Food insecurity: addressing a challenging social problem with supply chains and service ecosystems

Abstract

Purpose

Extant literature in supply chain management tends to address a portion of the product flow to make food accessible to clients in need. The authors present a broader view of food insecurity and present nuances relevant to appreciate the complexities of dealing with this social problem.

Design/methodology/approach

The authors conducted an inductive study to reveal the deep meaning of the context as managers of nonprofit organizations (NPO) define and address food insecurity. The focus was on a delimited geographic area for capturing interactions among NPOs which have not been described previously.

Findings

This study describes the role of supply chains collaborating in unexpected ways in the not-for-profit context, leading to interesting insights for the conceptual development of service ecosystems. This is relevant because the solution for the food insecure stems from the orchestration of assistance provided by the many supply chains for social assistance.

Research limitations/implications

The authors introduce two concepts: customer sharing and customer release. Customer sharing enables these supply chains behave like an ecosystem with no focal organization. Customer release is the opposite to customer retention, when the food insecure stops needing assistance.

Social implications

The authors describe the use of customer-centric measures of success such improved health measured. The solution to food insecurity for an individual is likely to be the result of the orchestration of assistance provided by several supply chains.

Originality/value

The authors started asking who the client is and how the NPOs define food insecurity, leading to discussing contrasts between food access and utilization, between hunger relief and nourishment, between assistance and solution of the problem, and between supply chains and ecosystems.

Citation

García-Dastugue, S.J., García-Contreras, R., Stauss, K., Milford, T. and Leuschner, R. (2024), "Food insecurity: addressing a challenging social problem with supply chains and service ecosystems", The International Journal of Logistics Management, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/IJLM-03-2023-0113

Introduction

Food insecurity is the “… lack of access, at times, to enough food for an active, healthy life for all household members and limited or uncertain availability of nutritionally adequate foods” (Carle and Rosenberg, 2018, para. 2). In 2019, the 76 low-to-middle income nations were home to 19.3% of the 3.8 billion food insecure people (Meade, 2019). While the proportion has decreased, at least 446 million in these low-to-middle-income nations are food insecure (Thome et al., 2019). Also, large numbers of food-insecure people struggle in developed nations; for example, in the United States, one in ten households experienced some level of food insecurity (Coleman-Jensen et al., 2020). While COVID-19 has intensified the problem (Blackmon et al., 2021), food insecurity preexisted and is likely to continue (Durisin, 2022), “As the cost of nearly everything surges, more Americans are turning to food banks to eat” (Gibson, 2022, para. 1). Food insecurity persists; for example, in the United States, it ranged between 10.5% and 11.9% since the 2000s, except for a spike above 14% in 2008, to decline back to the 10% level in 2019 (USDA-Economic Research Service, 2021). While food insecurity is a worldwide challenge, it shows idiosyncratic territorially bound characteristics (Berrone et al., 2016; Brenner, 1998).

Extant supply chain management (SCM) literature in food insecurity tends to focus on food distribution and redistribution (Ataseven et al., 2018). The food banking model implies distributing food through partners – food pantries, schools and shelters – (Ataseven et al., 2020; Blackmon et al., 2021; Feeding America, 2021). The study of food insecurity beyond food distribution is scant; less is known about what happens facing the client. Thus, there is an opportunity to further our understanding of nuances that are relevant to designing and managing supply chains (SC) for dealing with food insecurity, particularly with a focus on the end-customer or client (individual or head of the household in need).

Our motivation is to better understand food insecurity from a SCM point of view by exploring how different nonprofit organizations (NPO) define the problem and design their SC to respond. We asked the following: Who is the client and how is customer value defined? What do NPOs do to deliver that customer value to the client? We conducted a systematic inductive qualitative study. Following prior work (O'Connor et al., 2003; Sanders and Wagner, 2011; Wowak et al., 2016), and to account for the interdisciplinary nature of food insecurity (Pohl, 2011; Sanders et al., 2019) we adopted these research ideals: (1) use a multidisciplinary team (Sanders and Wagner, 2011) because it “… has greater potential for the development of more complete solutions to contemporary SCM problems” (Sanders et al., 2016, p. 108); (2) design the study to transcended organizational boundaries of the focal NPO; and (3) adopt a broad view of SCM to include product flow, processes, and SC design (Mentzer et al., 2001; Wowak et al., 2016).

Our study provides a two-way bridge between SCM and the management of NPOs in the context of food insecurity. First, it contributes to the dialogue from a SCM view of food insecurity by showing that it is a complex problem for SCM and that, at an aggregate level, requires a holistic solution in addition to distributing food. Second, we provide SC implications beyond the product-flow-centric view. Dealing with food insecurity requires service components that can only be described by adopting a service-centered logic of the social problem (Vargo and Lusch, 2004). Third, our data showed a level of interaction across NPOs unexpected in the for-profit context based on resource-sharing. Most interestingly, unlike current literature (Kapoor, 2018; Ketchen et al., 2014), customer-sharing reflected a pure service ecosystem where there is not a focal organization (Stolze et al., 2016; Tansakul et al., 2023).

Literature review

This study is focused on the NPOs dealing with food insecurity by integrating SCs; we review the literature around three major themes: defining food insecurity, providing food access and new SC designs beyond food access.

Defining food insecurity

Food security is defined as access “to safe and nutritious food and water adequate to sustain an active and healthy life with dignity” (FAO, 1996). Given that food security is unobservable (Barrett and Lentz, 2016), the antonym, food insecurity, is frequently used and is defined as “… lack of access, at all times, to enough food for an active, healthy life” (Raskind et al., 2019, p. 476). Food insecurity might result in individuals experiencing “disrupted eating patterns and reduced food intake” (Coleman-Jensen et al., 2020, p. 25) which risks affecting the individuals' active and healthy life with dignity.

