In the light of climate change, biodiversity loss, increasing demand for agricultural products, social inequalities, and economic pressures, a transformation toward sustainable, resilient and inclusive food systems is urgently needed (Pigford et al., 2018). Alternative food systems, including farmers' markets and shops, community-supported agriculture, food cooperatives and organic agriculture, are increasingly recognized as promising approaches to address these challenges (Renting et al., 2003; Ilbery and Maye, 2005; Forssell and Lankoski, 2015).

In particular, community-supported agriculture (CSA) offers opportunities to provide synergies between ecological, economic and social aspects that could contribute to a transformation toward sustainable food systems (Nost, 2014; Bloemmen et al., 2015; Schmutz et al., 2018; Haack et al., 2020). Linking farmers and consumers (members) through long-term partnerships including upfront payments to cover production costs is a core principle of CSA (Lamb, 1994; Cone and Myhre, 2000; Volz et al., 2016). Thus, both risks and the harvest are shared. CSA differs from other cooperative approaches, for example community gardens, where gardening is mainly carried by unpaid and nonprofessional volunteers, or food cooperatives, where risks are not shared by producers and consumers and products are not necessarily obtained from primary producers (Haack et al., 2020). CSA farms are highly heterogenous for example regarding size, products, member involvement and legal forms (Volz et al., 2016). In its simplest form, CSA is a contractual agreement between a farm and a group of members (Cone and Myhre, 2000), but also models exist where both farmers and consumers are organized in one legal entity (Strüber et al., 2023). Likewise, underlying motives include securing livelihoods of farms, spiritual-communal practices (e.g., connection to nature, emphasis on community) or a political tool for sociopolitical change by opposing the capitalistic system through decommodification of food (Blättel-Mink et al., 2017; Paech et al., 2020). Environmentally friendly farming practices however, are a main principle of most CSA farms (Volz et al., 2016; Cristiano et al., 2020; Netzwerk Solidarische Landwirtschaft e.V., 2022b). Accordingly, reducing synthetic inputs, closing nutrient cycles, improving soil properties and biodiversity are inherent goals of many CSAs (Haack et al., 2020; Cristiano, 2021). Regarding social aspects, many CSA aim at fair wages, transparency, knowledge exchange and participation (Schmutz et al., 2018; Diekmann et al., 2020). Economic security is fostered through holding members and guaranteed sales (Matzembacher and Meira, 2019; Paech et al., 2020).

During the past years, CSA has gained increasing attention and the number of CSA farms has grown in many world regions. In the United States over 12.000 CSA farms existed in 2017 (Woods et al., 2017; Samoggia et al., 2019). The first CSA in Europe was founded in 1978 in Switzerland, yet currently most European CSAs are located in France (>2,000) (Volz et al., 2016; Egartner et al., 2020). In Germany, only five CSA were founded between 1988 and 2010, but more than 400 farms are registered today (Diekmann, 2020; Netzwerk Solidarische Landwirtschaft e.V., 2022a). In China, the first CSA farm has been only founded in 2009, yet already 254 existed in 2016 (Tang et al., 2019). In Japan however, the interest in CSA has recently declined (Gugerell et al., 2021). Compared to the total number of farms [e.g., more than two millions in the US in 2017; (USDA National Agricultural Statistics Service, 2023)], CSA only provides a small proportion of total food production so far. Nevertheless, CSA is increasingly recognized in policy, civil society and academia, also during the COVID pandemic due to increased interest in regional food supply (Stephens et al., 2020; Enthoven and van den Broeck, 2021). In Germany, for example, it has been mentioned as a best-practice model with a catalytic impact in government documents and commissions (CDU et al., 2018; Zukunftskommission Landwirtschaft, 2021).

Despite the growing interest, to our knowledge, an overview of sustainability impacts of CSA is missing, which is essential, given the sustainability promises of CSA. In this study we provide a synthesis of ecological, social and economic sustainability outcomes based on quantitative empirical work in English and peer-reviewed literature. For this purpose, we (i) developed an analytical framework to assess sustainability outcomes of CSA at the farm level in a transdisciplinary process and (ii) reviewed the literature to systematically assess empirical and quantitative findings related to the different levels of the framework. On the one hand, this is an important baseline for well-informed and targeted policies and funding. On the other hand, this work will highlight relevant research gaps, given that research on CSA is still in an early stage (Cristiano, 2021).

Within a transdisciplinary research project, we developed an analytical framework for a comprehensive assessment of sustainability impacts of community-supported agriculture at the farm level. This framework consists of dimensions, categories, sub-categories and key performance indicators. We then conducted a literature review to systematically assess empirical and quantitative findings related to the different levels of the framework.

We found that < 30% of the quantitative studies included investigated ecological aspects and in around 40% of these studies relative effects compared to other farming systems were either unclear or not investigated, thus a reliable evaluation of the ecologic sustainability of CSA is not yet possible (Christensen et al., 2018; Wellner, 2018). Nevertheless, existing studies show a clear positive trend. In particular, regarding resource use efficiency and greenhouse gas emissions, investigated CSAs mostly performed better than conventional systems. Research on crop yields, crop and livestock diversity, and soil health remains incomplete or is not yet performed. Environmental effects have been investigated widely and more often than social and economic aspects for agri-food systems generally, but also with regard to organic agriculture and local food systems (Mundler and Laughrea, 2016; El Bilali et al., 2021). To which extent these findings are transferrable to the CSA context needs to be evaluated. With regard to slowly changing processes in complex ecosystems such as soil, changes in soil functions as response to altered management practices might be only detectable after decades (Nortcliff, 2002; Bai et al., 2018; Bünemann et al., 2018). In this case, the use of proxies and systemic modeling approaches combined with field data might be essential to evaluate altered management practices long-term sustainability (Bünemann et al., 2018; Rabot et al., 2018; Vogel et al., 2018).

