Introduction
Food loss and waste (FLW) is a pressing socioecological concern globally that has implications for both climate change and food security (Zhu et al., Reference Zhu, Luo, Sun, Li, Zhou, Wang, Fei, Tong and Yin2023; Gatto and Chepeliev, Reference Gatto and Chepeliev2024; Heydari, Reference Heydari2024). Recent evidence shows that up to 40% of food produced globally is lost or wasted (Goodwin and Lipinski, Reference Goodwin and Lipinski2024). This level of FLW is concerning in a world where nearly 2.3 billion people face some level of food insecurity (FAO, 2024). Aside from its contribution to food insecurity, food loss is a key source of greenhouse gas (GHG) emissions, with recent evidence showing it accounts for 50% of all GHG emissions from agriculture (Zhu et al., Reference Zhu, Luo, Sun, Li, Zhou, Wang, Fei, Tong and Yin2023), and 8%–10% of all global emissions (Goodwin and Lipinski, Reference Goodwin and Lipinski2024). Beyond GHGs, FLW contributes to environmental degradation through the use of land, freshwater resources, and agrochemicals in producing food that is ultimately not consumed. For instance, about 45 trillion gallons of water, amounting to one quarter of all water used in global agriculture, go to the production of food that is later wasted (Goodwin and Lipinski, Reference Goodwin and Lipinski2024).
The sustainability impacts of food loss are particularly pressing in sub-Saharan Africa (SSA), one of the world’s most vulnerable areas to food insecurity and climate change (Ayanlade et al., Reference Ayanlade, Oluwaranti, Ayanlade, Borderon, Sterly, Sakdapolrak, Jegede, Weldemariam and Ayinde2022; Kansanga et al., Reference Kansanga, Shanmugasundaram, Ledermann and Rain2025). It is estimated that PHL represents over $4 billion in lost value for farmers in SSA annually, a number that far exceeds the amount of international food aid the continent receives each year (Mmereki, David and Wreh Brownell, Reference Mmereki, David and Wreh Brownell2024). Recent evidence shows that FLW is responsible for nutritional losses of about 550 calories per capita per day in the region (Gatto and Chepeliev, Reference Gatto and Chepeliev2024). As aforementioned, FLW consumes water resources that are limited in rainfed agricultural areas, and the application of chemical pesticides and fertilizers to food that is wasted further depletes already poor soil, affecting the long-term sustainability of agriculture in the region (Kummu et al., Reference Kummu, De Moel, Porkka, Siebert, Varis and Ward2012).
Unlike high-income countries, where food waste is largely a product of consumer behavior and choice and food retail preferences and decisions by supermarkets, food loss in SSA typically occurs much earlier in the value chain (Kansanga et al., Reference Kansanga, Shanmugasundaram, Ledermann and Rain2025). Smallholder farmers frequently incur losses due to inadequate storage facilities, poor transportation networks, limited access to cold storage, and long distances to markets (Kansanga, Pienaah and Luginaah, Reference Kansanga, Pienaah and Luginaah2026). In these contexts, deteriorated produce is often left at markets, discarded near homesteads, or abandoned in open spaces because it can no longer be sold or consumed.
Food loss has received considerable policy attention globally, yet emphasis has largely been on preventing or reducing waste/losses at their source or at the consumer level (Kansanga et al., Reference Kansanga, Mohammed, Batung, Ansumah Saaka and Luginaah2023). While these FLW prevention/reduction efforts are timely, recovery/repurposing of deteriorated produce presents an extra opportunity to harness food waste for other socioeconomic uses. In practical terms, repurposing refers to the conversion of deteriorated produce into alternative productive uses such as livestock feed, compost, biochar, or dried products for later consumption or sale. Such practices can help farmers recover value from losses while simultaneously reducing environmental burdens associated with disposal. For smallholder farmers in Africa, repurposing food waste as livestock feed can support livelihood diversification beyond crop cultivation. This offers a viable option for poor households, given the high cost of formulated animal feed (Chisoro, Jaja and Assan, Reference Chisoro, Jaja and Assan2023). Repurposing also presents opportunities to convert food waste into compost and biochar, which are viable substitutes for synthetic fertilizer (Yeo et al., Reference Yeo, Dongo, Mertenat, Lüssenhop, Körner and Zurbrügg2020). In SSA, biochar is gaining popularity not just for valorizing food waste, but also for its ability to sequester carbon in the soil and support sustainable agriculture (Zanli et al., Reference Zanli, Gbossou, Tang, Kamoto and Chen2022; Koné and Galiegue, Reference Koné and Galiegue2023).
The ability of farmers to engage in these forms of food waste revalorization practices is not evenly distributed. Existing research suggests that demographic characteristics, particularly gender, can shape participation in sustainable agricultural practices through their influence on access to productive resources, extension services, transportation, land, and agricultural technologies (Peterman, Behrman and Quisumbing, Reference Peterman, Behrman and Quisumbing2014; Tsikata and Yaro, Reference Tsikata and Yaro2014; Quaye et al., Reference Quaye, Fuseini, Boadu and Asafu-Adjaye2019; Kansanga et al., Reference Kansanga, Luginaah, Bezner Kerr, Dakishoni and Lupafya2021). Women farmers often face greater constraints in accessing information, postharvest technologies, and market opportunities, all of which may influence their capacity to repurpose deteriorated produce (Kansanga et al., Reference Kansanga, Luginaah, Bezner Kerr, Dakishoni and Lupafya2021; Pienaah, Seidu and Issahaku, Reference Pienaah, Seidu and Issahaku2022). Similarly, socioeconomic factors such as wealth, access to credit, livelihood diversification, and land access can affect farmers’ ability to invest time, labor, and resources in food waste revalorization activities (Kansanga et al., Reference Kansanga, Luginaah, Bezner Kerr, Dakishoni and Lupafya2021; Pienaah et al., Reference Pienaah, Saaka, Yengnone, Molle and Luginaah2024b).
