Introduction

The effects of border closures on food trade during the COVID-19 pandemic are well documented in the media and to some extent in scientific studies (Aday and Aday 2020; Bouët and Laborde 2020; Montalvao and Van de Velde 2020; Price 2020). However, since primary data collection through physical presence in the field was constrained during the pandemic, few empirical studies document the impacts of border closures on the activities of informal cross-border food traders. This study is based on fieldwork conducted in autumn 2021– one and a half years after the Ghana-Togo border closed in March 2020 and half a year before it reopened in March 2022– and documents the impacts of the Ghana-Togo border closure on the livelihoods and food security of the cross-border fish traders, the majority of whom were women, as well as their strategies to deal with the difficult situation during the COVID-19 pandemic.

Given existing knowledge about the socially embedded nature of West African trade systems (Clark 1994, 2000; Walther 2015b; Overå et al. 2022), we expected to find that trust-based social networks played a crucial role in the cross-border traders’ crisis mitigating strategies. However, the long duration of the border closure and the severity of stress in terms of income loss and resultant food insecurity that this brought about provides a remarkably telling story about the importance of the still underrated qualities of the informal trade system (Hapke et al. 2023). By analysing how women traders’ activation of social relations worked in practice to manage a crisis, this study demonstrates how the social embeddedness of informal trade systems enhances their resilience. We aim to advance the understanding of informal traders’ importance for food security, both locally at the level of individual traders households and regionally at the level of consumers who rely on fish supplies across borders.

Fish represents a significant part of diets in West African countries such as Ghana and Togo (Ayilu et al. 2016; Hasselberg et al. 2020). The food source has considerable qualities in terms of nutrition (Béné and Heck 2005; FAO 2022). It is a source of high-quality bioavailable animal proteins, as well as micronutrients that are vital for infant and child development, particularly when eaten whole with the head, eyes, bones and viscera (FAO 2022). Whereas wealthier consumers can afford larger, more expensive fish, poorer households tend to buy small and cheap fish, which they often consume whole (Hasselberg et al. 2020). Fish consumption is therefore especially beneficial for food-insecure, mainly vegetarian households that cannot afford other protein-rich or nutrient-rich animal-sourced foods (ASF) (Kawarazuka and Béné 2011). In Ghana, fish represents 61% of Ghanaians’ expenditure on animal protein sourced food and provides 70% of their total animal protein intake (Hasselberg et al. 2020). Whereas in Ghana, the annual per capita fish consumption amounts to 27 kg, the levels of fish consumption are lower in Togo (about 12 kg per capita per year), but fish nevertheless contributes to 40% of total animal protein consumption (Ayilu et al. 2016). A large share of the fish consumed in Togo is imported from Ghanaian by cross-border traders. The traded fish is mainly harvested by Ghanaian small-scale fishers. The small-scale fisheries sector in Ghana is responsible for c.70% of the total domestic fish production of c.400,000 tons per year, of which about 70% of marine fish landings consists of small pelagic fish such as mackerel, sardines and anchovies (FAO 2016).

In Ghana, trade in food items (including fish) is mainly part of the informal economy in the sense that activities are largely unregistered and unregulated (Hansen and Vaa 2004), although traders pay fees when selling their fish in marketplaces, which generates revenue for public or private market administrations. Unlike other business activities, trade requires few investments, and can be undertaken by unskilled labourers (Béné and Heck 2005). In particular, women’s engagement in trading activities represents significant opportunities for generating incomes and securing livelihoods (Ayilu et al. 2016; Ayilu and Nyiawung 2022).

Of the 2.6 million people in Ghana whose livelihoods are supported by the fishery sector (roughly 10% of the country’s population), the majority are women who work in fish processing and trade (FAO 2016). In common with other types of trade in the country, fish trade is considered a gender-appropriate way for women to fulfil society’s expectations of them making an income and contributing to the household economy, in particular by providing food for their children (Clark 1994; Overå 1998; Overå et al. 2022). As Kawarazuka and Béné (2010) argue, households in which a wife or mother engages in fish processing or trading spend a large share of their income on food. Hence, when women’s fish trade is negatively affected, their household’s food security is threatened (Béné and Heck 2005). Thus, through their fish distribution activities to populations across borders, cross-border fish traders make a direct contribution to the populations’ nutrition from fish consumption, as well as an indirect contribution to their own food security through earning an income that enables them to improve their households’ diets (Kawarazuka and Béné 2010).

This study documents how COVID-19 pandemic related restrictions at the Ghana-Togo border impacted the activities of cross-border fish traders. We analyse how the traders’ capacity to activate and extend gendered trust-based trade networks across the closed border enhanced the resilience of the cross-border fish trade system, which to some extent cushioned the adverse effects of the two-year long border closure. We argue that the social embeddedness of traders’ informal economic activities strengthened the resilience of the cross-border fish trade system during the pandemic, which in turn enhanced the food security of the traders facing the crisis.

This paper is divided into seven sections. The second section sets the context for the study and presents the study area. The theoretical background is detailed in the third section , and then our methods are presented in the fourth section. The fifth section describes the findings of the study, which are discussed in the sixth section. We conclude this paper in a seventh sectionion by summarizing the main findings.

Cross-border fish trade between Ghana and both Togo and Benin

Flows of fish from Ghana to Togo and farther east to Benin are significant, with an estimated annual market value of USD 18.6 million (Ayilu et al. 2016). From the Denu market (7 km by road from the Ghana-Togo border, see Fig. 1 in Sect. The study area), an estimated 1851.52 tons was traded annually from Ghana to Togo and 819.21 tons to Benin (Ayilu et al. 2016). These trade volumes are undoubtedly underestimated, as cross-border fish trade (hereafter abbreviated as CBFT) is often unrecorded and overlooked by governments and organizations because of the informality of transactions (Ayilu et al. 2016). As a result, fish rarely features in trade policies in West Africa in comparison to other food products, and challenges faced by fish traders such as informal tariffs persist.

Most cross-border female traders engage in low-profit and highly competitive trade in agricultural and food products and they tend to sell their products in marketplaces or directly on the street (Desai 2009). Cross-border trade transactions take place through close person-to-person contact and most still involve cash-based payments (Luke et al. 2020).

Within households in West Africa, women are responsible for educational, health and dietary matters (Bennett 2005). Cooking and care duties may therefore prevent them from leaving home to engage in CBFT for a long period. Alternatively, the women may experience stress when they are away and wish to return home to their children. However, many women are assisted by female kin or other household members, which enables them to travel (Overå 1998; Desai 2009; Overå et al. 2022).

The study area

The Ghana-Togo border was selected as the study area (Fig. 1) because many women from communities on either side of the border engage in CBFT on a regular basis, with traders from Ghana mainly travelling to Togo and Benin, and to some extent to other countries in the region, such as Burkina Faso and other Sahelian countries (MoFA 2022). Although traders in focus in this study engaged in fish trade across borders, most of them could be characterized as small-scale fish traders as they lived close to the border and did not travel long distances.

