MEASURING THE REACH OF ASIAN REGIONAL FOOD REGIMES IN THE WTO ERA

abstract

Although the food regime concept has become an important approach for addressing the global food trade, most of its applications have utilized qualitative methods. This article applies a quantitative social network analysis (SNA) and a more-than-qualitative relational approach, drawing from the geography literature to explore regional food regimes in Asia in the World Trade Organization (WTO) era. Employing data from the Food and Agriculture Organization Corporate Statistical Database (FAOSTAT) and the Design of Trade Agreements (DESTA) project database, this paper’s analysis of the governance of Asian regional vegetable trade networks reveals the multiplicity inherent in food regimes under the WTO. In this context, the topologies of food regimes—including degree centrality, betweenness centrality, and closeness centrality—reveal the emergence of a multi-polar food system in Asia.

Keywords:

• Asian regional food regimes
• more-than-qualitative relational approach
• social network analysis
• territories
• vegetable networks
• WTO

Acknowledgments

I would like to thank those who provided insightful comments on earlier versions of this paper: Ryan Light, Alexander Murphy, and Daniel Buck. Members of the Duke Network Analysis Center are helpful in answering my questions on SNA. I appreciate comments from anonymous referees and the editor, David Kaplan, who provided constructive comments on an initial draft of this article. Any remaining errors or omissions are my own.

Notes

¹ Padgett and Ansell (1993) also contributed to developing SNA approaches employing an algebraic procedure called blockmodeling (also see Proto (2016) mentioned earlier in the paper) to conduct positional analysis as an approach to representing social structure. Padgett and Ansell’s (1993) paper identifies marriage and economic ties of “blocks” within the ruling elite of the Medicean political party and using the model to predict actual party membership of the Medici. The predictions from this approach have been recognized as being more accurate than those from standard categorical analysis.

² This study’s visualization of emerging trading blocks in Asia would be less effective at demonstrating the “structured agency” of international class alliances and class struggles that lies at one of the most crucial moments of the state–capital nexus of the food regime (Tilzey 2019).

³ URL: [https://www.designoftradeagreements.org/downloads/]

⁴ These measures rely on the geodesic measure of distance. Closeness, C(ni), for example, is the inverse of distance, in that, the shorter the distance between node i and other nodes, the more central node i is. Formally, C(ni) = [Σ j ^N d(ni,nj)]⁻¹, where d (ni,nj) is the geodesic (shortest path) between entities i and j in the network, and N is the network size. Betweenness, B (ni), measures the probability that a path from actor j to actor k takes a particular route through agent i, assuming that each one-step tie has equal weight and that interactions will occur through the shortest route. Formally, B(ni) = Σ j ^N gjk (ni)/gjk, where gjk (ni) is the total number of geodesics through i, and 1/gjk is the probability that a particular geodesic will be chosen.

⁵ For a weighted network, modularity is defined as: Q=$\frac{1}{2m}$∑_ij [A_ij −$\frac{kikj}{2m}$]δ (c_i,c_j), where Aij represents the edge weight between nodes i and j; ki and kj are the sum of the weights of the edges attached to nodes i and j, respectively; m is the sum of all of the edge weights in the graph; c_i and c_j are the communities of nodes; and δ is a simple delta function. In order to maximize this value efficiently, the LM has two phases that are repeated iteratively (Blondel and others 2008): First, each node in the network is assigned to its own community; then, for each node i, the change in modularity is calculated as i is removed from its own community and moved into the community of each neighbor j of i. This value is calculated using:

∆Q = ∑_ij [$\frac{in+ki,in}{2m}-\left(\frac{tot+ki}{2m}\right)$²]$-[\frac{in}{2m}-\left(\frac{tot}{2m}\right)$²$-\left(\frac{ki}{2m}\right)$²]

where ∑in is the sum of all the weights of the links inside the community that i is moving into; ∑tot is the sum of all the weights of the links to nodes in the community; ki is the weighted degree of i; ki,in is the sum of the weights of the links between i and other nodes in the community; and m is the sum of the weights of all links in the network. Once this value is calculated for all communities i is connected to, i is placed into the community that resulted in the greatest increase in modularity.

⁶ The network matrix of trade agreements is symmetric (composed of an undirected network) and contains entries of ij = 1 when there is a tie (or bond) between elements i and j, and ij = 0 in the absence of a tie.

Additional information

Funding

This work was supported by the Ministry of Science and Technology, and Academia Sinica, both at Taiwan.