A Conceptual Framework for Food Sharing as Collaborative Consumption

Food waste has increased significantly and become a global issue amidst a growing concern regarding famine in several countries. Food sharing constitutes the solution to the problem provided an appropriate framework is developed that guides its application. The sharing economy was touted as the appropriate framework, yet it is excessively macroscopic to be able to capture the dynamics of food sharing activities. A microscopic framework is required to overcome this problem, the concept of collaborative consumption with its focus on activity level being one potential solution. However, an investigation into how food sharing activities can be viewed as collaborative consumption should be completed. This paper presents an analysis of the relationship between food sharing activities and collaborative consumption. The authors employed a systematic literature review conducted by meta-analysis and content analysis to identify the commonalities between the two and the theories underlying them. The result is a conceptual framework of food sharing activities as a collaborative consumption practice. The framework highlights eight propositions that can explain the intention, performance, and continuity of food sharing activities. At the end of the paper, the authors outline the theoretical and managerial contributions and recommend future research activities.


Introduction
Changes in food consumption patterns have increased the amount of food waste, which is the amount of edible food discarded for various reasons at successive stages of the food supply chain, such as excessive quantity and poor quality [1]. Food waste has risen to 32% of global food production, or 1.3 billion tons per year [2]. Worldwide, those primarily responsible for food waste are the retail sector, food service providers, and consumers (40%); households (25%); and agriculture (35%) [2,3].
Food waste negatively affects the environment, society, and the economy [4][5][6]. The impact of food waste on the environment lies in its contribution to greenhouse gas emissions (GHG) and global warming [7]. In terms of social consequences, food poverty is growing rapidly [8]. Finally, the economic impact is the loss of numerous resources such as water and energy [1,9].
Food sharing, an initiative to reduce food waste by redistributing surpluses [6,10], represents a response to United Nations (UN) Sustainable Development Goals (SDGs) 2 and 12 that relate to zero hunger and responsible consumption [11,12]. The practice of food sharing was initially restricted to household or kinship relationships such as family, neighbours, or friends. Technological developments have extended food sharing practices beyond sharing with relatives to include strangers. Moreover, such practices can accommodate large food industry or food services surpluses to be shared with the needy [3].
Previous food sharing studies have mainly employed the sharing economy as a fundamental sharing model [13][14][15]. However, a significant drawback of using the broad

Materials and Methods
Food sharing represents an embryonic issue and a minor topic that requires the developing of a solid understanding of previous studies through a literature review. In reducing food waste, many studies on food sharing are dominated by macro-level studies such as socio-ecological, geographic, economic, and cultural [35][36][37][38]. However, it is not optimal in reducing food waste because the main problem is at the micro-level, such as behavior at all levels in the food supply chain and improper food management [1,6,38,39]. Therefore, a critical approach is needed in food sharing research to deconstruct a new framework at the micro-level, which is reflected in the context of collaborative consumption. This study focuses on two things: the behavior of actors and the role of Information and Communications Technology (ICT) in food sharing activities.
This research is comprised of two stages in order to develop a conceptual framework of food sharing from the collaborative consumption perspective. Firstly, the study compares food sharing and collaborative consumption as a means of investigating the characteristics of and the relationship between both concepts using a meta-analysis. Secondly, the content analysis explores the first stage results as the fundamental inputs for conceptual framework development.
This study employs a systematic literature review in order to promote an understanding of food sharing within a collaborative consumption context. The review, conducted by both metaanalysis (using statistical analysis of R-studio Bibliometrix) and content analysis, develops a conceptual framework. A systematic literature review analyzes the transparency, methodology, and reproducibility of previous empirical research [39]. The transparency of the study methods employed is achieved through three stages: data collection, selection and extraction, and, finally, analysis. The proposed conceptual framework is reproducible through the application of both qualitative and quantitative methods.
Data collection involves the gathering of information from the database source and determining the keywords related to the study, namely, food sharing and collaborative consumption. The second stage consists of the selection and extraction of the relevant data. Finally, the R studio tool and content analysis address the research objectives. The successive stages of the research are described in Figure 1 below: 1 Data Collection The study employed raw data from the Scopus database which, as the largest, most comprehensive, and robust database for citation, is the one preferred by the majority of scholars [40][41][42] The subsequent data collection stage identified the relevant keywords employed to select sample articles as the raw data [43], which was collected on 1 January 2022. Two keywords searched for separately in the titles and abstracts were "food sharing" and "collaborative consumption". The study set a specific time frame of 2000 to 2022 when identifying the most recent food sharing and collaborative consumption studies. The initial search retrieved 296 articles related to food sharing and 113 on collaborative consumption.

