Changing fish trade practices in Myanmar’s rapidly transforming food system

Abstract

Wholesale markets, wholesalers, and processors play critical but underappreciated roles in shaping food systems. Most conventional research on food provisioning analyzes value chains in terms of their structure, conduct, and performance. We contend that a practices lens can illuminate how food systems are produced and reproduced through the emergence of new wholesale practices relating to quality, trust, and risk, and provide more nuanced understandings of how markets and trading shape and are shaped by food system transformations. Applying a practices lens to the analysis of these changes can help identify new ways of steering food systems toward more sustainable outcomes.

Abstract (Burmese)

Video Abstract

Read the transcript

Watch the video on Vimeo

©2024 The Author(s). Published with license by Taylor&Francis Group, LLC

Keywords:

• fish
• food systems
• Myanmar
• social practices
• wholesale markets

Wholesale markets and the actors that populate them – wholesalers, brokers, processors, and retailers – play critical but underappreciated roles in shaping food systems and provisioning sustainable and nutritious foods (Ebata 2022; Gerber, Turner, and Milgram 2014; Subramanian et al. 2022; Tezzo et al. 2020). As argued by Bush and Oosterveer (2007), studying these actors’ shared practices makes it possible to decode lived experiences that are central to the social and cultural reproduction of food systems. But the roles that provisioning actors play in steering food system transformations are poorly understood (Béné et al. 2019; Reardon 2015; Veldhuizen et al. 2020). The “hidden middle” of the food system is transforming particularly quickly in Asia, where wholesale markets are proliferating to supply increasingly urban and affluent consumers (Ahmadi-Esfahani and Locke 1998; Alita, Dries, and Oosterveer 2021; Barrett et al. 2019; Hu et al. 2004; Reardon 2015; Reardon and Timmer, 2014).

Previous research on food provisioning has tended to analyze trade relations in terms of supply chain structure, and the conduct and performance of actors producing and distributing value along these chains (Bair 2009; Gereffi et al. 2001). Associated development strategies typically focus on identifying opportunities for chain actors to reduce economic and production risks and/or increase their efficiency and economic performance via processual or institutional “upgrading” (Gereffi 1999; Humphrey and Schmitz 2000). However, viewing agrifood value chains in terms of their economic logic can disembed them from their historical and sociocultural context and emphasize linear trajectories of development (though for exceptions see Gerber, Turner, and Milgram 2014; Goldman, Krider, and Ramaswami 1999; Lie 1997).

The role of traders in provisioning non-staple perishable foods such as fish, meat, and fruit has been analyzed elsewhere (e.g., Bestor 2001; Ebata 2022; Qanti, Reardon, and Iswariyadi 2017; Sharma et al. 2016; Steenbergen et al. 2019), but relatively few studies have characterized and situated the everyday experiences of “trade, traders and trading” in wet markets in Asia (Alexander 1987). Similarly, there has been little attention to how provisioning actors in wholesale markets effect and are affected by changes in the availability of and demand for food products, forms of production, and changing consumer cuisines and lifestyles (although see Fabinyi, Dressler, and Pido 2018; Fabinyi and Liu 2014; Fang and Fabinyi 2021; Tezzo et al. 2021). Following Bestor (2001), we contend that illuminating what provisioning actors do on an everyday basis, and why, can contribute to understanding how women and men in the middle of the chain “make the system what it is” (77).

In this article, we use a social practices approach (Reckwitz 2002; Schatzki 2002; Shove, Pantzar, and Watson 2012; Spaargaren, Lamers, and Weenink 2016) to examine the social embeddedness (Granovetter 1985) of aquatic food trade in the context of a rapid shift from wild-caught to aquaculture-based fish production. Aquatic food is comprised of diverse wild and farmed aquatic animals and plants that are central to diets, cuisines and nutrition across much of Asia (Fabinyi and Liu 2016; Ishige 1993; McIntyre 2002; Tezzo et al. 2021). These aquatic organisms are socially and culturally reproduced as food through a myriad of everyday and even mundane shared practices – that is, routinised sayings and doings (Spaargaren 2011; Spaargaren, Lamers, and Weenink 2016) that shape the performance of consumption, trade and production.

