Behavioural Compliance Theory

Following in the wake of its social science siblings, international law scholarship is experiencing a ‘behavioural turn’. One particularly fruitful area in which to explore the utility of behavioural insights is compliance. The contributions to the present symposium represent some of the ﬁrst efforts to explore how psychological inﬂuences may shape the compliance decision-making process and to study the theoretical and practical advantages offered by pursuing a behavioural analysis of compliance. This essay addresses several cross-cutting issues of importance. First, it outlines the epistemological stance that is implicit in behavioural approaches to compliance. It draws a parallel with economic theory, suggesting that the persistent dominance of rational choice theory in that ﬁeld belies a teleological difference between international law and economics. The essay then turns to suggest how psychological inﬂuences both interact with and build on the material and normative considerations that are posited in existing theories to determine actors’ compliance-related behaviour. It continues to address the limitations of adopting a behavioural approach to compliance, from both a methodological and normative point of view. The ﬁnal section provides a brief introduction to the pieces included in this symposium and outlines avenues for future research.


INTRODUCTION
Following in the wake of its social science siblings, international law scholarship is experiencing a 'behavioural turn'.Legal scholars are increasingly looking towards insights into psychological processes as a way of explaining actors' behaviour and evaluating the potential of norms and institutions.Whilst initial work staked the broad parameters of the field, 1 more recent scholarship has focused on conducting empirical studies, including laboratory experiments, to test the power of explanatory theories based on psychological factors. 2 One particularly fruitful area in which to explore the utility of behavioural insights is compliance. 3The theories of compliance that dominate the extant literature, such as realism, constructivism, and liberal theory, all rely explicitly or implicitly on psychological processes that take place on the individual level for their operation. 4ealism, for example, presupposes a calculation based on the perceived costs and benefits of a particular course of action, a calculation that can only take place if there is someone (or something) assessing those costs and benefits in the first place.Similarly, constructivism rests on the idea that continuous interaction with norms and other actors influences a State's behaviour.Yet, this iterative process depends very much on how the State views the actors and norms with which they interact; the likelihood that a State's identity will converge with institutions or norms that it perceives as being, for example, illegitimate or unfair would seem to be small.This, in turn, begs the question why States might have such perceptions, and this is where psychological insights might play a role.
Often the psychological processes that underpin compliance theories are not made explicit.Adopting a behavioural approach to compliance, then, rests on a twofold commitment: first, to deconstruct the psychological underpinnings of current theories of compliance; and, secondly, to explore how psychological influences may shape the compliance decision-making process, whether or not those influences neatly fall under the rubric of the well-known compliance theories.The articles in this symposium take the first steps in this direction in several ways: by challenging the myopic focus on the compliance 'moment' to better understand how psychological factors may influence decision-making; by understanding how different group identities affect the effectiveness of sanctions on compliance; and by exploring how accountability might be leveraged to increase compliance.
This framing essay to the symposium addresses, first, the epistemological stance that is implicit in behavioural approaches to compliance; put another way, what do these scholars think is worth knowing?It draws a parallel with economic theory, suggesting that the persistent dominance of rational choice theory in that field belies a teleological difference between international law and economics.The second section of the essay then turns to suggest how psychological influences both interact with and build on the material and normative considerations that are posited in existing theories to determine actors' compliance-related behaviour.The following section moves to outline the limitations of adopting a behavioural approach to compliance, from both a methodological and normative point of view.The final section provides a brief introduction to the pieces included in this symposium and outlines avenues for future research.
2. NAILING ONE'S COLOURS TO THE MAST Disagreements about theory are often disagreements about what the proper role of scholarship should be. 5Theoretical abstractions may have intellectual attraction, enabling the author to neatly frame a subject with aesthetically pleasing conciseness.But when they are measured up against the messy, confused reality of everyday life, their theoretical simplicity looks rather like an impediment.The rule of recognition might be the key to understanding what the law is,6 but identifying it 'on the ground' in a convincing manner is another thing.A self-interested State might, on paper, look like a good way to explain State action;7 in reality, focusing on material self-interest as the principal factor that motivates States to comply with law seems, at best, only partially true, and begs myriad follow-up questions (What does the State see as a cost/ benefit?How does it weigh those?).
