The Unbearable Lightness of Trust: Trade, conviviality, and the life-world of Indian export agents in Yiwu, China

Abstract Drawing on ethnographic data collected in China, the United Arab Emirates (UAE), and India, this article explores the life-world and practices of small-scale Indian export agents based in Yiwu, China, the world centre for the export of small commodities. It shows that in a market overdetermined by fast-moving goods, short-term gains, and low margins, export agents have to steer their way between acting with extreme caution or taking risks with their clients and suppliers. These apparently contradictory dispositions or orientations are negotiated by the judicious exercise of mistrust and suspicion. The article suggests not only that mistrust is valued and cultivated as an indispensable practical resource for success in Yiwu's export trade, but that contingent relations of trust between market players emerge at the interstices of a generalized mutual mistrust, via the mobilization of practices of hospitality, commensality, and masculine conviviality. Indeed, feelings of amity and mutuality elicited by the performance of modalities of social intimacy become the affective terrain upon which divergent economic interests might be reconciled and taken forward. That is, mistrust might not lead to generalized distrust, instead a situational or contingent trust might actually emerge through the judicious exercise of mistrust.


Introduction
'If I did business only with people I know and I can trust,' Gafoor once explained, 'I'd be a poor man. Trade in Yiwu is like gambling-skills and planning are not enough. You have to take chances, sometimes it is just about your luck, that's all!' The words of Gafoor-a 40-year-old small-scale Indian export agent based in Yiwu, China-reveal the predicaments of a striving export market which, over years, has brought to the city thousands of foreign buyers, making it the world centre for the commerce of cheap commodities. The nature and modalities of the export trade from Yiwu are such as to require market players to be prepared to do business with each other without prior connections, often on a one-off basis. Moreover, trade in cheap commodities, especially when it involves the export of relatively small volumes of goods-as in my respondents' case-tends to be semi-formal in nature. That is, while the shipping of goods requires appropriate, formal paperwork, the actual buying/ selling of commodities mediated by export agents is seldom secured by formal contracts, but is negotiated through informal agreements. In other words, Yiwu's expansive export trade is neither solely embedded in long-lasting, personalized relationships of trust as typically associated, for instance, with so-called bazaar economies, nor is it entirely safeguarded by the legal regulatory frameworks assumed to underpin contractual obligations in fully fledged modern markets. Reflecting on ethnographic data collected during fieldwork among Indian export agents who, like Gafoor, operate from Yiwu, in this article I explore ways in which contingent and contextual relations of trust in the export business emerge through the deployment of the techne and politics of mistrust. I suggest that everyday market interactions between export agents and their clients or suppliers are informed by suspicion and mistrust, while trust is built up out of the careful concealment and deployment of its absence. More than this, I will argue that it is the capacity to navigate or assuage mistrust itself-via the mobilization of practices of hospitality, commensality, and masculine conviviality-that provides the basis for success in Yiwu's export trade.
To support my discussion, I will focus on the work of Indian export agents in Yiwu as an assemblage of practices, dispositions, and technologies that requires the cultivation of specific skills and social networks (Mathew 2019), but which, at the same time, elicits uncertainty and ambiguity. As a result, the reliability and trustworthiness of those who participate in the export trade emerge as necessarily contingent and contextual, making it impossible to draw hard-and-fast categorical distinctions between honourable and trustworthy players or identify those who have a penchant for sharpness (Marsden 2020). Notoriety and reputation might not be easily disentangled from each other in the biographies of the export agents with whom I became acquainted. Reputation in itself might not be built simply on trustworthiness and integrity, but also upon the capacity to strike deals and generate profit, even at the cost of bending rules or navigating fuzzy boundaries between legality and illegality, the licit and the illicit (see Osella and Osella 2000b). A market player who is overly trusting is likely to be taken advantage of, if not cheated, by others; likewise, an excessively honest export agent might not be able to make the most of the market, and thereby become blocked from building the reputation for success which is necessary to attract customers with an eye to quick gains (cf. Osella andOsella 2000a and; see also Gambetta and Hamill 2005).
My respondents in China, India, and Dubai might seek to cultivate or elicit trust as a means of attracting business and exercising a degree of control over -or even exploit-the uncertainties of trade (see Harriss 2003), but the fact that they regularly bemoan the untrustworthiness of clients and partners alike suggests that trust, for them, is always in short supply. Building on recent historical (for example, Mathew 2019) and anthropological literature on trade (for example , Marsden 2016;Humphrey 2018; see also Hart 1988;Fanselow 1990), I will argue that the particular characteristics and organization of export trade from Yiwu are such as to require wrapping the careful management of mistrust and suspicion within a language of mutuality and amity, in which affective relations and economic calculation are folded onto each other (cf. Wolf 1966;Osburg 2013;Armytage 2015). Crucially, the apparent tension between mistrust and trust is neither resolved nor erased (cf. Gudeman 2009), but is worked through and navigated via the performance of forms of hospitality which seek to elicit a degree of amity and mutuality between market players.
The instances of hospitality which export agents extend to their clients are a broad set of practices that not only stretch across different nations-China, India, and the UAE-but also arch back to analogous modalities of male conviviality which have historically undergirded trade in port cities across the Indian ocean, and that I have discussed at length elsewhere (Osella 2012). While the notion of conviviality might evoke and elicit feelings of open-ended amity and sociality between strangers, in practice it is seldom divorced from mutual suspicion, self-interested instrumentality, and even coercion. 1 In turn, the explicitly gendered and eroticized hospitality I will discuss later chimes with Osburg's (2013) study of the role of feasting in the reproduction of business relationships in contemporary urban China (see also Allison 1994). Indeed, recent research in globalized Asian economies has underscored the significance of eroticized modalities of conviviality-enabled by an increasingly global sex industry-for the affirmation of hierarchies of entrepreneurial masculinity which sustain the working of expansive, aggressive markets (see, for example, Zheng 2006;Hoang 2015;Allison 1994). In Yiwu, as much as elsewhere, politics of eroticized male sociality, and the gender hierarchies the latter entail, are constitutive of the market and thus cannot be separated from the working, the aesthetics, and the imagination of the latter (Osella 2012;Osburg 2013). The ethnography I discuss suggests that what oils the wheels of trade among Indian export agents and their clients is the contingent and contextual nature of trust elicited out of mistrust, via the mobilization of gendered aesthetics and hierarchies, played out on the stage of hospitality.

The qualities and value of trust and mistrust
Trust is usually conceived of as an essential lubricant to economic action and social life at large (see, for example, Gambetta 1988a;Sztompka 1996;Cook et al. 2004), but a personal reputation for trustworthiness and reciprocal trust are deemed to be particularly important to the working of markets such as Yiwu, in which a degree of informality underpins relations between buyers and sellers. In these trade environments, the formal regulatory frameworks which underwrite fairness and trust in modern markets are believed to be either underdeveloped or altogether absent (see Geertz 1979;Hart 1988;Cook 2005). Instead, the circulation of information and knowledge about trade, the setting of prices, the extension of credit-and all the practices which enable the exchange of goods and commodities-take place within wider social networks in which reciprocal trust is embedded in ties of kinship, friendship, patronage, ethnicity, religious allegiances, and more (see, for example, Keshavarzian 2007;Anderson 2019a;Rabo 2005;for India, see Bayly 1983;Ray 1988 andYang 1999;Harriss 2003;Birla 2008). These social networks might be narrowly local in orientation-as in Clifford Geertz's classic studies (1968;1979) of markets in Indonesia and Morocco-or expansive, cosmopolitan, and transnational in shape and orientation, as revealed in scholarship on historical or contemporary cross-regional land and sea trade (see, for example, Gupta 1967;Meillassoux 1971;Dale 2002;Markovits 2000a, Sood 2016Marsden 2016;MacGaffey and Bazenguissa-Ganga 2000;Humphrey 2018).