Despite growing scholarly and policy interest in circular economy approaches, little empirical evidence exists on the factors that shape farmers’ participation in food waste revalorization within smallholder farming systems. In this contribution, we draw on survey data with smallholder vegetable farmers in Ghana to examine the determinants of farmers’ decisions to repurpose deteriorated food waste to other socioeconomic uses. By identifying the factors that shape the decision to repurpose, we identify entry points and make recommendations for promoting the sustainable repurposing of food waste in smallholder farming systems.
Conceptual background: wasted ecologies and the politics of revalorizing food waste
Dominant framings of waste often posit it as unwanted or garbage—a situation which has largely oriented sustainability policy toward the core goal of eliminating waste entirely from value chains through waste discarding strategies (Liboiron and Lepawsky, Reference Liboiron and Lepawsky2022). Specific to the contemporary food system, the widespread framing of deteriorated produce as dirt that needs to be eliminated forecloses opportunities to revalorize food waste for other productive uses (Mattila et al., Reference Mattila, Mesiranta, Närvänen, Koskinen and Sutinen2019; Lehtokunnas et al., Reference Lehtokunnas, Mattila, Närvänen and Mesiranta2022). But is waste always dirt that must be eliminated? The answer to this question requires a brief engagement with the ontological underpinnings of waste.
Consistent with Liboiron’s (Reference Liboiron2019) observation, framing waste and dirt as equivalent is a mistake. While dirt can be conceptualized as material that has no value, an existential threat with the potential to pose danger and challenge the functioning of a system, waste, on the other hand, belongs in a particular process of material ordering, where, if placed in context, for example, by sorting and recycling, it could assume a new material value (Lepawsky, Reference Lepawsky2023). Although it is probable that at some point, a material without use value in a given context (waste) can become dirt if no alternative uses are found (which can be dependent on the level of technology, underlying need, etc.), waste is not always dirt but material that can be revitalized, harmonized, and revalorized (Corvellec and Bevan, Reference Corvellec, Bevan and Corvellec2025).
The prevailing understanding of waste as an unwanted byproduct of growth and development that must be discarded is applicable to food waste. Food waste is built into the modern agri-food system in that, to meet consumer preferences, volumes of agricultural produce are discarded merely because they appear suboptimal, and therefore less likely to be bought by consumers. This “making” of waste based on quality and aesthetics forecloses opportunities for its revalorization, first as food, and second into other economically viable products (Gille, Reference Gille2012; Gascón, Reference Gascón2018). FLW does not simply need to be waste that ends up in a landfill. Rather, the end uses of FLW and waste more generally can be reimagined and repurposed as solutions to some of the ecological and social problems we face today. As argued by some scholars (Gille, Reference Gille2012; Slorach et al., Reference Slorach, Jeswani, Cuéllar-Franca and Azapagic2019; Turner, Reference Turner2019), moving beyond disposal as the key option for dealing with waste by opening opportunities to interact with it in more proactive ways that target reduction as opposed to elimination can unlock the competencies necessary for sustainably handling waste and an overall cultural positioning of food waste as a resource.
The concept of circular economy demonstrates how waste can be repurposed for other uses, limiting negative externalities and creating more sustainable systems of production and consumption. While the circular economy approach has broad implications, scholars have suggested it as an important model toward achieving sustainable food systems, from production to consumption (Ciccullo et al., Reference Ciccullo, Cagliano, Bartezzaghi and Perego2021; Santagata et al., Reference Santagata, Ripa, Genovese and Ulgiati2021; Lehtokunnas et al., Reference Lehtokunnas, Mattila, Närvänen and Mesiranta2022). Circular economy applications to the food system contend both the reduction of external farm inputs, including pesticides and fertilizers, and the reduction of food waste (Selvan et al., Reference Selvan, Panmei, Murasing, Guleria, Ramesh, Bhardwaj, Thakur, Kumar, Sharma, Digvijaysinh Umedsinh, Kayalvizhi and Deshmukh2023). One goal is finding other “end-of-life alternatives” for food waste, such as turning food into raw materials for new products or finding new markets for it (Diekmann and Germelmann, Reference Diekmann and Germelmann2023). Additionally, circular economy has been further specified to a circular bioeconomy, where food waste’s potential for nutrient cycling back into the environment is the focus, rather than its economic value (Fassio and Chirilli, Reference Fassio and Chirilli2023). A circular food economy also suggests production-side improvements and benefits to reduce waste, including the integration of AI and machine learning to improve supply chains (Gonçalves and Maximo, Reference Gonçalves and Maximo2023). Some scholars have critiqued the idea of the circular economy, including that it is depoliticized in the sense that it often frames environmental challenges as technical or managerial problems while paying insufficient attention to the broader political, economic, and power relations that shape patterns of production, consumption, and waste generation (Genovese and Pansera, Reference Genovese and Pansera2021; Pansera, Genovese and Ripa, Reference Pansera, Genovese and Ripa2021), hinging on several key neoliberal assumptions of never-ending growth, consumption, and profit, thereby promoting an endless cycle of waste (Valenzuela and Böhm, Reference Valenzuela and Böhm2017; Corvellec, Stowell and Johansson, Reference Corvellec, Stowell and Johansson2022; Dzhengiz et al., Reference Dzhengiz, Miller, Ovaska and Patala2023). What is not clear is whether these arguments apply to smallholder farming regions where the realities of production and FLW are quite different from the Global North, where overproduction and consumer-level waste are key (Messner, Johnson and, Reference Messner, Johnson and Richards2021; Ouro-Salim and Guarnieri, Reference Ouro-Salim and Guarnieri2022).