Fig. 1
figure 1

The main study area in Ghana and the Ghana-Togo border area. Specific locations of fishing communities where interviewed traders lived are omitted to ensure the anonymity of the interviewees

Full size image

The border between Ghana and Togo was arbitrarily established in 1890 by colonial powers in the middle of Ewe settlementsFootnote 1, dividing the Ewe people between both sides of the border (Akyeampong 2001; Nugent 2021). High profitable tariffs for the British colony of the Gold Coast (now Ghana) coupled with lower customs duties in German Togoland (now Togo) promoted smuggling into the Gold Coast and enabled cross-border trade to develop in the area. The historical relationship between the Ghanaian border town Aflao and Togo’s capital city Lomé, located adjacent to the border with Ghana (see Fig. 1), was therefore ‘fundamentally shaped by trading opportunities’ (Akyeampong 2001: 150).

Today, both the Denu market on the Ghanaian side of the border and the Assigamé market in Lomé are highly frequented by traders who have travelled long distances from Ghana, Togo, Benin, Nigeria, and Côte d’Ivoire (Bewiadzi Akakpo 2021; Nugent 2021). Situated on the Abidjan-Lagos corridor, the Aflao checkpoint on the Ghana-Togo border experiences heavy trade flows in all types of commodities (Bewiadzi Akakpo 2021). Such national borders stimulate intra-regional trade through incentives due to price differences between countries and import/export bans (Walther 2015b), and this is also the case for the Ghana-Togo border (Bewiadzi Akakpo 2021). While most cross-border traders use the main border crossing site at Aflao, some use unofficial crossings (footpaths called ‘beats’) located in neighbourhoods north of Aflao, in order to pay lower border crossing fees (Bewiadzi Akakpo 2021).

Most of the fish traders from either side of the border between Ghana and Togo share an Ewe ethnic identity and language (Akyeampong 2001; Nugent 2021). The traders who participated in our study mainly originated from fishing communities in south-eastern Ghana, in the Volta Region close to the border with Togo, which is the home of the Anlo-Ewe, a coastal subgroup of the Ewe (Akyeampong 2001; Nugent 2021). The Anlo-Ewe people constitute one of the main ethnic groups that engage in coastal fishing in GhanaFootnote 2 (Overå 1998; Akyeampong 2001; Kraan 2011). In common with coastal marine small-scale fisheries in Ghana, Anlo-Ewe fisheries are rooted in a gendered division of labour where men engage in fish harvesting whereas women typically occupy roles in fish processing and trading. Once the fishers have sold their fish on the beach, the fish ‘enter a female domain’ (Overå 1998: 101). Female wholesalers, who usually are canoe owners’ wives (doto srowo) and/or creditors, are entitled to sell the catch from the canoes to female fish processors and/or fish traders (Overå 1998: 245). Women trade in fresh, dried, smoked, fermented, salted, or fried fish. These processing methods allow for conservation that does not necessitate refrigeration (Belton et al. 2022). For example, the shelf-life of smoke-dried anchovies– popularly known as ‘Keta school-boys’ in Ghana, indicating that there are many of them in the area around Keta (Fig. 1)– can be up to one year (Overå et al. 2022). This makes processed fish a highly tradeable item that is affordable and practical to use for low-income consumers, who often do not have refrigeration facilities at home.

The COVID-19 pandemic outbreak and border closure

The outbreak of the coronavirus COVID-19 pandemic disrupted CBFT activities in many ways. Ghana started to implement measures to contain the pandemic on 15 March 2020 and the first step was a ban on public gatherings for four weeks (Asante and Mills 2020). Marketplaces were considered essential and therefore were not closed, but they were thoroughly disinfected. One week later, on 22 March, the Ghanaian government decided to close all land, sea and air borders to human traffic (Asante and Mills 2020). Goods (e.g., foodstuffs), supplies and cargo were not affected by that measure. Transport of food across the border by trucks was therefore still possible, but traders (or any other people apart from truck drivers) were not permitted to cross the border in vehicles or on foot. Containment measures in Ghana were particularly strict for a two-month period spanning March and May 2020 (Atkins et al. 2021). The various restrictions in the country were gradually lifted in June and the following months (Frimpong 2020), but the national land border remained closed. Officially, travellers could only enter Ghana through the Kotoka International Airport in Accra after 1 September 2020. In Togo, policy measures to limit the spread of the coronavirus did not include a lockdown but did include a curfew from 8 p.m. to 6 a.m. in the capital city of Lomé between 2 April and 8 June 2020, as well as a closure of all meeting places (Togo Revenue Authority 2020; Global Monitoring 2023).

After two years of closure, the Ghana-Togo border reopened on 28 March 2022 (Dossavi 2022). We conducted data collection for our study from August to October 2021, when the border had already been closed for one and a half years and remained closedFootnote 3. Despite the fish traders’ many coping strategies, the effects of border closure on their food security were starkly noticeable, as we show in the findings.

Conceptual framework: social embeddedness, agency and resilience of the CBFT system

Resilience is an important concept in the literature on informal trade systems (Clark 2000). Informal traders’ agency, strengthened by information sharing and trust cultivation in social networks (Clark 1994; Walther 2015b; Overå et al. 2022), contributes to making cross-border trade systems more resilient by reinforcing their capacities for adaptation and flexibility (Clark 2000).

Informal trade networks are well known to be resilient and they have been crucial for the distribution of food supplies during and after food crises (Walther 2015b; Olusola and Lere 2020). In development literature, resilience is broadly defined as ‘the ability to draw upon a set of capacities to deal with shocks and stressors before, during, and after a disturbance, in a way that maintains or improves well-being outcomes (such as food security or adequate nutrition)’ (Bryan et al. 2023: 240). Awareness of the local context is fundamental to resilience approaches in order to make sense of existing capacities and risks as well as local practices, opportunities and institutions. In the context of trade, Clark (2000) argues that the resilience characteristic of the market trade system emanates from the flexibility enabled by its gendered construction. As female informal traders operate on a relatively small scale, usually not facing strict regulations, and exchange market information extensively (Overå 2006), they have the capacity to ‘respond rapidly and creatively to the new opportunities’ (Clark 2000: 255).

For the majority of people in Africa informal cross-border trade is considered more important than formal markets for improving food security, both in terms of volume of traded goods and employment opportunities (Desai 2009). Trade networks are based on and developed through birth, kinship and ethnicity (Walther 2015b), as well as trust established through collaboration with trade partners over time. Such collegial relationships are often referred to in terms of friendship (Overå et al. 2022). Traders also often transfer their trade-related knowledge, skills and starting capital to their younger kin (Desai 2009). Client networks determine how successful a trader is; having regular and loyal customers helps to make a trader’s business more lucrative (Wrigley-Asante 2008). Studies have shown how communications technology, particularly mobile phones, has strengthened social ties between traders, as well as between traders and customers, whereby trust can be cultivated and the risk involved in transactions that take place over long distances (and across borders) can be reduced (Overå 2006; Wrigley-Asante and Agyemang 2019).

