red” [xiyanghong] would parade through the store on weekends carrying signs warning customers of pickpockets, while they chanted “Take care, take care!” Rural families in town for a day of shopping and working-class urbanites in blue or white work smocks enhanced the store’s populist and proletarian feel. Many people carried on other, non-store business activities inside the store as well – fruit sellers, telephone card peddlers, even a roving tailor and a friendly pornography salesman. Workers even extended their hospitality to animals, which customers might deposit atop sales counters, their pets’ antics a form of public entertainment.

Workers themselves were not so distinct from “the masses.” The work force, though largely women, was sprinkled with men across store departments. Salesclerks ranged in age from their mid-20s to their late 40s, and retired workers would regularly stop by for chats. Lax policies governing uniforms meant that through the winter months, workers were only required to don a blue blazer, allowing them to adapt their work clothing to their individual tastes and comfort. Permed and dyed hair sat piled atop heads, and women salesclerks skittered about in impossibly tall high-heeled shoes and stylish, tight pants. This diversity of appearance allowed the streets of Harbin to be reproduced within the store walls. As one store manager pointed out to me, the store was very dazhonghua, or “massified.” In this sense, a state-socialist institution remained an important social space for the city’s middle-aged and older working-class people.

Given that store management directed much of its energies towards the local state, managers expressed little interest in monitoring or scripting day-to-day sales interactions. From the day I arrived at Harbin No. X, I was struck by how little supervision workers were subjected to, relative to my expectations. Management was a rare presence on the floor, and when a manager did pass he or she was likely to focus upon superficial and incidental problems like a forgotten bucket of soapy water or a sloppy work area. Workers could run errands, visit friends at other counters, or even take a nap (if well concealed) without ever being discovered. During trips to the toilet, workers claimed they needed “a partner” along to chat with and on the way back to their sales areas would frequently take “the scenic route.” Some of this autonomy stemmed from managers’ sense that, in practice, they were powerless to fire a worker, even for poor service or work performance (unless that person was caught embezzling funds or committing some other dramatic crime). Worker autonomy was likely also a consequence of a long-standing, pre-reform organizational orientation that cast customer-worker interactions as a peripheral concern.Footnote 4

This autonomy did not mean, however, that the behavioral patterns of the planned economy era persisted in the form of surly, inattentive salesclerks abusive to customers, in large part because the incentive structure for workers has since shifted. They can be and are disciplined with fines. Most importantly, at Harbin No. X salesclerks earned wages on a commission basis, and this produced considerable incentives to please their customers. Most of the salesclerks believed that properly handling an interaction with a customer was central to making a sale. Nevertheless, both job security and worker autonomy helped foster a mutuality and reciprocity that characterized service interactions in the state-owned department store, in stark contrast to the private, luxury department store. Work at Harbin No. X was not organized to recognize customers’ class status. Service interactions were more likely to recognize forms of parity, reciprocity, and mutuality between customer and salesclerk than to produce social distinctions.

Service encounters among equals

What might be characterized as a “socialist” organization of work (Hanser 2006, 2007), then, provided the context in which working-class social interactions could be enacted. Without management dictating the contours of the customer–clerk relationship, Harbin No. X clerks and shoppers performed one of parity. The store’s customer base fell into the low- and middle-income ranges among urbanites, meaning that customers and workers came from similar socioeconomic strata and earned similar incomes – between 400 and 1,000 yuan a month, on average – and could consume many of the same things, be it clothing, food, or entertainment. In fact, workers and customers would often make explicit the lack of social distance between them. For example, when trying to sell a man’s down coat to a young Southern man, one salesclerk urged him to purchase navy blue, explaining that it had “laoban qi” and would give him the “air of a boss.” In response, the young man joked that he was not a boss at all, only a dagongzai (slang for “male worker”). This interaction was notable for how a lack of distance resulted. In another instance, a customer complained to us about how his son wanted a new winter coat every year in order to keep up with fashion. One of the clerks diagnosed this as “guizu bing,” or “aristocracy syndrome,” to which the man immediately replied that his own economic situation was not so good. At times, workers would even point out their own wages to customers as a way of establishing a sense of empathy or rapport. One clerk joked that if there were a problem with a coat after the customer washed it, she would give the customer the money for the coat. “Out of my wages!” she exclaimed, adding that she only earned 400 yuan a month (this being her pre-commission base wage, less than her average monthly take-home pay of about 600 yuan).

Based upon the social parity that characterized both the store itself as a social space and especially the interactions between shoppers and clerks, sales staff were able to establish themselves as both experienced and expert regarding the products they sold. The two sides did not, of course, enter into these interactions as complete equals, especially as salesclerks’ salaries were closely bound to customer purchases. However, while customers expected salesclerks to respond to their requests and needs, they did not expect clerks to enact extreme deference to customer wishes. Instead, salesclerks constructed a knowledge hierarchy that elevated them above customers, and from that elevated position clerks could and often did issue judgments and pronouncements on customers’ tastes, choices, and product knowledge.

