G8 Research Group
Help | Free Search | Search by Year | Search by Country | Search by Issue (Subject) | G8 Centre

Explaining Summit Success: Prospects for the Denver Summit

Remarks Delivered at the University of Colorado at Denver

19 June 1997

[Document Contents]


In her introduction, Lt. Governor Gail Schoettler, emphasized the importance of the annual summits to forward-looking international business communities such as Denver. As commerce becomes more global, few firms can escape the pressures of international competition. Policies shaped and negotiated in Denver over the next few days can affect the trading environment, the competitiveness of businesses, and the quality of life of the citizens, of the G7 nations, European Union members, and Russia.

Since 1975, the attention paid to these events by the media has escalated. At the first summit, there were approximately four hundred official press credentials issued and the leaders were isolated from the media. It is anticipated that the number of official press credentials issued at this summit will exceed five thousand. Hence, the statements and declarations of summit participants have received ever greater scrutiny.

Whether the summits really accomplish anything at all, however, is arguable. For the summits to be newsworthy, the summit process must be meaningful. That is, the policy commitments endorsed by the leaders and made public though the summit declaration should meet two criteria: They should be ambitious and they should be complied with.


von Furstenberg and Daniels (1991, 1992) and Daniels (1993) are the first efforts to quantify commitments made at the first 15 summits and evaluate the extent to which these commitments were fulfilled. This work, which centers on the economic communiqué only, establishes a uniform approach for gauging compliance.

2a Methodological Approach

The economic communiqué issued at the conclusion of the summit represents a form of legal contract, as the leaders endorse the commitments contained therein. This document is used as the sole source of the compliance study. We focus exclusively on the communiqué and ignore statements or press releases that may proceed or follow the release of the communiqué. Peter Hajnal's text The Seven Power Summit: Documents from the Summits of Industrialized Countries, 1975-1989 contains the official economic communiqué of these early summits. (We are pleased to share the podium with him today.)

Using the communiqué, we attempt to define the goals and objectives of the various commitments. This is not always so straightforward, as many of the commitments are couched in vague terms. This lack of specificity in policy commitments is understandable, because, as Lord Keynes stated, it is better to be partially right, as opposed to being dead wrong.

There are two types of commitments that are revealed in the communiqué. The first type is a policy measure, which is a commitment to deliver a specific legislative package, such as a balanced budget agreement. The second type of commitment is a policy outcome, which is a shift in an economic variable, such as reducing the deficit, or reducing inflation. We consider those commitments that are concrete enough to identify and quantify the intended policy measure or policy outcome.

The next step in monitoring compliance is to establish a scoring metric that assigns a number reflecting the degree to which a commitment was fulfilled. Our approach is to define a range of scores from -1 to 1, where a score of 1 reflects complete fulfillment. A score of -1, however, is assigned if the actual outcome were the opposite of that committed to. For example, if the rate of inflation was 8% at the time a commitment was made, and the commitment was to reduce inflation by 3%, then an outcome of 5% inflation would be assigned a score of 1. An outcome of 8% means that the summit commitment accomplished nothing, and is assigned a score of 0. An outcome of 11% inflation, on the other hand, is assigned a score of -1. This methodology enables us to identify and quantify the commitments and assess in a uniform manner the degree to which they were fulfilled.

2b Results

The summit communiqué reveals 209 commitments made at the first 15 summits. By applying the methodology outlined above, a score was assigned to each commitment. Having a sufficient number of commitments, conclusions are reached, in a statistical sense, about their ambition and degree of fulfillment. The resulting scores are also used to draw conclusions on some popular conjectures regarding international policy making.

The first conjecture is that joint economic commitments tend to be honored to a lesser degree than an individual commitment. The findings indicate no statistical difference between these two types of commitments.

The second conjecture we consider is that commitments that promise delivery of a policy measure or instrument that is under the direct control of policy makers would be honored to a higher degree than commitments promising an outcome for a policy target. The scores reveal the opposite: Commitments on direct policy measures receive a lower than average score than those on economic targets.