Asian aquatic food practices are changing rapidly as once abundant supplies of freshwater and marine aquatic animals are increasingly supplemented or supplanted by farmed products (Bush et al. 2019; Tezzo et al. 2020). Changes in the production, consumption and trade of farmed fish are often aligned to already existing aquatic food practices where farmed fish are substituted into culturally established dishes, but also lead to the emergence of new “urban” fish consumption practices (Tezzo et al. 2021). It remains unclear, however, how the social and cultural provisioning routines of aquatic food trade, traders and trading are affected by this transition, especially in “wet markets,” which are often characterized by highly routinised practices of trust, loyalty and patronage (e.g., Acciaioli 2000; Alexander 1987; Mele, Ng, and Chim 2015).

We examine changes to wholesale market provisioning practices in Myanmar, one of Southeast Asia’s poorest and least studied countries, located between Thailand, China and India. Myanmar underwent a remarkable decade of political and economic transition from 2011 to 2020,¹ accompanied by rapid transition from wild caught to farmed freshwater fish supply (Tezzo et al. 2018). We focus on San Pya, the largest fish wholesale market in Myanmar’s largest city, Yangon, where the share of farmed fish traded has grown rapidly. Drawing on Cadilhon et al. (2003) and Tezzo et al. (2021), we examine how provisioning practices in San Pya have been affected by the growth of aquaculture and fast-changing consumption practices in urban centers. We find that new provisioning practices related to quality, trust and risk have emerged, that shape the conduct and performance of production, trade and consumption. We argue that applying a practices lens to the analysis of these changes can support the identification of new ways of steering food systems toward more sustainable outcomes.

Our article is organized as follows. First, we explain the key elements of a social practice approach to food system transformation and its relevance to food provisioning. We then introduce Myanmar, describing the current transition in fish production and justifying the empirical focus on the country’s main wholesale fish market San Pya. We then describe our methodology, before presenting empirical results showing how practices of quality, trust and risk have been reconfigured in the context of growth of freshwater farmed fish production. Finally, we reflect on the implications for understanding and governing food systems in rapidly-transforming low and middle-income countries.

Understanding provisioning practices in food systems

Food systems are generally understood as a set of production, trade and consumption activities embedded within social, political, economic and environmental contexts that affect the availability, accessibility and utilization of resources necessary for food security, environmental sustainability and social welfare (Ericksen 2008; HLPE 2017). The concept of food systems has recently been used to examine the contribution of aquatic foods to environmental, social and nutritional outcomes (e.g., Golden et al. 2021; Simmance et al. 2022). But while the food systems “turn” has encouraged more thorough consideration of the relationships between these dimensions, the concept has relatively limited analytical power with respect to the social dimensions steering system change (e.g., Béné et al. 2020). In particular, the links between the micro-scale behavior of actors and macro-scale systemic change remain poorly understood, especially when behavior is embedded in complex trade relations.

In contrast, analyses of global value chains, commodity chains, and production networks have focused directly on the structure, conduct and performance of trading activities. Value chain research has variously examined market coordination by lead firms (Gereffi 2018), processes of innovation and upgrading (e.g., Pietrobelli and Rabellotti 2011), market-based sustainability governance (Bush et al. 2015) and the immanent development of new industries (e.g., Belton, Bush, and Little 2018). Recently, the structure, conduct and performance paradigm has been applied to understanding food systems transformation (e.g., Farmery et al. 2021; Reardon et al. 2019). However, much of the literature on value chain upgrading focuses on reducing the complexity of information, bargaining, and monitoring to improve efficiency and rationalization and enable higher volume, higher value and/or lower risk market exchange (Abebe, Bijman, and Royer 2016; Reardon, Liverpool-Tasie, and Minten 2021). We argue that adopting a food systems perspective opens up space for closer attention to the social embeddedness (Granovetter 1985) of provisioning actors involved in handling, processing and trading food, and can illuminate how provisioning actors’ conduct is shaped by and shapes food system change.

We use a social practice-based approach to explore the relationships between social embeddedness and food system transformations. By emphasizing practices as fundamental building blocks of markets and trade, practice theories draw attention to how commodities such as fish are shaped by the meanings and values assigned to them by producers, traders and consumers throughout the food system (Fine 2002). Understanding how actors negotiate these meanings and values in the context of highly routinised provisioning practices related to buying, handling, processing and selling offers opportunities to investigate how food is socially produced in changing social and material settings (Reckwitz 2002; Spaargaren, Lamers, and Weenink 2016).