For example, a clerk chided a customer who was being fussy about a loose thread on a coat. “It is impossible not to have a loose thread or two on these coats! The coat isn’t made with a single length of thread!” Clerks could also be quite aggressive about redirecting customers’ choices (often to fall in line with what we had in stock), and customers were sometimes cowed into accepting these authoritative instructions. For example, one of my colleagues exclaimed to a customer who wanted a smaller-sized coat that if the woman wore a size smaller her bottom would be hanging out (lou) and it would be unattractive (bu hao kan). The customer relented and purchased the recommended size. In many cases, clerks would guide customers’ choice of color or style, instructing them as to which colors were appropriate or inappropriate for their size, age, or skin color. One clerk put a woman customer through an elaborate series of inspections, having her put on different coats and step back from the counter so the clerk could get a good look. The clerk then told her light brown was the right color given the woman’s age, and suggested that she wear a black sweater under the coat for the best effect. As the woman made the purchase, the clerk, a middle-aged woman herself, continued to lecture enthusiastically about how dark brown was not appropriate for middle-aged women given their age and their skin color.

Clerks would repeatedly resort to such tactics, and customers rarely dared to openly challenge clerk interpretations of consumption norms. In fact, those that did not conform to clerk standards of customer and consumer behavior were sometimes described as deviant or even “sick-in-the-head.” An amusing exchange occurred when a clerk was introducing a coat with netting sewn into the lining to an older customer. The customer, unfamiliar with this innovation to allow for better ventilation, was suspicious. The clerk forcefully scolded her. “You don’t understand!” he exclaimed with exasperation. “But won’t dust get into the coat then?” the woman worried. “How much dust can you possibly have on your body?!” the clerk cried. “The coat does not have a problem, it’s your ideas (gainian) that are the problem (you wenti),” he added. In another case, a clerk became irritated with an elderly man who was confused by the distinction between down and feathers. After the clerk attempted a detailed description of the physical distinction between down and feathers, the customer was still baffled. The clerk threw up his hands in defeat and exclaimed, exasperated, “I don’t know how else to explain it to you, unless I pull out a dictionary and show you the words!” As the man shuffled off the clerk turned to me and complained loudly, “That guy is sick-in-the-head (you bing)! He understands less Chinese than a foreigner!”

Another example was a woman clerk’s response to a customer, a tiny woman, on whom the sleeves of every coat she tried on were too long. The clerk explained that the woman had narrow shoulders and told her to consider other styles, but the woman insisted she liked that particular style. After the woman had wandered off, the clerk turned to me and complained that the woman’s arms were “bu biaozhun” (non-standard) and suggested the woman had a problem acknowledging this fact. What the clerks were engaged in, then, was an enforcement of “universal” standards against which customers’ behaviors (and bodies) were judged, standards that were not meant to reflect upon the worker’s or the customer’s location in the socioeconomic hierarchy. Failure to conform was a mark of stupidity or even “old ideas,” but not of relative class location per se. The irony, of course, was that these “universal” standards had actually become class-specific – that is, specific to the middle- and low-income urbanites who frequented, and were employed in, the store.

The mutuality of disrespect

If, as I have argued earlier in this article, service workers’ expressions of deference are key to signaling class difference and recognizing customer claims to class privilege, then service interactions at Harbin No. X were especially notable for the lack of deference expressed by either side. In fact, the level playing ground upon which clerks and customers engaged one another was exemplified by the degree of give-and-take between actors and the doses of disrespect – mutual disrespect – they served up to one another. On the part of the clerk, disrespect at times took the form of gruff dismissal of fussy or troublesome customers. One time some customers responded with suspicion to a clerk’s claim about merchandise quality. The clerk rejoined, “Go have a look around! (Liu yi quar ba!) I tell you there is no need to doubt the quality and you don’t believe me. Go look elsewhere and if you find better, then buy that.”

Customers, of course, dished out such disrespect as well as they received it. The most common way of doing this was to simply ignore a clerk who was rattling off a sales pitch, but customers might do other things, like ask to see an item and then leave while the clerk went to fetch it. Sometimes they would even have us write out a receipt and package up the item while they went to the cashier to pay, only to never return. Disrespect was sometimes expressed as distrust in the clerk’s honesty. One customer cut a sale pitch short by exclaiming “La dao ba! Give it a rest!” The clerk responded evenly, “Do you mean that I am incorrect, or that you think that what I said is incorrect?”

Ironically, it was clear to me that both sides felt they should be treated with respect. For example, in response to a customer who had just described our merchandise as “ke’chen,” local dialect for “extremely ugly,” a coworker turned to me and said: “With the steady improvement of people’s quality (suzhi), this kind of language will disappear.” He continued: “Our treatment of customers is already very friendly. There is no need for customers to use such language with us.” At the same time, customers also expected a degree of respect from clerks (and likewise might attribute bad service to the poor “quality” of the individual). For example, when a clerk muttered something under her breath in response to a particularly annoying customer, he wheeled around and confronted her. “What was that?” She repeated what she had said: “I said, I wasn’t able to explain clearly to you.” The man, satisfied that he had not been insulted, moved off, at which point the clerk turned to me and said with irritation that the man was a “waihang,” or “non-expert.”