Importantly, these debates are highly relevant to food waste valorization in smallholder agriculture. In contexts where food loss results primarily from storage constraints, transportation challenges, and market access limitations rather than overproduction and overconsumption, repurposing deteriorated produce may serve as a livelihood-supporting strategy rather than a mechanism for sustaining cycles of accumulation. Consequently, understanding who participates in food waste revalorization and the factors that facilitate or constrain participation becomes central to assessing the practical relevance of circular economy principles in rural agricultural settings.
Indeed, an emerging body of literature views the concept of circular economy as a “floating signifier” whose meaning is contingent upon the context within which it is being applied and the motives of the actors (Knox et al., Reference Knox, Bacon, Beverley, Davies, De Laurentis, Barrios, Miller, Rudd and Walpole2023; Niskanen, Anshelm and McLaren, Reference Niskanen, Anshelm and McLaren2020; Rödl et al., Reference Rödl, Åhlvik, Bergeå, Hallgren and Böhm2022). Thus, in smallholder farming systems where the relations of production are not grounded in profit accumulation, repurposing may serve the socioeconomic needs of households through the promotion of new local income-generating activities made possible by the reuse value of waste (Boon and Anuga, Reference Boon and Anuga2020; Härri, Levänen and Koistinen, Reference Härri, Levänen and Koistinen2020; Apanovich, Reference Apanovich2025). A deployment of repurposing within a smallholder farming context must therefore pay attention to the material realities within which waste is (re)produced and the factors that could exploit opportunities to revalorize it for neoliberal interests (Härri, Levänen and Koistinen, Reference Härri, Levänen and Koistinen2020).
Guided by this perspective, we examine how demographic characteristics (e.g., gender and education), socioeconomic resources (e.g., wealth, credit access, land access, and livelihood diversification), and farm-level conditions shape farmers’ participation in food waste revalorization. We argue that while circular economy principles create opportunities for repurposing, farmers’ ability to take advantage of these opportunities depends on their access to resources, knowledge, infrastructure, and productive assets.
Consistent with this framing, we operationalized repurposing in this study as locally relevant ways of reusing food waste that have the potential to produce direct socioeconomic benefits for smallholder farming households, including the conversion of deteriorated food to animal feed, biochar, and compost, among other uses (Apanovich, Reference Apanovich2025).
Materials and methods
Study context
The data for this study were collected in five administrative districts in the northern savannah ecological zone of Ghana, specifically the Upper West Region (see Fig. 1). The region is bordered to the north and west by Burkina Faso, and the Upper East and Northeast regions to the East and the Savannah region to the South. The region has a land area of about 18,476km2, underlain by savannah ochrosol and lithosol soil types. This edaphic profile is rich in minerals that support the cultivation of a variety of crops, including cereals, legumes, and vegetables, as well as the maintenance of vast pastures for livestock (Asiamah, Reference Asiamah2008; Pienaah et al., Reference Pienaah, Kansanga, Baada and Luginaah2026).
Map of the Upper West Region showing the study districts (Authors construct).
Figure 1. Long description
The image consists of a main map and an inset map.
1. Inset Map (Top Right): Shows the country of Ghana within West Africa, bordered by Burkina Faso to the North, Togo to the East, Ivory Coast to the West, and the Gulf of Guinea to the South. A black square highlights the Upper West Region in the Northwest corner. The map includes ecological zones such as Coastal savanna, Guinea savanna, Sudan savanna, Deciduous forest, Moist evergreen, Transitional zone, and Wet evergreen, along with the Volta lake.
2. Main Map (Center): A detailed view of the Upper West Region. The regional boundary is a thick black line. Study districts shaded in grey include Sissala East in the Northeast, Wa East in the Southeast, Wa Municipal and Wa West in the Southwest, and Nadowli-Kaleo in the West. Other districts shown in white include Nandom, Lambussie Karni, Sissala West, Lawra, Jirapa, and Daffiama Bussie.
3. Features: Black triangles represent District Capitals. Dashed grey lines indicate Roads. Thin grey lines represent Rivers. Neighboring regions labeled include Upper East to the East, North East and Northern to the Southeast, and Savannah to the South. Burkina Faso borders the region to the North and Ivory Coast to the West.
4. Scale and Legend: A scale bar at the bottom center shows distances of 0, 28, 56, and 112 K m. The legend on the right defines District Capitals, Roads, Districts, Regional boundary, Study districts, Rivers, and Other Regions.
Owing to its semi-arid profile (Abdul-Razak and Kruse, Reference Abdul-Razak and Kruse2017; Dapilah and Nielsen, Reference Dapilah and Nielsen2019; Mohammed et al., Reference Mohammed, Batung, Kansanga, Nyantakyi-Frimpong and Luginaah2021), agriculture, the predominant economic activity, is possible over a few months, May to September (Asravor, Reference Asravor2018; Kansanga et al., Reference Kansanga, Andersen, Kpienbaareh, Mason-Renton, Atuoye, Sano, Antabe and Luginaah2019; Mohammed et al., Reference Mohammed, Batung, Saaka, Kansanga and Luginaah2023). This unique ecological profile has driven investment in dry season vegetable production, often to complement production in the major season. Traditionally, farmers construct hand-dug wells in valley floors to provide water for their vegetable plots in the dry season (Nakuja et al., Reference Nakuja, Sarpong, Kuwornu and Felix2012; Donkoh et al., Reference Donkoh, Kudadze, Adzawla and Ansah2016; Yiridomoh, Sullo and & Bonye, Reference Yiridomoh, Sullo and Bonye2020; Kansanga et al., Reference Kansanga, Konkor, Kpienbaareh, Mohammed, Batung, Nyantakyi-Frimpong, Kuuire and Luginaah2022). To augment these efforts, successive governments have implemented dedicated agricultural programs to improve smallholder agriculture, including the recently launched one-village-one dam initiative to support dry-season vegetable farming (Hausermann et al., Reference Hausermann, Warner, Roan, Adamu, Badumah Kolobire, Walder-Hoge and Taylor2025). Most farming households reside in rural villages and depend primarily on smallholder agriculture for their livelihoods. In addition to crop cultivation, which includes green leaves, tomatoes, peppers, and onions, households frequently engage in livestock rearing, shea nut processing, petty trading, seasonal migration, and the harvesting of wild foods to supplement household income and food supplies. Livelihood diversification is particularly important given the highly seasonal nature of agricultural production and recurrent climatic shocks characteristic of the region (Pienaah et al., Reference Pienaah, Baruah, Kansanga and Luginaah2024a).