Cross-border trade in West Africa involves ‘distant ties developed with foreign partners from a different origin, religion or culture’ in addition to relying on ‘trust and cooperation among local traders’ (Walther 2015b: 603). Social embeddedness theories originally suggested that social networks and proximate relations between actors positively affect economic activity (Granovetter 1985). More recent research has distinguished between two forms of social networks as structures of relationships, namely embeddedness and brokerageFootnote 4 (Burt 2005; Walther 2015a). Whereas embeddedness indicates the closeness of relationships between actors within a given group and who frequently interact, brokerage refers to relationships between actors of different groups. In the context of cross-border trade, social embeddedness materializes as social linkages between a trader and other actors in the same group or community, while social brokerage translates into ties between a trader and actors in other communities and countries. Balancing interaction through internal and external ties can improve a business if a trader both cultivates safety through trusted close-knit relations and builds relations of trust and cooperation with foreign partners, exchange with whom is more risky but potentially more economically rewarding (Kuépié et al. 2016). Embeddedness and brokerage through social networks thus facilitate access to credit, price information, and insight into the truthfulness of clients and suppliers (Walther 2015b). Interacting frequently with peers and clients builds trust and reduces the risks of a trader incurring losses. Social networks are particularly important when formal institutions cannot provide security and traders have to rely on relations of trust to conduct their business (Odera 2013). Trust in that sense is understood as an informal institution in which the different agents involved are expected to behave in certain ways and fulfil specific obligations. Interpersonal trust is therein the basis for social networks, which serve as a form of social capital that later can be converted into economic capital (Bourdieu 1986).

The agency of fish traders is to a large degree conditioned by the social embeddedness of their economic activities, and it contributes to their resilience. Kabeer (2021) defines agency as the capacity of individuals to act independently and to make free choices within the constraining or enabling structures that constitute and govern their livelihood strategies. In our study context, the traders’ agency was enabled by societal gender norms whereby trade is considered as an appropriate economic activity for women (Clark 2000; Overå 2007). Moreover, social embeddedness in the form in which social interaction (mainly between female traders) is central in the trade process, fosters agency through information sharing and the development of knowledge, skills, and trusted partnerships (Clark 1994; Overå et al. 2022).

Although female fish traders are able to exhibit a high degree of agency within the woman-dominated informal trade system, they are frequently constrained by gender discrimination (Overå et al. 2022). In fisheries, gender norms and power dynamics often limit women’s participation in decision-making and leads to their ‘limited organizational capacity and unequal access to usable assets, technology, finance, education and services’ (Zelasney et al. 2020: 2). Such gender constraints shape fish traders’ ability to sustain a livelihood and their households’ well-being, which also affects the agency of fish consumers, as their availability and access to nutritious fish is reduced (Overå et al. 2022). People with little agency lack the tools to manage potential risks, which can lead to their vulnerability to food insecurity (FAO 2008). Women have been found particularly vulnerable to food insecurity because gender inequalities tend to deepen during food crises (FSIN and Global Network Against Food Crises 2020). Within households, women are the first to limit their food rations so that their children and old members of the household can eat. Even in normal times, women consume less fish than men on average despite their higher micronutrient needs (Kawarazuka and Béné 2011).

While domestic traders face many challenges, such as lack of available and affordable transportation and low access to credit, cross-border fish traders face additional challenges when confronted with borders (Desai 2009; Wrigley-Asante and Agyemang 2019) which influences their agency. Rigid border controls and formal trade policies are ill-suited to the informal nature of cross-border traders’ business (Olusola and Lere 2020). Cross-border traders often face particular constraints associated with power relations at the border, involving actors who control the border by delimiting or managing it (Newman 2017). A border is a complex legal, cultural, social, and economic concept and institution (Mezzadra and Neilson 2013). Borders ‘regulate and structure the relations between capital, labor, law, subjects and political power’ through processes of blocking, confrontation, contact and passage (Mezzadra and Neilson 2013: 8). As a social institution, borders are characterized by tensions between practices of border crossings by individuals and border reinforcement by states.

Borders and cross-border trading networks are key components of the relationship between people and the state (Kusakabe et al. 2008). Borders constitutes fuzzy spaces because they are located between two states with different policies and different social and cultural norms (Kusakabe et al. 2008). Borderlands are considered dynamic when they take advantage of political, economic and fiscal disparities between states and are considered as partly exempt from public authorities’ control (Igué and Zinsou-Klassou 2010). On the one hand, the fuzzy space created gives room for people to manoeuvre and negotiate; on the other hand, it makes people more vulnerable to abrupt changes in border regulations (Kusakabe et al. 2008).

Different people are treated differently at borders depending on social, cultural, religious, linguistic and economic factors, which influence either positively or negatively the disparities between those people (van Houtum and Bueno Lacy 2017). People’s characteristics (e.g. skin colour, accent, material possessions) are politically interpreted by border officials and can lead to ‘expected, manageable, or even desirable elements of socialization [… or…] obstacles for people’s spatial and social movement’ (van Houtum and Bueno Lacy 2017: 2). For example, in West Africa, belonging to a specific ethnic community can facilitate the border crossing process, depending on how well the community is perceived in the country entered (Igué and Zinsou-Klassou 2010). Through their dual nature of connecting and dividing people, borders thus create differentiated forms of sociability and vulnerability (Mezzadra and Neilson 2013).

Borders are ‘technologies of rule to regulate a spatially delimited population of […] workers’ by states in what Campbell (2018: 7) describes as ‘border capitalism’. However, state regulations remain unable to grasp the particularity of regional labour regimes, resulting in context-insensitive policies (Campbell 2018). As borders are sites of state control, their governance is not always accepted by the affected populations and can give rise to border struggles and conflicts, notably in former colonial countries (Igué and Zinsou-Klassou 2010). Such tensions can reveal a tendency for borders to be tightly controlled and become rigid, which can result in the shutting down of avenues of possibilities and negotiation for people who wish to cross them (Mezzadra and Neilson 2013).

In this paper, we demonstrate that although the cross-border traders’ agency was impacted during the COVID-19 pandemic, they were able to mobilize trade networks and develop new social ties that enabled them to maintain a certain flow of fish across the closed border. We analyse how social embeddedness and brokerage to some degree enabled the traders to adapt to the abrupt constraints, which contributed to making the CBFT system more resilient and reduced their vulnerability to food insecurity.

Methods

Empirical data for this study were collected during eight weeks of fieldwork conducted by the first author from August to October 2021 in Ghana. The border closure made it impossible for her to enter Togo to do fieldwork there too, though it was possible to interact with participants crossing from Togo into Ghana. The study area consisted mainly of five rural coastal fishing communities in the Volta Region in south-eastern Ghana. Fieldwork was also conducted in the Denu market and in the area surrounding the border checkpoint in Aflao.

Qualitative and quantitative data were collected mainly through a survey and in-depth (follow-up) interviews. Observation and informal field discussions were used as additional methods. Secondary data were gathered before and after fieldwork to provide a context for the study and to validate the primary data obtained during fieldwork. We present quantitative data to reveal the background profiles of the traders and to evaluate the different impacts of the border closure during the COVID-19 pandemic, whereas the analysis of qualitative data reveals traders’ personal views, experiences, and strategies to deal with the crisis.

In total, 34 participants were interviewed (Table 1). Of the 34 participants, 30 were involved in the survey, while the remaining 4 participated in a group interview. A total of 26 Ghanaian traders were interviewed in fishing communities, in most cases on the beach, which is an important meeting place for traders and other fisherfolk; additionally, 8 Togolese traders who had entered Ghana to purchase fish were interviewed in the Denu marketplace and in one of the fishing communities. The traders were selected through snowball sampling, which also contributed to us to gain insights into their social networks. The selection criteria were that they were cross-border fish traders and had started their business before the pandemic. Most interviews were conducted with the assistance of an interpreter, who was a fisher from the area. The breadth of his social relations was of substantial help in identifying traders who fulfilled the selection criteria. Togolese and especially Beninese traders were harder to get in touch with than Ghanaian traders as they only travelled to Denu on market days (every fourth day), and often they had little time available for an interview as they needed to return to their respective home countries as soon as possible. Local Ghanaian traders were helpful in recruiting available foreign traders for an interview.