My claim that salesclerks in the state-owned department store did not engage in judgments of class and social position is subject to an important caveat, however. For while urbanites could expect to shop in the store free of status anxieties, clerks’ treatment of rural people was colored by the pronounced urban–rural status distinctions that have characterized state-socialist China (Solinger 1999). It was not uncommon for people who were clearly from non-urban backgrounds, especially migrant workers, to be summarily categorized as outside “the norm.” For example, the relation between a middle-aged man accompanied by an attractive young woman, both of whom were definitively labeled “peasants” by my colleagues, was debated intensely after they left. “Certainly not father–daughter nor husband–wife relation, and not relatives,” one clerk decided. “Er hun guanxi,” said another, “‘second’ marriage [meaning extramarital] relations.” “Whatever their relationship,” replied the first clerk, “it is not normal (bu zhengchang), and they’ve just come to Harbin to piss all of us off (chou de’se, local dialect).” In another case, a group of deeply tanned young men, probably casual laborers from the countryside, were quickly labeled deviant. “They aren’t like us normal people,” a clerk explained to me. “We feel uncomfortable just looking at them.” These “abnormal” rural shoppers, marginalized as they are in China’s urban centers, were neither directly confronted with these judgments nor did they challenge the reserved manner in which they were served.

As the presence of rural customers shows, the egalitarian nature of customer–clerk interactions in the state-owned department store was to some degree class-bounded. The department store itself made few efforts to impose a set of class-recognizing relations upon its workers and customers. As a result, clerks did not dole out deference to most customers because deference recognizes a class difference, and in most cases clerks and customers occupied similar socioeconomic positions in Harbin society. But when clerks were confronted with customers with whom there was a gap-rural people-the mutuality of disrespect was disrupted. As urbanites, clerks felt distinct from and distinctly elevated above these rural shoppers, and the mutuality of disrespect was converted into one-sided disdain and recognition of social difference.

Exclusive service in a luxury department store

When I took up a post selling cashmere sweaters in the upscale, privately owned Sunshine Department Store, I found myself in a dramatically different setting. The highly professional, well-trained management in this luxury store saw themselves as situated at the top of a competitive market hierarchy in which they must distinguish themselves to attract customers, and as a result they instituted strict service routines. These routines produced service interactions that recognized customers’ claims to elevated class positions. The key means for recognition was through salesclerks’ expressions of deference, and consequently deference became an integral part of what workers were expected to produce and customers expected to consume.

However, because the store was ostensibly open to anybody, and because salesclerks wished to recognize only those customers truly deserving of their distinction work, the service interaction was also opened up as a space for conflict. Indeed, one of my first and longest-lasting impressions of this store were the angry eruptions of customer dissatisfaction with salesclerk behavior, despite the more standardized and tightly managed provision of service. On a single day, I witnessed three such blowups in rapid succession. Angry customers would scream and yell, swear, call clerks names, and threaten to report them for their insolence. Below, I suggest that these outbursts were the unintended consequence of distinction work. Complex discounting practices and workers’ attempts to limit their work required workers to make evaluations of customers’ claims to class distinction. Most importantly, these angry outbursts dramatically highlight the embedded role of class recognition and distinctions in the store’s service interactions.

A space of distinction

The Sunshine Department Store was born of China’s economic reforms, opening its doors in the early 1990s. The luxury department store explicitly catered to Harbin’s wealthiest consumers, with one analysis of Harbin’s retail sector pronouncing that the store’s “most obvious characteristic is its clear market position...[its] target market being the top 3% income bracket” (Wang, no date, p. 11). The store did not compete on the basis of price or even on the exclusivity of its brand-name merchandise. Above all, Sunshine offered customers an environment of luxury and an atmosphere of attentive, deferential service as the mark of social distinction.

In sharp contrast to the management at Harbin No. X, Sunshine’s managers were acutely attuned to Harbin’s evolving social hierarchy and the retailers that increasingly mirrored it. Upon our very first meeting, a manager at the Sunshine Department Store assured me that I would find his store much better than Harbin No. X, suggesting that the salespeople there delivered poor service. The distinction between Sunshine and the state-owned store was in part necessary because the Sunshine Department Store was recognized by managers, workers, and customers alike through what it clearly was not. For whereas Harbin No. X was regarded as a relic from the socialist past, with an aging, undisciplined workforce and working-class clientele, Sunshine identified itself with a new China populated not by the “masses” but by newly rich and upwardly mobile groups served by an army of obedient, attractive young women.

