This article was written after the author joined a field visit to study the e-commerce industry in Donghai in May 2021, organized by Alibaba Group. Some author’s PhD peers also took part in the trip.
Donghai County is a major source of crystal, and in its early years it developed a crystal-processing industry on the basis of its natural resource endowment. Donghai crystal serves not only the domestic market. Crystal also carries religious and cultural significance for many foreign consumers, so a considerable share of Donghai’s crystal products is exported as well, though I do not know the exact proportion.
Donghai Crystal City was built largely with government funding. Its main business was to purchase crystal raw materials from abroad, carry out basic processing, and then sell them on, with overseas markets as its main destination. What people call “cross-border e-commerce” here often works in a very direct way: salespeople who can speak English find customers through platforms such as Facebook and Instagram, then conduct livestream-style sales through video calls, showing products to foreign buyers through the screen, quoting prices, and closing deals in real time. Once an order is settled, Crystal City ships the product out to the customer. In practice, these shopping guides are really sales agents for Crystal City, earning a commission on each order. During the pandemic, when cross-border travel became difficult, this screen-based way of selling took off.
Later, Donghai evolved from a center of crystal processing into a center for crystal distribution and trade, though I do not know exactly when this shift took place. The raw materials used in Donghai Crystal City still mainly come from imports, though again I do not know the exact share. And Crystal City no longer sells only crystal. It also sells spell books, interior decorations, divination items, and similar small goods. These things are not produced in Donghai, yet Crystal City has in effect become a trading center for what one might call a broader “magic industry.”
Image
At the same time, Donghai accumulated considerable know-how in the business of selling, and this became the starting point for its e-commerce industry. We also visited another town, where firms, after gaining e-commerce knowledge through selling crystal, began specializing in e-commerce and shifted into selling agricultural products, apparently with government support, though I do not know the exact form that support took. These firms selected seed varieties elsewhere, planted them locally, and then trained local villagers to become e-commerce influencers who could directly promote agricultural products. The place had the feel of a training base. For many e-commerce firms in Donghai, the development path seems to have been: making crystal → becoming a crystal distribution center → selling crystal → selling other things → teaching others how to sell.
The seafood e-commerce industry in Ganyu District, Lianyungang, followed a similar path. Fishermen came ashore and began selling seafood products on online platforms. Once business grew, they no longer relied only on local supply, but instead sourced seafood globally and brought it to Ganyu for sale. One might say that Lianyungang has a port advantage and is therefore well suited for the storage and distribution of imported products such as Ecuadorian shrimp. But a simple transport-cost advantage cannot explain why even crawfish tails from Wuhan are sent to Ganyu and then sold nationwide from there. The advantage of being a distribution hub is something distinct from a mere transport-cost advantage.
Seafood e-commerce requires a higher standard of supporting infrastructure than crystal does. A cold-chain warehousing base is essential. More generally, however, what e-commerce needs most is cheap logistics. If logistics can be made just RMB 2 cheaper per order, that is already a huge saving for e-commerce firms selling to consumers or small business buyers. For that reason, the government of Rizhao in neighboring Shandong spent heavily to subsidize the logistics industry, pulling many e-commerce firms away from Lianyungang and into Shandong.
When people talk about an “e-commerce cluster,” they usually mean a place where large numbers of people specialize in the supporting services that e-commerce requires. E-commerce, after all, must have something to sell, so its formation generally requires some underlying manufacturing or agricultural base. In its early stages, an e-commerce sector can often grow by relying on local industry and taking advantage of proximity, using lower transport costs to serve the local production base. But as Lianyungang shows, once sales techniques, commercial knowledge, and business networks have accumulated, some places can develop an e-commerce industry that becomes partially detached from the local production base and instead thrives on the advantages of being a distribution center.
Seen from this angle, e-commerce is simply one form of circulation and is not all that special. If “online” counts as e-commerce, then one might say that there was already something similar in the letter-based trading networks described by Greif among the Maghribi merchants. Before mobile internet, there were also passing trends such as telephone sales and television shopping, not to mention the geographic concentration of specialized markets in places like Yiwu and Guangzhou. What may distinguish the new e-commerce is that the new technologies involved are far more widely adopted. E-commerce, especially livestreaming, has achieved a level of digital penetration far beyond earlier media and has spread deep into lower-tier markets, while also extending commercial reach across global distances. E-commerce is a very large pie.
An important part of any industrial cluster is the large number of small and medium-sized entrepreneurs, and e-commerce clusters are no exception. As smartphones gave ordinary people the ability to speak and broadcast to others, e-commerce influencers sprang up everywhere. But the logic is the same: once too many people join the competition, the effect of e-commerce begins to cancel itself out. Consumer attention is limited, which means that traffic is always limited as well. In recent years, social feeds and short-video platforms have introduced information to consumers in fragmented form. In occupying consumers’ time, they have also created huge amounts of traffic, though many people have criticized this as well. But the gains in traffic from such formats are finite. In the end, influencers are pushed toward increasingly strange and attention-grabbing tactics. As the livestream e-commerce boom becomes crowded, grassroots influencers may well be replaced, just as grassroots entrepreneurs in industrial clusters are often replaced. E-commerce clusters, too, may run into a quality crisis. And it may come even faster.
Image
There is no denying the success of star influencers such as Viya and Li Jiaqi. On the contrary, we should learn from the experience of successful e-commerce operators. Good livestreaming and influencer-led sales show us that selling is a high-skill activity. In an economy of excess supply, it becomes even more important to study the demand side and the consumer, and then use that understanding to guide improvements on the production side. What happens before and after a consumer presses the payment button? How does one occupy a place in the consumer’s mental world? Answering these questions requires insights from marketing, psychology, sociology, and many other fields, and they are crucial for the future of e-commerce clusters as well. At the same time, the professional teams behind top livestreamers, Donghai’s boutique agricultural e-commerce base, and Ganyu’s Douyin seafood base all take on responsibilities such as product selection and quality monitoring. How to use the ability to sell things to help high-quality products fetch a good price, and how to solve the lemons problem, are both important questions in improving industrial quality.
rReflectio
Finally, a few reflections on doing fieldwork. On this trip, at each site, we were mostly just “taking a quick look around,” in the literal sense. Research is about answering questions, and to do that, one must first ask a good question. A good question usually begins with something puzzling or anomalous. Such anomalies come from observing facts in the field. The first step in fieldwork is to get the facts straight. Even if one cannot grasp every detail, one must repeatedly test one’s judgment about the core facts until it is more or less right.
If the process of understanding facts is like blind people feeling an elephant, or like assembling a puzzle, then one must first form a basic understanding of the overall picture of a place, for example, its economic structure and development history. Ideally, one should begin with a discussion meeting with local grassroots officials and ask them to introduce the general situation. Only then should one start piecing together the local details, constantly asking where each part fits into the whole and gradually refining the puzzle. If one rushes straight into isolated fragments of knowledge, one may end up saying things that will only make local people laugh.