Although agriculture is the primary livelihood, employing up to 80% of households, farm sizes are typically small, averaging less than 2.5 hectares (Kansanga et al., Reference Kansanga, Andersen, Kpienbaareh, Mason-Renton, Atuoye, Sano, Antabe and Luginaah2018). Poverty is disproportionately high in northern Ghana. Specific to the Upper West region, data from the Ghana Living Standards Survey of the Ghana Statistical Service shows approximately 9 in every 10 people live on less than a dollar a day, the highest across the country (Ghana Statistical Service, 2015). Food insecurity is equally predominant, with recent findings showing more than half of smallholder farming households transition to food insecurity less than 5 months after the major harvest (Kansanga et al., Reference Kansanga, Konkor, Kpienbaareh, Mohammed, Batung, Nyantakyi-Frimpong, Kuuire and Luginaah2022).
The socioeconomic categories used in this study should therefore be interpreted relative to local conditions. Households classified as relatively wealthy are not necessarily affluent by national or international standards; rather, they are more likely to possess productive assets such as livestock, bicycles or motorcycles, access to electricity, and comparatively larger areas of agricultural land. Conversely, poorer households often have fewer productive assets and face greater constraints in accessing agricultural inputs, transportation, and market opportunities.
Alongside high levels of poverty and food insecurity, food loss is also pervasive. Farmers lose up to 40% of their overall harvest, while losses among vegetable producers can exceed 50% under certain conditions (Kansanga et al., Reference Kansanga, Mohammed, Batung, Ansumah Saaka and Luginaah2023). These losses are typically driven by inadequate storage facilities, transportation constraints, poor road infrastructure, limited access to cold storage, and market gluts that leave farmers unable to sell highly perishable produce before deterioration occurs. Women farmers often bear a disproportionate share of these losses because of persistent inequalities in access to productive resources, transportation, and agricultural information.
Data collection
The analysis is based on data from a cross-sectional survey with smallholder vegetable farmers (n = 1001) in Ghana conducted in the summer of 2023. The broader objective of the survey was to understand food loss among vegetable producers and included questions on vegetable production and consumption, foraging, food loss rates, access to extension services, climate change resilience as well as demographic and socioeconomic factors.
We used a multistage sampling technique to identify participants. On the basis of information from the regional office of the Ministry of Food and Agriculture, we first used a non-probability purposive sampling technique to select five study districts where vegetable production is dominant (see Fig. 1). We conducted preliminary verification visits to these districts, which enabled us to further purposively sample communities where vegetable production was relatively dominant. Within each community, we randomly sampled every third vegetable farmer from the list of vegetable farmers to participate in the survey subject to participants’ interest and availability at the time of survey administration until the required sample size proportional to the population of farmers per community was reached.
Before administering the survey, the research team conducted a community entry visit during which we met with local leaders and farmer representatives to deliberate on a suitable date for data collection. Data collection was mostly on market days or religious days as most farmers rested on these designated days. All surveys were administered in private environments around the homes of respondents to ensure confidentiality. We received ethical clearance from the Institutional Review Board of the George Washington University.
Measures
The outcome variable in our analysis is “sustainable end use of food waste.” The variable was coded as 0 = no, representing farmers who discarded deteriorated produce, and 1 = yes, representing farmers who repurposed deteriorated produce to alternative productive uses. In the study context, discarding refers to situations where deteriorated produce was abandoned at markets, left in open spaces, disposed of near homesteads, or otherwise removed from productive use. Repurposing refers to the conversion of deteriorated produce into alternative uses such as livestock feed, compost, biochar, or drying for later consumption or sale. To ensure a common understanding of the concept, farmers were provided with examples of locally relevant repurposing practices, such as using as animal feed, drying and selling later at a lower price as dried ingredients, and composting, identified during reconnaissance visits prior to survey implementation.
We included a range of predictors that are conceptually relevant to the repurposing of food waste across three categories: farmer demographics, socioeconomic/capacity-related factors, and agricultural-related factors. Specific demographic factors include the respondent’s age (continuous variable), education (0 = no education, 1 = primary school, and 2 = secondary school and above), and gender (0 = man, 1 = woman). A range of socioeconomic variables with the potential to influence food waste repurposing were also introduced as predictors. These include household size, wealth index (computed based on ownership of a list of contextually relevant household assets and categorized into 0 = poorest, 1 = poorer, 2 = middle, 3 = richer, and 4 = richest), access to land (0 = had difficulty accessing land, 1 = had no difficulty accessing land), access to credit (0 = no, 1 = yes), and livelihood diversification (0 = only one livelihood activity, 1 = two livelihood activities, 2 = three livelihood activities and 3 = four or more livelihood activities). Agriculture and policy-level predictors included the type of vegetable farmers cultivated (0 = green leafy, 1 = fruit vegetables, 2 = marrow/tuberous/cruciferous), farm size, number of years cultivating vegetables (a proxy for experience), season of cultivation (0 = both wet and dry seasons, 1 = either wet or dry), yield (in kg), food loss (proportion of total harvest), ever received training on postharvest loss management (0 = no, 1 = yes), access to timely agricultural advisory services (0 = no, 1 = yes), distance to nearest market (km), prior adoption of sustainable land management practices (0 = no, 1 = yes).