Many an anthropologist has argued that when trust between traders, or between traders and their clients, cannot be established or secured by means of formal contracts and regulations-often by virtue of the specific conditions and nature of trade itself (see, for example , Hart 1988;Fanselow 1990;Humphrey 2018;MacGaffey and Bazenguissa-Ganga 2000;Roitman 2005;Harriss 2003)-affective relations are relied on as a means to shore up the potential risks and uncertainties of doing business with strangers. While relations of kinship, friendship, and shared ethnicity or religion might not in themselves secure trust, they are, nevertheless, presumed to constitute the terrain upon which practices of hospitality might be mobilized strategically-if not aggressively-to elicit degrees of mutual intelligibility and confidence between possible trade partners (Marsden 2016;Osburg 2013; see also Armytage 2015). To be sure, rather than being solely a feature of so-called informal, transitional, or emerging economies which might lack regulatory structures to ensure long-term compliance and reliability (see, for example, Hart 1988;Cook et al. 2004), the building of relational trust remains central, even in the everyday practices of fully fledged, formal markets (Cook et al. 2009). 2 Whether ascribed behavioural, cognitive, or sociocultural salience, we note that in the copious academic literature on trust, the latter is normally conceived of as a necessary constituent of human action (see, for example, Gambetta 1988a and1988b;Dunn 1988;Putman 1993;Fukuyama 1995;2 Recent ethnographic studies of stock markets and financial institutions, for instance, have underscored the role of personal connections in the working of contemporary financial markets (see, for example , Zaloom 2006;Ho 2009;Miyazaki 2013; see also Richard and Rudnyckyj 2009;Keane 2008). Informality and mediation, then, are not solely the domain of marginal or emerging economies, or of the economically dispossessed (Bayat 2010;cf Roy 2009;Björkman 2021), but undergird economic action at large. As such, rather than deploying them as analytics or to establish mutually exclusive typologies of economic practice according to a teleology of market rationalization and modernization, I use 'formal' and 'informal' simply as descriptive terms to indicate differences which are coeval and always a matter of degree rather than kind. Sztompka 1996 andCook 2005;Cook et al. 2004;Tilly 2005). Built upon evaluations of the actual or supposed trustworthiness of others-whether people or institutions-and the projection of oneself as trustworthy, the extension of trust elicits 'a benign loop of evocated trustworthiness and reciprocated trust' (Sztompka 1999: 28) between social actors. Although in practice a balance should be struck between naive trustfulness and outright distrustquite simply, one should trust those who are believed to be trustworthy, and distrust those who are not-the scale is always tilted towards the value of trust (see, for example, Sztompka 1996;Hardin 2002;Gambetta 1988a;Cook 2005). Whether driven by economic calculation (Hardin 2002) or appealed to in order to keep in check individual self-interest (Elyachar 2005), trust in much of the existing literature is naturalized as a positive moral value or disposition immanent in the affective qualities of relations of social intimacy, kinship, and friendship (see, for example, Wolf 1966;Hart 1988;Osburg 2013). That is, while relational trust might be superseded by formal regulations or institutions (see, for example, Zucker 1986), or indeed work alongside the latter (Granovetter 1985;Gudeman 2009;Harriss 2003), trust is conceived of as a fundamental (moral, cognitive, or practical) quality of social relations that can be recovered, replicated, and drawn from with relative ease to smooth the path of modern democracies and markets.
Inevitably, I argue, this attribution of quasi-ontological status to trust precludes an interrogation of the processes through which relations of trust might be obtained or elicited in practice. Here, I take heed of Geschiere's (2013: 28) critique of all-too-common assumptions-in anthropology and beyondabout the (naturally benign) qualities of social intimacy, in which he suggests that 'trust is never self-evident. It has to be studied as a product of specific historical circumstances, as a continuously new "event"' (see also Humphrey 2018;Carey 2017;Corsín Jiménez 2011;Krishnamurthy 2015). Following Geschiere, then, in order to unsettle the apparent taken-for-granted immanence or inevitability of trust, in the following pages I will explore ways in which relations of trust-ephemeral and transient as they may be-are constituted out of the management, and indeed concealment, of mistrust. To clarify, I take mistrust to indicate a disposition towards questioning the possible trustworthiness of others, which might lead to either trust or distrust on the basis of direct experience, available information, or haphazard assessments of the character and behaviour of exchange partners Gambetta and Hamill 2005). In other words, rather than conceiving of mistrust negatively as the absence of trust, and thus detrimental to economic action, I consider mistrust to be an individual quality and practical skill essential to the success of Yiwu-based export traders. Indeed, it is from the standpoint of mistrust that market players test and gauge each other's reliability and trustworthiness, making it the very grounds upon which reciprocal trust might be cultivated.

Indian export agents in Yiwu
Located in the Zhejiang province of eastern China, Yiwu is the world centre for the trade in and export of cheap commodities, even counterfeits. China's 'small commodities city' houses a number of specialized wholesale markets, including the sprawling Futian complex which contains more than 70,000 wholesale outlets, and the Huangyuan market with its 5,700 stalls. Building on an established tradition of 'walking traders' and informal farmers' markets, commodities trade in Yiwu took off in the early 1990s, supported by Deng Xiaoping's programme of economic reforms. At first catering predominantly for the Chinese market, by the late 1990s trade began to internationalize on the back of the lifting of export restrictions, as well as the increasing global demand for cheap commodities such as those sold in Yiwu's markets. By 2015, 65 per cent of all the commodities traded in Yiwu were exported (Rui 2018: 26). Crucial to this process of internationalization of trade were not only the entrepreneurial skills of Yiwu traders who promoted their goods across the Middle East, Central Asia, and beyond, but also a shift to singlecommodity trade and the clustering of trade into similar commodities within specific markets or streets (Rui 2018;Belguidoum and Pliez 2015). Foreign buyers who visit Yiwu looking for particular commodities or to discover novel goods can easily move between stalls in some ten specialized markets and 30 specialized clusters of streets, as well as in the Futian market's five 'districts', each one dedicated to the trade of related commodities-from fashion accessories and ornaments, to electric appliances and hardware, and much more. Within and around these trading centres, buyers can find the offices of forwarding and export agents-normally dedicated to sending goods to specific nations or geographical areas. Further away, near Yiwu's sprawling dry port-from where containers are transported by lorry to Nigbo's sea portthey can access the warehouses of expeditioners, often located in storerooms on the ground floor of residential apartment blocks. Yiwu's orientation towards global trade is inscribed in its urban environment, whereby different neighbourhoods' restaurants, hotels, travel agents, supermarkets, and so on all cater to the particular tastes and needs of diverse bodies of foreign residents and visitors (see Marsden and Ibañez Tirado 2018;Skvirskaja 2018). During various 'buying seasons' which anticipate increased consumer demand associated with festivities-such as Diwali, Eid, Christmas, or Valentine's Day-or annual events such as the beginning of the school year, Yiwu's trading centres and streets are animated by foreign buyers from the world over, lending an apparent cosmopolitan feel to the city.
The fast development of its expansive markets and trade centres has attracted not only a buoyant global trade, but also a substantial population of foreign residents who work as traders or export agents, either on behalf of clients or for themselves, often for both (Belguidoum and Pliez 2015;Marsden 2017). The 3,000 3 or so Indian agents and traders resident in Yiwu are perhaps less numerous than their African or Central and West Asian counterparts (Bodomo and Ma 2010;Cissé 2015;Marsden and Ibañez-Tirado 2018;Skvirskaja 2018;Anderson 2019b), and do not appear to be as prominent in trade as they are in other Chinese cities (Haitao 2015;Cheuk 2016). Yet they are well established in the commercial life of the city, as testified by the row of hotels, restaurants, and shops in Yiwu's 'little India', close to the Futian market, which caters for the resident Indian population and their Indian clients.