Table 1 The sampled traders by country and community of residence, location of interview, and their participation in follow-up interviews
Full size table

A survey using a semi-structured questionnaire was conducted among 30 of the 34 participants. The aim was to acquire information about the characteristics of cross-border traders, the impacts of the pandemic on their life and business activities, and their coping strategies towards CBFT since the border had closed approximately one and a half years earlier. The survey interviews lasted 20–60 min, depending on how comfortable or talkative the trader was and how much time she could spare. A group interview was conducted with four traders, one of whom was the local representative of a fish trader association. Nine participants from the survey were contacted for follow-up interviews; based on their answers to the survey questionnaire, an individual follow-up interview guide was created for each of those nine traders. Additionally, three of the four participants from the group interview were involved in a follow-up group interview. The follow-up interviews lasted about the same amount of time as survey interviews. All follow-up interviews were recorded (the survey interviews were not). Notes were taken during all interviews.

All of the interviewed traders were women of Ewe ethnicity who resided in either Ghana or Togo. Of the 30 survey participants, 4 were in the age group 20–35 years, 17 in the 36–50 years age group, and 14 in the 51–65 years age group; 26 were married and 5 were widowed. With regard to their education level, 18 traders had not received any schooling, 9 had attended primary school, 2 had attended junior secondary school, and 1 had a university education. The traders’ households generally included many members (Fig. 2a and b).

Fig. 2
figure 2

(A) Survey participants’ number of children. Source: Fieldwork, 2021. (B) Survey participants’ number of household members. Source: Fieldwork, 2021

Full size image

Although a few of the Ghanaian participants spoke some English, all interviews with Ghanaian traders were conducted in Ewe, the participants’ mother tongue, and translated into English by the interpreter (i.e. the fisherman). As the field researcher spoke French, the Togolese traders seemed more comfortable engaging with her directly in French.

Oral consent was obtained at the beginning of each interview. Participants were made aware of the nature and purpose of the study and that they could withdraw from the interview at any time. They were also informed that their names and the names of fishing communities would be anonymized to ensure their privacy. The participants’ consent was also required before photos were taken or interviews were recorded. Most participants were very interested in participating in the study and none of them declined to be interviewed or withdrew from the study. Ethical considerations were, however, carefully considered due to the harsh situation the traders faced as a consequence of the border closure. Some participants showed vulnerability and the field researcher thus avoided questions that could be perceived by the participants as sensitive or make them feel uncomfortable, especially regarding their (lack of) food intake.

Additional methods included observation and informal field discussions. Observation was conducted on the beaches where the caught fish were landed, before and/or after interviews, inside/outside some of the traders’ houses, as well as on Aflao beach in the vicinity of the border checkpoint. Informal discussions with people in the different communities and with the interpreter helped to deepen our understanding of the context and processes under study.

The quantitative data produced by the questionnaire were entered and analysed using Microsoft Excel. The qualitative data obtained by from the survey interviews and follow-up interviews were coded and categorized using NVivo 12.

Findings

The findings depict the many ways in which the crisis caused by the COVID-19 pandemic affected CBFT. In this section we show how the traders’ pre-existing social networks within their communities and across the Ghana-Togo border, although weakened in during the crisis, enabled over half of the interviewed traders to reach fish customers across the border through effective communication and mobilization of collaborators. Through adaptation of border crossing practices, they were also able to maintain a minimal level of income generation during the pandemic.

Characteristics of the cross-border fish traders

Of the 30 traders who participated in the questionnaire survey, 14 had started fish trading at a local market in their home country (e.g. in Denu or Keta in Ghana, and in Tsevie or Vogan in Togo, see Fig. 1) before engaging in CBFT after three to ten years in the business. Half of those traders had started directly in CBFT, while one trader had started both local trade and CBFT simultaneously.

The main reason for engaging in CBFT was economic attractiveness, which was mentioned by 15 of the survey participants. Exchange rates between the Ghanaian Cedi and the CFA Franc in Togo often favour trading in CFA currency. Furthermore, fish prices in Togo tend to be higher, as Togo is not a considerable fish producer. Thus, it is profitable to bring fish from Ghana to sell in the Togolese market. An overview of the average quantities that participants reported normally traded per trip across the Ghana-Togo border before the border closure is provided in Fig. 3, which shows that most of them traded fish in small to medium quantities, in units designated as baskets or barrelsFootnote 5.

Fig. 3
figure 3

Average quantities of fish traded by cross-border fish traders per trip before the COVID-19 pandemic.

Source: Fieldwork, 2021

Full size image

For 6 of the 26 Ghanaian traders who participated in the survey, a lack of customers or space in their local markets was evoked as a reason to trade abroad, usually in Lomé (Fig. 1). Many other reasons were cited, including taking over a relative’s cross-border trade business, having relatives in the destination country, and the advantages of accessing bigger markets in Togo and Benin, where they could also purchase household and food items cheaper than in Ghana.

Traders commonly defined themselves as being the head of their household, even when they were married, which was the case for 27 of the 30 survey participants. In total, these 27 household heads provided for 205 household members. Being the head of a household implied buying food, paying school fees and hospital bills, and taking care of children and domestic tasks. Their husbands often contributed, but not to the same extent. Three of the traders said their husbands were in charge of the aforementioned tasks, and therefore they did not define themselves as being the heads of their households.

Purchasing and preparing fish before crossing the border

The organization of pre-trade processes in Ghana was not affected by restrictions related to the COVID-19 pandemic. All 26 Ghanaian traders continued to purchase fish from fishers or wholesalers on the beach. They processed the fish themselves by sun drying and/or smoking, or more rarely by frying, which they often did together with other fish processors. Togolese traders typically purchased processed fish in the market in Denu. While most of the interviewed traders had their own businesses, one of the Ghanaian traders worked as a representative for a wholesaler’s business. She was one of two trusted representatives who were sent to the market, the other one being the wholesaler’s own daughter.

Small pelagic fish species were the most traded fish; 29 traders involved in the survey regularly traded in anchovies and 25 in sardinellas. Other commonly traded fish species were barracuda, tuna, and mackerel. More rarely mentioned fish species were rainbow runner, plain bonito, and shrimps. As fish were rarely available in large quantities, traders often purchased whatever species were caught by the fishers. Almost all interviewed traders mentioned that the 2021 fishing season had not yielded many fish. The main fishing season is normally in July and August and thus, in 2021, it preceded our period of data collection by a few weeks. A 59-year-old Ghanaian trader stated: ‘When we have fish in abundance, we keep some to sell later. This year, there are no fish in the house to sell later because the fish are not available’. Moreover, when catches are low, the available fish becomes expensive.

The quality of processed fish is an important factor for trade. Well-processed fish lasts longer and looks more attractive. Wholesale traders thus particularly depended on ‘the beauty of the fish’ to make sure their customers would buy in bulk.