We define provisioning practices as routinised doings and sayings that are reproduced by knowledgeable and capable actors with little discursive reflection on the material and social conditions that shape them (Spaargaren 2011; Spaargaren, Lamers, and Weenink 2016). Following Shove, Pantzar, and Watson (2012) and Schatzki (2002), we observe and analyze fish provisioning in terms of four integrated “elements” of practices. The first is meanings, made up of general understandings or values attributed to a doing or saying. The second is skills and competences required to perform a given practice. The third is material objects and infrastructures that enable the performance of a practice. The fourth is the goals or “teleoaffective structures” that give direction to the behavior of practitioners. Our focus is not on the role of each of these elements, but rather, building on Tezzo et al. (2021), we examine how they are reconfigured by and contribute to transformations within food systems.

We focus specifically on changes to provisioning practices for aquatic foods. Drawing on Cadilhon et al. (2003) and Tezzo et al. (2021), we start with the provisioning practices of buying, processing and selling. This enables us to identify the actors carrying these practices and understand how practices were reconfigured and/or reinforced by new materials, meanings, competences and goals introduced through the growing trade of farmed freshwater fish. In doing so, we deductively identified new bundles of practices related to quality, trust and risk that contribute to a more systemic understanding of the conduct of provisioning actors (building on Spaargaren, Oosterveer, and Loeber 2013; Welch 2017; Welch and Warde 2016). By reimagining food systems as continually produced and reproduced through new practices relating to quality, trust and risk, we offer a more nuanced understanding of how markets and trade shape and are shaped by food system transformation.

San Pya market

San Pya market is the largest fish wholesale market in Myanmar and an ideal site for exploring changes to provisioning practices associated with the transition from wild caught to farmed fish. Located on the Yangon River on the western side of Yangon, Myanmar’s largest city, the market is the main gateway for fish from the Ayeyarwady Delta where most of Myanmar’s wild and farmed freshwater fish are produced. San Pya was established in 1991 as the main landing and trading point for freshwater fish in Myanmar, during a period of partial economic liberalization as the country transitioned from a military-socialist to a “market reform military” rule following a coup in 1988. Prior to this time, private trading was not permitted (Turnell 2009).

San Pya is administered by the Markets Department of the Yangon City Development Committee (YCDC), and operates daily from late night until mid-morning. Fish arrives at the market by boat and truck, where it is aggregated and redistributed in and around Yangon and to the rest of the country. Most fish is delivered to San Pya by boat, because many aquaculture ponds in the Delta are accessible only by canal (Belton et al. 2015).

Fish are landed at six “jetties” on the river, operated by the six largest traders in the market – licensed commission agents who auction fish and exert most control over fish supply in San Pya (Figure 1). Around 300 licensed wholesale fish businesses operate stalls in three large buildings set back from the river, at the center of the market. Nearly all license holders are men, although some of the businesses are operated by both men and women family members. Approximately half of the licensed wholesalers are considered “large” or “medium” scale, and offer advance output-tied loans to farmers and/or fishing boat owners and traders to secure supplies of fish directly from them (Figures 1 and 5). The remaining 150 or so smaller licensed traders do not offer credit to their suppliers, and they obtain fish purchased at auctions from commission agents. A further estimated 150 small unlicensed wholesalers operate from buildings around the outskirts of the market and along the side streets leading to it; they buy fish from wholesalers and commission agents (Belton et al. 2015; Figure 2).

Figure 1.  

Figure 2.  

More than 50 small-scale fish processing businesses producing fish balls (nga chit in Burmese) are located around the outer fringes of the market (Figures 3 and 4). Their proprietors are mainly women, originating from a single village located across the river from the market. Nga chit production employs as many as 1,500 workers, the majority women. Around 2,000 male casual laborers provide loading and unloading services to wholesalers. Numerous other ancillary businesses are clustered around the periphery of the market, including transport companies and drivers, ice suppliers, small restaurants and tea stalls, and shops selling equipment such as weighing scales, chopping boards and knives (Belton et al. 2015).

Figure 3.  

Figure 4.  