Those that adopt a behavioural approach to compliance reject pure theoretical abstraction in favour of grounded empiricism.This is driven by a conviction that international law scholars need to know how and why States act the way they do before making suggestions for change or criticizing the status quo.8If we want to act constructively as scholars, we need an empirical basis on which to ground our work.If one adheres to this position, a corollary is that we should explore all influences that potentially impact decision-making regarding compliance, whether those be material, normative or psychological in nature, or something else.And while these might be tested within the framework of existing theories, the empirical impulse of behavioural scholars means that the construction of 'grand theories' of compliance will take a back seat until a sufficient body of evidence of the impact of psychology on State behaviour has formed.
In thinking about the interaction between international legal theory and behavioural insights, it is instructive to look to another field in which behavioural approaches have taken hold-indeed, the one in which they originated: economics.Exploring the psychological processes that challenge the rational choice assumptions that underpin homo economicus has a long tradition in economics, stretching back to the 1950s. 9Yet, despite the profound impact that behavioural economics has had both within and beyond the field of economics, the dominant paradigm for explaining individual and group action has remained rational choice theory.Why is this so?And, more importantly for our purposes, why should international lawyers-unlike many mainstream economists-move beyond existing theories of compliance towards more empirically grounded, behaviourally sensitive research?
Economists have advanced several arguments to explain the enduring persistence of rational choice theory.Some suggest that theories are not-and should not be-straightforward descriptions of the real-world;10 instead, the proper role of theories is to generate testable hypotheses, which may subsequently be subject to empirical verification.According to this approach (referred to as 'instrumentalism'), theories act as nothing more than 'codification schemes or scaffolding'11 that allow for the summary and classification of different hypotheses. 12As a consequence, the value of a theory is not measured by how well its assumptions reflect the real-world, but rather by the predictive capacity of the hypotheses it generates. 13he suggestion that theories do not-and are not intended to-represent the real world is well-developed in the economic literature, most prominently by Milton Friedman. 14In an influential 1953 essay, Friedman expounded a 'strong' variant of instrumentalism, arguing that it simply does not matter whether the assumptions underpinning economic theory are unlikely to be true if that theory generates hypotheses that are more likely to be predict real-world behaviour. 15ince Friedman's essay, however, the growth in experimental evidence that challenges rational choice assumptions has put pressure on the strong variant of instrumentalism that he adopted.More recent theorists adopt a toned-down version of instrumentalism (or 'neo-instrumentalism') that explains how theories operate in the reasoning of economists while nevertheless accepting the instrumentalist idea that theories need not reflect reality.Itzhak Gilboa and colleagues, for example, suggest that the persistence of rational choice assumptions in economics, such as the maximization of expected utility, may be explained by the fact that economists engage in analogical, as opposed to rule-based, reasoning.According to this account, theories are used by economists as a way to reason about new cases. 16A theory represents a new, hypothetical case (a Gedankenexperiment) that the observer can compare to the real-world problem that they are trying to understand, making a subjective judgment about the similarity and relevance of the theoretical case and their real-world problem.If sufficiently analogous, they might use the theoretical model to guide their 'prediction (or classification, diagnosis, or ethical or legal judgment) in the current case.' 17 Accordingly, a theory might not inform us of anything new or bring in novel empirical data, but-if successful-it should be insightful and change how we think about the issue being analysed. 18ccording another group of scholars-'scientific-realists' 19 -theories are tools that help researchers to isolate certain behaviour, allowing them to explore causal relationships that would be otherwise too complex to investigate directly.According to this school of thought, which most closely resembles the prevalent view in political science, 20 the role of theories is more than merely to act as an intellectual construct upon which to hang hypotheses or from which to draw analogies.Instead, theories are 'statements about real entities and processes, even unobservable ones, which affect natural and social phenomena'. 21A corollary is that useful theory must at least approximate reality, including the assumptions upon which those theories are based. 22Of course, few people nowadays maintain that theories can, or should, accurately reflect nature in all its complexity. 23Instead, the prevalent view is that theories should be pared-back representations of the real-world that allow us to better understand certain elements of reality without necessarily providing a mirror-image of it. 24By focusing on specific causal elements, actors, and relationships, theories lay 'bare the essential elements in play and indicate the necessary relations of cause and interdependency-or suggest where to look for them'. 25In this respect, theories are like maps: [b]oth aim to simplify a complex reality so we can grasp it better.A highway map of the United States, for example, might include major cities, roads, rivers, mountains, and lakes.But it would leave out many less prominent features, such as individual trees, buildings, or the rivets on the Golden Gate Bridge.Like a theory, a map is an abridged version of reality. 26o what extent is the instrumentalist justification for the persistence of rational choice theories of action applicable to international law?The answer to that question depends in turn on the extent to which the objectives of economic scholarship and those of international law scholarship align.And this is where we think that the epistemological ambitions of international law more closely match those of economic scientific-realists than instrumentalists.