Fig. 4
figure 4

(A) Baskets at the Denu market in Ghana, containing smoked and dried anchovies and sardinellas, which have been bought by Togolese and Beninois traders. Photo: Maïlys Rouillé, 2021 (B) Fisherman on a canoe holding a barrel used for transporting fish from the canoe to the landing site. Photo: Maïlys Rouillé, 2021

Full size image

After processing, traders packed their fish in either baskets or barrels (Fig. 4a and b), depending on the type of packaging used locally, and lined the containers with brown paper before filling them with smoked and sundried fish. Barrels are normally used by fishermen to transport fish from canoe to land, and in some communities, they may also be used by traders for fish packaging. Since barrels are bigger than baskets, it is difficult to compare volumes of fish between the different types of packaging.

Apart from fish purchase and fish processing, all other activities related to CBFT, such as communication, border crossing and travel practices, were strongly affected by restrictions due to the pandemic.

Foreign markets targeted

Before the COVID-19 pandemic, the majority (26) of the 34 interviewed Ghanaian and Togolese cross-border traders normally sold their fish in Lomé. The main markets they mentioned were Assigamé (‘Big Market’), Atikpodji, Agoe Assiyeye (‘New Market’), Kodomé, Adidogomé, and Hanoukope (Fig. 5). When on a trading trip across the border, the traders usually left home early in the morning and returned home the same evening.

Fig. 5
figure 5

Destination markets of cross-border fish traders in Lomé, Togo, before the COVID-19 pandemic.

Source: Fieldwork data 2021 and Google Maps

Full size image

Trading trips to the large market Comé in Benin (Fig. 6), which was a destination for two of the Ghanaian traders, normally took two days. Their journey started at 4 a.m., crossed the border around 8 a.m. and arrived in Benin at 5 p.m., where their customers were waiting for them. To save time, they slept together with other traders in the truck in which they travelled, and left at dawn the next morning to reach home around 7 p.m.

One of the Ghanaian traders targeted more distant inland markets in Togo, such as Kpalimé, Glei, Datcha, Anie, and Atakpamé, and even sometimes Dapaong in Northern Togo (Fig. 6). Her trips lasted between three and seven days, depending on the quantity of fish traded and anticipated profits in the different markets. She travelled from market to market and if fish prices were low in the first market, she would buy more fish there and travel to another market where prices were expected to be higher. Having relatives in Togo and a family house in Datcha made it easier and cheaper for her to stay abroad longer.

Fig. 6
figure 6

Destination markets in Togo and Benin targeted by cross-border fish traders before the COVID-19 pandemic.

Source: Fieldwork data 2021 and Google Maps

Full size image

During the border closure, the few traders who were still engaged in CBFT activities continued to trade in the same markets as before the pandemic. Many of them spent short periods abroad as before the pandemic, but some had to stay longer since they could not travel frequently and the border crossing process took more time and was more expensive than before the pandemic. It was difficult to estimate the quantities of fish they traded during the border closure because the quantities varied more than before the pandemic. However, some study participants reported that they were able to trade at best half of the number of baskets/barrels compared to before the pandemic.

Border crossing before and after the border closure

The traders said that before the COVID-19 pandemic, they had not faced any difficulties or challenges related to the crossing of the Ghana-Togo border. Traders had regularly crossed the border for CBFT purposes, usually two or three times per week in the case of the Ghanaian traders, while the Togolese traders mainly travelled to Denu on market days (every fourth day). Cross-border fish traders often travelled together with other traders from their local community. One elderly trader employed a representative to trade in her place, as she needed time to train her daughter in processing and trading practices to enable her to take over the business. Traders from Ghana travelled to markets in Togo by either taxi or rented car; those who travelled longer distances to Benin used bigger vehicles, such as pick-up trucks. Togolese traders used shared taxis, trucks and even motorcycles. The official border checkpoint in Aflao had been used by 26 of the 30 survey participants before the pandemic. Three traders said that before the pandemic they had also regularly used unofficial crossing routes up to 10 km north of Aflao.

When the Ghana-Togo border closed on 22 March 2020, most of the 34 interviewed traders ceased CBFT activities altogether; only 7 of them attempted to cross the border during the following weeks. Many were afraid of becoming infected by the coronavirus while travelling and visiting crowded markets, while others suspended their CBFT activities because of the frequent harassment and beatings of those who attempted to cross the border by some border officials at unofficial border posts.

When suspending their CBFT activities, 12 of the survey participants resorted to selling fish only in local markets within their home country. This alternative was not available to all, since one of the reasons they engaged in CBFT in the first place was the lack of space in their local marketplace (e.g. Denu market). Those who were unable to trade in the local market had few income generating alternatives. Two Ghanaian traders managed to send fish across the border with the help of truck drivers who were allowed to cross the border at the official checkpoint. Seven of the interviewed traders were out of work for several weeks after the border closure, with no income at all.

Of the seven traders who did not suspend their CBFT activities, three crossed the border a couple of times, while the other four continued to cross the border regularly during the first three weeks of border closure. The official border checkpoint in Aflao was only open to people traveling in a vehicle carrying goods in bulk (i.e. truck drivers and traders accompanying their goods), and border fees at Aflao increased after the border closed. To avoid this cost, most traders had to find new routes some kilometres farther north where they could cross the border on footpaths: ‘Since the lockdown came, you can’t pass the official border, but there are ‘beats’, illegal routes, to enter’ (Ghanaian trader, 59 years). Since ‘beats’ were not actual roads but footpaths in the bushes, traders would travel with their goods to an unofficial border crossing point in a vehicle, then unload their goods and cross by foot. Traders were assisted by carriersFootnote 6 when transporting their fish from one side of the border to the other. Some traders prearranged for a transporter to pick them up at the border to take them to the marketplace, while others, who lacked such social connections with truck or taxi drivers, had to find a vehicle themselves. This was not too difficult, as taxi drivers were aware of the use of ‘beats’ and often waited by the border in anticipation of traders, though finding transport was sometimes time-consuming. Instead of traveling with their goods, traders alternatively sent them with a truck driver through the official point at Aflao, while they themselves travelled separately along a ‘beat’.

Violence inflicted by border officials on traders who attempted to cross the unofficial crossing points culminated during the first weeks of the border closure. Gradually, as attempts at crossing the border became more common, the violence reduced. However, on the Aflao beach, where many people (including some of the interviewed fish traders) walked across between Ghana and Togo, the violence continued; the beach is located immediately behind the border control post, and people on the beach are often in sight of border officials. There, traders had to run c. 200 m to leave the border area quickly and avoid meeting officials:

We take a taxi to Aflao and on the beach, we have to walk around 200 m with the [fish] load that is really heavy. If people start saying ‘akayi akayi’,Footnote 7 we have to run back because it means that the border official who controls that area is coming. We will get arrested, [so] we must hide. (Ghanaian trader, 48 years)

As border crossing conditions improved over time, more and more traders resumed their CBFT activities (Fig. 7). While 14 of the 34 traders reported trading once or several times a week, a similar number of interviewed traders had still not resumed their CBFT 18 months after the border closure.

Fig. 7
figure 7

Frequency of border crossing 18 months after the closure of the Ghana-Togo border in March 2020.

