What we talk about when we talk about compliance

How to write about, or whether to even use, the concept of compliance is something I’ve struggled with in my own work. I was therefore thrilled to read Lisa Martin’s convincing piece on what she deems is a misguided use of compliance in IR (go here for an earlier version free to download).

Essentially, Martin argues that IR scholars have devoted time and resources to studying states’ compliance with international institutions, despite the fact that compliance is not what they are interested in. A central question for IR scholars is determining causal effects – answering the counterfactual of how real state behaviour would differ in the absence of the institution of interest. These interesting indicators are observed changes in behaviour, and the substantive content of state policies. A focus on compliance, she argues, could lead us to miss these entirely.

To illustrate this, Martin gives the example of a state with low levels of environmental regulation, ratifying an environmental treaty. Upon joining,

“The institution provides knowledge and capacity, and in response State B modestly improves its performance on emission indicators. However, it still falls short of institutional requirements. If compliance were measured dichotomously, State B would be found out of compliance.”


Thus, the researcher could fail to notice the change in State B’s behaviour, and might wrongly dismiss the institution’s effectiveness. At the same time, (echoing Downs, Rocke, and Barsoom) she argues that we could wrongly interpret another, more environmentally inclined state’s behaviour to the effectiveness institution, when in fact their substantive policies would not alter in its absence.

Even more interesting to me was her take on Dai’s work on domestic pressure groups and international compliance, which focuses on a domestic enforcement mechanisms for compliance – essentially interest groups who will win or lose based on a state’s compliance with an international agreement. However, as Martin asks, do states “have compliance strategies, or more directly strategies about substantive policies?” and “do domestic constituencies care about compliance per se, or about substantive policies?”

I think this is a key question. A lot of (admittedly earlier) work on compliance argued that democracies are more likely to be found in compliance with international agreements due to audience costs and a culture of rule of law. This always seemed an odd thesis to me, and Martin clearly articulates the problem with it here.

Of course technical compliance itself has symbolic value, and if we believe in the utility international law at all, it cannot be fully dismissed. However, from a political perspective, a focus on compliance glosses over more interesting nuances, and is may not be what we want to be talking about, when we talk about compliance. However, not everyone agrees with me. What do you think?

2 thoughts on “What we talk about when we talk about compliance

  1. I agree with Martin’s suggestions. It makes more sense to start with the measures that are more appropriately conceptualized and designed to elicit more valid inferences rather than to begin with the measures of compliance and attempt to modify them to the task of measuring causal effects. Otherwise the language of compliance may become rather restrictive and turn into a straightjacket for the political scientists.

Leave a Reply

Your email address will not be published. Required fields are marked *