Categorical variables such as education, wealth categories, livelihood diversification categories, vegetable type, and season of cultivation were entered into the regression model as dummy variables with clearly specified reference categories.
Analytical approach
We employed both descriptive and inferential statistical analyses to explore the correlates of food waste repurposing among farmers. Specifically, we used descriptive statistics to understand the distribution of the sample and further employed binary logistic regression analysis to explore the different predictor variables and sustainable repurposing of food waste. We further visualized these findings in Stata using multiple logistic regression to generate a chart of the relationships. This multivariate model was an extra step to ascertain if some of the predictors observed at the bivariate level were consistent in the presence of other conceptually relevant potential determinants of food waste repurposing in a smallholder farming context.
where
$ \pi $
is the probability that a farmer’s response falls in the affirmative category of the dichotomous variable Y, on repurposing of deteriorated food (i.e., 1 = yes), exp is the exponential function.
$ \beta 0 $
is the intercept,
$ \beta 1 $
is the coefficient of the key predictor variable, and
$ \beta k $
is the coefficient of the last predictor variable. Categorical predictors were specified using indicator (dummy) coding, allowing estimated effects to be interpreted relative to designated reference categories rather than as continuous or ratio-scaled variables. Results are reported as odds ratios, where odds ratios above one (OR > 1) denote a higher likelihood of repurposing deteriorated food to other economic uses and odds ratios below one (OR < 1) indicate a lower likelihood of repurposing. We performed all the statistical analyses in Stata version 15.
Results
Descriptive statistics of the sample
Table 1 presents findings on the distribution of the sample across relevant demographic, socioeconomic, and agricultural factors that may shape food waste repurposing among smallholder farmers. About 62% of the sample reported repurposing their deteriorated agricultural produce to other socioeconomic uses, while 38% reported discarding it. The average age of sampled farmers was 39 years, while the majority (60%) were women, reflecting the important role women play in vegetable production across many parts of Ghana and Africa more broadly. Most farmers (65%) reported having no formal education, while the average household size was eight members.
Sample characteristics
Table 1. Long description
The table consists of two columns: Variable and Percent/mean.
Demographics and Socio-economics:
- Repurposing of food waste: 62 percent Yes, 38 percent No.
- Age: Mean of 39.
- Gender: 40 percent Man, 60 percent Woman.
- Education: 65 percent No formal education, 17 percent Primary, 18 percent Secondary and above.
- Household size: Mean of 8.
- Accessing land: 54 percent Had difficulty, 46 percent No difficulty.
- Wealth: 10 percent Poorest, 12 percent Poorer, 10 percent Middle, 15 percent Richer, 53 percent Richest.
- Access to credit: 72 percent No, 28 percent Yes.
- Livelihood diversification: 62 percent One livelihood, 20 percent Two activities, 12 percent Three activities, 6 percent Four activities.
Farming and Market Characteristics:
- Vegetable type: 42 percent Green Leafy, 39 percent Fruit vegetables, 19 percent Marrow/tuberous/Cruciferous.
- Years cultivating vegetables: Mean of 8.65.
- Season: 69 percent Wet season, 9 percent Dry season, 22 percent Both.
- Farm size: Mean of 2 acres.
- Yield: Mean of 558 kg.
- Postharvest loss: Mean of 25.
- Postharvest management training: 26 percent Yes, 74 percent No.
- Access to agricultural advisory services: 20 percent Yes, 80 percent No.
- Distance to market: Mean of 21 km.
- Prior adoption of S L M P: 79 percent Yes, 21 percent No.
The total sample size is 1,001.
† Mean reported.
Half of the sample reported experiencing difficulties accessing land for vegetable cultivation, particularly plots located near reliable water sources that support dry-season production. In terms of relative wealth, measured using ownership of locally relevant assets such as livestock, motorcycles, bicycles, aluminum-roofed housing, and access to basic household amenities, the sample was not evenly distributed across wealth categories. Specifically, 53% of respondents were classified in the wealthiest category, while approximately 10% and 12% were categorized as poorest and poorer, respectively. It is important to note that these wealth categories are relative measures of socioeconomic status within the study area. Thus, households classified as wealthier are not necessarily affluent by national or international standards but generally possess more productive assets and improved access to services than poorer households.
Access to credit for agricultural development was extremely limited, with 72% of respondents reporting no access to formal or informal credit. Similarly, livelihood diversification was relatively low. Most farmers (62%) depended solely on vegetable farming as their primary livelihood activity, while only 6% reported engaging in four or more income-generating activities. This pattern has important implications for food waste repurposing because complementary activities such as livestock rearing may create opportunities to convert deteriorated produce into animal feed.
Specific to potential agricultural predictors, green leafy and fruity vegetables (tomatoes and peppers) were popular among farmers as major crops, with most farmers (69%) producing in just the wet season and only 22% cultivating in both the wet and dry seasons. The average number of years cultivating vegetables was about 9, implying most farmers may have extensive experience in vegetable cultivation and food loss management. Farm sizes are small, averaging 2 acres (0.8 hectares) compared to the general average for cereal and legume production in this context.
The average vegetable yield was 558 kg, and farmers reported losing approximately 25% of total production after harvest. These losses were largely associated with the highly perishable nature of vegetables, limited storage facilities, transportation challenges, and difficulties accessing markets. In some cases, losses reached as high as 100% of harvested produce.