Source: Fieldwork 2021

Full size image

Before the pandemic, traders crossing the Ghana-Togo border were accustomed to paying a ‘border fee’ of around 10 GHSFootnote 8 or more to border officials: ‘The more fish you carry, the more expensive [the fee]’ (Ghanaian trader, 50 years). The unofficial ‘fee’ can be characterized as a bribe. After the border closed, some border officials demanded the double amount and crossing in one direction cost on average 21 GHS/2000 CFAFootnote 9. The ‘fee’ was still not standardized and depended on several factors: the quantity of fish traders carried, the trader’s personal ability to bargain, their familiarity with individual border officials, and other subjective criteria decided by the border officials. One participant recounted the following:

When you cross, you have the fish in baskets, usually small fish like anchovies and herrings. The [customs] official sees fish, he knows the trader and her business, and he asks her to pay. Sometimes it’s expensive, so we have to bargain […]. But we struggle bargaining, it takes a lot of technique. Border officials recognize traders, and traders also recognize border officials. […] We can ask who is there, but we feel threatened if the person working is not one that we like. (Ghanaian trader, 50 years)

The border ‘fees’ were expensive, but as one interviewee said, ‘We cannot waste time negotiating how much [we] pay at the border, [otherwise] the goods might get lost. You have to follow the carriers so that your goods do not get lost’ (Ghanaian trader, 61 years).

The main border checkpoint in Aflao was open for fish traders if they accompanied their goods in the same vehicle but due to the high bribes required it was more expensive than crossing at unofficial sites, and traders thus increasingly resorted to using the ‘beats’. In general, the inflated border ‘fees’ made traders reluctant to engage in CBFT activities altogether. Traders who operated through truck drivers and did not accompany their goods themselves were less affected by the inflated ‘fees’. One Ghanaian trader (61 years) explained that, ‘The vehicle owner has some sort of arrangement with the border officials.’ Togolese traders in Ghana hired truck drivers to carry their goods from the Denu market to Togo through the official border checkpoint.

The experience of harassment by border officials became both more common and severe after the border closure, especially during events of ‘border fee’ payment. According to one trader, many officials regarded female traders as inferior due to their gender. The officials, most of whom were male, therefore often took advantage of their position of power and harassed women, depending on the degree of their collaboration over illegal border crossing. However, being acquainted with border officials could reduce the risk of traders facing high bribes and harassment. Likewise, if the border official was an Ewe, sharing Ewe as a common language facilitated the communication and relationship: ‘We are all like one family’ (Ghanaian trader, 59 years).

The required payment of elevated bribes, which for female traders can be characterized as economic abuse enforced through male aggression by government officials, represented a barrier to border crossing during the pandemic. The cross-border traders’ only available strategy to surmount this barrier was to activate their social networks and social skills to sustain at least a minimal level of business during the crisis.

The importance of maintaining social relations to continue trading

Despite making very meagre profits, traders simply had to continue their activities, as they could not afford to stop trading altogether: ‘As the head of household you need money to eat, [so] you have to find a way’ (Togolese trader, 55 years). To avoid wasting money while travelling, it was important for the cross-border traders to cultivate social relations in order to access information about fish availability and market prices of fish, and to ensure that customersFootnote 10 showed up at the market as prearranged. However, many traders had limited social networks, especially outside their local community or country. Of the 30 traders who participated in the survey, 18 did not have relatives or close contacts abroad before the COVID-19 pandemic, 5 reported having relatives abroad, and 8 mentioned having contact persons abroad. Contact persons were mainly regular fish customers or fellow traders, both local and foreign, who travelled to the same markets (in the case of Ghanaian traders) or, in the case of Togolese traders, they were other traders from whom they regularly purchased fish. After the border closed, it proved challenging for several of the traders to maintain their relationships with their contacts. Face-to-face communication, and, in the first couple of weeks after the border closed, mobile phone communication, occurred more rarely between traders across the border, and this led to more uncertainty concerning current prices and demand for fish in the market.

During the pandemic, fish buyers were reluctant to visit marketplaces due to the risk of infection in crowded places. Their expectations of fish availability in the market also influenced their attendance: ‘When customers hear that there is fish, they show up at the market. If they hear that there’s a lack of fish, they don’t come’ (Ghanaian trader, 39 years). Since their regular customers no longer attended the market, many traders had to find new fish buyers elsewhere. As a result, the cross-border traders developed and perfected their communication and trading methods. In particular, the use of mobile phone communication increased in an attempt to bridge the gap in face-to-face communication between cross-border traders and customers. Calling customers before travelling to negotiate and agree on the price, quantity, and where to discharge the fish therefore became much more common. The initial interaction with new customers was, however, only possible through face-to-face communication in the marketplace (after having crossed the border), as customers needed to evaluate the quality of the fish before buying it. Exchanging phone numbers when selling fish to a new customer became a common practice for traders in order for them to be able to make arrangements by mobile phone in the future. The cross-border traders thus used mobile phone communication to maintain existing customer relations as well as to build relationships with new customers.

Maintaining relationships with customers was the main reason cross-border fish traders resumed or did not suspend their CBFT after the border closed. For Ghanaian traders for instance, it remained easier and more profitable to sell their goods in Togo or Benin to the few customers they were in contact with, rather than having to establish new customer relations in Ghanaian markets that were unfamiliar to them and to build social networks from scratch. The interviewed traders often said ‘we are used to that’, meaning that CBFT was the main economic activity they were familiar with and of which they had experience and knowledge.

Impacts on cross-border traders’ livelihoods and food security

The interviewed traders frequently pointed out that the most prevalent impacts of the COVID-19 pandemic on their lives were of an economic nature. The small profits they made were immediately spent on food, which quickly eroded their working capital. When the 30 survey participants were asked how their income was impacted during the first two months of the pandemic, 29 indicated a decrease, 22 of whom specified it was a strong decrease.

Selling fish on credit was a common practice before the pandemic. Therefore, when the border closed, some traders were still owed money by their customers. Consequently, the traders who suspended their CBFT activities struggled to recoup that credit, and the few who managed to continue CBFT stopped selling on credit. One trader who had discontinued her CBFT explained how, little by little, she had managed to retrieve the credit she had given to customers in Togo through the assistance of other cross-border traders from her home-community who continued travelling to Lomé.

To cover the increased travel and border fee expenses at least partly, some of the traders who continued trading across the border had to increase their selling price. One trader mentioned that she only made one-quarter of her normal market profits compared to before the pandemic, due to the reduced margin and increased travel expenses.

Although selling at a loss sometimes happened before the pandemic, it became much more frequent after the border closed. Despite the risks involved, many traders continued crossing the border in order to maintain relationships with customers and ensure their continued loyalty. Despite increased trading expenses, the traders could only increase their fish prices moderately because otherwise, they risked long-standing customers buying fish from other traders in the future. However, all interviewees asserted that CBFT nevertheless remained more profitable than local trade. Trading in large markets in Togo and Benin guaranteed the sale of their goods and with higher profit margins than in smaller local markets.

One of the Ghanaian traders found a new strategy to deal with the economic impacts of the crisis. When the border closed, she quickly shifted her focus from CBFT in Togo to trade in a new market destination within Ghana. She followed the recommendation of a friend (a trading colleague) to sell fried fish in Kadjebi, a town 250 km north of Keta. As a result, her profits quadrupled and she advised other traders to do the same, but they were reluctant to change their strategies.