The biographies of some Indian export agents offer a unique insight into their involvement in small-scale trade in Yiwu, and provide a context for agents' entanglements with wider historical and contemporary trading networks. 4 As is the case elsewhere in China (Haitao 2015;Cheuk 2016), the majority of Indian export agents in Yiwu belong to castes or communities with long traditions and histories of trade, with roots in Sindh, Gujarat, Punjab, and Rajasthan in northwest India (see Tripathi 1984;Cadene and Vidal 1997). While some of these export agents came to Yiwu directly from the subcontinent, others have more mobile biographies, having either grown up in the Indian trading diaspora (cf. Rudner 1994;Falzon 2004;Amrith 2013) or having worked in different locations outside India before settling in Yiwu. Take Mr Sheth, for instance, whose family originates from Sindh. Initially, his father moved to Bombay in 1947, when Sindh became part of Pakistan, and from there he emigrated onwards to Singapore to manage a textile business. After some years, he started an export agency, which in time was passed on to his four sons (Mr Sheth and his brothers). Eventually, the four brothers moved out of Singapore, but remained in the export business: two of them are now in Dubai, and the other two in China-Mr Sheth in Yiwu and his younger brother in Shenzhen. Mr Sheth left Singapore for Taiwan some 30 years ago, first to expand the family firm and, later, to establish his own export business. As trade from Taiwan became less profitable, he took his chances in South Korea; for similar reasons and in the same way, he eventually landed in China-first in Shanghai, and then Yiwu. At 65 years old, Mr Shirish has a similarly complex and mobile biography. He has been in Yiwu for the last 15 years, where he began by working for an established Indian export agent and eventually started his own export business. Now he has two offices in the same complex: in one, his son caters for Indian clients; in the other, Mr Shirish deals with buyers from Sudan and East Africa. His family has roots in Rajkot (in Gujarat, a state in eastern India), but Mr Shirish's father was taken to Sudan as a young man by his eldest brother, who worked for a British firm which exported water pumps from India to East Africa. Mr Shirish was born in Khartoum and, like his father, worked as an import/export agent, and eventually married a woman from Rajkot, with whom he had a son and a daughter. When the children reached secondary school level, they were sent back to Gujarat with their mother as 'the political situation had made schooling in Sudan too difficult'. When his wife passed away, Mr Shirish joined his children in Rajkot and tried to continue his business activities there, but he could not adjust to the new life. He wanted to go back to Khartoum-'in my heart, I missed Sudan too much'. But his son and daughter opposed the move: 'they thought of Sudan as backward and less developed than India'. He jumped at the chance to come to Yiwu to work for a family acquaintance, and now his son has joined him there.
For others, moving to Yiwu was simply a matter of trying to make the most of an expansive market for Chinese goods in the Indian economy's post-1991 economic liberalization. Take Mr Girdar, a chartered accountant by training and one of the first Indian export agents to settle in Yiwu. He was born in Calcutta to a family of Rajasthani Marwaris who had moved to Bengal during colonial times to trade in silk and gold, and who had eventually shifted into sourcing raw jute for Calcutta factories. With the post-independence decline of jute factories, young Girdar shifted back to the export/import business, selling Indian wool and cotton yarn to Bangladesh, and soon becoming the main export agent for three major Indian cotton mills. As this trade waned, he used his experience and knowledge to import ready-made garments from China. Girdar finally decided to move to Yiwu 18 years ago to expand his trade activities. Nowadays, his business focuses on exporting machinery on behalf of Indian corporate clients in the garment sector, but he also continues to source a variety of goods from his two-floor office-where he runs the business along with his wife and younger brother, supported by ten Chinese staff. Mr Girdar also has offices in Calcutta and Dubai (run by partners) as well as an export company in Hong Kong.
Other export agents had different trajectories of mobility and work. Some came with little or no experience of the export business and had been brought to Yiwu by relatives or acquaintances to learn the trade by working for established export companies. Many have slowly worked their way up from the position of employees to that of independent agents or are in the process of doing so. Among this group, I met Rajiv, a 35-year-old man who came to China seven years ago from his native Mumbai. In Mumbai, he had worked in his father's wholesale haberdashery business, which specialized in supplying embroidery stones to the garment industry. But by Rajiv's own admission, politics was his biggest passion. And it was politics that brought him to China. He had become so mixed up with violent vigilantism that his father decided that the further away Rajiv was from Mumbai, the safer he would be. He asked his suppliers of imported Chinese embroidery stones to find a job for his son in this apparently thriving and potentially lucrative business. A reluctant Rajiv was packed off to Yiwu, where he joined a Chinese-owned export agent's office which dealt predominantly with Indian clients. He learnt the ropes of the work he was tasked with very quickly-finding the most appropriate suppliers for Indian clients and organizing the shipping of the goods they purchased. To his own surprise, Rajiv discovered that he had a flair for languages. Within a year, he had become fluent enough in Mandarin to be able to mediate between buyers and suppliers to everyone's satisfaction, and thereby established his own network of clients. After seven years, Rajiv is close to acquiring the local government licence necessary to trade independently, and has accumulated enough capital to set up and sustain his own export business.
Yet another group of export agents I worked with-composed almost exclusively of Muslims from Kerala, South India 5 -arrived in Yiwu via the Gulf states of West Asia (especially Dubai). These agents had started their careers as accountants, stock traders, or shipping agents for international companies. Almost all are educated to degree level. Some were offered the chance of a transfer to the Chinese offices of their companies. Others simply learned about the economic opportunities offered by the expanding commodity export business and utilized personal or familial contacts, as well as financial resources drawn from friends and business partners, to move to China, eventually finding their way to Yiwu. Faisal was born into a landowning family in Malappuram district in North Kerala. In the 1970s, Faisal's father, with a newly acquired engineering diploma, left Malappuram for Saudi Arabia where he spent all his working life employed in the thriving construction sector in the Arabian peninsula, sending remittances home to guarantee his family a comfortable life, including university education for all his sons and daughters. Faisal received a good English-medium school education and an accountancy degree. Right after graduation, he joined his elder brother at an international investment bank in Bangalore. After one year, he was transferred to the Dubai office, where he became well acquainted with the director, a Malayali who had worked for many years in South Africa before moving to the UAE. The manager advised Faisal that as a young man 'he should move around and experience the world' and offered him a transfer to the Guangzhou office in China. Faisal followed his manager's advice and moved to China, but soon 'grew bored of office life'. He learned the basics of the export trade from various Kerala export agents whom he had befriended in Guangzhou, and decided to move to Yiwu (where some old college friends from back home were working as agents). With some capital invested by relatives, Faisal was able to enter the export trade. He now caters for clients found through his extensive network of contacts in both Kerala and Dubai. He rents a small office in one of the many tower blocks leased almost entirely by export and shipping agents, where he employs a clerk and a couple of young English-speaking Chinese women as translators.
A common thread in the biographies of these Indian origin export agents I came to know in Yiwu is not just their history of mobility, but also their capacity to create connections and exploit chances built on their acquired familiarity and by managing diverse social and economic environments. They reminded me many times that their entrepreneurial spirit came down to having 'trade and business in their DNA', ensuring they are ready to go wherever opportunities arise, in order to make the best of expansive economies. Export agents have deployed both know-how and know-who to exploit the chances offered by economic liberalization in both China and India, and the expansion in these countries of commodity production and consumption respectively.
Indian agents facilitate the export of whatever goods in the Yiwu wholesale markets their clients acquire-from household goods and school stationery, to embroidery beads and power tools-and, according to the size of their business, normally dispatch between one to five containers per week. The agents I met were mostly below the age of 50, usually married (but with wives and children living in India 6 ), and might have Chinese lovers or girlfriends in Yiwu. A handful of these men have married Chinese women with whom they now run their export business (see Lan 2015;Joseph et al. 2017;Sha 2020). Agents combine regular business trips to India with visits to their home and family, but while in Yiwu, most of them lead what they describe as a 'bachelor life' (Bristol-Rhys and Osella 2016). Indian export agents, while not necessarily partnering directly in the export trade, exchange information and expertise, share clients and contacts, and lend each other practical support, more often than not on the basis of shared regional identities. This includes the lending of money to support export deals and also-most importantly-providing the means to facilitate the (formal or informal) transfer and exchange of foreign currency. These men socialize frequently during work and free time, meeting regularly to play badminton or football together, but also to enjoy each other's company around 'home food' and a few drinks. I will discuss modalities of work and sociality later: for now, I simply note that Indian export agents rely on the support of extensive networks of 'connections', usually consisting of 'partnerships' with people-relatives or friends-also engaged in the export business, in India or the Gulf countries. Partners not only introduce potential clients and check the latter's trustworthiness and sincerity but also facilitate clients' payments and the all-important customs clearance of containers in the port of arrival.