About 30 unlicensed traders without fixed business premises act as brokers, buying fish from licensed wholesalers for distribution to other parts of the country. Retailers and small wholesalers from Yangon and surrounding areas (mainly women) visit San Pya daily to buy fish, often using transport rented collectively with other traders from their locality (Figure 2). Approximately half of the fish traded at San Pya is distributed by truck or bus to domestic markets outside Yangon (Belton et al. 2015). Trade of fish from San Pya to the rest of the country boomed in the wake of economic liberalization initiated by the quasi-civilian Union Solidarity and Development Party (USDP) government in 2011, deepening following the election of the National League for Democracy (NLD) in 2015. Key reforms included removal of fuel rationing and vehicle import restrictions, deregulation of private transport companies, the banking sector and telecommunications, and the construction of a new highway linking Yangon to the country’s second largest city, Mandalay. This spurred a decade of rapid economic growth that greatly increased domestic trade, mobility and consumer incomes (Belton et al. 2015, Belton, Marschke, and Vandergeest 2019).

San Pya is an ideal location to study everyday provisioning practices. The diversity of provisioning actors, selling to a range of wholesale and retail market channels (Figure 2; Table 1), is indicative of the importance of fish – both wild caught and farmed – in the diets of consumers in Yangon and throughout Myanmar (Scott et al. 2023). San Pya’s location on the edge of Ayeyarwady Delta and its transport connections to the rest of the country also make it the ideal site to examine changing practices associated with the transition from capture to farmed fish. When the market was established, fish originating from capture fisheries in the Delta accounted for the vast majority of aquatic food traded, and farmed fish a small share. These proportions are now inverted, with farmed fish accounting for an estimated 70% in 2014 (Belton et al. 2015); a share which has grown since.

Table 1. Summary information on sampled mid-chain provisioning actors.

N	Gender	San Pya XP (years)	Midstream actor role	Fish traded
				Volumes (kg/day)	Wild (%)	Farmed (%)
1	Male	25	Unlicensed processor	1,500	10	90
2	Male	20	Licensed commission agent	3,500	90	10
3	Male	17	Unlicensed domestic distributor	700	20	80
4	Male	25	Licensed large trader	1,700	0	100
5	Male	18	Unlicensed domestic distributor	800	30	70
6	Female	40	Unlicensed processor	1,200	5	95
7	Male	33	Unlicensed processor	800	0	100
8	Female	8	Unlicensed processor	500	0	100
9	Male	8	Farmer & licensed large trader	32,000	0	100
10	Female	15	Unlicensed retailer	80	25	75
11	Female	20	Unlicensed retailer	25	0	100
12	Female	24	Unlicensed retailer	50	10	90

Most of Myanmar’s fish farms lie within a 50 km radius of Yangon, extending westward into the Ayeyarwady Delta, which is also the location of the most important inland fishery in the country (Tezzo et al. 2018). Fish farms in the Delta expanded rapidly from the 1990s following the policy of the market-reform military government to allocate large land concessions to individuals and companies linked to the regime. This policy was intended to promote export-oriented industrial-scale agriculture and aquaculture, with the aim of generating foreign exchange. Aquaculture grew steadily from the 1990s until the coup in 2021, driven by the continued growth of very large farms and the emergence of numerous smaller ones (Mark and Belton 2020).

During the same period, the productivity of the Delta’s once abundant inland and marine capture fisheries declined sharply due to factors including over-fishing, land use change, and habitat degradation associated with agricultural intensification, the establishment of water control infrastructure, and the enclosure of aquatic commons for use in aquaculture (Mark and Belton 2020). The growth during the reform period of an increasingly affluent population in Yangon and other cities in Myanmar stimulated demand for fish sourced through San Pya (Tezzo et al. 2021) and oriented most farmed fish production toward the domestic market. Similar patterns of aquaculture development have occurred throughout much of South and Southeast Asia over the past 30 years, resulting in a wider “aquaculture transition” in the region’s food systems (Belton and Thilsted 2014; Bush and Marschke 2014; Garlock et al. 2020).

Methodology

We studied changes in provisioning practices at San Pya via a mix of direct observation, narrative description from visits, and quick interviews during peak evening and morning trading hours (n = 35), followed by in-depth interviews with a smaller selection of respondents outside market hours (n = 12). Respondents were selected through purposive sampling to encompass a broad mix of small, medium, and large-scale operators, including un/licensed traders, retailers, laborers, domestic distributors, and fish processors who sold both wild and farmed fish (Table 1).