With varying degrees of success, the traders made huge efforts to continue pursuing their CBFT in order to earn an income and fulfil their responsibility of purchasing food for their households. However, they claimed that the pandemic was the worst period they had ever faced in terms of trade. Incomes declined and food prices increased, and therefore they struggled to purchase enough food: ‘There is no more money. We stayed home and we ate everything’ (Togolese trader, 45 years).

The traders’ reduced incomes resulted in household dietary changes. Even before the pandemic, a relatively large share of their profits was always spent on food. Of the 30 survey participants, 19 estimated that they used to spend at least half of their profits on food, 5 (16.7%) estimated that this share used to reach 70%, and 3 claimed that they spent 100% of their profits on food (meaning that they always had to purchase fish for their trade on credit). Thus, the larger the share of their income normally being spent on food, the more vulnerable they were to food insecurity when their incomes declined. Another consequence of the reduced incomes was that traders were either unable or struggled to contribute to rotating saving ringsFootnote 11. Without that economic safety net, they were more vulnerable to food insecurity.

All traders stated that they generally ate well and had a varied diet before the pandemic, and that fish was an important part of their diet since they usually kept some of the fish they purchased for consumption. Almost all (29 of the 30 traders in the survey sample) reported that they had noticed a negative change, and for the most vulnerable a severely negative change, in their diets during the pandemic. Their own diets became less varied and meals per day less frequent. Many traders only had two meals per day during the pandemic, compared to three to four meals per day previously. Many said that they ate the same foods every day for breakfast (porridge or bread) and lunch/dinner (mostly akple/bankuFootnote 12, yam, rice, maize, or beans). They still generally consumed fish every day, but in smaller portions: ‘You have to reduce your fish consumption or you will eat your capital’ (Togolese trader, 55 years). Other animal-sourced foods (ASF), usually chicken, were consumed on average twice per month before the pandemic. Since chicken and other sources of meat were more expensive than fish, traders’ ASF consumption reduced during the pandemic and many could only afford to consume ASF sources about once per month or every other month. Other household members reduced their food consumption too, but in line with expectations of them as women. Moreover, to cater for their responsibilities as household heads, the traders would eat last and let others have a larger share of the meal and were thus, according to their own accounts, the most affected household members and reduced their food consumption to a larger extent.

Discussion

The crisis caused by the COVID-19 pandemic and related border closure profoundly affected traders’ social networks, both those involving ties with actors in the CBFT system within their communities (i.e. social embeddedness) and those involving ties beyond their communities and country (i.e. brokerage) (Walther 2015b; Kuépié et al. 2016). While relations and interaction with traders and other people from their home communities did not seem to be affected by the crisis, exchange through ties with fish customers on the other side of the closed Ghana-Togo border was abruptly and deeply disrupted. As time elapsed, despite severe challenges, traders increasingly picked up cross-border trade again and were able to develop ties with new customers. As in the case of the trade networks that were partly lost, ethnicity played a major role in shaping the new relationships. Traders could communicate in the Ewe language with customers within what we may call the ‘Ewe region’, which spans the national borders between Ghana, Togo and Benin and functions as a unified geographical entity in which populations across borders are connected by cultural factors (Igué and Zinsou-Klassou 2010).

However, the quality of the new ties was different from the more established trade networks that traders had cultivated before the pandemic. Before the crisis, relationships between actors in the trade system were mainly based on trust (Walther 2015b; Overå et al. 2022), as is often the case in informal economies (Odera 2013). However, as customers who feared being infected by the coronavirus stopped going to the market and failed to repay their credit to traders, trust between traders and customers collapsed.

To bridge the gap in face-to-face interaction across the border and to build/rebuild trust with ‘old’ and ‘new’ customers, cross-border traders gradually strengthened the ties through frequent mobile phone communication. It has been observed that transaction costs associated with the many risks involved when operating in informal personalized trust-based economies can be reduced through frequent mobile phone communication to diminish information asymmetries and enable monitoring of trading partners’ movements and actions, which can contribute to building trust (Overå 2006). During the pandemic, traders exhibited agency as they creatively replaced in-person interaction with intensified mobile phone communication and flexibly adapted trade practices to maintain their businesses. Trust develops slowly as it is tested over time (Odera 2013). The trust between traders and customers in our study was confirmed when they repeatedly shared reliable information and when traders moderated their fish price increases or even sold at a loss. They considered cultivating trust and maintaining relationships with customers as a necessary investment to be able to trade during the crisis and to keep trading in the future. Interpersonal trust is often the only safety net traders can rely on in informal economies (Odera 2013). Selling at a loss to build and maintain trust with customers thus appeared as a temporary strategy for the traders to try and limit the long-term effects of the crisis. They expected the border to reopen shortly, and therefore employed this risky and costly strategy to be able to get back in business quickly and even to have a competitive advantage after the crisis.

According to Bryan et al. (2023), resilience is the ability to draw upon capacities before, during and after a crisis. It can be suggested, as has been the case in similar contexts (Bouët and Laborde 2020), that traders had little time to prepare before the crisis and they had very limited savings that could serve as a buffer. The traders nevertheless exhibited the ability to draw on their different resources through social networks and agency during the crisis, which translated into quite strong resilience capacities. However, while Bryan et al. (2023) argue that resilience leads to maintained or improved well-being, we found that compared with before the border closure, the studied cross-border traders’ well-being and food security during the border closure was barely maintained, and drastically deteriorated for the most vulnerable traders and their households. The period after the pandemic could not be assessed in our study because data collection took place only during the crisis. Nevertheless, based on our understanding of how the CBFT system works, we believe that it is likely that the traders gradually were able to resume their CBFT activities to a similar frequency as before the pandemic, restoring old social networks while also engaging the new social networks acquired during the pandemic.

The traders’ lack of agency, which impacted their resilience capacities, appeared to stem from their lack of relationships with border agents, making them vulnerable to the agents’ unethical regulatory practices implemented during the pandemic. Traders showed adaptability when finding unofficial routes to avoid expensive fees at the checkpoint in Aflao, but their ability to cross the border depended on actors at the border. In centralized states (such as Ghana) border policies are conceived at the national level and therefore tend to overlook local practices (Mezzadra and Neilson 2013). In this case, local informal practices such as cross-border trade were not considered when national authorities decided to close the border. Border officials were thus able to assert their power reflecting national policies, and to abuse that power for personal gain by forcing traders to pay bribes. Abuse of power by law enforcement officials is not uncommon in many regions of Africa when exceptional measures are taken and is often indicated by instances of bribe collection at trade corridor checkpoints (Bouët and Laborde 2020). According to Bouët and Laborde (2020), trade corridors in West Africa faced an increase in bribe collection of 30% between March and April 2020. Moreover, instances of violence and rough treatment by officials were reported in urban markets in Ghana, where security officers brutalized not only people who travelled to markets outside their neighbourhoods, a practice not allowed during the lockdown, but also traders who were actually permitted to enter the markets (Asante and Mills 2020).