Data Selection
The study identified and removed irrelevant articles leading to the extracting of appropriate data from 41 articles on food sharing, and 13 covering collaborative consumption. The final data extraction consisted of 255 articles on food sharing and 100 articles concerning collaborative consumption for subsequent examination. The main "food sharing" and "collaborative consumption" data is shown in Table 1. The food sharing research period is more extended than collaborative consumption in the 2000s and the 2010s. The term "collaborative consumption" was first mentioned in a Felson and Spaeth article of 1978, but the term fell out of currency and was not employed again until the 2010s, when technology developed rapidly. Sharing activities using technology increased the volume of collaborative consumption research in the early 2010s. The 255 documents related to food sharing feature 692 authors, 742 author keywords, 11,335 references, and 183 sources. In comparison, 100 collaborative consumption documents contained 256 authors, 335 author keywords, 5583 references, and 65 sources. 3 Data Analysis The study has two research objectives. First, through meta-analysis, to recognize and map food sharing characteristics appropriate to a collaborative consumption context. A meta-analysis is a statistical method of integrating results from various studies as a means of identifying and comparing the patterns, disagreements, or relationships that appear in multiple studies on the same topic [39]. A meta-analysis employs Bibliometrix R-package tools which focus on keyword analysis, while Bibliometrix is an open-source software valuable for performing meta-analyses [44]. Secondly, the study's objective is to develop a conceptual framework for food sharing from a collaborative consumption perspective through comprehensive content analysis and further research. The content analysis elaborates the thematic evolution of the discipline while drawing on the authors' expertise [44].

Food Sharing
The keyword analysis describes the research trend, identifies the research gap and highlights future potential research [45]. Word Treemap contains the R-package Bibliometrix results that collate, extract and count the high-frequency keywords from the articles reviewed [46]. The Treemap reflects the characteristics of the field of studies and the production of concise and intuitive articles [47]. The study utilized keyword analysis in order to identify characteristics of food sharing and collaborative consumption studies. There are two types of keywords analysis. First, intensive keywords refer to keywords that are often used in the research topic. These keywords construct the core concept of the field. Secondly, episodic keywords mean that keywords are only mentioned one to three times in keyword analysis. The episodic keywords have a broad scope of context to describe a comprehensive trend of a research topic [48]. This study uses intensive and episodic keywords to convey a clear and comprehensive coverage. The first Treemap contains a keyword analysis of food sharing studies (see Figure 2 below).

1.
Of the fifty terms derived from the keyword analysis of food sharing studies, two duplicating terms: (1) "food sharing" and "food-sharing"; (2) "social networks" and "social network analysis" were identified. 2.
"Reciprocity", "reciprocal altruism", and "exchange" are the terms considered to explain food sharing motivation and are also similar to the basic principles of social exchange theory [29,49,50,53]. 5.
"Motivation" is a keyword frequently investigated in food sharing research to understand why individuals participate in food sharing [51,52]. 6.
8. "Canada", "Ethiopia", "Belgium", and "Berlin" refer to the country, region and city where food sharing studies are most widely conducted. The keyword "cities" indicates where most food sharing activities are undertaken in urban areas [60][61][62]. 9.