Addressing food insecurity may seem straightforward, if the problem is lack of access to food, the solution is providing access to food. However, several paradoxes depict how challenging this social issue is (Richards et al., 2021). The overarching one is that even though global food production is sufficient to feed the entire population of the world, a significant number of people lack access to the necessary food products (FAO, 2011). Second, the wealth of an economy does not guarantee the ability to provide access to food for everybody (Irani and Sharif, 2018). A third paradox is that even though many individuals lack access to the necessary nourishment to maintain an active and healthy life, food waste is substantial (Moates et al., 2016). The fourth is that lack of access to food is associated with obesity (Dhurandhar, 2016; Muzigaba et al., 2016). This indicates that access to food and access to nourishment are two distinct problems; that is, one can have access to food and still be unhealthy or undernourished. Ultimately, it should be noted that social problems related to food are varied and complex (Ataseven et al., 2020; Long and Wood, 1995; Wills, 2017).

The causes of food insecurity are varied, complex to determine, and probably impossible to articulate an agreed-upon comprehensive list (Godfray and Robinson, 2015; Irani and Sharif, 2018). First, food insecurity can happen in any geography and in any economy (Araújo et al., 2018; Borch and Kjærnes, 2016; Hanson and Connor, 2014; Irani and Sharif, 2018). Second, unlike disaster relief, there is no beginning and end; that is, food insecurity is chronic (Cason, 1999; McDevitt, 2018; Wills, 2017). Social workers explain that a food insecure individual “does not know where the next meal is coming from.” The condition of food insecurity precedes the state of hunger (Cason, 1999, p. 49). Lack of access to food happens because excess food is not where it is needed when it is needed (Bals and Tate, 2018; Barrett, 2010; Long and Wood, 1995).

Providing food access

In the United States, Feeding America is the largest network of food banks (Ataseven et al., 2020) with 200 food banks across the country (Feeding America, 2021). Downstream, they operate through partner agencies such as food pantries, kitchens, schools and shelters (Ataseven et al., 2020; Blackmon et al., 2021) who deliver and, thus, have a relationship with the client. When compared to a contemporary SC of a business, this structure resembles a traditional, long-distribution channel with suppliers (producers/manufacturers), distributors (e.g. Feeding America), wholesalers (food banks), retailers (food pantries) and end-customer (clients). While not all food banks in the United States are partners of a network, the food banking/food pantry model is the most frequent. The food banking model attracted most research attention in SCM. Blackmon et al. (2021) focused on the operation of a food bank (box preparation). Some use food banks, though, their focus resembled that of an operational study in any [for-profit] context. Solak et al. (2014) studied a location-routing problem in the context of food distribution at food banks. Similarly, Biswal et al. (2018) studied RFID adoption to improve warehouse operations in a food bank.

SCM studies about food insecurity, either in the context of food banks or not, primarily focus on product flow operations. This is reflected on the performance measures used; they predominantly are process-centric in contrast to customer-centric measures. Ataseven et al. (2020) use the amount of food distributed per food insecure individual; Blackmon et al. (2021) use the number of cases delivered from the food bank to food pantries; Davis et al. (2016) center on forecast accuracy of prediction of food donations. In sum, in SCM, research dealing with food insecurity focuses on food access to the food insecure, and tends to focus on the operational aspects of managing the product flow.

Beyond food access: new supply chain designs

Some believe that dealing with or solving the chronic aspect of food insecurity requires more than providing food access. Specifically, it requires “[…] long-lasting, systemic solutions involving additional programs such as child nutrition, nutrition education, and job training for the unemployed” (Mendoza-Abarca and Gras, 2019, p. 988). This underscores the difference between access to food and to nutrition, and is relevant because there is a strong bidirectional causal relationship between poverty and malnutrition (Lentz and Barrett, 2013). Likely, this systemic solution called for by Mendoza-Abarca and Grass (2019) will require SCs to deal with issues beyond managing the product flow (distribution of boxes of foods). Annossi et al. (2021) highlight the re-conceptualization of SC designs to leverage the use of digital technologies to prevent food waste. Sundgren (2020) found new actors adopting new roles in closed-loop food SCs. Along the same lines, Ciulli et al. contend “circularity brokers” (2020, p. 299) are needed to connect waste generators and receivers. In sum, dealing with the social problem of food insecurity likely requires a more complex solution that transcends providing access to food. For example, Wills, supporting the “criticism of the dominant food bank model” (2017, p. 62), calls for the development of social enterprises that complement existing efforts. Our literature review shows that little is known about what happens at the customer interface. Also, while, strictly speaking, the definition of food insecurity is based on nourishment, in our field, studies tend to overlook the distinction between food and nourishment (e.g. Martins et al., 2019).

Methodology

We used inductive research based on qualitative data to examine food insecurity from the point of view of SCM. This required a detailed examination of the data using the language and mindset of NPO managers to reveal the deep meaning of the context as they define and address food insecurity (Kirchoff et al., 2016; Lincoln and Guba, 1985). We approached the data without pre-determined models and used interpretation and inductive association. This methodology is suitable for exploring questions “involving inter-firm integration/collaboration/relationship issues”; that is, “questions involving complex social interaction within and between firms that can benefit from a deeper understanding of underlying concepts and their interrelationships” (Mello and Flint, 2009). We followed the tenets of systematic inductive research to guide our procedures (Gioia et al., 2013; Quarshie and Leuschner, 2020; Villena and Gioia, 2018). In addition, we interactively reviewed literature, and collected and analyzed data (Glaser et al., 2013; Glaser and Strauss, 2017; Kirchoff et al., 2016). Data collection, coding and analysis happened jointly (Cohen et al., 1969; Holton, 2010). At each iterations, using theoretical sampling, we decided what data to collect next (Cohen et al., 1969; Corbin and Strauss, 1990).

