Week 9 Young (2002)

Global Environmental Change 12 (2002) 73–77

Institutions for Global Environmental Change Monitoring the effectiveness of international regimes is now an academic industry in its own right. Regimes have come of age. For the main global environmental change issues such as climate change and biodiversity, they are the big game in town. In his thoughtful and comprehensive review of the recent work on regime effectiveness, Oran Young indicates that it is possible to monitor how far regimes are changing policy, shifting behaviour, and inclining nation states towards sustainability. What is particularly interesting is that regimes appear to hold, even when pummelled by belligerent national leaders. They are robust because collectively we appear to care for the well-being of the planet, at least to some degree. What is less edifying is that all regimes are

inadequate in creating the conditions needed to achieve sustainability. Admittedly, none was designed with sustainability specifically in mind. But an ‘effective’ regime should be capable of being tested on this measure. A sustainability test would also require regimes to be examined on issues of justice and equity. May be the next round of effectiveness analysis will address these deficiencies. Tim O’Riordan and Andrew Jordana a School of Environmental Sciences University of East Anglia CSERGE Norwich NR4 7TJ, UK E-mail address: [email protected]

0959-3780/02/$ - see front matter r 2002 Published by Elsevier Science Ltd. PII: S 0 9 5 9 - 3 7 8 0 ( 0 1 ) 0 0 0 1 6 - 4

Evaluating the success of international environmental regimes: where are we now? Oran R. Young* Dartmouth College, Institute on International Environment Governance, 6214 Fairchild, Hanover, NH 03755, USA

Writing in these pages in 1996 [Vol. 6 (4), pp. 394– 397], Marc Levy observed that research on the consequences of international environmental regimes has become a growth industry among students of international institutions, and he offered a preliminary assessment of approaches to the study of regime effectiveness highlighting compliance, behavioural change, and policy suitability. What has happened in this field of study during the intervening years? Interest has continued to rise, and we have made significant progress in devising analytic techniques appropriate to an analysis of the effectiveness of international environmental regimes. Yet major challenges lie ahead in this important area of study, especially when we seek to separate the effects of institutions from the impacts of other driving forces in international society. In the *Tel.: +1-603-646-1253; fax: +1-603-646-1279. E-mail address: [email protected] (O.R. Young).

following paragraphs, I describe recent developments in this field and comment on some of the most promising research initiatives currently underway.

1. Defining and measuring the dependent variable Most research on the consequences of international regimes focuses on regime effectiveness as the dependent variable (Underdal and Young, forthcoming). In essence, the resultant stream of analysis asks how the actual state of the world differs from what it would have been in the absence of a specific regime and treats this difference as a measure of the effectiveness of the relevant regime. While this procedure has obvious attractions, it raises two major issues that have come to occupy much of the attention of researchers in this field.

0959-3780/02/$ - see front matter r 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 9 5 9 - 3 7 8 0 ( 0 1 ) 0 0 0 1 7 - 6

74

Institutions / Global Environmental Change 12 (2002) 73–77

The first of these issues is conceptual in nature and focuses on different ways to think about changes in the state of the world. Adopting a distinction that originated among scholars concerned with the effectiveness of public policies, the mainstream of research on regime effectiveness rests on a distinction among the outputs, outcomes, and impacts of institutions (Underdal, forthcoming). Roughly speaking, outputs encompass those procedures and arrangements (e.g. ratification of international agreements, promulgation of appropriate regulations, creation of relevant organizations) needed to transform a regime from a paper arrangement into a going concern. Outcomes are behavioural consequences arising from the creation and operation of a regime. They include compliance and conformance on the part of regime members as well as other behavioural changes occasioned by the operation of a regime. The idea of impacts, by contrast, refers to problem solving in the sense of measurable changes in the status of the problem a regime is designed to address (e.g. reductions in emissions of ozone-depleting chemicals, improvements in the state of fish stocks) that can be attributed to the operation of the regime. Clearly, these categories of consequences are not mutually exclusive; effective regimes typically have consequences that fall into all three categories. But note that a regime can produce outputs without generating either outcomes or impacts and that the production of outputs and outcomes together still does not ensure success with regard to problem solving. Even if the Kyoto Protocol were to enter into force and be fully implemented, for instance, the problem of climate change would not go away. Since problem solving constitutes the ultimate measure of regime effectiveness, it is well to bear in mind that success in the realm of outputs and outcomes does not guarantee effectiveness at the level of impacts. These observations lead to a second problem that is more analytic in nature. How can we demonstrate the existence of causal connections between the operation of a regime and the outputs, outcomes, and impacts thought to be regime consequences (Young, 1999b; Miles et al., 2001)? If a regime arises in response to a problemFsuch as the depletion of stratospheric ozoneFand the problem subsequently disappears or subsides, can we be certain that this achievement is properly interpreted as a consequence of the regime? As experienced researchers know, there is always a danger in such situations that apparent connections will turn out to be spurious correlations. But note that there is an important relationship between the character of a regime’s consequences and the difficulty of making persuasive claims about their causal significance. If the negotiation of an international agreement (e.g. the 1987 Montreal Protocol) leads to outputs like treaty ratification and the promulgation of implementing regulations within individual member countries, for example, it is