Finally, it is often argued that policy makers of smaller nations would scrupulously honor their commitments so as to provide political leverage over the policy makers of the lager and more powerful nations. Though Britain and Canada received the highest scores, and the United States second lowest, the scores do not reveal any systematic pattern based on economy size.

The overall average score for the 209 commitments was 0.317, or 32 percent, meaning that, roughly one-third of what was promised was actually delivered by the policy makers. The null hypothesis of "no summit ambition" and "no summit effect" was then tested. Though the overall score was low, we are able to reject, in a statistical sense, the null hypothesis.

2c Conclusion: 1975 through 1989

The first fifteen summits delivered a large number of commitments on economic policy making. The extent to which these commitments were fulfilled, however, was disappointingly low. Promises were ambitious, rejecting the notion that policy makers "do not go out on a limb" at the summits. We must conclude, therefore, that the summits are indeed worth something, but perhaps not as much as that indicated by the media attention.


Our research project at the University of Toronto identifies the patterns,

explains the causes and explores the processes of compliance within the G7 process by analyzing compliance with the summit's environment and development commitments from 1988 to 1995. This period provides an era of sustained summit attention to, and important action on these issues and one over which summit attention and ambition has varied. This combination of eight years, two countries (the most and least powerful within the group, the United States and Canada) and four issue areas (biodiversity, climate change, debt reduction and assistance to Russia), including 83 specific commitments, offers enough cases to identify compliance patterns and isolate key causal factors. This study thus provides the first systematic review of the record of compliance with G7 decisions in the post-cold war, globalizing system of the 1990s, and in regard to the prominent themes that have flourished within this period.

3a Results

The empirical findings reveal that during its third summit cycle, the G7 has produced a large number of specific and often ambitious commitments within these particular issue areas, with Canadian and US compliance being generally positive, with a net score of +72 over the 166 commitments accepted (for an overall score of 43%). Wide variations appear, however, by country, issue area and time period. Canada's overall score of +44 over 83 commitments (or 53%) contrasts with the US overall score of +28 over 83 commitments (or 34%). Moreover, compliance is much higher in regard to developing country debt and assistance to Russia than for climate change and biodiversity. What we do find, however, is that compliance is lower for both countries in the environment domain in the pre-Rio period of 1988-1991 than in the post-Rio period of 1992-1995.

The study thus concludes that over time, the summit has become more active in generating agreements that are specific, identifiable and measurable, that compliance with summit commitments has been positive, and that positive compliance is a wide-spread phenomena - in other words, the most powerful US, and least powerful Canada, comply with their summit commitments. Why then, is there a sustained, and in fact, increasing level of compliance with summit commitments? How do we account for and explain these patterns of summit compliance?

3b Institutional Variables and Regimes

The findings in this study indicate that there is considerable support for institutional variables and the role of regimes. On the environment side of the agenda, there is no net compliance during the period 1989-1991 in either climate change or biodiversity. Beginning in 1992 and onwards, however, there is a high level of sustained environmental compliance. This is primarily due to two factors. First, 1992 marked the launching of a new era in environmental diplomacy with the convening of the United Nations Conference on Environment and Development (UNCED) in Rio de Janeiro, Brazil. Convening just three weeks later in Munich for their annual Summit meeting, the leaders endorsed the Rio conventions and agreed that Rio represented a landmark in heightening the consciousness of global environmental challenges and in giving new impetus to the process of creating a worldwide partnership on development and the environment.

But along with the success of UNCED, came the recognition that if the Earth Summit was to have any lasting significance, the Summit seven had to act collectively to implement the conventions embraced at Rio, thereby setting a model and incentive for the rest of the international community to follow.

Yet there is an additional institutional factor at work, grounded in the development not of the UN system, but of the G7 itself. The 1991/92 "Rio Rise" was also coincident with the institutionalization of the G7 environment ministerials, beginning in Germany just prior to the 1992 Munich Summit, and then continuing in Florence, Italy in 1994, Hamilton, Canada in 1995, Cabourg, France in 1996, and Miami in 1997. Along with the institutionalization of the G7 environment ministerials, has come the recognition of the importance of keeping the "spirit of Rio alive" by continuing with the sustained implementation of the Rio commitments on climate change.