Collaborative Consumption
However, of the fifty terms mentioned in the collaborative consumption analysis, the following keywords were identified:
Netnography and case study methods are commonly used in collaborative consumption research [27,99,100]. 7.
8. E-government, fashion sharing, online car-hailing, car sharing, carpooling, and online food ordering are the most frequently explored units of analysis in collaborative consumption [99,100,103,[107][108][109][110]. Transportation has been widely discussed in collaborative consumption, while traffic conditions and Google Maps are considered in car-sharing [111]. 9.

Food Sharing and Collaborative Consumption
This section divides the food sharing and collaborative consumption keyword analysis into six categories: characteristics, research objects, methods, theoretical background, unit analysis, and actors (Table 2).

Characteristics
There are similar characteristics between food sharing and collaborative consumption: sharing, sustainability, and the sharing economy; the latter of which is an umbrella concept for the sharing movement [16,114]. In most previous studies, the sharing economy and collaborative consumption are often defined. However, in others, collaborative consumption constitutes a subset of the sharing economy. It can, therefore, be interpreted that the sharing economy has a broader reach [12,16,115]. At the same time, the context of collaborative consumption is more specific. This finding supports our earlier statement on the scope of the sharing economy and collaborative consumption. Most food sharing research uses the sharing economy, rather than a collaborative consumption context, which means that food sharing studies fail to use this context in their microanalysis. Collaborative consumption is a sustainable business model which aims to create value from three factors: economic, social, and environmental [116], in line with food sharing activities that optimize food surplus to improve economic and social welfare and reduce food waste. Sharing is the primary activity of collaborative consumption and food sharing research. These similarities of food sharing and collaborative consumption characteristics indicate that future food sharing research can potentially be investigated within a collaborative consumption context. The differences between food sharing and collaborative consumption lie in the technology category and the application of the peer-to-peer system. Collaborative consumption is solidly in character with interactions between actors via the platform, while the prominent food sharing characters are those of "care" and "cooperation" (Figure 4).

Research Objects
Collaborative consumption involves a broader scope of sharing activity than in food sharing research, which focuses on food as the object. Transportation, fashion, egovernment, and food are the objects of collaborative consumption research. Hospitality and accommodation also represent collaborative consumption research subjects not captured by the keyword analysis. Even though the word "food" appears in the collaborative consumption keyword analysis, the context is potluck, gathering, and online food delivery activities [92,109,111,117]. It means that the context of food sharing in optimizing food surplus remains limited [10,[30][31][32]34]. The objective of collaborative consumption research is to investigate profit-oriented rather than non-profit models. Therefore, non-monetary or non-profit oriented collaborative consumption requires further investigation [114]. Food sharing forms part of the non-profit based sharing practice, together with commensality and social eating [118,119].

3.
Methods Qualitative methods dominate food sharing and collaborative consumption analysis. Collaborative consumption and food sharing analysis in the digital era is still at an early stage and, therefore, requires a more in-depth analysis. These two studies lack quantitative methods [120,121]. Quantitative methods support the validation of the qualitative results, primarily through netnography [122]. However, if explored more fully, quantitative methods are also widely employed. Both quantitative and qualitative methods have functions that mutually reinforce research results. The method employed depends on the research objectives to be achieved.

4.
Theoretical Background Reciprocal altruism and exchange are motivations highlighted in food sharing research. These two terms relate to Social Exchange Theory (SET) foundations. In SET, the interaction between two or more individuals constitutes a sustainable relationship if the benefits outweigh the costs that it entails. However, if the opposite is the case then the interaction will cease. Therefore, SET is more renowned as a cost and benefit analysis [123][124][125]. There are three categories of food sharing: for money, charity, and community [13]. Sharing for charity is a non-profit form of food sharing. Indeed, although it will not directly produce economic benefits, social and environmental ones will ensue [32]. Three theories are captured in collaborative consumption research: the Theory of Planned Behaviour (TPB), Social Practice Theory (SPT), and Social Capital Theory (SCT). Several studies that use TPB as a theoretical basis are usually analysed using quantitative methods [77,110]. Most of the other articles combine TPB with other theories such as the Technological Acceptance Model, Self-Determination Theory, Norm Activation Theory, Value Belief Norm Theory, and Trait Theory [73,78,94,96,126]. All dimensions of TPB in combination represent theories that explore individual behaviour. SPT is the theoretical basis of qualitative analysis applied in several arti-cles [127][128][129]. SPT has three dimensions: Meaning, Material, and Competence [127] which describe the practice of an organization. In contrast to TPB, which captures the individual aspect, SPT analyses the organizational role. As reflected in its multidimensionality, social capital theory also investigates structural, relational, and cognitive capital [98] from an organizational perspective.