reasonable to conclude that these occurrences are consequences of the creation of the regime. The causal chain in such cases is both direct and short. If a regime arises to deal with a problem like ozone depletion and the production of ozone-depleting chemicals declines during the ensuing years, on the other hand, the claim that the regime can be credited with success in the realm of problem solving is a far more difficult one to justify persuasively. The causal chain under such conditions is longer, and there is more room for other causal forces (i.e. independent variables) to come into play. Not surprisingly, therefore, the development of analytic techniques designed to reveal causal connections between institutions and their consequences constitutes a major focus of interest among students of regime effectiveness (Young, 2001a). Some researchers have sought to carry this line of thinking a step further by aggregating the consequences of regimes in such a way as to produce an overall index of regime effectiveness. The most sophisticated effort of this sort treats effectiveness (E) as the proportion of the distance between the no-regime counterfactual (NR) and some measure of the collective optimum (CO) covered by the actual performance (AP) of the regime (Sprinz and Helm, 1999). This approach yields the formula E ¼ AP-NR=CO-NR:

ð1Þ

This formula has the attractive properties that it produces an overall score for the effectiveness of any given regime falling into the interval between 0 and 1 and that these scores are comparable across regimes. But note that the approach does not solve the problems referred to in the previous paragraphs. There is no procedure built into this index for combining various types of outputs, outcomes, and impacts to generate a composite measure of actual performance or AP. Above all, this approach does not solve the problem of demonstrating causal connections (Young, 2001a). In effect, the issue of causality is embedded in the noregime counterfactual which is a measure of what would have happened in the relevant issue area if the regime had not come into existence. These difficulties have led some students of regime effectiveness to propose an alternative approach to the subject (Mitchell, forthcoming). This alternative starts by taking some empirically tractable phenomenon (e.g. trends in the production and consumption of ozonedepleting chemicals) as the dependent variable and proceeds to inquire into the role of a number of independent variables, including institutions or aspects of institutions, as determinants of the value of the dependent variable selected. This way of thinking frees the measurement of the dependent variable from the need to establish causal connections and allows the analyst to focus on the proportion of the variance in

Institutions / Global Environmental Change 12 (2002) 73–77

some measurable variable(s) attributable to the operation of a number of independent variables, including but clearly not limited to institutional factors. For the most part, those who think in these terms assume, either explicitly or implicitly, that analysts will make use of some form of multivariate statistics (e.g. multiple regression), although there are other techniques of analysis that may prove relevant in this context (Ragin, 1987). It is unlikely that this way of thinking will displace analyses that treat regime effectiveness as the dependent variable. But as a means of supplementing mainstream efforts to explain variations in the effectiveness of regimes, this alternative seems promising.

2. Explaining regime effectiveness Assuming we can find satisfactory methods for measuring variations in success or effectiveness both across regimes and through time with respect to individual regimes, the next challenge is to develop explanations that can account for these variations. Mainstream efforts to respond to this challenge divide into two branches, one examining the relative importance of factors relating to power, interests, and knowledge as determinants of regime effective and the other directing attention to the relative importance of problem structure, institutional attributes, and socioeconomic settings in this context. In both cases, research has focused so far on the formulation of generalizations that spell out the role of specific factors as determinants of regime effectiveness. Accounts dealing with the role of power, interests, and knowledge are typically rooted in the larger intellectual constructs that students of international relations employ (Baldwin, 1993; Hasenclever et al., 1997). The views of those who emphasize the importance of power reflect the perspectives of neo-realism, a way of thinking that produces hypotheses of the following type: the success of specific regimes requires the active participation of a hegemon or dominant power; the ability of regimes to solve specific problems will change as a function of shifts in the underlying configuration of power among the member states (Keohane, 1986). Neo-institutionalists, by contrast, approach the effectiveness of regimes in terms of utilitarian perspectives that emphasize configurations of interests among key actors. They articulate hypotheses to the effect that regimes will prove successful to the extent that they reduce transaction costs associated with interactions among the relevant actors; transparency regarding the degree to which actors comply with regime rules or fulfill institutional commitments is a key determinant of the success of regimes (Oye, 1986). For their part, those who focus on the role of knowledge tend to espouse social constructivist modes of analysis;