Context and sample

The context for this study is four counties in Northwest Arkansas. The US state of Arkansas, the fifth highest food insecure population (17.3%) and the second highest population for children (23.1%) (“Map the Meal Gap. Food Insecurity in the United States”, 2020). All four had food insecurity rates higher than the national average of 11.5% in 2018: Benton (11.6%), Carroll (13.8%), Madison (15.7%) and Washington (14.4%) (Carle and Rosenberg, 2018). Initially, we planned to identify NPOs with novel SC designs in any geography. However, after interviewing the first two organizations, it emerged that “competing” NPOs interacted in unanticipated ways. For this reason, and consistent with studies in this domain (Sundgren, 2020) and the constructivist research tradition, we adapted the data collection strategy to focus on a limited geographic area. This approach has the benefit of understanding a concept deeply, without introducing additional noise in the data, at the expense of broader generalizability.

Our sample included food banking and all organizations that designed and implemented a SC solution different than the food banking model. In this geography, there were another 154 active organizations focused on food insecurity. These organizations were not included because they were the distribution channel of the food banking model: 101 are food pantries, and 47 organizations providing free meals (soup kitchens or meal delivery services). From the remaining six, four were community gardens, one delivered groceries (not at scale), and one recovered and re-distributed food products that otherwise would go to waste. Table 1 shows there were 14 interviewees representing 10 organizations. While the sample might seem low, it is above other studies in a similar research context (e.g. Sundgren, 2020).

The research study began in 2019 by reviewing publicly available supporting material and having initial contact with potential participants. Interviews happened between April and October 2020. All interviewees were conducted via videoconference and lasted between 75 and 120 min. While all NPOs were experiencing the impacts of COVID-19, all interviewees were specifically asked to focus on their experiences before the pandemic. Before each interview, we collected and studied secondary information from each organization, including publicly available material, reports and presentations used by NPO management previously, and interviews available in trade magazines and social media. Most of these organizations were small; thus, we had access to the founding or top management team. On the flip side, the printed material (usually available in qualitative research on large corporations) was limited. At least two researchers participated in each interview. Interviews were recorded and transcribed, and notes were taken. Coding started after the first interview and team debriefing meetings were held regularly. There were four main researchers and two graduate assistants.

Through a constructivist lens, researchers encouraged interviewees to reflect on their experiences as leaders and ascribe meaning to the structure, purpose, and strategies implemented. An interview guide was developed around five dimensions: (1) customer value creation (Bals and Tate, 2018), (2) relationships internal (Ataseven et al., 2018; Kahn and Mentzer, 1998), (3) relationships external (Ataseven et al., 2018; Gulati, 1998; Lambert et al., 1999), (4) execution (Patel et al., 2013) and (5) performance measurement (Randall et al., 2015). The interview guide is shown as Appendix A in Supplemental Online Material. The interview guide served to maintain consistency across all interviewees. Having interviewed all NPOs directly involved in integrating a SC to deal with food insecurity in the geography, we reached saturation.

Data analysis

The initial step was a first-order analysis (Gioia et al., 2013) coding of the interview transcripts. Three team members independently read and coded transcripts; one used manual coding, two used Atlas.ti Version 8. The researcher conducting the manual analysis read each transcript twice searching for reoccurring themes. The other two researchers did so, too, and used software features to identify reoccurring themes. Strauss and Corbin (1994) suggest comparing each new narrative to prior interviews. Researchers reviewed the results and discussed discrepancies. The second step was to formulate second-order themes (Gioia et al., 2013), which are “researcher-induced concepts cast at a more abstract level” (Wowak et al., 2016). We developed coding networks (Atlas.ti) rooted in the interview guide (see Appendix A in online supplemental material) to identify relationships among codes and second-order themes emerged. Second-order themes were grouped in third-order aggregate dimensions (Gehman et al., 2018; Quarshie and Leuschner, 2020). The outcome of this process is the data structure in Figure 1. Aiming to balance between how much data to show and our interpretation of these data (Pratt, 2009) and following established practices in our field (Quarshie and Leuschner, 2020), proof quotes supporting each first-order concept in the data structure are provided as Online Supplemental Material Table S1.

During the analysis, we returned to the literature on management, operations management, SCM, social work and supply networks. Sundgren (2020) underscores the value of focusing on a limited geographic area to study food insecurity; this is a specific example of the value of going back to the literature during analysis. Most interviewees had limited formal SCM expertise; however, we relied on accepted terminology in the SCM field to describe findings.

Findings

Each NPO defined food insecurity uniquely, which in turn guided their work, resulting in each organization having a SC that reflected their conceptualization of food insecurity. Even the discussion about who the client was (recipient of the value created), was not unanimous. Some surmised the clients were the individuals or households that experience food insecurity, while others argued that the client was the community at large as an indirect recipient of the benefits associated with working on reducing food insecurity. For this study, we chose to define the client as a food-insecure individual who requires assistance for herself/himself or for the household. This was chosen because statistics on the issue frequently count at the individual level. The description of the findings followed the data structure shown in Figure 1, organized in the two aggregate dimensions that emerged from the data: (1) problem definition and (2) SC design (Ketchen et al., 2014; Pathak et al., 2014; Stolze et al., 2016; Tansakul et al., 2023).