All respondents were interviewed using a semi-structured interview guide focused on: (1) the life history of respondents and their experiences in San Pya; (2) everyday sourcing behavior; (3) stocking and processing activities, where relevant; and (4) selling practices. Interviews were typically conducted with the main stallholder or their righthand person (i.e., manager, spouse). Depending on availability, additional staff were interviewed to complement the information. Whenever possible, the respondents were also observed during their trading activities (n = 10) to confirm some of the information collected in interviews, with ad hoc questions to respondents aimed at eliciting their immediate reactions, meanings and choices attached to the practices performed.

All interviews were conducted by the first and second author using a mix of Burmese and English. These interviews were audio-recorded and lasted between one and two hours. All data were then transcribed in English and coded in NVivo 11 software. A codebook was developed based on the theoretical framework: the main code categories corresponded to two analytical dimensions, namely (1) types of fish traded (i.e., wild/farmed) and (2) provisioning practices (i.e., sourcing/stocking and processing/selling). Secondary codes were added inductively to explore new themes emerging from the analysis.

Analysis of provisioning practices

In the following section, we examine how increasing scarcity of wild fish, the growth of aquaculture, and the commoditization of fish ball production has led to new practices for (1) assessing quality, (2) consolidating or developing new long-term and trust-based relationships, and (3) mitigating new economic risks associated with growing disparities between large traders and their smaller counterparts.

Quality

Faced with the declining availability of wild freshwater fish and the rapid expansion of aquaculture in Myanmar, provisioning actors have adapted how they determine, maintain and/or cope with the different qualities of wild and farmed fish. Provisioning actors perceive farmed fish to have higher fat content, lower firmness, and less appealing flesh and flavor than wild caught species (reflecting observations in other Asian markets, Bestor 2001; Saguin 2014). However, despite these differences, “most customers cannot tell the actual difference [between wild and farmed fish],” in the words of one trader. Consequently, provisioning actors have established and leveraged constructions of quality that go beyond the materiality of the fish they sell. For farmed fish, emphasis is placed on affordability and year-round availability to meet the demands of urban consumers. At the same time, increasingly scarce wild fish fill new high-value niches associated with ruralness and family cuisine (Tezzo et al. 2021).

New practices are also emerging around the provision of fish balls that were traditionally made with specific wild species, but are now mass-produced to capitalize on the influx of “lower quality” farmed fish. Fish balls consist of the scraped or minced flesh of fish (Figure 4). Said to have Chinese origins, the processing of fish balls is accepted as a means of providing urban consumers with low-value marine fish across Southeast Asia (Heng and Eong 2005; Saguin 2014; Siriraksophon, Pangsorn, and Laong-Manee 2009). In Myanmar, fish balls were traditionally made using a wild caught freshwater fish called featherback (Notopterus notopterus), which has become increasingly expensive, and so fish balls now commonly use cheaper farmed rohu carp (Labeo rohita). As recounted by a processor whose family had been in the business for over 40 years,

At the time of my grandmother and mother, we were only processing featherback fish balls … Since 1995, we mostly process rohu fish balls because their demand keeps increasing. Featherback fish balls are now only made on order by few customers, typically for special occasions.

The processing of farmed fish balls has enabled these provisioning actors to broaden demand. When visiting San Pya market, the scale of farmed fish ball processing is immediately evident, with small processing plants – often operating without a license – in and around the market, serving not only Yangon but also other expanding urban centers countrywide (Figure 3).

The increased volume of fish balls is related to the qualification of “first” and “second” grade fish in the market, based on tangible product qualities such as freshness, shelf-life, and new intangible credence qualities (Wessells 2002) associated with the origin of the fish and (farm) production processes. The fish balls produced using wild caught fish are unquestioned as of higher quality over those made from farmed fish – in some cases proving just as important as other attributes such as freshness. The quality of fish balls made from farmed fish is downgraded further if the fish come from (or are perceived to come from) more intensive farms or integrated chicken-fish farming systems. Provisioning actors have enabled this seemingly intangible quality by ascribing new meanings to observable product attributes. As one respondent explained:

The stomach should not be full of feces so we always check for the black color around the cloaca …The size of the head is another indicator of quality: it should not be too big. What is more difficult to check before slicing the fish is the quality of the flesh. For instance, if farmers feed too much peanut cake, the flesh tends to be whiter and very oily… A quality flesh should be slightly pink.

By creating new grades of fish balls from farmed fish based on the farming process, provisioning actors have been able to increase the value of both wild fish and farmed fish from more extensive farming systems. These new definitions of quality are also deployed to bargain down the price of fish originating from less preferred farming systems. In the process, traders have positioned themselves as key arbiters in determining quality grades, setting new incentives for suppliers and influencing which fish is made available to urban consumers and at what price.