Nearly 90% of the quantitative studies included addressed social sustainability aspects. Given that CSA is a social innovation centered around a community of producers and consumers, their relationship, solidarity, cooperation, trust, engagement and participation, this is hardly surprising (Jarosz, 2000). Moreover, in contrast to many ecological variables that require continuous and long-term effort for data collection, a wide range of social aspects can be captured with one-time (online) surveys or interviews. However, the majority of studies within the social dimension assessed socio-economic variables of CSA members including gender, age, income, ethnicity and education. Accordingly, CSA members do not represent the average population, for example with regard to low income households (Galt et al., 2017). Besides various internal mechanisms to offset costs, e.g., membership fees related to income or anonymous bidding rounds, externally subsidized memberships could improve access to CSA and related health benefits (Izumi et al., 2018). Apart from the limited diversity of CSA members, most studies found positive social effects, particularly with regard to health and sustainability behavior, for example, including dietary changes, which can substantially reduce environmental impacts of food systems (Willett et al., 2019).

More than half of the studies included covered economic aspects. However, knowledge on actual economic performance compared to other systems is still very limited. Nevertheless, first studies indicate high economic viability of CSA farms. If this pattern is generalizable, CSA could provide a suitable alternative to farm growth or termination in the light of economic competition in the agricultural sector (Paech et al., 2019). In this context, diffusion of CSA is another emerging topic in the CSA literature screened here. Amongst others, diffusion relies on institutional support, access to affordable land and sufficient demand (Doernberg et al., 2016; Pisarn et al., 2020; Zoll et al., 2021). For example, Diekmann and Theuvsen (2019a) found that 27% of non-participants would be interested in CSA membership in Germany, i.e., much more than the proportion of the German population currently organized in a CSA, which is < 1% (own estimation based on Netzwerk Solidarische Landwirtschaft e.V., 2022a). However, members need to be hold, e.g., by increasing crop diversity and cooperation among different CSA farms (Galt et al., 2019). In this context, understanding and fostering the values and preferences of CSA members is crucial to establish long term relationships (Chen, 2013a,b; Diekmann and Theuvsen, 2019b). Reducing fluctuation is particularly relevant in the context of current inflation rates, which will also provide insights on the economic resilience of CSA.

Besides the observed bias in current CSA literature toward social variables, we also observed a clear geographical imbalance. While most studies analyzed in this paper focused on the US, the country with the largest number of CSA farms, no study investigated French CSA farms, where the vast majority of European CSA farms are located (Samoggia et al., 2019; Egartner et al., 2020), yet this might also be related to our search strategy (see below). Moreover, besides Brazil, China and Turkey no countries from the Global South or emerging economies were considered. In the light of contrasting socio-economic conditions, knowledge from different types of countries is crucial. To obtain an overview of the current status of CSA, the international CSA network Urgenci (urgenci.net) is launching a global census in 2023, which will provide an important baseline for future research. Many of the studies considered only cover a limited number of CSA farms, focus on few sustainability aspects and often remain descriptive.

While our study offers an important overview of sustainability-related effects of CSA, it faces two major limitations. First, we only included English articles indexed in Web of Science, which leads to a bias toward articles published in US journals and particularly neglects research from the Global South (Gibbs, 1995; Lund, 2022). Given that CSA research is often geographically limited and closely related to practice, many studies are likely to be published in other languages. This could also explain, why despite the large number of CSAs, no study from France was included. In Germany (and probably also elsewhere), the number of theses on CSA is rapidly increasing. While most of them are in German and not published in scientific journals, many of them offer valuable insights that should be considered, if certain quality criteria are fulfilled. While CSA is the most common term in international literature, including country-specific terms, for example AMAP (Association pour le Maintien d'une Agriculture Paysanne) in France and teikei in Japan could increase the literature base and the geographic scope. Second, we only included quantitative findings related to our framework and without applying any criteria regarding the statistics of the selected paper. In particular regarding social aspects, qualitative studies offer additional valuable insights, for example related to underlying values, identity and social practices (Diekmann and Theuvsen, 2019b; Neulinger et al., 2020; Zoll et al., 2021). Consequently, our literature review is only a first step and more literature need to be included to draw broader conclusions on existing findings of the sustainability of CSA.

Based on our review we suggest that more studies investigating actual sustainability effects of CSA farms following standardized protocols are needed, in particular including comparisons across different farming systems and with regard to their ecological effects. As empirical work in this context might require high efforts, proxies and findings related to organic farming could be considered. More research is essential for CSA stabilization, development and scaling, as well as the development of policies to support CSA. This also includes approaches to reduce the fluctuation of CSA members, to increase their diversity and to understand and utilize future consumer potentials. Moreover, existing knowledge should be better integrated and publicly accessible. The German CSA network for example, collects bibliographic information on existing theses. Ideally, the central findings of such research would be integrated into one structured and international database to simplify access and knowledge transfer. Finally, more funding is needed to support these efforts. In the light of the high sustainability promises of CSA shown so far and the urgent need to transform food systems toward sustainability, such funding would be well placed to support this goal. Therefore, obstacles for CSA to gain access to existing programs and funds (e.g. common agricultural policy in the EU) could be reduced and new programs could be established.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

JR identified the relevant studies for the review. LE performed the analyses and wrote the paper in collaboration with JR and JP. All authors designed the study and revised multiple versions of the manuscript.