5.
Unit of Analysis Food sharing research is dominated by the macro context depicted by the sharing economy as the basis for more general and macro research, primarily into cultural and health issues that significantly influence the implementation of food sharing. Aspects of daily food consumption such as type of food, taste, and consumption patterns form part of the local culture. Food is also closely related to health since this is influenced by what is consumed [6,14,130]. Meanwhile, collaborative consumption research explores this issue more from the micro side. The behaviour of individuals or organizations is more widely discussed in the literature on collaborative consumption. Motivation, trust, attitude and intention represent the majority of research topics relating to collaborative consumption [18,73,78,94]. 6.
Actors Before discussing food sharing actors, it is necessary to discuss food supply chain actors because there is a relationship between the two. Food supply chain actors consist of agriculture, manufacturing, wholesale, retail and the consumer [38]. Food waste can occur at all levels of the food supply chain. The factors causing food waste include: (1) the agriculture and manufacturing level: overproduction, aesthetic values, and storage [6,131]; (2) the wholesale and retail level: freshness, expiry date, aesthetics, poor management [1,38]; and (3) the consumer level: overconsumption, incorrect storage of food, meal planning, consumer behaviour [6,11,132]. The most preferred strategy for managing food waste is donating food to needy people [133]. The distribution of surplus food from the food supply chain is managed by food mediators such as food banks or food sharing communities that use digital applications. Therefore, it can be said that the food supply chain actor is part of the food sharing actor as a food contributor (Figure 4). Three actor roles are involved in collaborative consumption: the customer, the service provider, and the platform provider [85,134]. Food sharing also has three actors: donors, recipients, and mediators [36,93,[135][136][137][138]. Food donors fulfil the role of providing services or products, which is the same as that of service providers. Meanwhile, food recipients are actors who receive products or services from service providers either directly or indirectly through food mediators. The latter act as intermediaries between food donors and food recipients whose role is almost the same as that of platform providers. The difference lies in using a technology platform and whether or not this food mediator can use it.

Problem-Solving Analysis
One of the units of analysis studied deals with the constraints faced in cases of collaborative consumption ( Table 2). The research employs problem-solving analysis to address the second research objective of developing a food sharing conceptual framework within a collaborative consumption context. The problems encountered during the practice of food sharing are listed and described in Table 3. Explanation of these problems is essential to map and solve problems in the background environment. The conceptual framework in this study was developed against the background of the problems faced by the actors in food sharing (food donors, food sharing mediators, food recipients) in conformity with the research objectives of this paper. Most of the obstacles that arise are due to a lack of experience with food sharing and limited familiarity with certain foods within a collaborative consumption context [10], especially the maturity of food sharing as a mediator. Therefore, the conceptual framework focuses on food sharing actors as mediators in this research. The theoretical basis of the food sharing conceptual framework employs social practice theory (SPT), which has three dimensions: competence, material, and meaning [146]. There are three reasons for adopting SPT as a theoretical framework in this research: Firstly, SPT can analyse alternatives when considering 'problem' behaviour which enables the accommodating of problems that arise in food sharing practice [147]. Secondly, most studies that use SPT as a theoretical framework analyse sustainable consumption in line with the goal of food sharing within a collaborative consumption context [147][148][149]. Thirdly, practice is the smallest unit of analysis that can form a pattern of social practice on the basis of individual behaviour. The use of SPT in developing a food sharing conceptual framework can investigate in microanalysis, which is still rarely studied in food sharing research [147].
The efforts to reduce the obstacles encountered in food sharing activities are contained in Table 4. The problem-solving indicators in the table can be used as dependent ones and the impact of problem-solving in reducing the constraints can be used as an independent indicator. The problem-solving analysis meets the criteria of rigor and exactness, parsimony, and logical consistency in creating the quality of indicators [150].
The contents of Table 5 form the basis of reference and identify indicators for designing a food sharing research framework from a social practice theory perspective. The findings of the problem-solving analysis are explained in the following sub-chapter in the form of a comprehensive food sharing framework. Table 4. Problems-solving analysis from a Social Practice Theory perspective.