Aggregate dimension 1: problem definition

Three themes emerged to describe problem definition: (1) Food, (2) Access and (3) Other Service Components. Understanding the way these organizations view food insecurity enables a deeper understanding of the associated SC complexities. Food involves two concepts that pose distinct challenges: (1) hunger and (2) nutrition. The basic response to ameliorate hunger is to provide food. Some perceived food insecurity as a hunger problem, others, consistent with FAO’s definition, perceived it as a nutrition deficit. This contrast has substantial implications for the customer and the SC. Shelf-stable foods ameliorate hunger while easing operational complexity. But most shelf-stable foods are of lower overall nutritional value (nutrients as they relate to, for example, carbohydrates, fats, sodium and sugars). Of the organizations interviewed, some explicitly described the main problem as lack of nutritious food, even when food (though less nutritious) in general can be reasonably available.

Access includes three concepts. The first two concepts, Place Utility and Time Utility, from a SCM point of view are connected. Regarding Place Utility, some clients are home-bound; thus, they do not have the means to go to a food pantry (part of the food banking distribution channel). Concerning Time Utility, many food pantries are open during hours when volunteers are available to provide their services, but inconvenient for working clients. For a business, this could be an easy fix based on a revenue-minus-cost. For NPOs relying on volunteers, personnel operating a food pantry are often retired or have limited availability. As shared by one interviewee, “… most of them are all volunteers, and the volunteer base [here], as well as the rest of the world, is aging out and there is not the next age of volunteers coming through.” Additionally, operating hours are important because, as one interviewee asserted, “[i]n most cases, the food pantry client is working at least one job, perhaps two.”

The last concept of Access is Food Deserts, which, to our interviewees, refers to geographic areas where providing access to food is challenging for an NPO. Frequently, food deserts imply minimal access to grocery stores and/or that availability of healthy food to the customer is limited (Walker et al., 2010). In this study, because of a combination of lower population density, longer distance to the NPOs’ locations, and a lack of food pantries (frequently associated with houses of worship), there are some geographic areas where it is more difficult to provide accessible food.

The third theme was Other Service Components. From our findings, this theme included several factors that added depth to the understanding of the complexity of food insecurity. Within this theme, the challenges that frequently emerged were Age, Social Isolation, Education and Stigma. The first are connected; however, social isolation can relate to other – sometimes more challenging – issues to manage such as chronic illnesses or mental health. Age and isolation resulted in additional complexity to access.

Education was manifested in two ways. One was the challenge faced by everyone. This includes knowing what to eat and being aware of the health implications of eating processed foods – usually with higher contents of carbohydrates, fats, sodium and sugar (Gearhardt and Hebebrand, 2021). Participants reflected there was a need to educate clients on the fresh produce available and how to cook these items. One interviewee described:

We realized there were different items that people were unfamiliar with, like maybe eggplant or bok-choy. So, we started coming up with recipes with pictures so that you could see that looks delicious even though they’ve never used it. We made sure all of our recipes were simple, still tasty, and very affordable to make.

The second education challenge was how to access programs such as SNAP (the Supplemental Nutrition Assistant Program run by the USDA). NPOs frequently assisted individuals in navigating the systems to access to available governmental programs.

Stigma was the last concept in Other Service Components. Two distinct aspects of stigma emerged. The first included the feelings from the client. This refers to the human tendency to not seek help which might be the result of fears and cultural judgments. Understanding stigma is relevant to dealing with the challenges of Time and Place Utility. Also, stigma was related to fundraising and donors. It is important to keep in mind that these organizations do not receive revenue from clients but fund their operations via donations and grants. Stigma ranged from being unaware of the problem of food insecurity to the belief that food-insecure individuals were abusing the system (see Table S1 in Supplemental Online Material).

Aggregate dimension 2: SC design

SC Design describes the SC NPOs designed to deal with food insecurity. Our purpose for this study was to further our understanding of food insecurity from the SCM point of view. Thus, the aggregate dimension SC Design reflects the structure of the interview guide: (1) Value Creation, (2) Relationships External, (3) Execution and (4) Performance Measurement. The fifth dimension, relationships internal, did not result in rich data. We believe that the fact that the participating organizations were small, in terms of the number of individuals working full-time, internal coordination was done informally.

The first theme, Value Proposition included (1) Target Customer, (2) Product, (3) Time and Place Utility and (4) Customer Experience. The food insecure individual or household was present in all discussions about for whom the NPO creates value. Defining who the target customer was could be accompanied by a greater level of specificity which represented a market segment or a target market. For example, some NPOs target the elderly food insecure individual while others define their target customer as “everybody.” One NPO manager replied, “You know, that’s the $64,000 question in our business”; another said, “so, that’s a multilayered answer.” Regarding the Product that was distributed, the physical product was a standard bundle of food (shelf-stable, fresh produce and frozen protein). To this, a manager stated: “you want to give them food that they won’t necessarily have to cook like cucumbers or peppers or tomatoes or fruits.”

In response to the 2nd order theme Access (part of Aggregate Dimension #1), Time and Place Utility was connected to Value Proposition. For instance, NPOs use food pantries or design alternative methods for direct delivery (direct-to-consumer in business). Everyone mentioned that, for strategy development, NPOs consider how their value proposition will complement that of existing NPOs.

Interestingly, the last concept was Customer Experience. Some sought to create an experience in addition to delivering food. For example, social isolation is a problem for some populations such as the elderly. After receiving assistance from one of the NPSs, they showed interest in becoming a volunteer, either coordinating activities or acting as an impromptu volunteer. Also, mobile food pantries created a socializing experience as the vehicle tended to remove the stigma felt by clients. When the goal of the NPO is to raise awareness about nutrition, the focus became creating an educational experience.