Store managers engineered Sunshine to be a physical space that conveyed that distinction through many symbols of wealth and privilege. The store itself was a glistening structure located in the city’s center. Sunshine dominated a newly evolving shopping plaza where both the rich and style-conscious liked to shop. Inside, the floors were laid with colorful marble, and mirrored pillars punctuated the sales floors. Like the state-owned store, this store was also constructed in a ring, but here the center formed a closed, atrium space into which customers could gaze as they glided upwards and downwards on paired escalators. Sales areas were designed in “boutique” style, and many offered seats and even an occasional table at which weary shoppers might rest while clerks tended to their wants. In the morning, store workers stood at military-style attention along the aisles as customers entered the store to Bizet’s theme from “Carmen.”

This aura of luxury and privilege was borne upon the bodies of the shoppers who moved about the store. Many customers, women and men, arrived clad in stylish fur or fur-trimmed winter coats. During the winter, the season when I worked at the store, many customers wore clothing that made clear they had arrived either by private car or taxi, such as high-heeled shoes and boots, and light-weight or light-colored clothing, none of which were any match for Harbin’s sub-zero temperatures, brown, slushy winter streets and crowded, grimy public buses.

Managers ensured that the store’s workplace disciplinary regime produced, often through the bodies of its workers, the markers of class privilege. The department store closely monitored staff appearance, accentuating the distinctions between customer and worker to create a formal environment that clearly delineated staff duties and customer privileges and emphasized gaps between the two. As Paules (1991) notes, uniforms eliminate clues to worker’s non-work status, facilitating the extraction of deference. At Sunshine, uniforms were just that: highly uniform and strictly regulated. Security officers who policed the store wore army-green uniforms, and store management was clad in professional blue suits and gray button-down shirts. Floor sweepers wore crisp white overcoats, a departure from the proletarian blue work smocks that similar workers wore in more modest retail settings.

Given the centrality of the service interaction to successful distinction-making, salesclerks’ appearance was especially important to the creation of a high-class retail space. As a result, they were, of all workers, subject to the most intense regulation. Salesclerks were selected from a large pool of candidates and had to meet strict requirements in terms of age (25 or younger at time of hire), education (high school or higher) and appearance. They wore store-issued, regulation trousers, shirts, vests, ties and blazers, and shoes were restricted to black leather shoes with flat, black soles. At times I felt as though store uniforms made sales staff indistinguishable and almost invisible to customers (until they deemed us useful or necessary), a point exemplified by customers’ reactions to me. Despite the anomalous combination of my European features and my store uniform, passing shoppers would regularly ignore me even after I had spoken a few polite words to them. In numerous cases, customers would be well into an interaction with me before they noticed that I was not Chinese. As a number of scholars have argued, “invisibility” is a sign of the devalued labor and low status of service workers (Rivas 2003; Rollins 1996).

In fact, almost every aspect of clerk appearance was subject to some kind of regulation-the length of your fingernails, the number of rings on your fingers or necklaces around your neck, the length, color, and styling of your hair (hair longer than shoulder length must be pulled up above the shirt collar), even the color and type of hair clip you used. All this assumed that you had already met the height requirement for the job of 1.6 m. Combined with the fact that the store only allowed young women to continue working as clerks up to the age of 30, management had created a workforce that appeared neat, clean, and highly obedient. All this distinguished Sunshine – for the benefit of its clientele – from other retail outlets in the city, especially the disheveled Harbin No. X.

Many restrictions served as a means to extract what Judith Rollins (1985, p. 171), borrowing from Goffman (1967), calls “spatial deference” directed towards the store’s exclusive clientele. Clerks were expected to strike certain poses while waiting for customers, and their movement about the sales floor was theoretically limited to trips to storage areas and the bathrooms. Salesclerks were also prohibited from riding store escalators in their uniforms (though management and security were free to do so), so workers on the fourth, fifth and sixth floors would grumblingly trudge up dingy, and unheated, stairwells. Management would monitor compliance with all of these rules by circulating the sales floors equipped with walkie-talkies.

These forms of control were not absolute, and over time, I learned that clerks had numerous strategies to circumvent restrictions on their movements and behavior (for US and Taiwan parallels, see Benson 1986 and Lan 2003). They would wear long coats over their uniforms (to conceal their worker status) and then take the escalators anyway. They would don a sweater or jacket over their uniforms so that they could unobtrusively sit down for a rest, and they were quick to warn one another of an approaching manager. When a coworker had slipped out on a personal errand, we would tell management that she had “gone to the toilet” and would be “right back.” Even so, these rules and restrictions set the stage for the unequal service exchanges carried out with customers.