The mass processing of fish balls has also been affected by the standardization of trading lots. Commission agents and large traders are responsible for sorting lots of wild fish based on a wide range of sizes and species – with a generally higher demand for larger size classes of fish (as elsewhere Gates 1974; Sjöberg 2015). However, the growth of aquaculture has meant that these commission agents, who control the wholesale movement of fish to San Pya, receive lots made up of uniform sized fish (see Figure 4). At the same time fish ball processing has redefined the relationship between quality and size – with demand shifting to lots made up middle size fish. As one processor explained, higher quality rohu fish balls are made using “specimens … about a hand size (i.e., 600 g). If much smaller, the flesh is too soft and if much bigger, the bones make it harder to scrape the flesh.” One trader in San Pya explained even further qualification:

There are three sizes of rohu and we have specific names for them: Gaw refers to fish below 600g; Chit nga is the standard 600g size and finally Khon koe koe tit for fish larger than 600g. Most processors buy chit nga, but I typically target the ones that are slightly bigger because they are cheaper and tend to better travel long distance.

The standardization of grading has created feedback between production and trade. Compared to wild caught fish supply, the control over size grading by farmers and the larger and more consistent lots that commissioning agents can deliver to San Pya has driven the practice of mass fish ball processing. As processing volumes have increased, processors have developed new practices that have created new product categories for consumers while also requiring new harvesting and trade practices upstream in the chain.

Trust

Supply from capture fisheries is more geographically dispersed than supply from aquaculture, and involves more complex trade networks that rely heavily on long-term quasi-credit relations to secure intermittent supplies of fish (as elsewhere in Southeast Asia and beyond, Bush 2004; O’Neill and Crona 2017; Platteau and Abraham 1987; Ruddle 2011). In contrast, aquaculture is geographically concentrated within a 25–50 km radius from Yangon city, with production dominated by large farms (Belton et al. 2015, Belton, Bush, and Little 2018). These farms typically have production volumes large enough to bypass local fish collectors, so farmed fish supply tends to be aggregated at San Pya (see Figure 2). This consolidation is exacerbated by the fact that upstream actors in the wholesale market often extend loans to large farms – the largest of which are several thousand hectares in size – to secure fish supply. These differences in the structure of wild and farmed fish supply chains have implications for how trust between provisioning actors is reproduced, and lead to new emerging practices.

Commission agents and large traders tend to buy directly from individual producers more than is the case with wild freshwater fish. The rapid adoption of mobile phones and mobile banking in Myanmar, which expanded dramatically between 2012 and 2020 (Rieffel 2016), have been important in enabling increasing numbers of sourcing transactions. While increasing the ease of communication along the chain, digital technologies have also increased the need for new ways of creating trust. As one buyer trader explained enthusiastically:

I don’t even need to go the market anymore: I simply place my order by phone a day ahead and confirm quantity, price, and quality… trust is all the more important between us and our suppliers because we usually do not see the fish lots before buying.

The success of sourcing is dependent on both retaining reliable suppliers and reaching out to new suppliers, often through familial networks. Alternatively, very large traders have sought to vertically integrate to circumvent trust issues around their farmed fish supply. In doing so they can better align harvest and buying, as one trader explained, “according to the demand,” enabling them to “transport … fish daily from the pond to the market.”

Provisioning actors have also developed new ways of building trust to cope with weaker transparency in supply chains associated with a growing number of intermediaries. Here the case of farmed fish balls serves as a good illustration. As one processor explained, “some of my customers now buy my product to process it further. They mix it with salt, MSG, coloring powder, and baking soda to make it a bit stickier and ready-to-cook.” Other processors explained how, under increasing competition, it had become common practice to pass fish balls through a blender to get rid of bones. They also explained how this practice made it possible to conceal the quality of the fish being processed.

Overall, additional stages of processing render adulteration easier (see also Fine 2002). In San Pya, there were reported instances where “retailers trick their customers by labelling their products as (wild) featherback fish balls when they actually use a combination of featherback and (farmed) rohu fish.” As a result, fish ball supply chains are now tinged with suspicion, increasingly leading buyers to request that their fish balls be processed in front of them at time of purchase. Even when they already have some processed product for sale, fish ball processors increasingly process new product on demand from prospective customers as a demonstration of their integrity.