Problem-Solving Indicators
Code of the Problem Solved Meaning 1.
Transform and re-position value from commodity to gift, from poverty to prosperity, from waste to conservation [8,144,151]; 2.
Social value: food sharing creates a social space in which to meet new people and demonstrate caring behaviour [3,56]; 3.
Economic value: improving CSR value and business sustainability for food donors [151], reducing spending on food [152]. Ability to assess the quality of food donations [141]; 3.

Conceptual Framework
This conceptual framework is structured on the basis of the four stages of Dubin's theorybuilding method [154]: (1) Unit: identifies the basic theoretical concepts; (2) Law of interaction: explains the interaction and interrelationships between theoretical units; (3) Boundaries: defines the boundaries of the conceptual framework; (4) System states: the condition of the system in which the theoretical units interact and operate differently ( Figure 5).

Units of Analysis
The unit analysis of this study employs the three dimensions of SPT proposed by Shove (2012) and the extended attitude dimension from the Theory of Reasoned Action (TRA). The attitude as an extended dimension assists in clarifying the impacts of the three dimensions of social practice. Attitudes refer to an individual perceiving specific behaviour as favourable or unfavourable. Intentions, in turn, are assumed to capture the motivational factors that influence such behaviour [74]. Practice theory covers the dynamics of daily life and what people actually do [155]. The three elements of social practice theory are meaning, material, and competence [146]. Meaning refers to the interpretation, perspective, and characteristics of the practices. Material constitutes all the tangible elements that support practices, including technologies, tools, and hardware. The material can also be divided into three topics: infrastructure, device, and resources. Competence is defined as the expertise, understanding, skill, and knowledge necessary to implement the practice [155][156][157]. 2 Law of Interaction This law covers the combination of the three components of social practice theory (meaning, competence and material) in a unified whole, interrelated and supporting one another to implement a social practice [146]. 3 Boundaries There are three models of food sharing: for profit, for the community, and for charity [3]. This proposed framework is limited to non-monetary collaborative consumption that is specifically related to food sharing for charity. The previous research into collaborative consumption generally discussed fashion, accommodation, and transportation as producing monetary compensation, while devoting less attention to food sharing as non-monetary collaborative consumption [10,[30][31][32][33][34]109]. The conceptual framework focused on the food sharing mediator that affects the mediator and other actors, such as food donors and food recipients. 4 System States The system states of this proposed framework include food sharing in the urban area. Food sharing communities are generally established in conurbations for three reasons. First, inhabitants of rural areas tend to be more satisfied with simple plantbased dishes, while city dwellers enjoy delicious food [135]. This fact, allied with the behaviour of urban communities and various other reasons [143], causes more food waste in urban areas. Second, the major characteristic of urban communities is individualistic, which results in social isolation. Consequently, food sharing is necessary in urban areas to promote social inclusion [135,151]. Third, the high incidence of poverty and hunger in urban areas necessitates food sharing in order to reach those individuals requiring food [158]. Besides urban areas, food sharing constitutes an increasing phenomenon in the global socio-economic crisis [159]. The analysis of problem-solving, the meaning of food sharing, competencies required by food sharing mediators, as well as the tools and materials that can support all activities carried out by food sharing mediators can be generalized and described as follows ( Figure 6): The three dimensions of social practice theory are described through the indicators resulting from the content analysis. First, the meaning dimension: food sharing mediators need to re-position their values through three sustainability values, namely; economic, social, and environmental in order reduce the stigma of shame on the part of food recipients who receive assistance. Second, the competence dimension: skill and capabilities in the areas of technology and food management increase the trust of food recipients in the quality of food distribution and optimize the role of ICT in enhancing the performance of food sharing mediators. Knowledge of marketing communications for non-profit organizations provides the information required by food donors to increase food sharing participation and to educate the public to improve sustainable consumption behaviour. Third, the material dimension tangibly supports food sharing practice. The prerequisite materials of food sharing practice comprise the ICT infrastructure to increase food sharing performance; storage infrastructure to maintain the quality of perishable food; appropriate transportation to cover long distances and accelerated food donation distribution to prevent their accumulation and from becoming food waste in storage areas; modifying the distribution channel through cooperating with local communities to distribute food donations that are difficult to access by food sharing mediators; and a legal tool to increase participants' trust in food sharing mediators. Figure 6 shows that ICT is an essential factor in food sharing and handling food waste. There are three crucial roles of ICT in reducing food waste [38], First, digital platform technologies serve as a liaison between food donors and recipients and can expand the food stakeholder network. The digital platform is also an interactive and responsive communication tool, making it easier for all food-sharing actors to interact [160]. Second, RFID (Radio Frequency Identification) is beneficial for more controlled inventory management, more structured food category management and better food layout management. Third, monitoring food quality, humidity, and temperature according to the type of food and managing expiration dates can use the IoT (Internet of Things).
From the content of Figure 5, it can be concluded that social practice dimensions affect the level of trust of food recipients and donors, sustainable consumption behaviour, the performance of food sharing mediators, and the intention to participate in food sharing activities, as depicted in Figure 7. Based on the framework it contains, there are eight propositions:  These eight propositions are theoretical contributions for further research. The framework in this research can be analyzed using qualitative and quantitative methods to obtain empirical results that can improve research findings.