The customer is always right

The key opportunity for recognizing customers’ class privilege, however, was through the service interaction, interactions exemplified by the lack of reciprocity between customer and clerk. Store management went to great lengths to set and clarify for Sunshine salesclerks the acceptable parameters for such interactions. In training sessions, future clerks were exhorted to take every precaution possible to avoid angering or offending the customer. Directions should be given with an open hand, not a pointed finger; when taking a drink of water, workers were instructed to turn their back to customers, lest the customer take offense. Crossing one’s arms was dismissed as looking uncooperative, and at one point, the instructor dove her hands into her pockets, and asked us, “What kind of feeling does this posture give you?” She remarked that customers, seeing a clerk with her hands stuffed in her pockets, would be sure to think, “What on earth! What [lack of] class (cengci)!” One’s body was not to smell of unpleasant things (smoke, alcohol, onions, garlic), and picking one’s nose or ears, sneezing, and even yawning were all dismissed as “dirty.” “These things might make you comfortable, but they are unpleasant for those around you. Sales staff should not create feelings of repulsion in customers (yinqi guke de fangan),” one instructor explained.Footnote 5

Workers were explicitly trained that deference and courtesy moved in one direction in the salesclerk–customer relationship. One instructor bluntly informed students that, “salesclerks do not have reciprocity” (chuefa duidengxing) and characterized this as one of the key principles of sales work. “Remember, the customer is always right (guke yongyuan shi dui de)...this is an important part of the principles of the relationship between customers and salespeople (guke he yingyeyuan zhijian de faze). Of course there are always some unreasonable customers, but the only way for you to deal with them is with patience...everyone has an individual personality (gexing), but you are not to express it in this context.”

This golden rule of salesclerking – never assert yourself with a customer – was reinforced with the iron reality that clerks who did so might lose their jobs. One of the training sessions I attended began by progressing through a series of parables about how workers’ mistakes had caused them to lose their jobs (xiagang). For a woman high-school graduate in Harbin, these were good jobs to lose, as commissions on luxury products offered workers relatively high pay; my coworkers in cashmere regularly earned between 1,200 and 1,400 yuan per month. Managers and trainers made workers keenly aware of the consequences of openly defying a customer, and as a result clerks felt unable to react to mistreatment or verbal abuse. As one coworker told me after she had endured a customer’s tirade in silence, “In ordinary circumstances, I definitely would have lost my temper with him, but in this job you can’t. If they yell at you, all you can do is not respond, because if you start to get angry with them then the fault becomes yours.” Indeed, while salesclerks engaged in subtle forms of resistance against store discipline, some mentioned above, this rarely extended to interactions with customers.

Store-enforced work routines meant to ensure that salesclerks recognized customer claims to deferential treatment were buttressed by “real world” differences between clerks and customers in terms of income, lifestyle, and consumption patterns. As a general rule, clerks could not afford to buy the merchandise they were selling; this was especially true of more expensive merchandise like fur coats, cashmere sweaters, jewelry, and exclusive brand clothing. Even though clerks earned decent wages compared to the average city resident, they still inhabited a lower socioeconomic class than the customers whose needs they tended. Even so, management clearly felt disciplinary measures, such as fines and especially the possibility of being fired, were necessary to ensure that the proper class roles were adhered to.

Shoppers at Sunshine would freely express their expectations of deferential service. Customers exerted a considerable degree of authority over clerks, and some customers would brusquely issue commands and requests.Footnote 6 One customer chastised my coworkers by noting “We [as in ‘Chinese society’] still haven’t realized ‘the customer is God’ (guke shi shangdi).” Another customer, insulted by a clerk’s sullenness at his returning of some very expensive merchandise (which affected her commission) said to me, “She has a bad attitude, I should report her. It should be warm service (reqing fuwu), service with a smile...shouldn’t it?”

A more subtle reflection of customer authority lay in the supplicating manner in which salesclerks offered suggestions and advice to many customers. As one salesclerk explained to me, offering advice to a customer was a delicate matter that involved reacting to a customer’s choice and telling them what they wanted to hear. “So if they ask you, ‘Do I look stylish in this?” you tell them, ‘Yes, you really look stylish in that,’ and they go away feeling very good about their purchase.” Another clerk explained to me that if you explain too much to customers, some get irritated and leave. “Many of them are kind of like experts themselves, knowing what is and is not good cashmere” through their experience buying and wearing it. Sometimes customers would even make a show of their expertise in cashmere, announcing to friends as they entered the sales area, “You don’t have to worry about [the quality of] this brand, this is ‘Goat King.’ It’s a famous brand (mingpai).”

A very different sort of knowledge hierarchy was in operation, then, than the one constructed and maintained by workers in the state-owned department store. In the luxury store, clerks were expected to be knowledgeable, but not too expert, about merchandise, lest they challenge a customer’s own expertise in the consumption of luxury goods. In some cases, customers would even challenge a clerk’s knowledge, especially with regards to wool quality, brand reputation, and what were and were not the most current styles. In fact, much of the advice clerks dispensed was not about their own knowledge of consumption and the merchandise per se, but rather knowledge about how other people consumed such things. In this way, salesclerks produced a form of deferential service that affirmed customer expectations for class recognition.