Unlike other foreign residents who participate directly in the export business as traders, the role, work processes, and practices of Indian export agents in Yiwu resonate with those of the 'commission agents' who brokered trade during the colonial and early post-colonial period (Osella and Osella 2011). 7 Too quickly dismissed as rent-seeking profiteers (see, for example, Neild-Basu 1984;cf. Geertz 1960), historians have underscored the centrality of commission agents and brokers in the integration and expansion of (continental and trans-continental) South Asian trade, as well as in the growth of indigenous capital (see, for example, Bayly 1983;Allen 1981;Siddiqi 1982;Ray 1988;Rudner 1994;Birla 2008;Roy 2014;Nomura 2014;Mathew 2019;Aldous 2017;cf. 6 Given the absence of affordable English-medium schools in Yiwu, women who might have joined their husbands in China return to India once children reach primary school age. 7 Extensive research has underscored the pervasiveness of various modalities of mediation or brokerage in South Asia at large, whose influence was expanded by unfolding colonial and postcolonial political and economic relations (see, for example, Srinivas 1976;Chakrabarty 1989;Chatterjee 2004;Chandavarkar 1994 andRoy 2008, Breman andGuérin 2009;Piliavsky 2014;Björkman 2021). Davies 1952;Chapman 1979;Dethloff 1982). 8 These agents did so by providing not only credit facilities and logistical infrastructures to connect coastal and inland markets but also-and most importantly-information and contacts which facilitated access to local markets. 9 In the same way, export agents in Yiwu bring together Indian buyers and Chinese suppliers, offering the former information concerning prices, quality, and availability of goods, while guaranteeing payments to the latter.
As we shall see, while Indian export agents work to moderate the possible risks and uncertainties associated with the physical and/or social distance between buyers and suppliers, the volatile and competitive nature of Yiwu's export trade is such as to foster suspicion and mistrust between market players (cf. Birla 2008: 319ff). At the same time, while Indian export agents take advantage of the need for mediation in small commodity export trade from Yiwuand, more generally, of the centrality of mediation and brokerage within Indian society at large (Björkman 2021)-the outcome of this is that they themselves then also become the object of the scrutiny that targets all those who, by acting as middle-men, use their ingenuity and connections to manage, assuage-or exploit-tensions and contradictions within and between diverse social, political, or economic environments (see, for example, Neild-Basu 1984;Neale 1983;Osella 2014;Björkman 2021: 35ff). This is what Björkman (2021: 16) aptly describes as the 'paradox of the necessary-yet-suspect' nature of mediation and brokerage-a paradox which informs not only public confabulations, but also many academic discussions about vernacular brokers. 10 However, the scrutiny to which export agents are subjected in Yiwu is not driven by judgements concerning the legitimacy (or even morality) of their apparently opportunistic means of value creation. On the contrary, scrutiny is exercised upon the actual and context-specific behaviour of individual middle-men, in particular whether or not they deliver the services for which they demand and receive fees. I shall show that participants in Yiwu's export trade-agents, buyers, suppliers-not only operate and compete within the structure of the market, but also seek to make the best of their chances by maintaining differential access to information, exploiting networks and-if needs be-changing the terms of trade. The paradox here, to paraphrase Björkman's words, is the necessary-yet-elusive nature of trust between market players, whereby trust relations (cultivated as a means to mitigate the dangers of possible defaults) might be undermined when they are perceived to be a fetter to profit. In other words, in Yiwu's export trade there are no hard-and-fast means to determine in advance whether or not any particular market player would prioritize relationships based on mutuality and trust over short-term gains. 11 If export agents share many of the broad characteristics and business orientations of colonial or post-colonial 'commission agents', they also experience similar conditions of precarity and instability (Mathew 2019). Against the predictions that informal mediation would wither away once both capital and democracy reached 'maturity', recent research has shown that in postliberalization India, different modalities of brokerage have flourished (see, for example, Picherit 2009;Roesch 2009;Björkman 2021;cf. James 2011). And yet, the fortunes of commission agents wax and wane, as the result of markets' cyclical propensity to reduce costs by eliminating the work of mediation from transactions (Roy 2014: 287). In the case of export trade from Yiwu, the price of Chinese commodities might increase to the point of making them uncompetitive in the Indian market, while novel taxation regimes and protectionist import duties in India diminish margins and returns. Crucially, with the fast introduction of e-commerce in Yiwu, the export trade itself might make the mediation of export agents altogether redundant. As such, agents are constantly looking for new markets and opportunities-from Africa and Southeast Asia, to Latin America. They know, not only that Yiwu might be only one stop in their mobile careers, but also that they might even have to change occupation or trade altogether along the way (see Haugen 2018).

Doing business in Yiwu
Indian export agents in Yiwu mediate the brokering of trade deals between suppliers and buyers, for which they demand a commission (which can vary between 1 per cent to 5 per cent of the value of the deal), the fee normally being charged to the buyers. Buyers might be clients with whom export agents have had past dealings, who have visited Yiwu on previous occasions, and who then continue to place orders and negotiate further deals long-distance, using texts or video calls. In those cases, it is the agents who approach suppliers on 11 In exploring the ambiguities and unpredictable outcomes of mediation, I follow Kate Meagher's (2005: 226) warning against the essentialism and cultural determinism of 'abstract models of solidarity and connectivity', and focus instead 'on the specificities of how particular types of networks operate'. behalf of buyers and carry out all the negotiations concerning the quality and price of commodities, as well as delivery dates. Such negotiations are punctuated by repeated-often tense-WeChat 12 exchanges between buyers and agents to ensure that goods purchased satisfy the needs and wallet of the clients.
Take Rajiv, mentioned earlier, whose clients are all Mumbai-based and generally introduced to him by his father. He receives most client orders by mobile phone, with conversations about the type, quality, quantity, and price of the goods to be acquired, followed by photos of similar goods purchased in the past or available on the Mumbai market. One day, Rajiv was asked to purchase 400 cartons of nail varnish sets with six different colours, the client specifying not only the price per piece he was prepared to pay, but also the range of colours and packaging of the sets. After a few phone calls with other agents and known suppliers, Rajiv sets off to the area of Futian market specializing in cosmetics, moving from one wholesaler to the next, until he finds one who has nail varnish sets on display close to his client's specifications. He begins by enquiring (in English) about the range of colours available and the price per set, taking pictures with his phone of what the seller shows him and forwarding them to the Mumbai buyer.
A back and forth negotiation then begins: the buyer wants six specific colours for each set, packaged and labelled 'rainbow brand' (which is currently selling well in Mumbai); he also sets a specific price per set. The wholesaler, in turn, calls a manufacturer to check for availability and price of the goods. After some quick calculations, he shows Rajiv a calculator displaying a price, assuring him that labelling and packaging can be made to look almost identical to that already being sold in Mumbai. Rajiv does his own calculations to determine the price-on the basis of the current exchange rate between Renminbi, US dollars, and Indian rupees, and adding to that a percentage to cover all the expedition costs as well as his commission-and calls back his client. The latter is all but pleased and begins to shout on the phone that the price is far too high, accusing Rajiv of being useless, if not a cheat, and threatening never to do any business with him ever again. The wholesaler is also getting annoyed with Rajiv, telling him that for the low price demanded by the buyer, no one in the Futian market could provide that range of nail varnish colours and packaging.