Relationships External included five first-order concepts. The first two were Suppliers and Channels of Distribution. The most distinct characteristic of management of these relationships were reported by most interviewees as the presence of an aspirational component that drives SC partners. For example, farmers know the value of eating healthier foods and put extra effort to make produce available rather than disposing of the produce because they do not want to waste their production (see quotes in Online Supplemental Material Table S1).

Additionally, literature underscores the importance of having a Board for guidance and networking (Lecy et al., 2012). Interviewees did describe the importance of active board with technical, strategic, and relationship development skills. Other Supporting Organizations with which the NPOs interacted include government agencies at the city or state level and the chamber of commerce. Successful social innovation initiatives usually involve such cross-sector collaborations (Von Jacobi and Chiappero-Martinetti, 2017). Thus, while these organizations were labeled as supporting, they frequently provided pivotal influence on how the problem of food insecurity was defined and how the SC was designed. Interviewees reported while there was a stigma surrounding food insecurity, increasingly Other Organizations, such as companies and governmental agencies, got involved. For example, Kraft Heinz committed a donation to Feed America during the COVID-19 outbreak (Mullen and Galia, 2020). One interviewee specifically described the challenge of coordination across governmental agencies:

… for example, let's just stick with the National School Lunch Program […] the complexity in understanding all the rules and regulations … I mean, this is a career in and of itself.

Execution of flows was the third theme. As expected from a product-based SC, the first two concepts that emerged were Product Flow and Information Flow. Interviewees recognized the value of information visibility. As in the previous literature (e.g. Ataseven et al., 2020; Blackmon et al., 2021; Martins et al., 2019; Ssennoga et al., 2019; Thapa Karki et al., 2021), inventory management, order fulfillment and demand management were concepts that emerged from the data. Regulations, traditionally, become liability barriers for companies to donate; for example, food that was near the use-by day. For this reason, Regulations emerged as a first-order concept. Interviewees agreed that the right regulations often facilitated the availability of foods to redistribute.

The last two first-order concepts in Execution were Resource Sharing and Customer Sharing as they showed how NPOs collaborated as an ecosystem. Resource sharing refers to inventory (food) available that needs to be allocated. It also refers to assets needed to store, transform or distribute available food. In the for-profit context, this is like Walmart and Amazon sharing a warehouse or Procter and Gamble and Colgate-Palmolive sharing manufacturing capacity. Noteworthy, resource sharing is used to serve customers as NPOs value more the ecosystem as a whole being able to respond than who does it. In the for-profit context, this is referred to as co-opetition, competing and collaborating at the same time (Pathak et al., 2014). For these interview participants, this was their norm, and there was no indication that the NPOs competed. However, like Ashley and Faulk (2010), they do compete for grants and donations.

Customer Sharing also emerged as another novel concept. Frequently, clients experienced challenges not handled by the NPO such as housing or work skills. For this reason, NPOs collaborated with other NPOs to provide the client the necessary bundle of assistance. This resembles the concept of supply ecosystems (Pathak et al., 2014) or service ecosystems (Stolze et al., 2016). Beyond this sharing among non-competitors, an NPO might share a customer with a competitor. In for-profit contexts, customer retention is central to obtain repeat business. In our context, the aspirational goal of assisting the food insecure was so prevalent that customer sharing became a resource to fulfill the mission of the NPO.

In the last theme, Performance Measurement, three concepts emerged. Previous research used product-flow-centric measures, such as pounds of food delivered (e.g. Blackmon et al., 2021), rather than customer-centric measures. Almost all NPOs reported using operational indicators including pounds of food delivered, number of volunteers involved or number of visits. One NPO used people-centered indicators. Given their focus on nutrition, managers were using health-related metrics, such as weight loss or lab tests, to measure value provided.

Performance Measurement also includes Financial Indicators. The contrast between for-profit and not-for-profit management emerged here. When addressing financial performance indicators, most of the interviewees also addressed funding. External and Self-Generated funding emerged. External funding included donations and grants; these sources were not without limitations or costs. Donors often impose how funds can be spent (Rossouw, 2006). Grants also require substantial administration and reporting. For example, one NPO who focused on managing fresh produce described how during winter – low season – they handle administrative tasks including reporting. Because of these burdens, many NPOs sought to self-generate a revenue stream to decouple from other sources of funding. One interviewee asserted: “we’re really just trying to make a revenue stream so that we aren’t so dependent on grants and the rat race that that is.” Subsequently, measuring social impact is complex, multilayered and context-specific (Von Jacobi and Chiappero-Martinetti, 2017).

Discussion

Our purpose for this study was to investigate how different nonprofit organizations (NPOs) addressed the social problem of food insecurity. Specifically, we inquired how customer value was defined, and what NPOs do to deliver said value. SCM literature tends to focus on the distribution and re-distribution of food to clients to address the problem. The literature review underscored the need to consider additional SC approaches to address food insecurity; other scholars have identified the need for new SC roles (Ciulli et al., 2020; Sundgren, 2020), the need for additional connectedness among NPOs (Annosi et al., 2021), and the need for developing complementary solutions (Wills, 2017). Initially, we had the intuition NPOs collaborated. We found a mesh of independent SCs that, from the point of view of the client (the recipient of assistance), behaved like a service ecosystem. Vargo and Lusch proposed service-centered dominant logic and asserted that “new perspectives have emerged that have a revised logic focused on intangible resources, the cocreation of value, and relationships” (2004, p. 1). Additionally, service ecosystems received increased attention (Adner and Kapoor, 2010; Ketchen et al., 2014; Stolze et al., 2016). In fact, Stolze et al. concluded “[i]ntegrating S-D Logic and service ecosystems thinking will provide a unique approach to integrating marketing and SC thought into a singular focus on shopper experiences within the service provision of the networked members of the ecosystem” (2016, p. 194).