The contradictions of distinction work

All this created a highly regulated, but also highly stratified, space that did not make all customers feel equally welcome. This was one of the key contradictions of the luxury department store: the space was semipublic and open to all shoppers, yet it clearly catered to, and wished to focus its workers’ energies upon, the wealthy. Because the store had no way to separate the true store patron from the merely curious (or envious) shopper, this task fell to the salesclerks who had the most direct contract with customers. Given the expensive nature of the merchandise and the likelihood that many customers were “just looking,” workers would try to judge (and then ration) how much effort they would accord each customer (for parallels, see Donovan 1930). Efforts aimed at the “conservation of energy” were greatly exacerbated by store pricing and discount policies, which allowed discounts to be somewhat negotiable. Because clerks were discouraged from offering too many discounts, and also because their pay was largely made up of sales commissions, they engaged in a slippery game of judging a customer’s ability and willingness to pay. The need to evaluate customers’ class status in a luxury setting served as a source of anxiety for shoppers and produced conflicts over whether a class gap existed between “real” customers and any given individual.

The sheer cost of the merchandise was sufficient to make some customers sweat. The cheapest sweater in our sales area, before discounting, was a very simple cashmere crewneck costing 980 yuan; the most expensive item, a cashmere dress-and-jacket set, cost over 4,000 yuan (compared to an average monthly wage in Harbin of 580 yuan; Statistical Yearbook of Harbin 2001). As a result, from the moment I began working at the store I found myself sizing up almost every customer who entered our sales area, trying to gauge whether they were capable of purchasing one of our cashmere sweaters. It was clear that my coworkers were engaged in a similar calculus of evaluating customer class positions. For instance, customers who purchased our brand of sweater from sale carts on the store’s first floor were coldly received when they approached the clerk in the fourth floor boutique for gift boxes. People buying sale items were clearly lower on the customer totem pole, and salesclerks would sometimes even refuse to make eye contact with “tejia” or “sale price” customers. Another time, a salesclerk managed to convince a customer who had just bought another brand of cashmere sweater to purchase one of ours instead. The customer, claiming that she needed to return the other sweater first, left without a purchase. When a second salesclerk expressed dismay that the woman had not purchased a sweater, the first clerk murmured something about how the woman clearly was not so rich that she could buy a second sweater without first returning the one she had already bought. Customers who exclaimed with shock at the prices when they entered the sales area were especially safe to ignore.

Store management exacerbated the situation by institutionalizing differential treatment meant to reward especially moneyed customers, special treatment that some customers wore like a badge of honor. Sunshine granted “VIP cards” or “gold cards” to customers who spent over 5,000 yuan in a single day or over 10,000 yuan in the course of a year. These customers not only got small rebates on their annual purchases, they were also granted a 10% discount on all their purchases. Customers who were especially well connected could get even greater discounts through store management (something that was a rare practice in the state-owned department store), and “return customers” (huitou ke) expected all sorts of special treatment (such as extra discounts and more solicitous service) that less moneyed, less experienced customers never dared request. Less experienced customers felt less certain of their status in such an expensive shopping environment. On two occasions customers apologized for being underdressed, one woman even promising to return another day in better clothing so that she could try on a sweater. Most customers were fully conscious that they were subjected to class evaluations upon entering Sunshine.

Of course, salesclerks themselves had no intention of treating all customers equally, because customers who were not able or likely to make a purchase were, quite frankly, not worth the effort. Given that department stores are semipublic spaces open to browsers as much as to serious shoppers, workers would ration their energies and their deference, devoting considerably more time and effort to customers they thought would actually make a purchase. The rationing of effort was, in part, tied to the piece-rate nature of salesclerk’s pay. This was illustrated by the fact that clerks would disregard customers who seemed clearly unable to purchase a cashmere sweater, judgments usually based on clothing or customer remarks.

The process of evaluating a customer and gauging how much energy to invest in him or her was further complicated by the store’s discounting policies and the dynamics of granting discounts, over which salesclerks had some discretionary power. For example, a customer with a VIP card could legitimately be granted a 10% discount, but sometimes the manufacturer was willing to shave another 5% off if it meant the difference between making and not making a sale. Salesclerks also had techniques for granting regular, non-card carrying customers 10% discounts by using borrowed or fake gold card numbers. This was a strategy that was only tacitly approved by the store, and so store clerks had to limit the use of such practices to cases where it would make or break a sale. Clerks could offer no discount or a 5% discount if the customer could be convinced that the clerk did not have authority to grant something larger. The delicate balance between granting and refusing discounts obeyed another logic as well. Each cashmere company wanted to conceal discounting practices from its competitors, both to improve sales and to avoid price wars. It was also necessary to conceal discounts from scouts sent out by other department stores where the manufacturer operated sales areas, because if those stores discovered higher discounts being granted elsewhere they would demand the same discounts be granted by the manufacturer at their store as well. Granting discounts, then, was fraught with difficulties and required considerable negotiating skills of both clerk and customer. Most importantly, it demanded that clerks make a judgment about a customer’s ability and willingness to make a purchase before entering into lengthy, often tedious, price negotiations.