While technical knowledge is important for food waste repurposing, postharvest management training and access to agricultural advisory services were relatively uncommon. Only 26% of farmers reported ever receiving training on postharvest management, while just 20% reported having timely access to agricultural advisory services. The average distance to the nearest market was 21 km. Given the predominance of feeder roads and limited transportation infrastructure in the study area, such distances can significantly increase the likelihood of produce deteriorating before sale. Long travel distances may also discourage farmers from transporting deteriorated produce back to their communities for alternative uses such as composting or livestock feeding, increasing the likelihood of disposal.
Prior adoption of sustainable land management practices was relatively high, with 79% of respondents reporting the implementation of at least one sustainable land management practice on their farms. This is noteworthy because prior engagement with environmentally sustainable practices may increase farmers’ willingness to adopt other resource-conserving strategies, including the repurposing of food waste. The high prevalence of prior use of sustainable land management practices further suggests some farmers are already familiar with approaches that emphasize resource efficiency.
Predictors of food waste repurposing
Table 2 presents the results of the binary logistic regression analysis examining factors associated with the sustainable end use of food waste among smallholder farmers. Several demographic, socioeconomic, and agricultural variables emerged as significant predictors of food waste repurposing.
Logistic regression analysis of the predictors of food waste repurposing
Table 2. Long description
The table contains four columns: Variable, Odds ratio (S E), and two columns for the 95% C I. Significant predictors include:
* Gender: Women have an odds ratio of 0.654 (S E 0.089, p < 0.01).
* Household size: 0.973 (S E 0.013, p < 0.05).
* Land access difficulty: 0.516 (S E 0.069, p < 0.001).
* Wealth Index: Compared to the poorest, the richest have an odds ratio of 0.095 (S E 0.033, p < 0.001).
* Access to credit: 1.505 (S E 0.216, p < 0.01).
* Livelihood diversification: Odds ratios decrease as activities increase, with four activities at 0.254 (S E 0.073, p < 0.001).
* Vegetable type: Fruit vegetables have a high odds ratio of 2.81 (S E 0.0442, p < 0.001), while marrow/tuberous/cruciferous is 0.582 (S E 0.103, p < 0.01).
* Yield: 1.00 (S E 0.00, p < 0.01).
* Postharvest loss: 0.982 (S E 0.003, p < 0.001).
* No access to advisory services: 0.559 (S E 0.096, p < 0.001).
* Distance to market: 0.967 (S E 0.004, p < 0.001).
* Prior adoption of S L M P: 2.791 (S E 0.446, p < 0.001).
Non-significant variables include Age, Education, Years of cultivation, Season, Farm size, and Postharvest management training.
***p < 0.001, **p < 0.01, *p < 0.05; standard errors (SE) are in parentheses.
† = continuous variable.
At the demographic and socioeconomic level, gender, household size, land access, wealth status, access to credit, and livelihood diversification were significantly associated with repurposing behavior. Women farmers were significantly less likely (OR = 0.654; p < 0.01) to repurpose deteriorated produce than men. Household size was negatively associated with repurposing (OR = 0.973; p < 0.05), suggesting a modest decline in the likelihood of repurposing as household size increased. Farmers who reported difficulties accessing land were significantly less likely (OR = 0.516; p < 0.001) to repurpose food waste compared to those who did not experience land access constraints.
Consistent with the categorical specification of the model, wealth categories were assessed relative to the reference category of the poorest households. Compared to the poorest category, farmers classified as poorer (OR = 0.301; p < 0.01), richer (OR = 0.372; p < 0.05), and richest (OR = 0.095; p < 0.001) were significantly less likely to repurpose deteriorated produce. Farmers with access to credit were significantly more likely (OR = 1.505; p < 0.01) to repurpose food waste than those without access to credit. Livelihood diversification was negatively associated with repurposing. Relative to farmers engaged exclusively in vegetable farming, those involved in two (OR = 0.360; p < 0.001), three (OR = 0.166; p < 0.001), and four or more income-generating activities (OR = 0.254; p < 0.001) were significantly less likely to repurpose food waste.
Several agricultural and farm-level variables also emerged as significant predictors of food waste repurposing. Vegetable categories were similarly estimated relative to the reference category of green leafy vegetables. Farmers cultivating fruit vegetables were about 2.8 times more likely (OR = 2.810; p < 0.001) to repurpose deteriorated produce than those cultivating green leafy vegetables. Conversely, farmers cultivating marrow, tuberous, and cruciferous vegetables were significantly less likely (OR = 0.582; p < 0.01) to repurpose food waste.
Yield was positively associated with repurposing (OR = 1.00; p < 0.01), while increasing postharvest loss was negatively associated with repurposing (OR = 0.982; p < 0.001). Farmers who lacked access to timely agricultural advisory services were 44% less likely (OR = 0.559; p < 0.001) to repurpose deteriorated produce than those who had access to such services. Distance to the nearest market was also significantly associated with repurposing behavior. As market distance increased, the likelihood of repurposing declined (OR = 0.967; p < 0.001).
Given the relatively long distances that many farmers travel to access markets, greater market distance may increase the practical difficulties associated with transporting deteriorated produce for alternative uses. Prior adoption of sustainable land management practices significantly predicted food waste repurposing. Farmers who had previously implemented sustainable land management practices were nearly three times more likely (OR = 2.791; p < 0.001) to repurpose food waste than those who had not adopted such practices.
Discussion
Overall, our findings indicate that many smallholder farmers are already engaging in the repurposing of deteriorated produce for alternative productive uses. Nevertheless, a substantial proportion (38%) continue to discard deteriorated produce despite the potential socioeconomic and environmental benefits associated with food waste revalorization. Consistent with the conceptual framing advanced earlier, our results suggest that participation in food waste revalorization is shaped by a combination of demographic characteristics, socioeconomic resources, and farm-level conditions. These findings reinforce the argument that the practical realization of circular economy strategies within smallholder farming systems depends on farmers’ access to resources, knowledge, and relevant infrastructure.