75

they formulate hypotheses about the importance of a common discourse or the presence of an epistemic community to the success of specific regimes (Litfin, 1994; Haas, 1997). The individual hypotheses flowing from these divergent perspectives are not necessarily contradictory or mutually exclusive. But they rest on different models of social processes, a fact that often makes it difficult to compare and contrast them in a meaningful fashion. The other branch of this stream of analysis directs attention to factors involving the attributes of regimes and the conditions under which they operate in contrast to alternative models of social action (Wettestad, 1999). One influential strand of this mode of analysis focuses on problem structure and seeks to rank problems on a scale ranging from benign to malign (Miles et al., 2001). The implication is that malign problems are particularly hard to solve and that we can expect regime effectiveness to vary inversely with the degree of malignancy of the problem structure. Yet others take the view that regime design matters and argue that the effectiveness of individual regimes will be determined, at least in part, by their institutional attributes and especially by the fit between these attributes and the nature of the problem to be solved (Mitchell, 1994). Even hard problems can be solved, on this account, if the regimes created to cope with them are well-suited to the essential features of the problems they address. Beyond this lies the observation that the socio-economic settings within which regimes operate are important determinants of the success of these arrangements (Young, 1999a). A regime that links actors that share a tradition of effective cooperation in other issue areas or that operates in a setting characterized by economic prosperity, for instance, is more likely to succeed than a similar arrangement that must build bridges among antagonistic actors or function during a period of economic recession. Again, these arguments are not mutually exclusive. But they do point to fundamental differences regarding the extent to which institutions are independent drivers or arrangements whose success is determined by external forces. Taken together, these explanatory hypotheses represent a significant advance in our understanding of the sources of regime effectiveness. Nevertheless, they leave a good deal to be desired as contributions to a theory of effectiveness. It is easy to find prominent examples that run counter to virtually every generalization about the roles of power, interests, and knowledge as determinants of institutional success. Some regimes succeed in the absence of a hegemon; transparency is not always needed to ensure effectiveness; the absence of an epistemic community need not be fatal to the operation of specific regimes. Similar comments are in order about generalizations relating to problem structure, regime design, and setting. Although some problems are undoubtedly harder to solve than others, we lack a

76

Institutions / Global Environmental Change 12 (2002) 73–77

convincing procedure for operationalizing the benign/ malign continuum (Young, 1999a). Analyses of the consequences of regime design almost always focus on the details of specific arrangements; they do not lend themselves to the formulation of powerful generalizations pertaining to matters of design. The idea of socio-economic setting encompasses a wide range of factors that are not easy to cast in the form of meaningful hypotheses. What is to be done? Certainly, there is much to be said for continuing the mainstream effort to explore the roles of power, interests, and knowledge along with the effects of problem structure, regime design, and socio-economic setting in this realm. But an alternative approach is now emerging that takes the view that regimes must fulfill certain basic functions to succeed as problem-solving devices but that there are many different ways to handle these functional requirements in specific cases (Young, 2001b). To illustrate, it seems clear that successful regimes establish behavioural mechanisms that are able to channel the behaviour of key actors, give rise to robust social practices that engage actors and capture their loyalties, and include steering systems that monitor progress toward desired ends and adjust regulatory and other provisions as needed to achieve these ends. An examination of the various ways in which individual regimes fulfill these functional requirements is not likely to lead to simple generalizations of the sort considered earlier in this section. Yet it is reasonable to expect that this way of thinking can contribute substantially to the development of inventories of best practices relating to the design and operation of environmental regimes (Honkanen et al., 1999). With regard to steering systems, to take a concrete example, the repertoire of best practices might include information pertaining to systems of implementation review (SIRs), noncompliance procedures, multilateral funding mechanisms, the design of incentive-based policy instruments, and procedures for adjusting regulatory rules without triggering a need for ratification on the part of individual member states. Of course, it will never do to think of these inventories as cookbooks containing simple recipes guaranteed to produce successful results under a wide range of conditions. Even so, such compilations of best practices may prove useful to those responsible for designing and operating regimes to deal with a variety of specific problems.