Theories are important in international law because we understand them to explain-inevitably, in shorthand form-how, in whole or in part, international law works.They are understood to purport to depict reality, not to act simply as scaffolding within which to construct testable hypotheses. 27This is borne out by the claims that particular compliance theories provide 'better and more nuanced explanations of state behaviour',28 that they aim 'to offer a persuasive explanation'29 of why and when international law works, or to provide a theory that 'fully resonate[s] with the contemporary practice of international law-making and application'. 30A theory that had no pretence of describing reality would not only be wildly unsuccessful but also fail in helping us make sense of the world.
Why is this so?In our view, this approach to theory is connected to what might be called the 'applied nature' of much international law scholarship.International law scholars are often intimately linked to the practice that they observe, and relatively frequently exert a direct influence over the shape of international law.For example, international law scholars frequently serve on the international bench,31 as members of a State's delegation to multilateral law-making conferences,32 or as members of influential groups that frame the international law-making agenda. 33This means that the theories upon which international lawyers draw, in order to prove useful, must broadly reflect the practice that is seen in daily life.
To engage with, and to inform, the practice of States, international organizations and courts and tribunals, international legal theory (including theories of compliance) must, therefore, at least purport to depict reality.Theoretical parsimony should not be privileged at the expense of descriptive accuracy.Accordingly, it is imperative for compliance theory to engage with-and, where relevant, integrate-behavioural insights in order to continue to serve its purpose.

THE THREE PILLARS OF COMPLIANCE
Over the past 60 years, theories of compliance have proliferated, resulting in a rich and multifaceted literature that attempts to explain State behaviour.It is not the purpose of this essay to provide an overview of the extant literature on compliance, something which has been done well elsewhere. 34Rather, we want to sketch how we think behavioural insights complement, and add to, existing theories of compliance.For the purposes of this essay, we divide the factors that have been identified in the existing literature as influencing compliance into two pillars: material considerations and normative considerations. 35Material considerations may be tangible, such as the increased costs associated with compliance, or intangible, such as concern for the reputational impact caused by non-compliance. 36Normative considerations are those that influence a State to comply with a legal obligation because it feels obliged to obey the law because it is the law, or because the legal norm has been internalized within the State by, for example, the process of socialization. 37We think that behavioural insights can interact with these pillars in two different ways.