Thus, we frame our discussion from the service-centered dominant logic lens. Table 2 provides a description of the dimensions used by Vargo and Lusch (2004, p. 7) to distinguish the good-centered dominant logic (GD) and the service-centered dominant logic (SD) and serves to guide our discussion. We want to stress that, based on our study, the two dominant logics coexist and, more importantly, complement each other in novel ways. For this reason, Table 2 includes an additional dimension to address SC design and service ecosystems. As additional online material, Table S2 shows the proof quotes provided in Table S1 reorganized in the goods- and service-centered dominant logics.

Unit of Exchange

Food insecurity is based on three pillars: availability, access and utilization of food (Barrett and Lentz, 2016) which present a hierarchical relationship (Webb et al., 2006). For a food item to serve as adequate nourishment, it must be utilized; for this to happen, it has to be accessible to the food insecure; and, for that to happen, food has to be available. Availability represents the supply side (Barrett and Lentz, 2016). Statements such as “[e]nough food is produced today to feed everyone on the planet, but hunger is on the rise in some parts of the world, and having 821 million people ‘chronically undernourished’” (UN, 2019, para. 1) speak about availability, and tacitly overlooks access and utilization. It is important to underscore that the availability of food (the supply side) seems to be considered less of a challenge than providing access (the demand side) to the food insecurity problem. In the words of one interviewee: “… we feel like we have access to all the product necessary to serve the public, but I think it’s a distribution … the term the last mile comes in to play a lot and I think that’s where we’re short, […] we are not getting it into the hands of the people.”

Thus, the distinction between access (product) and utilization (the service rendered by the product-in-use) remains to be discussed, and it fits well to reflect on the two dominant logics. These quotes exemplify the focus of both logics (all quotes also are part of the data structure shown in Figure 1 and are provided as Table S1 available as supplementary online material).

• GD: food items

Our goal, our vision, is that by the year 2025 – and we have an estimate of how many pounds of food that is – everybody who is food insecure in the area will have reasonable access to that food.

• SD: nutrition

What we started hearing is that not only was there hunger that’s being faced, but within that there was a lack of nutritional value.

Role of Goods

The role of goods is important in both logics (Lusch, 2011). Under GD, the focus is not only on the goods but on logistics services (Vargo and Lusch, 2004). Access is the output of logistics: time and place utility. For this reason, many assert that hunger happens because available food is not where it is needed when it is needed (Bals and Tate, 2018; Barrett, 2010; Wood et al., 1995). Under SD, goods are intermediate “products” used as appliance in value creation-processes. Thus, the focus turns to utilization, the third pillar on which food insecurity is based. These quotes reflect this distinction.

• GD: access to food items

What we found along the way was that a huge barrier for people is transportation, it’s mobility ….

• SD: utilization of food to improve nutrition

The ultimate goal is we want people to want to eat this [nutritious] food and incorporate it as an ongoing part of their diet.

NPOs using long distribution channels face the challenge that partner agencies are volunteers. Thus, it is not unusual for food pantries to operate on limited days and times. For example, one states: “Pantry Hours: Wednesday 9:00 a.m.–11:00 p.m. We serve all who come” (FoodPantries.org, 2022, para. 1). Thus, NPOs, like businesses, seek to implement direct-to-consumer access with mobile pantries. The CEO of one NPO stated: “Well, we don’t have a lot of direct distribution to the client, that’s where the dilemma comes in. There are about 65,000 food insecure people in the four-county area. We have 160 partner agencies […] Now the caveat to that is that we do have 13 mobile pantry locations that we distribute ourselves.”

Role of Customers

Operating at the utilization level involves helping individuals improving eating habits. A longer distribution channel allows for reaching more clients (Ataseven et al., 2020). However, NPOs focused on utilization require customer intimacy, but they do so at the expense of scale. Accordingly, the role of customers in GD is to take part in the product flow as the output is access. In SD, the client becomes a coproducer by applying received knowledge and transforming the provided food items to fit specifically the client’s health needs.

• GD: recipient of goods

[ours is] a client-choice pantry where they [clients] actually walk through like a grocery store and say I want this, this, and this … Some of the others are not client-choice, they pre-bag and this is it, “Here’s your bag thanks very much” and this is what you’re going to get.

• SD: the client is a coproducer

So, we are primarily focused on nutrition education with our partner organizations because often food insecure folks are more at higher risk for health issues and diet related health issues and often the food that food pantry isn't always the most nutrient dense.

Determination of Value

In GD, value is determined by the producer (Vargo and Lusch, 2004). Contrastingly, in SD, value is determined when using the product. Under GD, the approach is focused on maximum output and efficiency, including product standardization. In contrast, the determination of value under SD, customization and education to support food utilization (the third pillar of food insecurity) is part of the value proposition, which materializes once the client transforms that proposition into value-in-use.

• GD: access to a standardized bundle of food items

… we’ll serve in a two-hour window between 250–300 families and give them a box of shelf-stable food, a bag of fresh produce, and a bag of frozen protein.

• SD: food choices, cooking healthy

Someone could get access to food, … but maybe it wasn’t the food that they necessarily needed or not all of what they needed. We found that while a lot of the food is not something that they are familiar with, we will have recipe boxes ….

NPO-Client Interaction

Under GD, interactions are transactions; in each, a standard unit is the locus of the transaction. Under the SD logic, NPO-client interactions are customized to the client, who, in turn, is actively involved as a coproduced. In the quotes below, this contrast is shown between the recipient of a standardized transaction, the 30-to-40-pound box, and the level of customization implemented under SD to match client’s needs and preferences.