The growth of fish farming has multiplied the number of intermediaries that traders at San Pya have to deal with downstream. As one processor complained, “it is getting more and more difficult to establish trusted relationships with our customers because they keep changing and they are becoming more and more demanding … sometimes, we have to operate at loss just to keep the business going.” In response, medium and small-scale traders in San Pya have had to invest in marketing themselves to secure business with retailers. This is visible around the processing of fish balls where it has become the norm for actors to display their product on trays in front of their shop to showcase the quality of their products (see Figure 4). With all the weight of her 40 years’ experience processing fish balls, one processor regretted “Even though the demand has kept increasing, we do not make as much profit as we used to because the competition is much higher and we sell smaller volumes individually.”

It is increasingly difficult for downstream provisioning actors to create loyalty among customers given the simultaneous proliferation of competitors and potential buyers. As a result, marketing efforts and price increasingly shape transactions for the multitude of small and medium actors operating downstream in San Pya.

Risk

The rapid growth of aquaculture in Myanmar has opened new opportunities for people previously excluded from fish trade; with them, new economic risks were introduced. In contrast to the trade of wild fish, controlled by a smaller number of well-established actors (Miñarro et al. 2016; Ruddle 2011), trade in farmed fish has opened up livelihood opportunities for laborers working for more established patron traders. After a few years of loyal service, some workers are able to secure the support of their patron to start their own business trading fish balls and/or consolidating and shipping farmed fish to new consumption hubs across the country (see Figure 5). The growing number of entrants, many of whom operate without a license, has led to strong competition and associated economic risks. In response, new practices have emerged that both mitigate these risks and/or shift them to others in the chain.

Figure 5.  

First, traders in San Pya have expanded the practice of selling on credit. It is common for large provisioning actors selling to major buyers such as supermarkets and international export-oriented processing plants to provide these privileged customers with secure and fast delivery of large volumes of fish and accept delayed payment (as elsewhere in Southeast Asia, see Boselie, Henson, and Weatherspoon 2003; Reardon, Timmer, and Minten 2012). However, selling on credit is also becoming common for smaller traders who are expected by their own buyers to offer the same service. As one fish ball processor described, “(selling on credit) used to be something you would only do with trusted customers, but if every other processor starts taking this risk, we end up having no choice … Nowadays most retailers expect to buy our product on credit.” She went on to explain how providing credit to new buyers has rendered her business riskier:

We lost a lot of money recently … that retailer bought a lot of our fish balls on credit and she never came back. There is another retailer who contracted a credit of USD 900 in only 2 months. I still deal with her but we ask her to pay every order cash … We must keep the business running because we have some outstanding debt ourselves.

The expanded practice of trade credit has enabled smaller traders to reduce their own cash flow risks when sourcing fish but, simultaneously, this has exposed them to new risks as credit providers. With increased competition for buyers in San Pya, these credit relations, once reserved for well-established trading partners and often secured through patron–client relations, have become marketized, with credit provided without a requisite level of trust or the kind of leverage that clientelism affords. However, compared with the older practice of output-tied credit that binds clients and their patrons together in exclusive relationships over long periods, the normalization of selling on credit has facilitated the accelerated circulation of capital.

Increased volumes of farmed fish in the wholesale market also mean that traders exceed their cold storage capacity more often than in the past, leading to spoilage. This is particularly true for smaller provisioning actors who tend to have more limited storage capacity. As a domestic distributor trading both wild and farmed fish explained, “(freshness) mostly concerns farmed fish because the volumes being traded are just too big and people always have some unsold or poorly stored fish that they try to palm off” (Figure 4). To mitigate this risk, provisioning actors in San Pya have developed more systematic purchase orders that make use of the new mobile banking infrastructure to optimize their storage operations. This involves, as one trader explained, carefully planning the orders to match their storage capacity:

A typical day starts the night before around 7pm when I start collecting orders from my customers over the phone. From 8pm to 9pm I usually go around the market and check for the best deals. Often, I would also call some of my customers to get additional orders based on the available supply. Until around 3am, I would then buy the fish according to my purchase orders.

Other traders reported completely bypassing the need for storage by further optimizing their use of mobile communication and shifting to order-based delivery. For instance, some distributors “do not buy fish unless they have a customer order” meaning that in practice they “immediately pack and dispatch upon receiving a purchase order.”

