Essay: Unraveling Agro-food Network(s)

This was written as a response paper for a course on social networks. We were asked to write three essays critiquing network research in our area of interest at the micro (people), meso (organization/community/infrastructure), and macro (nation scale) levels.

Generally, we chose essays that used structural network analysis themselves; in this case, I chose a paper that adopted a network (or relational) way of looking at the world, but didn’t use these formal methods. Structural network methods are a set of (mostly) quantitative approaches that (as their name implies)  describe the structure of relationships (ties) between different actors (nodes) or the position of a particular actor within this structure.

For more on network analysis, here’s a pretty good simple overview of some basic concepts.

_____________________________________________________________

Unraveling Agro-food Network(s)

Drawing inspiration from Granovetter’s (1985) seminal work on embeddedness, food systems researchers in the late 1990s began to integrate economic, social, and political approaches to food systems into a network-based ontology.  Rather than look at global food systems as structurally ossified “regimes,” linear commodity chains, or markets made up of rational, disconnected actors, researchers re-imagined food systems as complex webs of actors linked by social, political, economic, and physical ties. Despite the popularity of the network metaphor, there are still few examples of researchers employing formal network methods to describe the structure of agro-food networks.

Raynolds’ (2004) is no exception. Her study of organic agro-food networks falls within a family of research that has blossomed in the last decade, which focuses on “alternative” agrifood networks (e.g. local and regional, fair trade, artisanal, etc).  She employs commodity network analysis to examine consolidation in global organic networks focusing on network governance, or the mechanisms that underlie the development of network ties. She demonstrates that certification standards play a major role in determining and maintaining an inequitable structure of relations between organic food actors in periphery and core (South-North) nations, but stops short of explicitly specifying and measuring this structure. Finally she observes that there is a “bifurcation” in organic agro-food networks between this “globalized system of formally regulated trade” and networks based in “alternative movement conventions,” and suggests that these alternative networks may offer opportunity to upend the reproduction of traditional South-North inequities, as well as inequities between large and smaller scale firms (Raynolds, 2004:725).

By design, the commodity network approach looks at multiple dimensions of global organic networks simultaneously. It describes social, political, cultural, and economic ties. Nodes aren’t limited to one type, but at times are hemispheres, at times, nations, firms, and individual consumers. The boundaries of analysis shift at times from a North-centered organic processing and distribution network to a movement network of consumers directly connected to local farmers to a global exchange between North-South nations.  What might we might learn by focusing in and using formal network methods to measure the observable interactions between a specified set of actors? In the following paragraphs, I unravel three of the many networks that Raynolds (2004) invokes, specify the nodes, ties, and boundaries, and use her analysis to make guesses at network measures like degree, density, and centrality. Then I describe how network analysis might be used specifically to add depth to Raynolds’ final conclusion about the “bifurcation” between mainstream and alternative organic agro-food networks.

The main thrust of the argument takes place at the macro-level, looking at the relationships between periphery-core nations, specifically between Southern countries (especially in Latin America) and Northern markets (especially in the United States, Canada, Western Europe, and Japan). In this network, the nodes are countries, the ties are imports and exports, and the boundaries are (mostly) limited to Latin America and the major markets described above. From this, we can infer that Northern countries will tend to have higher in-degree centrality than Southern countries (hence their “core” status).  Raynolds also describes a robust “inter-core” trade “dominated by US exports to Europe and Japan, trade between European nations, and exports from Australia, New Zealand, and South Africa to the top markets” (p. 725). Considering this, and that products might flow more than one step (e.g. organic tomatoes produced in Chile, processed and canned in the US, and sold in Japan; peanuts grown and shelled in Canada, included in mueslix in Germany, shipped to the UK), we might meaningfully measure betweenness and closeness centrality. This might help to identify particular Latin American countries as “bridges” that are serving as a gateway between Southern producers and Northern markets; certain Northern countries (the US, for example) with high betweenness scores might also be brokers with more power to set the global organic agenda. These measures would require data measuring the flow of some subset of organic products (all edible organic products, organic fruits and vegetables, all processed products etc.) between each country dyad. With this data, we could also compare a network of actual trade with a network of trade that we might estimate based on a gravity model based on “distance” as measured by cost of transport between countries, and “size” as measured by number of organic hectares, length of growing season, and total population. The differences between the actual and estimated networks would shed light on political, social, and cultural structures that intervene in the network. Though the data required for this analysis would not be easy to compile, it might be possible to get at by combining a variety of sources and using estimates, and the result could a more nuanced view of North-South organic agro-food trade dynamics.