Gender emerged as one of the strongest demographic predictors of food waste repurposing, supporting existing literature that links access to productive resources and agricultural information to participation in sustainable agricultural practices (Peterman, Behrman and Quisumbing, Reference Peterman, Behrman and Quisumbing2014; Tsikata and Yaro, Reference Tsikata and Yaro2014; Quaye et al., Reference Quaye, Fuseini, Boadu and Asafu-Adjaye2019). Women farmers were significantly less likely to repurpose deteriorated produce than men, suggesting that structural inequalities continue to shape participation in circular agricultural activities.
The gendered nature of food waste repurposing reflects broader inequalities in access to transportation, agricultural technologies, extension services, and productive assets. In the study area, as in many smallholder farming systems across Africa, women frequently transport vegetables to markets through head portage, whereas men are more likely to have access to bicycles, motorcycles, or other transportation assets (Abass et al., Reference Abass, Ndunguru, Mamiro, Alenkhe, Mlingi and Bekunda2014). Carrying vegetables over long distances not only increases physical labor burdens but may also expose highly perishable produce to heat and sunlight, thereby increasing the risk of postharvest deterioration (Nordhagen, Reference Nordhagen2021). These findings are particularly important in the context of food waste revalorization because repurposing often requires transporting deteriorated produce from markets back to households or other users. In the study area, deteriorated vegetables may be used as livestock feed, composted, dried for later use, or processed into other products. Such activities require access to transportation, labor, and technical knowledge, resources that women often access less readily than men. Previous research has similarly shown that women frequently face barriers to postharvest technologies and information, reducing their ability to participate in value-recovery activities (Adejumo et al., Reference Adejumo, Okoruwa, Abass and Salman2020; Snider et al., Reference Snider, Adraki, Lolig and McNamara2023). The observed relationship between market distance and repurposing further reinforces this interpretation. As distance to market increased, the likelihood of repurposing declined. For many farmers, especially women, transporting deteriorated produce over long distances on poor road networks may simply be impractical, making disposal a more convenient option despite the potential value embedded in the produce.
Socioeconomic status also emerged as a significant predictor of food waste repurposing. The declining likelihood of repurposing across wealth categories suggests that poorer farmers may place greater value on recovering resources from deteriorated produce. This finding aligns with evidence that cost-saving considerations often shape postharvest management decisions among smallholder farmers (Janousek, Markey and Roseland, Reference Janousek, Markey and Roseland2018). For resource-constrained households, repurposing deteriorated produce may represent an important strategy for recovering value from lost harvests. From a circular economy perspective, this finding demonstrates that participation in food waste revalorization is influenced not only by environmental motivations but also by economic necessity. Poorer farmers with relatively smaller harvests and fewer alternative income sources may be more motivated to convert deteriorated produce into useful products rather than absorb the loss entirely.
Although poorer households were more likely to repurpose deteriorated produce, farmers with access to credit also exhibited a higher likelihood of repurposing. These findings are not contradictory. While relative poverty may increase the incentive to recover value from deteriorated produce because poorer households are less able to absorb postharvest losses, repurposing itself often requires complementary resources, including transportation, labor, livestock, composting facilities, or simple processing equipment. Farmers with access to credit are better positioned to acquire or finance these productive resources, thereby increasing their capacity to convert deteriorated produce into alternative uses rather than discard it. Consequently, our findings suggest that both economic necessity and productive capacity jointly shape farmers’ decisions to engage in food waste revalorization.
The negative relationship between livelihood diversification and repurposing further supports this interpretation. Farmers engaged in multiple income-generating activities may be less dependent on recovering value from deteriorated produce because they possess alternative livelihood options. Conversely, households that depend primarily on farming may have stronger incentives to extract value from food losses. At the household level, repurposing frequently occurs through informal arrangements in which deteriorated vegetables are transferred to livestock keepers or retained for use by farmers who rear animals themselves. In the study area, livestock systems commonly include goats, sheep, cattle, and pigs, all of which can create opportunities for utilizing certain forms of deteriorated produce. These localized exchanges illustrate how food waste revalorization can contribute to circular resource flows within rural communities.
The findings further demonstrate the importance of knowledge and institutional support in promoting food waste revalorization. Farmers without access to agricultural advisory services were significantly less likely to repurpose deteriorated produce. This is unsurprising given that many repurposing activities require technical knowledge regarding safe handling, compost production, animal feeding practices, and other forms of resource recovery. Agricultural extension services remain one of the most important mechanisms for disseminating knowledge on sustainable agricultural practices (Asante et al., Reference Asante, Ma, Prah and Temoso2024). Our findings therefore suggest that extension systems have an important role to play in translating circular economy principles into practical on-farm actions. The strong positive association between prior adoption of sustainable land management practices and food waste repurposing further supports the importance of knowledge, experience, and environmental awareness in shaping farmers’ decisions.
Extension services in many smallholder settings continue to focus predominantly on production-related challenges while devoting limited attention to postharvest management (Kitinoja and Barrett, Reference Kitinoja and Barrett2015). There is therefore a need to broaden agricultural advisory services to include training on food waste repurposing and value-recovery strategies. As Stathers et al. (Reference Stathers, Holcroft, Kitinoja, Mvumi, English, Omotilewa, Kocher, Ault and Torero2020) argue, such efforts should extend beyond farmers to include other actors across agricultural value chains. Particular attention should be paid to farmers living farther from major roads and market centers, who often face greater challenges in accessing extension services and market opportunities.