3. Next steps The ranks of environmental problems continue to grow. Today, we are increasingly concerned with global problems, such as ozone depletion, climate change, and the loss of biological diversity, that have risen to prominence alongside more traditional issues like the

management of marine fisheries, international waterways, and transboundary pollution. As the ranks of problems have grown, so too have the ranks of international regimes designed to solve or manage these problems. Some of the resultant regimes are more successful than others by any standard. As a result, we face an urgent need to learn more about what factors determine the relative effectiveness of individual regimes and whether there are opportunities to increase success through well-informed efforts to design institutional arrangements to fit the contours of specific problems. Recent years have witnessed a sharp rise of interest in this research challenge, and there is clear evidence that we are making progress in this field of study. Still, much remains to be done before we can even begin to talk about the development of a theory of regime effectiveness. Among the greatest challenges in this field are the need to find persuasive methods to demonstrate the causal links between regimes and their consequences and the need to move beyond simple hypotheses connecting regime effectiveness to individual factors relating to power, interests, and knowledge. These are major challenges; they cannot be met in the absence of a concerted effort on the part of the research community. Yet the good news is that we have not begun to exhaust the range of analytic techniques that can be brought to bear on the problem of understanding regime effectiveness.

References Baldwin, D.A. (Ed.), 1993. Neorealism and Neoliberalism: the contemporary debate. Columbia University Press, New York. Haas, P.M. (Ed.), 1997. Knowledge, Power, and International Policy Coordination. University of South Carolina Press, Columbia. Hasenclever, A., Mayer, P., Rittberger, V., 1997. Theories of International Regimes. Cambridge University Press, Cambridge. . Honkanen, M.L., von Moltke, K., Hisschemoller, M., 1999. Report of the Concerted Action on the Effectiveness of International Environmental Agreements, Noordwijk. Workshop, October 15–18, 1998. Report number R-99/05, Institute for Environmental Studies, Vrije Universiteit. of Amsterdam. Keohane, R.O. (Ed.), 1986. Neorealism and Its Critics. Columbia University Press, New York. Ltifin, K.T., 1994. Ozone Discourses: Science and Politics in Global Environmental Cooperation. Columbia University Press, New York. Miles, E.L., et al., 2001. Explaining Regime Effectiveness: Confronting Theory with Evidence. MIT Press, Cambridge. Mitchell, R.B., 1994. Regime design matters: intentional oil pollution and treaty compliance. International Organization 48, 425–458. Mitchell, R.B. Quantitative analysis in international environmental politics: toward a theory of relative effectiveness. In: Underdal, A., Young, O.R. (Eds.), Regime Consequences: Methodological Challenges and Research Strategies. (Forthcoming). Oye, K.A. (Ed.), 1986. Cooperation under Anarchy. Princeton University Press, Princeton.

Institutions / Global Environmental Change 12 (2002) 73–77 Ragin, C.C., 1987. The Comparative Method: Moving beyond Qualitative and Quantitative Strategies. University of California Press, Berkeley. Sprinz, D.F., Helm, C., 1999. The effect of global environmental regimes: a measurement concept. International Political Science Review 20, 359–369. Underdal, A. Methodological challenges in the study of regime effectiveness. In: Underdal, A., Young, O.R. (Eds.), Regime Consequences: Methodological Challenges and Research Strategies. (Forthcoming). Wettestad, J., 1999. Designing effective environmental regimes: the key conditions. Edward Elgar, Cheltenham.

77

Young, O.R., 1999a. Governance in World Affairs. Cornell University Press, Ithaca. Young, O.R. (Ed.), 1999b. The Effectiveness of International Regimes: Causal Connections and Behavioral Mechanisms. MIT Press, Cambridge. Young, O.R., 2001a. Inferences and indices: evaluating the effectiveness of international environmental regimes. Global Environmental Politics 1, 99–121. Young, O.R., 2001b. Can new institutions solve atmospheric problems? confronting acid rain, ozone depletion, and climate change. Paper prepared for the Global Change Open Science Conference, Amsterdam, the Netherlands, 10–13 July.