First, they can inform our understanding of how both material and normative considerations affect compliance-related behaviour by providing a more fine-grained, descriptively accurate account of how these factors are perceived by actors, and how they interact to result in compliance behaviour.To do this, we need to take a wide-angle view of compliance decision-making; a point that is elaborated in more detail by Daniel Peat in his article in this symposium.Suffice here to note here that behavioural insights could play a role in at least two distinct junctures in the compliance decisionmaking process, whether that process be driven by material considerations, normative considerations, or both.The first is how the State (and individual decision-makers within the State) perceives the inputs into that decision-making process.For example, there is abundant literature in behavioural economics that shows the systematic and predictable inability of individuals to estimate the costs and benefits of future trajectories accurately, instead giving outsize importance to, for example, highly salient pieces of information. 38The second juncture at which behavioural insights may play a role is when a State weighs up different courses of action once the inputs have been identified.Behavioural literature suggests, for example, that costs and benefits do not carry the same weight in decision-making, 39 and that some 'taboo' costs may override any benefit, no matter how large the latter promises to be. 40econdly, behavioural insights could add to the factors that influence compliance decision-making, standing alongside material and normative considerations as a third 'pillar' of compliance.This will be the case for psychological processes that may affect decision-making but which would be ill-placed under either material or normative pillars.Take the role of emotion in decision-making as an example.Over the past 20 years, scholars in international relations and political science recognized emotion to be a key factor in understanding human behaviour, 41 something that is only now starting to make inroads into international law scholarship. 42Emotions are gatekeepers, acting as lens through which individuals perceive and assess new information. 43eliefs about other States may, for example, be motivated by emotions, and may be held with more certainty than the available evidence warrants. 44An 'emotional belief' that a State is a untrustworthy partner or a vengeful actor may, one might hypothesize, affect not only the conclusion of agreements with that State but also subsequent compliance with those agreements.More than that, however, emotions themselves may cause States to act in certain ways.A feeling of fear or being cornered may, for example, cause a State to lash out, paying less heed to its international legal obligations than it might otherwise have done. 45e do not believe, then, that international lawyers should feel constrained to work within the framework of existing compliance theories when exploring what behavioural insights might bring to the table.They should see psychological processes as both enriching the accounts given by the normative and material pillars of compliance theory, as well as operating alongside these pillars to add to our global understanding of State behaviour.

THE LIMITS OF BEHAVIOURAL COMPLIANCE THEORY
So far, this essay has argued that international lawyers should pay attention to the psychological processes that influence State behaviour, as they open the doors to a more accurate, detailed description of how and why States act the way that they do.There are, however, limitations to what a behavioural approach can tell us as well as problems that must be overcome to realize the full potential of behavioural insights. 46These can be divided into two categories. 47he first is methodological limitations: how can we isolate and study psychological influences that affect State behaviour?Here, we run into issues caused by the methodological individualism that a behavioural approach necessarily implies.It is highly unlikely, for example, that we would be able to gain access to the elite decision-makers that actually decide whether a State complies or not with its international law obligations. 48Can a laboratory experiment conducted on members of the general public tell us something about how psychological processes may affect elite decision-makers?Or are the latter too different to the former to be able to extrapolate findings between groups? 49 related point is how do psychological processes that occur on the individual level affect State behaviour?One might say that State action is determined by relatively few elite decision-makers, such presidents or foreign ministers, and that the psychological processes that affect these individuals directly affect State action. 50owever, it is clear that some State action is the result of a process of group decision-making, or the culmination of a long series of individual decisions, for which it may be impossible to single out the impact of psychological factors.How should we approach studying such behaviour?Should we renounce any attempt to explore psychological factors because of the difficulties encountered?Or should we nevertheless explore how psychological factors might be tempered, exacerbated or aggregated from the individual to the State level?We think that research must take the latter direction, but international lawyers will need to be innovative in the methodologies they adopt to overcome this problem.This may entail adopting qualitative methodologies, such as process tracing and interviews that go beyond the lab experiments that are characteristic of behavioural research to date.
Finally, reference to psychological processes may help to explain actors' behaviour, but often so too might other factors, such as rational choice considerations, capacity constraints, or normative influences.For example, is the use of boilerplate clauses in bilateral investment treaties a result of the status quo bias, or is it simply the most efficient manner to create a new treaty that regulates the same subjectmatter as previous treaties? 51Might the outsize impact of judgments of international courts in the evolution of the law of the sea be explained by rational updating or the availability bias? 52The methodology that one chooses might be better or worse at isolating the effects of behavioural influences on compliance behaviour.Researchers need to reflect on what external indicators are specific to behavioural influences, as opposed to other non-psychological factors, and design their research projects accordingly.
There are also certain methodological issues that remain specific to compliance research.The most pressing of these is to define what we mean by compliance. 53Is late compliance still compliance, even years after the initial breach? 54How about partial compliance, for example, if a State pays only a certain percentage of the compensation awarded by an investment arbitral tribunal? 55And, relatedly, are the considerations that motivate compliance with legal obligations ('primary' compliance) the same, or similar, as those that motivate compliance with the judgments or awards of international courts or tribunals ('secondary' compliance)? 56Moving forward, scholars should be specific about the conception of compliance that they adopt, and explain why that conception is most appropriate for the purposes of their research.The clear insights into actors' behaviour that the behavioural research agenda aims to provide should not be obscured by conceptual opacity.
