In lieu of an abstract, here is a brief excerpt of the content:

Reviewed by:
  • Disaggregating International Regimes: A New Approach to Evaluation and Comparison by Olav Schram Stokke
  • James Hollway
Stokke, Olav Schram. 2012. Disaggregating International Regimes: A New Approach to Evaluation and Comparison. Cambridge, MA: The MIT Press.

In his latest book, Olav Schram Stokke proposes a “disaggregate approach” to evaluating regime effectiveness that involves “[p]artitioning the problem” (p. 265) of effectiveness by three regime functions. The book is well structured: three chapters introduce the approach and the Barents Sea fisheries regime case; three chapters model the regime’s effectiveness for each function in turn; and finally, two chapters model aggregate effectiveness and discuss general lessons. This is a dense, thorough text that leans “heavily on the fuzzy set version” (p. 57) of quality comparative analysis (fsQCA), yet nonetheless is very readable for those unfamiliar with the method. It is not a textbook substitute, however, and could benefit from earlier and more frequent use of explanatory diagrams and maps.

Stokke begins with a mainstream, positive statement that effectiveness is the degree to which regimes fulfill the mandate for which they were established. For the Barents Sea fisheries regime, Stokke considers the overall mandate to be the moving (though not multiple) target of balancing catch and conservation. But his most distinctive contribution is how he “disaggregates” the regime and its effectiveness by three functions: converging cognitional representations of the problem, regulating quotas appropriately, and improving behavioral compliance. Disaggregating, Stokke argues, allows a more nuanced assessment that includes partial and countervailing effectiveness. The core of the book deals with these functions. This analytic move is worthwhile, though since Stokke never sufficiently elucidates the expected relationship between these functions they appear as somewhat separate intellectual questions.

The first of these is perhaps the most interesting and innovative. Unlike the typical examination of how scientific advice informs regulation, chapter 4 investigates how the Barents Sea fisheries regime facilitates scientific advice. Stokke reemphasizes cognitional processes throughout the book and, although the relationship between effectiveness and the legitimacy gained from scientific credibility could benefit from elaboration, this section serves as an excellent first pass at the issue.

When it comes to the methodology of evaluating effectiveness on these three regime functions, Stokke endorses the Oslo-Potsdam formula, which uses a difficult-to-measure no-regime counterfactual (NR) as a baseline for both the actual performance (AP) numerator and collective optimum denominator. Stokke estimates NR for each function by entering scores on counterfactual conditions established through process-tracing into reliable configurative causal models for observed regime success and failure built inductively using fsQCA. Therefore the quality of NR—crucial to the goal of effectiveness evaluation—depends on the quality of the causal model. [End Page 131]

Unfortunately, the utility of Stokke’s models for this purpose is hampered by several issues. One is that the models’ rather low (joint) coverage scores in explaining AP suggests we cannot expect them to model NR well. Given these low scores and the demanding goal of the study to estimate NR, Stokke’s exploitation of only four causal mechanisms for each model seems odd. While commendable in a hypothetico-deductive study explaining AP where parsimony is prized, this limitation seems obstructive given the considerable challenge of modeling a counterfactual world. Indeed, Stokke could have made use of his rather sudden turn to inferential statistics in chapter 6 to consider other potential variables, especially state-level variables that also could have been extended to other chapters.

Another issue is that Stokke apparently considers it unproblematic to decompose the single-regime evidence into twenty-five regime-year “cases” for chapters 4 and 5 (his primary justification for using fsQCA) and again into state-year “cases” for chapter 6 (allowing inferential statistics). However, there is really only one case: the Barents Sea fisheries regime. While single case strategies are appropriate where one is not aiming to generalize beyond a particular regime, this strategy nonetheless creates issues here because Stokke does not take into account inter-”case” dependencies that may bias the causal models from which NR is estimated. For example, high regime performance in 2004 might be dependent on there having been an instructive crisis in 1994 within this or another regime affecting both...

pdf

Share