At that impasse point-and to the great surprise of the wholesaler-Rajiv shifts from English to Mandarin. Looking for a compromise, he asks if the price can be lowered by including in each set some cheaper colours that are close enough to what the client wants. After consulting with the manufacturer and another round of calculations, Rajiv calls back his client. 'Uncle,' he says, 'this is the last price, this is all I can do. You know me and my father, you know I am sincere, but we all have to eat: please agree!' The client agrees to a slightly different range of colours, but wants the price reduced further. By then, the negotiation has been going on for more than an hour, and Rajiv turns to the wholesaler, saying, 'If you agree with this price, I can transfer the money by phone now, immediately.' Although none too pleased with the outcome, the wholesaler nods, a delivery date is set, and the deal is concluded. As we leave, Rajiv comments, He [the buyer] doesn't care about quality, he just wants the biggest margin possible for himself, and screw everyone else! When some new product comes up and sells well, everyone wants to copy, flooding the market with similar but slightly cheaper goods, until that item is killed off. They are more concerned with ruining each other than with making a successful business for themselves. All Indian buyers are like that, and they don't understand China. Chinese suppliers work on very small margins, and so when a customer asks for big discounts or low prices, they lower the quality of the product they supply, and if they can't, they just pull out (cf. Cheuk 2016).
Proud of his business acumen, Rajiv continues, 'He [the buyer] wanted to screw me, to make money for free! At the end, he agreed to buy nail varnish of the lowest quality, but then he doesn't care about quality!' Rajiv's reflections not only reveal the complexities and peculiarities of export trade from Yiwu, but foreground the significance of dispositions such as improvisation, expediency, and, indeed, suspicion in steering relationships between agents and their clients.
Unlike Rajiv's clients, other buyers come in person to Yiwu-sometimes several times a year, in anticipation of India's various 'shopping seasons' (associated with major religious festivals)-to search for new commodities or to strike bargains. In this case, agents mobilize their contacts and knowledge of the local market to provide logistical support to the buyers: facilitating access to suppliers or manufacturers (in person or through their Chinese employees), helping with negotiations, or simply providing language translation. Often, clients come the export agents' way via personal or business contacts back in India or the Gulf, where agents might have a 'branch' run by partners. In many instances, though, buyers simply contact export agents on arrival in Yiwu, on the basis of advice given to them in Indian hotels and restaurants, or responding to adverts publicizing the services of Indian 'export agents' seen in Yiwu's 'little India' or in the back of taxis. Sometimes, it is simply chance encounters in a restaurant or club that bring together agents and clients, as in the case of Binu and Faisal.
One day, Faisal recounted, first-timer Binu was sitting with a couple of friends who had accompanied him to Yiwu in a restaurant-cum-shisha bar in a downtown area of the city popular with foreign visitors and residents. They were recovering after a day spent walking up and down the sprawling Futian market and strategizing on how to strike good deals in a trading environment of which they knew next to nothing. Faisal and Gafoor (two Malayali 13 export agents who regularly extend information and support to each other) 13 People from Kerala speak Malayalam, and hence they are called Malayalis.
could not help but notice the three customers who, as well as looking somewhat lost and bewildered, were clearly recognizable as Malayalis by their mundu-and-white-shirt attire. Drawn by such an unusual sight in Yiwu, the agents joined the strangers at their table and, to cut a long story short, Binu became one of Faisal's regular clients. Binu had come to Yiwu on the hoof, drawn by the growing reputation of the city as a market for cheap commodities that was already visited by many of his business competitors in Kerala. Now he comes to Yiwu to place orders at least twice a year, always accompanied by other buyers who, in their turn, take advantage of Faisal's intermediary services. Over the years, Binu has shifted from buying small quantities of 'household goods' and stationery in Futian market to importing full containers of goods bought directly from manufacturers, which he then distributes through his wholesale shops in Kerala and the UAE. As a regular and reliable client, Binu and his friends get Faisal's full attention and care. They are hosted and entertained throughout their visits, and are driven by Faisal in his spacious BMW SUV to suitable factories around Yiwu.
I will return to export agents' hospitality later. For the moment, I want to underscore that, whether they are newcomers or old Yiwu hands, whether they purchase goods directly from manufacturers or via wholesalers in Futian market, all buyers need the support offered by these export agents. If they are buying large quantities of the same goods (which normally entails at least one full 40-foot container), buyers approach a manufacturer-rather than a wholesaler in Futian market-either in person or via an agent. Besides providing the initial contacts, the services provided here by agents might be limited to ensuring quality control and monitoring delivery dates, as well as organizing the actual shipment. However, the success of Yiwu as an export centre is due not only to the variety and low price of goods found in its markets (as well as the ready availability of counterfeits) but also to the possibility of exporting mixed goods in the same container or renting part of a container to send smaller quantities of commodities. For these shipments of mixed or part containers, the services of export agents are essential: they organize the packaging of goods, receive and store goods purchased by buyers, check the quality and quantity of the goods, organize land transport to the nearest port and shipment to the final destination, provide the documentation necessary to export goods, and-most importantly-they manage the billing and payment of goods.
Normally, buyers are asked to pay a 30 per cent deposit on the cost of goods when the purchase is agreed, and they should settle the balance of the bill either once the goods are loaded on a ship or it reaches the port of final destination, against receipt of the bill of lading 14 necessary to get shippers to release containers. In many instances, however, the more experienced and entrepreneurial export agents can make money on top of their commission by convincing suppliers to waive the initial deposit and/or by extracting a deferral for the payment of the remaining balance, while asking the buyer 14 The bill of lading is a receipt issued by the carrier once containers are loaded onto a ship. It is used for customs and insurance purposes, as well as a proof of completing commercial transactions. to make timely payments for both deposit and final balance. In this way, agents will have in their hands a substantial amount of cash which can be invested in other export deals, turning a cash flow into the investment capital necessary to secure or expand their business. At the same time, because agents sometimes need to secure the custom of a good buyer or to extend their circle of clients, they might take upon themselves the risks of export deals by advancing the costs of the goods-both for the initial advance and for the final payments. By advancing payments to suppliers from their own pockets, agents thereby become, to all effects, partners in the deals.
In this complex trading environment, agents and their clients have to act with some caution-if not outright suspicion and mistrust. Buyers, for instance, are extremely wary of agents, suspecting them of inflating shipping costs and commissions, of failing to negotiate hard enough with suppliers for a good price-and even of getting into cahoots with wholesalers (either to keep prices up or to source goods of a quality inferior to those agreed upon). Of late, those large Indian importers that I met in both Dubai and Kerala prefer to buy goods directly from factories, to hire the services of specialized companies to conduct quality controls, and, more generally, are establishing their own supply chains to move goods directly out of China with the support of registered shipping agents based in India or the Gulf itself. Once they find an appropriate and reliable partner, these importers tend to stick with the same suppliers, in a relationship that is often described not simply in terms of efficiency and profit, but through the language of familiarity and friendship. 15 As we shall see, intimacy and affect also loom large in the imaginary of Indian export agents in Yiwu. But here, the uncertain and competitive nature of the trade they mediateoverdetermined as it is by fast-moving goods, short-term gains, and low margins, and buttressed by informal practices often straddling the licit and illicit (see Humphrey 2018; Marsden 2016)-gets in the way of consolidating longterm relationships of trust such as those which emerge in the shadow (and relative safety) of the formal contracts utilized in high-volume trade. 16 The dream of small importers is to model themselves on their larger counterparts, cutting out mediators and buying directly from producers. At the very least, importers often seek to influence negotiations by approaching wholesalers directly (via electronic media or in person) to gather information on quality and prices offered by different suppliers. By these means, importers can limit export agents' input into the minimum of organizing the collection, storage, and shipping of goods. At the same time, Chinese export agents have become extremely active not only in the Gulf, but also-and increasingly so-in 15 Many an Indian importer-from plywood and kitchens, to industrial machines-explained to me how, over the years, a relationship which was initially cultivated with a particular supplier or factory in the name of mutual economic interest with an eye to mitigating so-called transaction costs (see Williamson 1975;Lorenz 1988) developed into bonds of friendship, sustained by reciprocal gift giving and visits to each other's hometowns-in China and India (see Osburg 2013). 16 A similar argument is put forward by Fanselow (1990) with reference to the way in which the selling of either branded or unbranded goods shapes the nature of relations between traders and buyers in Tamil Nadu bazaars. See also Parry (1999) on the ways in which particular modalities of industrial production sustain specific relationships between workers. India, looking for opportunities to participate directly in the international circulation of commodities. Buyers try to 'steal' contacts and knowledge from agents, and internet platforms such as Alibaba and the fast development of e-commerce in Yiwu have the capacity to bring buyers closer to suppliers. But still, export agents and their commissions cannot be entirely erased from market transactions, because of the important role they play in facilitating navigation through complex regimes of taxation and currency transfer. Here, export agents are able to arrange for buyers to make considerable savings. This is done, for instance, by under-billing the cost of goods exported. This allows buyers to pay fewer import duties and sale taxes, as well as to pay bank charges on a smaller amount of foreign currency for the (underbilled) costs. The remaining balance of the full, real cost of the goods is transferred to China via much cheaper informal means-the well-known hawala system (see Martin 2009;Ballard 2014). 17 While Indian buyers seek to increase profit margins by limiting as much as possible (if not altogether excluding) the mediation of export agents, the agents themselves try to retain their centrality in the market by controlling access to information concerning trade-for instance, by repackaging goods so as to withhold the identity of suppliers-and by providing the means to circumvent, one way or another, those trade regulations which inflate the costs of imports. In India, as much as in Dubai, Yiwu-based agents mobilize personal connections in destination ports to smooth the timely release of containers, a process which often entails the discreet distribution of 'donations' to various customs or port officials. And yet, even the most well-established and reliable of connections can become easily unsettled and ineffectual as a result of novel import regimes and regulations. 18 In recent years, for instance, India implemented a number of measures-from demonetization and currency controls, to the introduction of the Goods and Service Tax (GST) and the curtailment of foreign imports. For several months, this had the effect of tightening customs controls in all the main Indian ports, to the point where local officials had to stop any informal arrangements with agents. As a result, the release of containers suffered major delays and the payment of the resulting increased costs and fees caused substantial friction between agents and their clients. To negotiate this tricky terrain around the facilitation of currency transfer, custom clearance, and tax regulations, export agents have to rely not only on their knowledge and familiarity with the ever-changing circumstances of international trade, but also on their capacity to mobilize personal connections. This enables them to leverage various degrees of partnership or collaboration beyond Yiwu and China.