A second limitation is normative: once we start to understand the behavioural processes that affect compliance, should we capitalize on this to bring about better compliance? 57This point mirrors a common objection in the domestic behavioural law and economics literature; namely, that designing institutions and norms based on psychological processes in some way manifests an unjustifiable 'legal paternalism'. 58But we think this objection holds less weight on the international plane.Unlike on the domestic level, there is not a hierarchical relationship akin to that of the citizen/State on the international level.If behavioural insights have been integrated into treaty design, it is because States have decided themselves that this would be best course of action.Similarly, if international organizations or international courts and tribunals decide to integrate behavioural insights into their work, then they are accountable to their stakeholders, which, in most cases, will ultimately be the member States of the relevant organization.In contrast to the domestic plane, the actors who will be affected by behaviourally designed institutions are also the ones that have the power to change those institutions.
The preceding paragraph proceeds on the assumption that the actor whose compliance we want to change is the State.But that need not always be the case.The focus of some international rules may be the action of individuals, such as certain rules of international humanitarian law. 59In that case, one would have thought that many of the same considerations arise here as in the domestic context.But, here again, we see a difference.On the domestic plane, the normative objection to using behavioural insights to inform law or policy design rests on an assumption of a lack of oversight or accountability of the law-maker.On the international level, however, that position is significantly less clear: if built into a treaty, the multilateral lawmaking process provides other States a chance to flag what they might conceive to be objectionable uses of behavioural insights; if implemented by an international organization, Member States would be in the position to challenge the same.The horizontal dimension of the international system may play, at least in theory, a check against the unwarranted, objectionable or improper uses of behavioural insights.

STEPPING INTO THE FIELD
The three papers in this symposium aim to make the first tentative steps into exploring how behavioural insights might inform our understanding of compliance.The first paper, by Daniel Peat, challenges the idea that international lawyers should focus on the 'moment' of compliance, arguing that, in order to understand where behavioural influences might affect compliance behaviour, we first need to expand our conception of the compliance decision-making process.Drawing on political science and international relations literature, he argues that behavioural compliance research should be focused around two pillars: first, how elite decision-makers perceive new information; and, secondly, how the process of decision-making might affect the outcome.Against this theoretical backdrop, he uses publicly available data from multilateral development banks and a sentiment analysis of Argentinian parliamentary debates to explore whether behavioural explanations might have purchase in one particular context: Argentina's partial compliance in October 2013 with a series of investment treaty arbitration awards.Peat concludes that behavioural insights might be particularly useful in explaining how and why reputation shaped Argentina's response to these adverse arbitral awards.
The second paper in the symposium, by Niccolo `Ridi and Veronika Fikfak, tackles the question of compliance by focusing one of the ways in which it can be induced: sanctions.Drawing on behavioural and international relations theory, as well as on traditional doctrinal legal scholarship, the contribution problematizes the concept of sanctioning, laying emphasis on the purposive nature of the act as a means to induce behavioural change-an aspect which has been, barring few exceptions, largely neglected in international law literature.In so doing, the contribution also sheds light on the structure of the relationship between sanctioning and the ultimate goal of compliance by focusing on the triggers, actors, and strategies characterizing it, as well as on the normative limits imposed by international law.Ridi and Fikfak conclude by tackling the question of whether and to what extent sanctioning should be a pillar of compliance, calling for further reflection on this problematic relationship.
The final paper in the symposium is by Sophie Duroy, who employs behavioural theory to develop a 'bounded rationality equation' that predicts when States will comply with its international law obligations in intelligence activities.Duroy argues that her equation not only helps international lawyers to understand when States will comply with international norms, but also suggests that it helps to identify why they do so-the increased likelihood of being held effectively accountable.Accordingly, she calls for international lawyers to focus not on the normative framework that regulates intelligence activities, but on the mechanisms through which violators may be held accountable.
Behavioural Compliance Theory 177

2
See, for example, D Peat, 'The Tyranny of Choice and the Interpretation of Standards: Why the European