• GD: transactions

We basically tell our end customer, each of the residents, we're going to bring you a 30-to-40-pound box of perishable and non-perishable product. So, you're going to get canned goods, you're going to get spaghetti, beans. You're going to get some level of produce and some kind of meat.

• SD: relational, customization, customer release

Customization: She’s from West Africa, Nigeria. She got tears in her eyes and said, ‘this is the food of my homeland.’ From then on, we exchanged phone numbers an any time I get some sweet potato vines I call her up and we meet somewhere in town, and I just go straight to her because I know that her and her family members are just going to love it and use it.

Customer release: … success for me would be not meeting the need but preventing the need in the first place. So, providing individuals with enough resources to get them out of the situation they are in. … I mean, I would like to think success would be not needing the services anymore ….

Customer release differentiates the NPO and the for-profit contexts. While some food insecure individuals may be dependent on community resources, some clients could become food secure by their own means. In business, customer retention, repurchase intention and lifetime value of the customer are associated with success. Contrastingly, in the context of food insecurity, success is customer release, that is, helping individuals develop the necessary resources to stop needing community assistance.

Supply Chain Design

Consistent with the contrast provided with the multiple dimensions described above, GD favors thinking about the management of flows in the SC. The management of both flows, products and information, emerges from the data. The next proof quotes represent well the relevance of the management of both flows.

• GD: Management of flows in the SC

Product flow: We serve much like a wholesale grocer where we collect food, store food, and distribute it through our partner agencies.

I believe we have a logistics and distribution problem more than a procurement problem.

Information flow: … out of our partner agencies we have thirty-four that […] subscribe to the software […]. … You [the client] are logged into the system with your name, address, how many dependents, it builds a profile for you …, we can take and mine data, […]. We can also identify if you’re having to go to multiple locations. […].

NPOs in this study achieved a degree of collaboration that leads to reflect on service ecosystems. Consistent with extant literature (Mendoza-Abarca and Gras, 2019), our data show that many food-insecure clients also experience other challenges such as housing- or job-related. For this reason, the level of collaboration across NPOs that provide complementary services is substantial.

• SD: Ecosystem

Collaboration: We work very hard at collaborating with other [NPOs] … [Affordable-Housing-NPO] job is to build that house. But I recognize that if somebody has to have help to build a house, more than likely they are food insecure. So, when [a client] moves into her new house, I’ll be there to stock her cabinets for the first time. […] We’re working with [NPO] trying to find a way that those students who are single parents don’t have to worry about food.

Resource Sharing: No, I don’t own any of those things. I mean, [COMPETING NPO] owns like fridges and freezers and they have a trailer. So, they have resources that I have access to.

Customer Sharing: On Wednesdays, we work with [this other NPO who works in food insecurity too], we’ve worked with them for 9 years now, and they actually distribute to [a couple of] places where […] people are living on a fixed income and have very little access to food.

Given the lack of competition for customer revenue, NPOs in this context collaborating with competitors for the sake of responding to client’s needs lead to sharing customers if needed. When further inquired about this level of collaboration, an interviewee shared a vision of the level of interaction [and integration] of the multiple independent organizations that pool resources to deliver value to clients:

… we’re kind of thinking about that as the four corners: a patient is connected within their [health care] provider and then referred to the nearest partner organization [NPO] that has access to the foods the patient will need [not just any food, but the one that meet the patient’s nutritional needs], the order is fulfilled as a [an order for our NPO], at [another NPO], and then the order is kicked out to be picked up by either a volunteer or driver, or at scale I think probably [a ride-hailing service], […] and then finally to the patient. I think with the patient that feedback loop is going to be really important.

NPOs achieved a degree of collaboration that leads to reflect on ecosystems. In this context, sharing clients and resources transforms the multiple independent SCs into a mesh of SCs which becomes an ecosystem with no central organization owning customer relationships. Conclusively, this study provides initial support to describe a duality of SCs and ecosystems. “[The] SC perspectives reflect the knowledge and operational capabilities through information coordination and collaboration across organizations throughout the service ecosystem” (Stolze et al., 2016, p. 191). For the customer in our study, these knowledge and operational capabilities are combined in whatever manner necessary. At ground level, each organization manages a SC; at 10 thousand feet, the mesh of SCs forms the ecosystem.

A last consideration regarding ecosystems versus SCs is the point of view for the analysis. The management team of each NPO manages its own SC to serve its target clients. However, resources and customers are shared among NPOs. From the client’s perspective approaching an NPO, the ecosystem responds by delivering the needed assistance. Whether that assistance is provided by the NPO initially contacted or any other is probably irrelevant to the client (customer-sharing). Whether the assistance is provided using the resources owned by the NPO initially contacted or owned by any other NPO is also irrelevant to the client (resource sharing).

Implications for practice

This study underscores how differently NPOs approach food insecurity. This is important to highlight because in SCM, most studies are in the context of food banks (Ataseven et al., 2020; Blackmon et al., 2021; Martins et al., 2019), and this might lead to the thought that this is the only way to address food insecurity. To achieve scale, the operation is simplified by, for example, standardizing the product and processes and by developing IT capabilities to connect with the partners and clients (Vargo and Lusch, 2008). Some NPOs adopt a service-centered view by developing a deep knowledge of the client to tailor their offering as needed. In practice, how food insecurity is defined and the approach to reach the client will dictate the design of the SC.