Conclusions
This research has two objectives. The first is to recognize the characteristics of food sharing that are appropriate to a collaborative consumption context. The second is to develop a conceptual framework for food sharing within a collaborative consumption context. The first result is that the characteristics of food sharing emphasize the social aspects of care and cooperation, while collaborative consumption prioritizes the use of technology platforms and peer-to-peer interactions. The similarities between the characteristics of food sharing and collaborative consumption is that they have the goal of sustainability, sharing activities, and the sharing economy as their umbrella concept. This similarity of characteristics is used to develop a food sharing conceptual framework from a micro-analysis perspective within the context of collaborative consumption. Previous research has discussed many macro-levels such as socio-ecological, economic and cultural. However, food waste is still high because the main problems are at the micro-level such as consumer behaviour and food handling management. Therefore, this study focuses on micro-units of analysis to optimize strategies for reducing food waste through food sharing. Secondly, indicators for each dimension of social practice theory in food sharing practice and eight propositions from the conceptual framework developed in the theory of social practice perspective were identified.
This study has several limitations, the first being that the research framework constructed in this study only covers food-sharing mediators. At the same time, there are two other actors in food sharing activities: food donors and food recipients. Therefore, further research is encouraged to develop a social practice conceptual framework for food donors and food recipients because the meaning, competence, and materials will be different for those actors [163]. Secondly, the research constitutes a literature review analysis. Therefore, an empirical analysis involving qualitative or quantitative methods, which are more relevant, is required to clarify the conceptual framework. Thirdly, research on food sharing is predominantly carried out in developed countries [61,62]. Therefore, further research can be carried out in developing countries to compare and improve social practice within food-sharing contexts [156]. Fourthly, there are three models of food sharing: food sharing for profit, for charity, and for community [3]. This study develops a conceptual framework of food sharing for charity. Future research into the social practice framework relating to food sharing for profit and for community can be conducted. Data Availability Statement: The study did not report any data.