Export trade and the politics of mistrust
Indian export agents often find themselves stuck between a rock and a hard place: not only having to deal with the predicaments of changing international trade regulations, but also squeezed between demanding clients and potentially unreliable suppliers. Agents complain continually about clients whose hard-bargain tactics force them to spend unreasonable amounts of time hunting down acceptable deals, who aggressively query every shipping charge, delay payments, and eventually reduce their profits to a minimum. And yet, agents cannot be too choosy with clients. They decide to take on demanding or potentially untrustworthy clients for the sake of increasing profit or maintaining business in a very competitive market-or as return favours to friends and relatives. At the same time, agents know that suppliers who are prepared to lower prices without much bargaining are also more likely to be unreliable in terms of deliveries or in the quality of goods they sell. Predictably, the greater the risks export agents are prepared to take for the sake of securing a potentially lucrative deal, the higher the chances are not only of making good profits but also of incurring substantial losses.
In 2018, for instance, one of the most established and able Indian export agents resident in Yiwu had to deal with the consequences of the reduction of trade to India due to the introduction of the GST mentioned above. Faced with a substantial decline in the volume of container traffic, he decided to take on an order from an Oman-based department store (a new client), for which he agreed to ship $200,000 worth of clothing, in time to be sold for the rather short Omani winter season. Normally extremely cautious and risk-adverse, this export agent agreed to advance the costs of the goods and shipment against a much higher than the usual commission fee. He had been warned by friends in Oman that this department store had a dubious reputation with suppliers, and yet business was so bad at the time that he decided to throw caution to the wind. Eventually, this turned out to be an unfortunate decision: once the two containers arrived in Oman, the client failed to pay and take on the delivery, leaving the agent to foot not only the costs of the goods and shipment, but also increasing demurrage fees. After many tense and unsuccessful trips to Oman, the agent had no choice but to dispose of the goods himself-at a heavily discounted price. Girdar, a wellestablished Bengali export agent, explained to me that this all-too-common peril of being cheated by clients steered his own business strategy towards dealing mostly with 'corporate clients who care for reputation and have no interest in making quick gains'. 'They are too prominent to be caught cheating,' he explained, 'and if they cheat, they do so for millions of dollars.' Just as important as the value of personal or corporate reputation, however, are the formal and legally enforceable contracts which are routinely signed for high-volume commodities or expensive machinery, such as the industrial looms exported by Girdar.
The businesses of most of the agents I encountered are far more precarious, small-scale, and informal than Girdar's, and so misadventures like the Oman story above are recounted frequently. On the one hand, they are passed on as tales in which agents portray themselves as the victims of unscrupulous, money-grabbing clients (cf. Rabo 2005;Keshavarzian 2007). On the other, those recounting these stories chide the incompetence or lack of judgement of those agents who take excessive risks too lightly. Most importantly, though, these stories underscore the impossibility in many circumstances of determining the trustworthiness of others, thus emphasizing the value of mistrust as a means of navigating uncertainty and risk (cf Harriss 2003). 'The way export trade works in Yiwu is such that agents have to trust their clients,' Faisal pondered one evening, 'and clients must trust agents; but both have to trust the Chinese suppliers!' This thought was shared by Gafoor, who added, 'there is cheating on both sides [both Indian clients and Chinese suppliers], but in India there is even more cheating. You have to be very careful who you trade with!' 19 In the world of Indian export agents and their clients, 'cheating' covers a wide range of events, from plain swindling to failure to deliver on promises of quality, quantity, or punctuality. This in no way maps neatly onto a continuum between 'trusted familiars' and 'risky strangers'. Not only might ties of reciprocity and mutuality-expected within kinship and friendship, for instance-turn into deception and exploitation, but the person who cheats someone may at the same time be doing honest business with somebody else. The same person might behave in both ways in different moments, so that it is the nature of the moment-the particular assemblage of circumstances and actors involved-that will flavour the event. While these cautionary tales might be based on actual events or ill-fated personal experiences, they become a rhetorical means through which agents imagine and reflect upon their engagement with the particular risky nature of small-commodities export trade (cf. Gupta 2005).
Confronted with the uncertainties of an extremely volatile market, Indian export agents operating in Yiwu need to contain the potentially expansive risks of international trade. This is pursued by various means, from building expertise on the workings of the export market, conducting repeat transactions with those with whom one has established familiarity, relying on personal contacts and recommendations in attempts to ascertain the reliability of market players, or simply by preferring to do business with buyers with a shared ethnic or religious background. Typically, the first exchanges with a client who enters an export agent's office will seek to establish not only whether the agent's services have been recommended by someone known, but also whether there are social affinities between agent and client-on the familiar grounds, for instance, of place of origin or community belonging. Rather than establishing trustworthiness, these conversations allow agents and their clients to find a means of evaluating each other's background, to place strangers within one's own social maps, and thereby to gauge the shape and limits of a possible deal (see Cook 2005;Gambetta and Hamill 2005).
This, however, is a conservative strategy. Limiting risks and uncertainties also constrains the possibility of maximizing profits in the face of increasing costs, restrictive export regulations, and fierce competition. Often, it will be those who might turn out to be unscrupulous or untrustworthy who secure better profits for agents: cunning Indian buyers ready to pay higher commission with an eye to defaulting on final payments or unreliable suppliers who offer discounts and bargain deals to conceal the low quality of goods they sell or their incapacity to meet delivery schedules. Likewise, sharp agents are often those who are the most adept at circumventing the costs and constrains of trade regulations and bureaucracy, ensuring higher returns for clients. In other words, in a market driven by intense competition and small profit margins, such as Yiwu's, trust might be elicited at the interstices ofand thus remain in tension with-the unpredictable or ambiguous behaviour of the players who are brought together by the export trade.