The negotiation of a discount on a cashmere sweater in the luxury department store always began the same way: the customer would ask “Da bu da zhe (Is there a discount)?” The clerk would then size up the customer and either say, “No discount” or “10% off with a VIP card.” If the customer remained interested, then the clerk would continue sizing up the customer – how small a discount would be enough? For example, one woman without a VIP card asked if she could get the 10% discount if she bought three sweaters; when the clerk cautiously agreed, the woman thought she had received a special deal, even though it was our regular practice to grant 10% off to customers with and without cards, regardless of the size of the purchase; a three-sweater purchase regularly garnered a 15% discount. When I accidentally revealed this to a customer who did have a VIP card, she angrily demanded “But you’ll give 10% even without the card...what good is the card?” The negotiating of discounts was by far the most nerve-wracking part of the job for me, and I found myself constantly concealing and revealing discounts when I was not supposed to.

In the end, it was the customers who looked or acted wealthy, and who could convince the salesclerk that they would only buy with a discount, who would be granted the most generous price reductions. For example, a woman entered our sales area and asked if there were any discount. The clerk, wanting to test the customer’s resolve, hesitated a moment and then said, “If you like the style, why not try it on?” The woman repeated herself somewhat sharply: “Da bu da zhe?” The clerk responded hurriedly, “10% off with a VIP card.” The customer was savvy. “I haven’t brought my gold card with me,” she said, and continued to pressure for a discount. The smartly dressed customer said that she wanted to buy two sweaters, and that she was leaving the city in just 2 h. The clerk then offered the woman a 20% discount, indicating this not verbally but with a calculator.

Given the tension clerks experienced between wooing those shoppers able, willing and likely to make a purchase and ignoring or at least neglecting browsers and pretenders, customers at times would actively try to lay justified claims to the level of clerk attention and deference they expected. Some customers would boldly announce to us that they had purchased cashmere sweaters before or they would make detailed comments on the previous year’s styles. Claims of class status could be delivered icily if the customers felt the clerk had already failed to grant them due deference. A clear example was a pair of customers who wished to exchange one sweater for another. The clerk was initially quite sharp with them, refusing to look at them and demanding to see their receipt. The man then said forcefully that there was no need to be in such a hurry, let the woman first choose the item she was going to switch the sweater for. The clerk quickly altered her tone, and in the course of the interaction it became clear that this couple was, despite their casual appearance, quite wealthy. The woman spoke some English to me, and then revealed to the other clerk that she already possessed some “20 or 30” cashmere sweaters, one of which she was wearing. “I wear them for a while and then I don’t like them anymore,” she said lightly. The couple had forcefully established themselves worthy of the clerk’s attentions without resorting to open conflict.

But a clerk’s perceived misreading of a customer’s class status was not always settled so peacefully, and this was when clerks might find themselves subject to a barrage of verbal abuse. In a case I observed closely, it was clear from the customer’s behavior that he felt both wronged and disrespected. He wanted to get an exchange on a sweater he had purchased over 6 months before. The sweater had developed a hole, he claimed, and therefore should be replaced; he was from outside Harbin (a source of status anxiety in itself), and this was his second trip to try and resolve the matter. The clerk, acting on manufacturer instructions, told him she could only offer him a repair. The customer exploded.

“I paid more than a thousand yuan for this and all you will do is bu xian (re-stitch)?? Look at this, what’s the difference if you do it or I do?” He then accused the clerk of being insolent with him and told her that he had a mind to report her bad behavior. The clerk hurriedly made a phone call to the manufacturer’s representative, who arrived almost immediately. The customer continued to rant until a manufacturer sales rep arrived and led him off to the floor manager’s office. After he left, the salesclerk, whose only response to the abuse heaped upon her had been to cower in a corner of the sales area, turned and said to me, “I didn’t think I’d been short with him. I’m so angry! (Qisi wo!) You can tell how low status this job is when people treat you that way.”

That same day, another argument broke out in the neighboring underwear area. A woman customer, feeling the salesclerk had treated her rudely, began yelling angrily at the clerk. This clerk exchanged a few sharp words with the customer, and soon a manager arrived to lead the clerk away. The customer, now being tended to by a different clerk, hollered after the departing worker, “Sai lian! Sai lian!” (A local term used to scold misbehaving children). In another case, in the neighboring sales area, a woman customer erupted when the salesclerk informed her that she would not be allowed to exchange an item. The woman suggested the clerk was trying to take advantage of her (and that she was not familiar with store policy) and fetched a manager. A compromise was reached, but as the customer left she yelled, “That salesclerk’s attitude is the worst, I tell you!” On another day, a man’s angry screams came from the sleepwear department.