Raynolds (2004) also considers a meso-level network of organic agro-food firms. In this case, the nodes are all organic firms (including farmers, aggregators, distribution companies, processors, and retailers) and the ties could be any type of business relationship (e.g. sales between firms). Raynolds describes a change from a “loosely coordinated local network of producers and consumers to a globalized system of formally regulated trade which links socially and spatially distant sites of production and consumption” (p. 725). The trend is towards greater spatial distance between nodes (which would not necessarily be captured in the network I specified above), and also towards consolidation: in network terms, a decrease in the overall size of the network and increased density. Howard (2009) documents this trend in his visualization of consolidation in the North American organic industry, but his study also does not employ formal network measures. Again, data is difficult to obtain on relationships between organic firms, especially given such a broad boundary; however, it is possible to limit the boundary to a particular commodity or limit the type of firm (e.g. only farmers and distributors) to get at a particular aspect of this broader network. This would make it possible to identify more “powerful” firms, not just in terms of endogenous characteristics like size, but also in terms of their position in the network.

Finally, operating beside the macro and meso-level networks is a micro/meso-level network of movement actors and industry groups that shape and challenge norms within the organic movement as well as certification standards. This network could be operationalized as a two-mode network of organic food movement organizations, industry groups and policy-making bodies like the USDA tied by common individuals (e.g. Deputy Secretary Kathleen Merrigan of the USDA formerly staff at the National Sustainable Agriculture Coalition, or former president of the Organic Farming Research Foundation currently head of an organic department at USDA); it could be a two-mode network of movement organizations tied by association with broader trade associations or participation in specific political campaigns; or it could be a network of social movement organizations tied by some other indicator of collaboration. Specifying and mapping these relationships would allow us to see more clearly which agencies, industry groups, and movement organizations occupy more influential positions in the network. If we notice particular clusters of groups, we might look to see if shared norms exist within these clusters and if they specific capacity for collective action. We might also be able to characterize more clearly the “conflict” that Raynolds (2004) describes between movement actors and industry groups in determining certification standards. If we were able to measure this over time, we might also see whether movement advocates like Fred Kirschenmann (2007), who have advocated for a more harmonious marriage between the organic industry and the organic movement into a more integrated organic community, have had any effect.

The article ends by reasserting this conflict between two parts of the organic agro-food network: the one that is governed by organic certification standards driven by commercial and industrial conventions that privilege economies of scale and efficiency, versus the one that is governed by domestic and civic conventions of trust, tradition, and overall good to society. Raynolds (2004) bases this on her observations of “alternatives” to “mainstream” organic networks that represent the “theoretically important […] contested terrain negotiated within and between commodity networks” (p. 738). This dichotomization of movement-based “alternative” networks versus “mainstream” or “industrial” networks is typical of contemporary food systems studies, yet little research has been done to examine these supposedly different networks systematically to compare their structures and ask whether they are really as “bifurcated” as theory assumes.

To systematically analyze this assumed separation, we might choose a particular organic product within a given geography that we believe has strong “mainstream” and “alternative” networks of production and consumption; say, for example, organic berries in the Pacific Northwest which might be produced by small local farms and sold at Farmers Markets and through Community Supported Agriculture schemes or produced in Latin American countries, imported, and sold at larger retailers. We could set the nodes as all firms that participate in production, aggregation, processing, and sale of the particular product, and stipulate ties as total volume of transactions between firms. The data could be collected through a mix of interviews, publicly available data, and estimates based on observations. With this data, we could do a better job answering questions like: Are “mainstream” and “alternative” networks really so bifurcated, or do firms actually overlap (we might expect, for example, some overlap in mid-sized producers who sell both at farmers markets and to larger supermarkets)? If two separate cliques of firms do emerge, are they different structurally: More or less dense? More or less centralized? Which firms have power in each clique? Are norms really different in each clique? How so?

To date, food systems researchers have not yet embraced structural network analysis despite a network-based ontology that recognizes the relational aspect of both industrial and alternative food chains. For one thing, as in the case of the above examples in the organic agro-food sector, data can be difficult to collect. In network analysis, missing data has particularly strong negative consequences on the statistical validity of the data. Even where it is possible to collect data, social network methods can seem inaccessible and overly technical. Yet these methods have the potential to bring more clarity to specific questions about how global, organizational, and individual actors connect to one another to both uphold and upend our current systems of producing, processing, distributing, selling, and consuming food.

_____________________________________

Works Cited

Granovetter, M. 1985. “Economic action and social structure: the problem of embeddedness.” American journal of sociology 481–510.

Howard, Philip. 2009. “Consolidation in the North American Organic Food Processing Sector, 1997 to 2007.” International Journal of Sociology of Agriculture and Food, 30.

Kirschenmann, Fred. 2007. “Guest Feature: Beyond Organic, What’s Really At Stake?”

Raynolds, Laura T. 2004. “The Globalization of Organic Agro-Food Networks.” World Development 32(5):725-743. Retrieved April 17, 2012.

0 comments

There are no comments yet...

Kick things off by filling out the form below.

Leave a Comment