Extant literature and our fieldwork show the contrast between “feeding the line” and “shortening the line” (Mendoza-Abarca and Gras, 2019). The former refers to providing food to food-insecure clients. Shortening the line references helping clients develop the necessary resources to diminish the need for assistance. While providing food items to the food insecure may assist to transit a rough moment, it seems there is agreement that to shorten the line, assistance from NPOs that work in other spaces than food insecurity is needed. Experiencing food insecurity seems to be a symptom of other challenges (Rosenberg et al., 2018) and as such needs to be addressed holistically (Zivkovic, 2017). Thus, the coordination of efforts across NPOs both in food insecurity and others such as housing and job training seems to be central to “shortening the line.” Other scholars have identified the need to complement existing solution approaches, new SC roles and a higher degree of information sharing (Annosi et al., 2021; Ciulli et al., 2020; Sundgren, 2020; Wills, 2017). Thus, policymakers could consider mapping by geography a mesh of NPOs, and services provided.

Implications for research

We address research implications from two viewpoints: the socioeconomic problem and SC research broadly. From the point of view of the problem, while existing literature in SCM tends to focus on a product-centric view, one in which at the center is the distribution or re-distribution of food products, there are multiple SC approaches to addressing food insecurity. As an explorative study, we have described numerous nuances associated with food insecurity that have received scarce attention from the SCM standpoint. Food insecurity is not solved by providing food (feeding the line) but by providing clients other services to enable themselves to become food secure (shortening the line). While this idea is present in the literature (Mendoza-Abarca and Gras, 2019), we have provided depth to support the further study of what happens at the client interface and the need for complementary services.

Addressing SC research broadly, the number of studies focused on food banks indicate that much of the research efforts adopt a good-centered dominant logic (Ataseven et al., 2018, 2020; Blackmon et al., 2021; Davis et al., 2016; Elmes et al., 2016; Hasnain et al., 2021; Martins et al., 2019; Mendoza-Abarca and Gras, 2019; Solak et al., 2014). While the food banking model is the one operating at the largest scale (reaching the largest number of food insecurity individuals and households), food banks act as distributors in a long distribution channel. Observing this discussion through the lens of the food availability-access-utilization hierarchy (Barrett and Lentz, 2016), utilization virtually remains unexplored in SCM. Future research could explore the customer interface by combining last-mile delivery, consumer behavior and customer well-being. Much of what happens at the customer interface in food insecurity is largely unexplored from the SCM point of view. Ultimately, we believe that a key concept is customer release, conceptually the opposite of customer retention. This is fundamental in our understanding of value and value proposition among socially oriented initiatives and social innovation strategies.

We observed a high level of interaction and adaptability among what would be described as competing organizations. This interaction resulted not only in sharing resources and information but, most interestingly, in sharing customers. Past research in ecosystems highlights the need for the supply ecosystem to adapt to the needs of the customer (Stolze et al., 2016) and the value of resource orchestration and information sharing for the ecosystem as a whole to respond (Ketchen et al., 2014). However, in the for-profit context, the importance of customer satisfaction and customer retention seems inevitable. In light of this competition, some members of the ecosystem might view their capabilities more positively than those of others (Ketchen et al., 2014), leading to tension among the ecosystem members. Extant research in the business ecosystem usually includes a “focal firm” (Adner and Kapoor, 2010, p. 306), something that resembles a SC. The context of NPOs providing social assistance may provide the adequate context to study pure ecosystems, that is, an ecosystem without a central organization that owns the relationship with the end customer.

Limitations

As with any study, limitations always exist. Food insecurity is a complex problem that unlikely will be solved. While our study sheds some light into understanding food insecurity from the SCM point of view, further research is needed. This study is based on a limited number of interviews, though, this number is above other studies in a similar research context (e.g. Sundgren, 2020). We were able to explore more deeply at the expense of generalization. It is possible that the research context we found can be described as a phenomenon (Groenewald, 2004), but that does not make it less interesting, in fact, we argue that our detailed approach allow us for a deeper understanding of what it takes to generate impact or imagine solutions to complex problems.

Figures

Figure 1

Data structure

Table 1

Interviewees and participating organizations

Interviewee	Organization	SC role	Size	Employees	Founded
1	A (*)	Focal	Large	34	1988
2	B	Focal	Small	2	2012
3
4	C	Focal	Small	5	2017
5	D	Focal	Small	3	2011
6	E	Focal	Small	3	2020
7	F (*)	Supporting	Large	16	1965
8	G	Supporting	Large	37	1988
9
10	H	Supporting	Small	14	2016
11
12
13	I	Focal	Small	14	2009
14	J	Supporting	Independent/Activist

Note(s): (*) Organization with national presence

Interviewees and organizations are listed in the chronological order the interviews took place

Source(s): Authors’ own work

Table 2

Goods- and service-centered dominant logics of food insecurity

	Goods-centered dominant logic	Service-centered dominant logic
Unit of exchange	People exchange for goods	People exchange to acquire knowledge and skills, or services
	Food items	Nutrition, experience
Role of goods	Change in form, place, time and possession	Goods are intermediate “products” used as appliance in value creation-processes
	Access	Utilization
Role of customer	The customer is the recipient of goods	The customer is a coproducer
	Recipient of food items	Applied knowledge (food choices, recipes) Develops community (reduce stigma and loneliness)
Determination of value	Value is determined by the producer	Value is based on “value in use”
	Access to a standardized bundle of food items	Food choices, cooking healthy
NPO-client interaction	Transactional exchanges	Relational exchanges and coproduction
	Transactions	Relational: customization, customer release
Supply chain design	Management of flows: product and information	Ecosystem: collaboration, and resource and customer sharing

Note(s): In italics, the original definition of the dimension. In non-italic, the summary of how the discussion fits with the goods- and service-centered dominant logics of food insecurity

The original table includes an additional row, namely source of economic growth

Source(s): Based on Vargo and Lusch (2004), Table 2, p. 7

Supplementary Material

The supplementary material for this article can be found online.