In an environment where clerks were constantly engaged in judgments of customers, some customers received less deferential treatment than they believed themselves entitled to claim. Usually a customer outburst, buttressed by store policies, was sufficient to restore recognition of the customer’s demands for deference. Customer outbursts were potentially destabilizing in cases where the dissatisfied customer was not a high-class customer but rather a low class one, revealing the accepted link between extraction of deference and class difference. One last conflict illustrates this. In this instance, the angry customer was a Chinese translator for a group of Russians. He had given a sales clerk money to pay for an item and then claimed she had shorted him 50 yuan on his change. The clerk was lividly angry at this accusation and the two shouted heatedly at each other. A manager came to deal with the situation and it became clear that the worker would be forced to pay for the “missing” money. I passed this clerk when I got off work, as she stood sullenly in her sales area, her face burning with anger.

This conflict is instructive because the clerk did dare to stand up to the customer, and vigorously so. The customer, a translator who worked in a subterranean wholesale market, was of undeniably low class status and worked in a retail setting of equally low status. These types of men were almost universally viewed with distaste and suspicion by Harbin residents and even by merchants in the wholesale market itself-there could be no misreading of his social standing. In such a case, then, the clerk felt she could justifiably stand up for herself – and deny the man deference.

Given distinction-producing nature of service interactions in the store, the irony is not that the clerk defended herself but rather that because of store policies she lost the dispute. Indeed, although explicit store policies appeared to make customer claims to deference reducible to purchasing power, in fact the extraction of deferential service was only sustainable when symbolic or cultural class boundaries were upheld.

Conclusion

Almost all service interactions require the service provider – the salesclerk, the waitress, the attendant, the desk clerk – to make some kind of assessment of the service recipient and his or her needs, wants, and the like. In retail settings in China, where salesclerks both work on commission and encounter very large numbers of customers in a single day, evaluations of customers and of how to best interact with them are especially important. However, as I have endeavored to show, not all service interactions are equal. New constructions of service work, and service workers, have contributed to an increasingly lopsided relationship between salesclerks and customers.

In fact, China’s context allows us to disrupt what often seems, in a Western context of service work, to be commonsensical: expressions of deference, especially from salesclerks. Through a comparison of service environments, I have argued here that service work can be organized around the production and consumption of class meanings. Distinction work, or the organization of work to produce class meanings, can also serve as a powerful public buttress broader social inequalities: she (or he) who pays is always right, and always respected. She who serves (and it is a she) may not dawdle in the bathroom, must turn away when she takes a sip of water, and must never talk back.

The institutional context is crucial – in a wider context in which wealthy elites are viewed as the legitimate and appropriate recipients of esteem, respect, and deference, organizations like department stores become invested in the production of social distinction in order to attract customers. Indeed, relations between workers and customers in a luxury retail setting are stripped of most forms of reciprocity and mutuality. A belief in the power of markets to regulate service interactions-that customers will simply take their business elsewhere if they are not adequately catered to-suggests that a customer’s money buys not only merchandise but also polite and attentive service. That money is also seen as rightfully buying a worker’s deference and submission. New department stores in China enforce this ideology in the workplace, hiring young women believed to be more adept at displaying deference and then managing worker behaviors in order to guarantee such displays. In this way, organizations like department stores ostensibly in pursuit only of profit actually become invested in recognizing and reinforcing a wider structure of social entitlement.

This is not to suggest that that state socialist institutions in China or even reformed ones, like Harbin No. X, represented a workers’ paradise that housed a working-class culture in which everyone treated everyone else with equal doses of (dis)respect. As Andrew Walder’s (1986) study of the Chinese workplace testifies, the state socialist work unit in China was an institution in which intense competition for scarce rewards and resources created coercive, divided workplaces. Yet the fact remains that when I carried out this study, service workers in state-owned department stores like Harbin No. X conducted their work with levels of dignity and respect that find no equivalents in more “modern” models of service work in urban China. My coworkers could, and sometimes did, opt out of an interaction that represented an unacceptable challenge to their dignity. So while state socialist organizations like Harbin No. X. did not actually produce an egalitarian, working-class culture in urban China, they nevertheless have served as a key site for its enactment.

I want to suggest further that worker disempowerment is closely tied to a broad acceptance in China of new levels and forms of inequality that are often seen as unavoidable accompaniments to economic reforms (Wang 2001, 2003). To be sure, the model of service work pursued by China’s most elite retailers invests in a new structure of entitlement that not only dictates who deserves a job on the sales floor but also who is entitled to the levels of deference and respect distributed there. Service interactions that involve such expressions of deference must be located in a context much broader than the worker–customer dyad or even the manager–worker–customer triad. Rituals of class recognition embedded in service interactions are indicative as well as constitutive of these new class hierarchies and help create a culture of differentiated privileges and entitlements in urban China. In China’s new department stores, service interactions are the very “stuff” of inequality, revealing, in Arthur Stinchcombe’s words, “what powerful people can get others to do” (Stinchcombe 1965, p. 180). New service work regimes are helping to create, as Barbara Ehrenreich has said of service work in the US context, “not just an economy but a culture of...inequality” (Ehrenreich 2001, p. 212).