In such an atmosphere of uncertainty and suspicion, the language and aesthetics of amity provide a means through which market players seek to create bonds of intimacy conducive to lessening the risk that mistrust might turn into generalized distrust, which would make trade all but impossible. Amity is fostered and elicited, for instance, through practices of hospitality and entertainment which export agents extend to visiting buyers and that the latter expect as a central feature of their 'Yiwu experience'. Malayali export agents I got to know routinely provide various forms of hospitality to their Indian clients, from offering accommodation, to organizing get-togethers in their houses around food, copious alcohol consumption, card-playing, and, towards the end of the evening, taking turns to sing popular Malayalam romantic songs or even recite poetry. 20 We meet for one of these get-togethers-a party-in Gafoor's flat, a small three-bedroomed apartment in a four-storeyed residential block near Futian market. Furnished with cheap or second-hand furniture inherited from friends who moved away from Yiwu, in every room, cardboard boxes are piled up against the walls-stuff which Gafoor is planning to take back to Kerala on his next visit or samples of goods for his clients. Although he lives alone, Gafoor might sublet the spare bedrooms to Malayali friends needing accommodation while waiting to rent their own apartment. At least once a year, during school holidays, his wife and two children join him in Yiwu for a month or so. Most of the time, however, he has room to put up clients from Kerala, and tonight there are five of them who are in Yiwu to buy school stationery. We are joined by another three Malayali export agents (Gafoor's friends), who arrive together, each carrying a bottle of vodka and some Indian snacks. 'This is our tharavadu,' 21 they joke, 'our meeting place in Yiwu!' A game of 20 My respondents also regularly entertained local Chinese suppliers, typically by treating them to drinks and meals in Yiwu's most upmarket Indian restaurants. For reasons of space, I will focus here solely on social events involving Indian export agents and their Indian clients. 21 Traditional Kerala property-owning, ancestral house associated to a joint family (see Moore 1985). rummy starts while we wait for dinner. Sambar, boiled rice, and beef fry (Kerala Christian food in honour of the five visiting clients) is cooked and served by one of Gafoor's friends. As the vodka bottles empty, more are brought up from a convenience store nearby. After eating, card gambling is replaced by heated and somewhat incoherent discussions about Kerala politics, by raucous reciprocal banter, and by Gafoor's soulful singing of Malayali folk tunes. Eventually, at 2am the party is over, and to avoid being caught drink-driving, Gafoor's friends call a Didi Daijia (chauffeur) service via mobile app, to safely drive them and their car back home. The atmosphere of parties like this is generally light-hearted, although the flow of alcohol might result in occasional tense confrontations, immediately doused by those present. Usually, conversations run from everyday events in the home country to political gossip, but seldom touch on business matters. 'We are here for enjoyment,' my respondents explained, 'not for business!' Indeed, the ability to develop camaraderie and amity with clients-whether buyers or suppliers-is as important a skill as is familiarity with the complexities of export trade. In Dubai, I accompanied a Malayali export agent, Anilkumar, who operates in both Yiwu and the UAE for several days on his routine calls to his regular and potential clients. Invariably, after exchanging pleasantries, conversations veered to any topic but business itself, centring instead on news and gossip from the homeland, with the occasional reference to the everyday predicaments of life in a foreign country. When I eventually asked him why he hardly ever discussed business during these visits-involving business deals which, from previous conversations with Anilkumar, I knew to be often complex and even controversial-he explained: there are so many agents here in Dubai, competition is fierce, and it is always about business, business, business. But clients want to feel friendship, a connection [in Malayalam ; translit. bandham] which is not just about making money. If you show that you are friends and connect at this level, they [the clients] will call you the next day and give you a good contract. If you visit only to talk about business, they will go to someone else.
Extended as an expression of amity, the practices of conviviality mobilized by agents in Yiwu to entertain their clients for the sake of hospitality and camaraderie seek to produce a degree of connection or attachment between participants-as entailed by the semantic field of the Malayalam word bandham, rooted in notions of relatedness and reciprocity. What becomes significant about these parties, though, is the means by which connection is established: by eliciting intimacy and complicity via the eroticization of sociality. If the mood is right and the gathering has not yet collapsed in a drunken slumber, parties shift to one of the many night clubs which animate Yiwu's night life and have become commonplace around commercial hubs in post-Maoist China (see Shen 2008;Zheng 2006 andYuk-Ha Tsang 2018;Osburg 2013;cf. Allison 1994;Hoang 2015).
In the clubs, export agents and their visiting guests can continue drinking in the company of female escorts-who hail from China as well as Central and Eastern Europe and Latin America-who vie for their attentions and custom, singing and dancing with them, playing dice games, or simply titillating them with the close proximity of their alluringly dressed bodies. Different establishments cater for the diverse pockets and tastes of potential clients of various nationalities. The most common set-up consists of tables set around a dance floor, where the main entertainment is music and dancing. As small groups of men arrive, they are seated and invited to order drinks (mostly international brands of beer and spirits). On the dance floor, groups of hostesses take turns to dance and attract the attention of the men at the tables. A DJ plays Chinese pop music, alternating it with popular Bollywood or other songs requested by clients; a microphone is passed from table to table and clients are invited to join in with the singing. When a client decides that he likes a dancer, he asks the waiter to take over one of the bouquets of flowers prearranged at the edge of the dance floor, thus inviting the dancer to join the group of men he is sitting with. At the table, the hostess makes sure that the alcohol keeps flowing, flirting and dancing with the clients to keep them interested and entertained.
Upmarket KTV (karaoke television) clubs are an alternative for those with deeper pockets-and who don't want to be seen drinking in public and in the company of female escorts. I visit one such club with Gafoor and Faisal, accompanied by one of Gafoor's new Malayali clients, Sreekumar, and a young Chinese wholesaler known as Dan, with whom the pair had done business on several occasions. The evening began at Gafoor's apartment, where, after a takeaway dinner ordered from one of the restaurants in Yiwu's 'little India' and a couple of bottles of vodka, Faisal started to talk about his 'new Chinese girlfriend', challenging Dan to prove that the mistress the latter regularly boasted about actually existed. To save face, Dan agreed, but on condition that Faisal did likewise, and it was decided to meet the two women outside one of the KTV clubs. In contrast to the nightclubs described above, here groups of male clients (in our case now also including the two Chinese girlfriends) are ushered into discreet, dimly lit private rooms, plushly furnished with comfortable sofas, coffee tables, and a small stage. While drinks are ordered, young Chinese hostesses dressed in inexpensive but fashionable clothes are paraded on the small stage, and the men in our group are invited to choose which hostess will entertain them for the rest of the evening. Girlfriendless Gafoor and Sreekumar respond enthusiastically, asking two hostesses to join them on a sofa. The focus here is on karaoke, with music played loudly and the text of Indian film songs and various Chinese and international hit songs displayed on a big screen. Faisal, Gafoor, and Dan take turns on stage to perform their favourite tunes, alone or accompanied by a hostess. Punctuated by explicit flirting and attempts at sexualized touch, a night in KTV club might lead onto discreet negotiations for more intimate 'services'. However, intense police monitoring of hotels and clubs-as well as the fact that by then Gafoor and Sreekumar felt the worse for wear after the copious drinking-relegates any more explicitly sexual advances to the realm of wishful thinking. Indeed, a visibly drunk Sreekumar starts to complain loudly that his hostess is 'trying to cheat him', asking him to be paid again for her company. The bouncers quickly intervene, but Dan mediates, offering to pay the extra costs himself. Faisal leads a somewhat bedraggled Sreekumar out of the club to compose himself, and the party is over. As we wait for the Didi Daijia service to arrive, Gafoor explains: it wasn't like this before … so many clubs, bars, massage parlours with girls ready to please you. But now the Yiwu corporation has enforced a clamp down, even massage parlours don't do any more happy-ending service! They [the Yiwu corporation] want to give a new face to the city, shifting to e-commerce only. No more need for clubs in Yiwu! Part and parcel of an all-too-common modality of male sociality-especially in the state of Kerala where several of my Yiwu respondents have their roots (see Osella 2012 and 2015)-parties are reciprocated when agents visit their clients back in India or Dubai. Here, they might have a far more explicit sexual flavour and outcome. Chiming with the politics of 'ritualized leisure' among Chinese businessmen discussed by Osburg (2013; see also Gold et al. 2002), these practices are directed towards creating a social environment for the cultivation of mutually advantageous, personalized business relationships. Here, reciprocal mistrust and suspicion are not dissolved, but put to the test through the expression of intimacy, amity, and conviviality. By providing entertainment and distraction from the busy-and often tense-routines of trade, these parties evoke a common ground of sociality, in which economic calculation and instrumentality are mediated by, or altogether dissolved in, shared experiences of open-ended mutuality and complicity. 22 By seeking to elicit trust out of mistrust, if the parties or entertainment extended to visitors might be read as an instrumental means to measure or manage what economists call 'counterparty risk', outcomes are all but predictable. Happenstance encounters might be turned into long-lasting trading relations between agents and their clients, as in Binu's case above. Over time, Binu's joviality and gregarious disposition were matched by reliability and punctuality in both orders and payments. This did not happen in Sreekumar's case. Unlike Binu, he turned out to be a 'difficult' client, who not only insisted on receiving credit and discounts from Gafoor, but also asked to renegotiate the specifications of the goods he had agreed to purchase and eventually delayed honouring scheduled payments. However, once established, Sreekumar's capriciousness-if not outright untrustworthiness-could be managed or contained, for instance, by reducing to a minimum further credit and discounts on commissions. This made him, in the short term, a more profitable client than Binu.