Ever since the Ghana-Togo border was established by colonial powers more than a century ago, fish traders have crossed on a regular basis, but they clearly still lack recognition from border officials. Unlike Desai’s (2009) claim that most border officials are pro-traders because women in their families are traders, the traders in our study were seldom directly acquainted with officials, and their gender was clearly a factor that made them a target for harassment and economic abuse. Gender-based harassment towards cross-border fish traders was also experienced at border checkpoints in other countries during the pandemic and was sometimes even described as sexual harassment (Mudege et al. 2022).

The traders’ agency does not only depend on their established social networks with other traders; their familiarity with border officials and their ability to form social networks with them plays an important role too. According to Kusakabe et al. (2008), traders’ ability to bargain is a form of social network building with border officials, which strengthens their agency. However, small-scale traders with limited social networks often have low negotiation power with authorities compared with larger-scale traders, a factor which further marginalizes them (Kusakabe et al. 2008). Lack of negotiation power was definitely a problem for traders in our study, as most of them had to pay high border fees to border officials to be allowed entry into the neighbouring country. The border officials’ increased opportunity to exercise power, which can be seen in their ability to demand high checkpoint fees, coupled with the traders’ limited negotiation power, constrained traders’ agency. Cross-border traders appeared vulnerable to the changes in border policies when the border closed, as is often the case for people operating in fuzzy spaces such as borderlands (Kusakabe et al. 2008). However, unlike what is usual in such fuzzy spaces, the traders in our study had very little room to manoeuvre around the regulatory practices to make their way into the neighbouring country. The act of closing the border in March 2020 made the border locally very tangible, and the event showed how borders can become rigid (Mezzadra and Neilson 2013).

The traders’ restricted agency can also be assessed in terms of border capitalism, as traders try to resist the technologies of rule implemented by states. Although border capitalism is socially produced (Campbell 2018) and traders participate in producing it by responding to regulations (e.g. through bargaining), the degree to which they can actually influence the regulations is limited. Their determination to cross the border in times of hardship and their constant bargaining appears as a way of protesting against regulations and practices at the border, which can be seen as ‘everyday forms of resistance’ against those who are in control (Scott 1985). Traders fight a constant struggle, notably over their work and autonomy, by using what Scott (1985) calls ‘the weapons of the weak’; in our case, the ‘weapons’ were collectively performed small acts of bargaining. The traders’ responses to regulations can explain, at least partly, how local regulatory practices were later adjusted with time (Campbell 2018), such as with the decline in violence by border officials. However, the traders’ influence remained limited, as they played a part in influencing local practices but their response did not lead to changes in national border regulations.

Traders’ lack of recognition from the government and other formal institutions also constrains them from accessing safety nets in times of hardship. It has been proven that traders’ purchasing power and food security can be improved by safety nets and economic assistance (FAO 2013), whether from an official source (e.g. loans) or an unofficial source (e.g. saving groups). Traders in our study did not, however, benefit from any assistance from governments or banks during the pandemic, and traders’ own informal economic arrangements were impossible to resume as all of them were struggling financially. In addition to limited financial resources, traders lacked education and information that could enable them to invest in another business, thereby limiting their agency. Informal traders are said to be part of the ‘missed middle’, a category of workers that are usually above the poverty line and therefore not supported by formal safety nets, but lack social protection when vulnerable to falling below the poverty line (Resnick et al. 2020; Guven et al. 2021). Informal traders have a weak social contract with the state, preventing them from receiving pensions, health insurance and other forms of social protection, and Resnick et al. (2020) argue that this made informal workers vulnerable when facing the restrictive COVID-19 preventive policies. Individual traders, and often even their organizations, do not have much political agency as they are frequently neglected in local policymaking in relation to government representatives and regional organizations, such as the Economic Community of West African States (ECOWAS) and the African Continental Free Trade Area (AfCFTA), which are prevailing actors in improving and facilitating regional trade (Walther 2015b; Olusola and Lere 2020). Lack of intervention from major governance actors undermines the functioning of trade systems (including cross-border trade systems), which risks further marginalizing not only the informal fish traders themselves but also the fishing communities and fish consumers, who rely on the fish distributed by the traders. Therefore, as Hapke et al. (2023: 51) conclude, ‘what happens to small-scale traders is of concern because it has a direct bearing on food security’.

Female traders’ household responsibilities represent a heavy burden and limit their resilience in times of bad trade business (Béné and Heck 2005). This dependency burden, as well as the lack of alternative sources of income for fish traders to diversify their livelihoods as a strategy to cope with the COVID-19 crisis, constrained their ability to make choices– in other words, their agency (Kabeer 2021). Those who before the border closure had spent a large share of their income on food for households were therefore the least resilient when the crisis hit, and were particularly vulnerable to food insecurity.

In addition to limiting their engagement in business and thus reducing their profits, the inability of some traders to cross the border during the pandemic considerably reduced the amounts of fish made available to people across the border. The closed border had effects not only on the traders’ indirect food security but also on the direct food security of populations that benefitted from the consumption of cheap fish that used to be available in large quantities (Kawarazuka and Béné 2010).

Containment measures against the spread of the COVID-19 pandemic completely overlooked the significance of informal trade in fish (and other foods) and focused mainly on formal exports. This became evident in our study as only vehicles transporting goods in bulk were allowed to cross the border and higher border fees were demanded. This made it difficult for small-scale traders with limited capital to operate their businesses. There is a need to shift the policy focus from promoting formal fish exports towards recognizing the important contribution that small-scale informal fish trade makes towards fish consumption and income generation (Kawarazuka and Béné 2011). The lack of this recognition is not new (Béné et al. 2010), but became highly evident during the pandemic when the restrictions proved to be directly undermining and damaging for the functions of the CBFT for both traders and consumers. To enhance informal trade processes, projects such as the FishTrade initiative by WorldFish (Keeley et al. 2020) can play a role in highlighting the importance of intra-regional trade in sub-Saharan Africa, informing decision-makers, adapting trade regulations, and contributing to capacity building among cross-border fish traders. Although the FishTrade project did not include the Abidjan-Lagos corridor where the Aflao border post is located, its breadth has raised awareness about the topic in similar contexts.

Conclusion

From this study we conclude that although the overall quantity and quality of relationships between traders and other actors largely deteriorated during the COVID-19 pandemic compared to before it, trust building through social networks contributed to the resilience of the cross-border fish trade system. The informal and flexible nature of the processes allowed traders to re-engage in trade using unofficial crossing points and to develop new ties remarkably swiftly, thereby strengthening the traders’ resilience. The resilience of the CBFT system was, however, constrained by the trader’s lack of agency and vulnerability emanating from their limited ties with government officials at the border. Traders were economically and physically abused by border officials endowed with power endorsed by the border authorities. Opportunities for power abuse intensified because of the ‘state of emergency’ caused by the pandemic measures. This affected the negotiation power– in other words, the agency– of small-scale traders, and reduced their possibility and motivation to undertake CBFT activities. Declining trade profits due to the costs associated with transporting fish over longer distances using cumbersome travel routes, as well as corrupt border fee collection practices, combined with high costs of purchasing food, extensively impacted livelihoods and threatened the traders’ food security during the pandemic. The lack of public respect for informal cross-border fish traders and recognition of their work at local, national and regional scales led to violence being inflicted by border officials and a lack of economic assistance and social protection. The border closure experience during the COVID-19 pandemic thus confirmed the general disregard and neglect of women cross-border fish traders in decision-making regarding national and regional policy measures concerning food trade and food security.