The growing trade volumes of farmed fish are also leading larger provisioning actors to engage in speculation. The expansion of aquaculture together with the development of mobile banking infrastructure means that more sales are transacted remotely. Taking advantage of new circumstances under which auctions are held 24/7, commission agents and brokers increasingly speculate on farmed fish supply, causing increased price volatility throughout the whole chain. The practice was described by a retailer in these words “(Farmed) fish price is set by the big guys according to supply and demand … At times when fish supply gets low, they typically purchase all of it before selling it back to us at a higher price.” From the perspective of the few large upstream actors who can afford to pull these strings, the advent of aquaculture is welcomed with great enthusiasm: “For my business, farmed fish is more profitable … The appeal of being a broker in San Pya is the capacity of making money quick and easily. Right now, I could easily generate USD 100 profit without doing much,” Yet, for the myriad of smaller actors downstream who endure associated price fluctuations, the picture is less cheerful. One small domestic distributor who had been in business for 17 years expressed a sense of despair: “even though I have been dealing with increasing volumes, my business is now barely making any profit.”

Discussion

Our results provide insight on how provisioning actors at the San Pya fish market affect and are affected by changes in the food system related to the transition from wild to farmed fish. By drawing attention to provisioning actors, we show how food systems are shaped by everyday practices related to sourcing, processing, buying and selling, and integrative social practices around the formation of quality, trust and risk (Figure 5). Following Bestor (2001), identifying how these practices emerge and evolve allows us to better understand food system dynamics.

We contend that conceptualizing food systems as a set of interrelated social practices offers opportunities to steer food system transformations in rapidly transforming low and middle income countries toward normative goals such as nutrition, sustainability or equity (Béné et al. 2020; Farmery et al. 2021). Drawing on the analysis above, we outline two broad ways in which this approach can help to reimagine food value chain dynamics and food system governance.

First, using a practices lens to study food systems in low and middle income countries such as Myanmar and, specifically, the “hidden middle” of trade, traders and trading, demands explicit recognition that value chains are more than the sum of sequential economic transactions; they are deeply socially embedded institutions. Building on the work of economic sociologists (e.g., Fine 2002; Granovetter 1985; Platteau 1994), this is an important corrective to largely asocial conventional accounts of value chains and food systems (e.g., GLOPAN 2016; HLPE 2017). Rethinking trade and trading in relation to the social construction of quality, trust and risk, and conceptualizing food provisioning as a bundle of practices, adds nuance to accounts of food systems that reduce actor conduct to the calculus of individual economic efficiency and rationality (e.g., Gabre-Madhin 2001).

Second, a practices lens reveals that provisioning actors in wholesale markets are active agents in shaping food systems. Evolving provisioning practices accommodate changes in the conditions of upstream production and downstream consumption by creating new meanings of quality, trust and risk, with the potential to reshape value chain structure and performance. Policy makers, development practitioners and researchers must be more cognizant of the roles of provisioning actors and their practices in shaping food systems outcomes, if they wish to design interventions that are effective in guiding food systems toward normative goals such as providing sustainable and healthy diets. As demonstrated in other contexts (Harris 2016; Lauer 2012; Simmons 2010), applied practice-based approaches can open up new ways of exploring situated processes of change, and potential context-specific and appropriate ways of addressing these.

Ethics statement

This study did not require ethics approval as no dilemna was identified by the Wageningen School of Social Sciences (WASS) committee who approved this research. We, authors, declare that we do not have any conflict of interest of any sort in publishing this manuscript.

Acknowledgments

This paper is an output of the CGIAR Initiative on Securing the Food Systems of Asian Mega‐Deltas for Climate and Livelihood Resilience (INIT-18), which is carried out with support from funders through their contributions to the CGIAR Trust Fund. For details please visit: https://www.cgiar.org/funders/. The paper also draws in research supported by the USAID funded “Food Security Policy Project” (Associate Award No. AID-482-LA-14-00003), and the Livelihoods and Food Security Trust Fund (LIFT) Grant Support Agreement Number: R1.4/029/2014 for the project “Agrifood Value Chain Development in Myanmar.” The contents of the paper are the responsibility of the authors and do not necessarily reflect the views of CGIAR, USAID, nor LIFT.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The reform period came to an abrupt and tragic end in 2021 when the democratically elected government was ousted in a bloody military coup.