Limitations
This study provides important insights into the determinants of farmers’ decisions to repurpose deteriorated produce, but several limitations should be acknowledged. First, the study relies on cross-sectional survey data, which limits the ability to draw causal inferences about the relationships between socioeconomic, demographic, and farm-level factors and repurposing behavior. While logistic regression allows us to identify statistically significant associations, longitudinal or panel data would provide stronger evidence on how farmers’ repurposing decisions evolve over time and in response to changing market and environmental conditions. Second, the analysis is based on farmers’ self-reported data on food loss levels and repurposing practices. Such data may be affected by recall errors or social desirability bias, potentially influencing estimates of both food loss and repurposing activities. Third, the study focuses on smallholder vegetable farmers within a specific regional context in Ghana. Although the findings provide valuable insights for similar smallholder systems across SSA, contextual differences in infrastructure, markets, institutional support, and cultural practices may limit the direct generalizability of the results to other settings. Finally, while the study captures several key socioeconomic and farm-level determinants, it does not fully examine value chain dynamics, such as the roles of traders, processors, and livestock keepers in facilitating the repurposing of deteriorated produce. Future research incorporating broader value chain perspectives would deepen understanding of the opportunities and constraints associated with food waste revalorization.
Conclusions
This study examined the determinants of smallholder farmers’ decisions to repurpose deteriorated agricultural produce using survey data from vegetable farmers in Ghana. The findings demonstrate that food waste revalorization is shaped by a combination of demographic, socioeconomic, and farm-level factors. Gender, wealth status, land access, credit access, livelihood diversification, crop type, food loss levels, market distance, access to agricultural advisory services, and prior adoption of sustainable agricultural practices all significantly influence farmers’ repurposing decisions.
The study contributes to emerging scholarship on circular economies and food waste valorization by demonstrating that participation in food waste revalorization is deeply embedded within broader social, economic, and institutional contexts. While repurposing offers an important pathway for improving resource efficiency and recovering value from food losses, farmers’ ability to engage in such practices depends on access to resources, knowledge, infrastructure, and markets.
In smallholder farming systems, where deteriorated produce is frequently abandoned due to storage, transportation, and market constraints, repurposing offers a practical opportunity to recover value through livestock feed, composting, biochar production, and other productive uses. Strengthening support for these activities can contribute to more resilient livelihoods, more sustainable food systems, and locally grounded circular economy transitions.
Direction for policy and practice
Overall, our findings identify several important entry points for promoting food waste revalorization within smallholder farming systems. Agricultural policy should prioritize strengthening extension services, reducing gender inequalities in access to productive resources, improving rural infrastructure, and expanding access to credit.
The findings also suggest that food waste revalorization should be integrated into broader discussions of circular economy transitions in agriculture. Rather than viewing deteriorated produce solely as a problem to be eliminated, policies should recognize its potential as a resource that can generate environmental and socioeconomic benefits.
Gender-sensitive extension programs are particularly important given the significant barriers faced by women farmers. Such programs should focus on improving women’s access to transportation, agricultural information, postharvest technologies, and productive assets. Investments in roads and transportation infrastructure could simultaneously reduce postharvest losses and facilitate the movement of deteriorated produce for alternative uses.
Policies should also support stronger linkages between vegetable farmers and other potential users of deteriorated produce, including livestock keepers, compost producers, and local processing enterprises. Creating such connections would help transform deteriorated produce from a disposal challenge into a valuable input within localized circular economy systems.
Directions for future research
Future research should expand the evidence base on FLW revalorization within smallholder farming systems. Longitudinal studies would be particularly valuable in capturing temporal and seasonal dynamics in food loss and repurposing behavior, as well as assessing how farmers adapt their strategies over time. Further research is needed to examine the economic feasibility and scalability of different repurposing pathways, including converting deteriorated produce into livestock feed, compost, biochar, and other value-added products. Such analyses would help determine whether repurposing can evolve from an informal coping strategy into a viable livelihood or enterprise opportunity.
Future studies should also explore how demographic and socioeconomic factors, particularly gender, shape participation in different food waste revalorization pathways and the extent to which these pathways contribute to livelihood resilience and resource efficiency.
In addition, future studies should adopt a broader food systems perspective by examining the role of other actors in the agricultural value chain, such as traders, transporters, food vendors, livestock producers, and processors, in facilitating or constraining the circulation of deteriorated produce. Understanding these interactions would help identify opportunities to develop structured markets for repurposed food waste and to strengthen localized circular economy systems.
Given the importance of context in shaping food waste practices, comparative studies across different regions and farming systems would also be valuable in assessing how infrastructure, market access, cultural norms, and production systems influence the adoption of food waste revalorization strategies.
Finally, gender-focused research is needed to better understand the structural and institutional barriers that limit women’s participation in repurposing activities and to identify targeted interventions that could improve gender equity in postharvest management practices.
Data availability statement
The data that support the findings of this study are available from the lead author, [anonymized for review], upon reasonable request.
Acknowledgments
The authors would like to thank all the smallholder farmers who participated in this study for their time and valuable insights. We are also grateful to the field enumerators and local facilitators who supported the data collection process. Their contributions were essential to the successful completion of this research. The authors confirm that no artificial intelligence (AI) tools were used in the writing, analysis, or preparation of this manuscript.
Author contributions
M.M.K.: Conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, resources, software, supervision, writing—original draft, writing—review and editing; N.S.: Data curation, validation, visualization, writing—original draft, writing—review and editing; C.K.A.P.: Methodology, project administration, software, validation, visualization, writing—review and editing; L.S.: Validation, visualization, writing—original draft, writing—review and editing; I.B.: Data curation, investigation, validation, visualization, writing—review and editing.
Funding statement
This work was funded through the seed grant program of the Global Food Institute.
Competing interests
The authors have no competing interests to declare.
Open access