Conclusion
These short vignettes of the hospitality which animates export trade suggest that long-term loyalties and mutual trust between export agents, as well as with their clients and suppliers, are an essential feature of everyday business life in Yiwu, and as such they are cultivated assiduously. However, these relationships might not lead inevitably to successful partnerships in business, but, as we have seen, they can also limit the expansive ambitions of those involved in international export trade. What is significant about the parties I described is that they are modalities of explicitly gendered-and even eroticized-conviviality, in which an atmosphere of light-hearted sociality and entertainment works towards establishing a common ground of intimacy in which participants can gauge each other's character and social skills, as well as testing the boundaries of their respective business orientations (see Zheng 2006: 162ff;Osburg 2013). Parties are marked out and experienced as events which straddle, or indeed expand, the boundaries between the licit and illicit. These parties, and their extension into night clubs, allow participants to perform different registers of male sociality. On the one hand, they reveal a disposition to break the conventions of (Indian) public morality-by enjoying drinking, gambling, or the company of female escorts-which chimes with the willingness, if needs be, to bend rules for the sake of securing maximum profit from trade. On the other hand, while this moral lightness emphasizes the wisdom of mistrust-as a skill and resource which affords, to use Caroline Humphrey's (2018:15) words, '[a] reflection not only on the motives of others but also on oneself as a person who is also likely to be acting in an untrustworthy way'-it simultaneously gestures towards an aesthetic of male mutuality and amity within which that trust is subsumed. It is precisely the simultaneous merging of, and tension between, affective and instrumental registers of sociality which constitute the moral economy 23 of (low volume and semi-formal) export trade from Yiwu (see also Osburg 2013;Armytage 2015;Allison 1994;cf Wolf 1966: 15). Here, amity and mutuality can coexist with, and indeed sustain, expansive individual economic interests, because such a moral economy rests not simply on the quality of relationships between men, but also on the abjection of the bodies and lives of those women who work in bars and clubs as entertainers and 'escorts' (Mayblin 2019;cf. Granovetter 1985: Ho 2009). Crucially, it is the mobilization of these gendered modalities of sociality that allows for the emergence of a contingent and contextual trust from an everyday politics of mistrust (see Mayblin 2019; see also Zheng 2006;Hoang 2015;Allison 1994;Osburg 2013).
By eliciting complicity around participation in well-rehearsed modalities and aesthetics of male sociality-which in India, as much as in the Indian 23 In academic writing, the notion of moral economy subsumes two related meanings-its economic aspect and its moral content. In the first, moral economy refers to the embeddedness of economic practice, the degree to which economic activity is shaped by and carried out amid social and cultural relations that are found everywhere in the society. In the second, moral economy draws attention to relations between social actors based on mutual obligations and expectations of reciprocity (Osella and Rudnyckyj 2017;Fassin 2009). Here, I draw attention to both meanings. diaspora, are normally obscured by politics of public morality-hospitality becomes a prosthetic to trade. And it does so, I argue, exactly because it produces bonds between buyers and export agents-contextual and ephemeral as these might be-which do not resolve or conceal, but rather convey, mimic, or rehearse the moral ambiguity and contingency of Yiwu's international export trade. To be sure, here ambitions for expansive personal gains and mutuality do not stand for opposite or incompatible poles of sociality and economic practice (Hart 1988;Wolf 1966) but are experienced and accepted as two sides of the same coin (Parry 1986)-a coin minted out of male sociality. Whether agents or buyers, my respondents neither yearned nostalgically for a time in which trust was unquestioned and plentiful, nor wished for stronger regulations to rid the market of sharp players and practices. They simply sought to exploit, as smoothly and as profitably as possible, the chances offered by the market, taking great risks or acting with extreme circumspection according to circumstances.
What to make, then, of the politics of mistrust I have discussed thus far? Recent scholarship has taken to task assumptions in the social sciences which normally postulate the presence of trust between social actors as a necessary condition for the smooth reproduction of social relations and institutions-whether in their 'traditional' or 'modern' guises. In these critical studies, mistrust is not defined naively by the absence of trust, but has been conceptualized either as a legitimate political tool of the subaltern (Krishnamurthy 2015; see also Elyachar 2005;cf Hardin 2004) or as a disposition embedded in culturally specific understandings of the inherent opacity, or incoherence and autonomy, of the self (Robbins 2008;Carey 2017; see also Corsín Jiménez 2011). While the ethnography I discussed does not warrant the attribution of either ethical value or ontological status to mistrust in itself, it does suggest that trust is not (always, necessarily) immanent in the qualities of social relations, but in certain circumstances it might be elicited by the management and deployment of mistrust (see also Geschiere 2013; Marsden 2016; Anderson 2019a). I am not suggesting here that the life-world of Indian export agents and traders in Yiwu-and also of those in India or the Gulf-is a Hobbesian dystopia or an instance of amoral familism or even an expression of unbridled self-interested instrumentality. Nor does it constitute or express 'communities of mistrust' such as those reported, for instance, in the transitional economies of post-socialist Mongolia (see, for example, Buyandelger 2013; Højer 2019; Højer and Pedersen 2019). On the contrary, we have seen that amity and mutuality are not just a gloss painted over the cracks of a sociality overdetermined by intense competition, uncertain markets, and expansive personal ambitions, but rather that they are the affective terrain upon which economic interests might be cultivated. I draw here on a time-honoured genealogy in anthropological analyses of the politics of exchange to suggest not only that mistrust might not lead to generalized distrust (cf. Hardin 2004;Sztompka 1999), but that a situational or contingent trust might actually emerge through the judicious exercise of mistrust. Among Indian export traders in Yiwu, then, mistrust is valued and cultivated as an indispensable practical resource which allows the negotiation of the vagaries and volatilities of international trade under conditions of intense competition and what are experienced as restrictive regulations. Reflecting on the contentious business partnership between an Indian merchant and his Somali agent in early twentieth-century Aden, historian Johan Mathew (2019: 245) cogently writes that, 'Trust is often experienced as a "gut feeling". It does not itself mitigate risk, but it does make that risk palatable.' In the case of Indian export agents in Yiwu, however, what is elicited might not be simply a 'gut feeling' about the trustworthiness of others, but, as Gafoor told us at the beginning of this article, the trust in one's own luck and capacity to navigate successfully the vagaries and risks of trade. However, it remains to be seen whether Indian export agents will have the wherewithal to navigate transformations of Yiwu's trade environment, driven by the expansion of e-commerce and the recent introduction of Chinese government policies directed towards privileging the production of high-value commodities. In the coming years, Indian export agents' resilience and capacity to adapt to new circumstances will be tested to the full.