The findings on the debt side and assistance to Russia indicate that there has been a sustained, and in fact high level of compliance by both Canada and the US during the period examined. Once again, this reflects the important institutional variables at work. Within both of these issue areas, the implementation of summit resolutions occurs through long-established departments (Treasury and Finance) possessing well-defined domestic implementation responsibilities, but also manifesting strong institutional links to powerful multilateral organizations.

Given that there is a well-defined and clearly-established process with Treasury and Finance, this serves to guarantee a systematic operationalization of the communique. Moreover, such long-standing departments possess well-established institutional links to the Paris Club, IMF and World Bank - the bodies primarily responsible for the global implementation of such issues. Because the G7 are major shareholders within these institutions -due to their financial contributions and controlling weight - they are able to formulate final decisions and forge a consensus on the implementation of the debt strategy within these broader groups.

On the other hand, international environmental commitments, where compliance is lower, are administered through the specialized agencies of the United Nations (UNEP/UNDP) where the G7 does not possess overwhelming controlling strength due to both institutional characteristics (one-country-one-vote) and underlying issue-specific distributions. Finally, the G7 Finance Ministers and Finance Deputies process has existed since 1986 and is thus more institutionally entrenched than the newer and more "embryonic" G7 environment ministerial forum that began only in 1992.

3c Political Control

In addition to regimes, we find that the element of political control also offers explanations for compliance with summit resolutions. The fact that leaders themselves are present at the Summit table, does seem to ensure that the decisions they reach, and the commitments they make, carry added weight. There are no higher-level bureaucrats at home to whom their decisions are deferred. As a result, when a head of state becomes personally associated with a summit commitment, it is somewhat different than if they would have been arrived at by a group of ministers; when the Prime Minister and President are directly involved, that has a major impact on policy and the priority of policy on the home front. Moreover, when the head of state attaches a high degree of personal importance and commitment to certain issues, the degree of implementation is even higher.

This political control variable also takes into account the social-psychological dimension of a leader's individual personality and the importance he/she places on international institutions and agreements more generally. For example, is a head of state demonstrates an attachment to sustainable development initiatives, consistently advances these themes at the annual Summits, demonstrates a commitment to multilateralism, we find that the effects of this attitudinal dimension also serve to affect compliance levels.

Yet because G7 heads are not merely leaders, but democratically-elected ones, their ability to impose their implementing will within their government is constrained by their political standing within society at large. When leaders and their parties enjoy high approval ratings and popularity, their ability to implement is increased. In addition, when domestic public opinion favours a particular issue area - such as the environment in both Canada and the US - even unpopular leaders at the time, facing a likely electoral defeat (such as Prime Minister Mulroney and President Bush) will comply with their communique commitments. This is primarily because they recognize the effects of public opinion and political pressure in areas important to their electorate

3d Conclusion: 1988 through 1995 Summits

To summarize, the results of this research project indicate an overall level of summit compliance in the positive range by both Canada and the US with environment and development issues during the last cycle of summitry from 1988-1995. The study finds that institutional variables, coupled with political control variables, best account for summit compliance during the last cycle of summitry in the environment and development spheres.


More recently, the G8 Research Group at the University of Toronto has begun monitoring compliance with summit undertakings on an annual basis. The Group's monitoring scheme covers all areas of policy, ranging from traditional economic issues to national and global security. Their scores for the Lyon Summit are remarkably high. Germany and Canada received aggregate scores well over 50 percent, the United States and France received scores in the 40 percent range, with Japan receiving a score just under 40 percent.


The compliance report on the Lyon Summit, as with the Kirton and Kokotsis, indicates an increase in the scores, as compared to the earlier studies. There are a variety of reasons that might explain why recent compliance scores have risen.


It is undeniable that when the heads of state get together, there is no such thing as "just talk." However, when the agenda is filled with idle time and cocktail parties, as the summits have become, little can be expected of the annual summits. The compliance scores indicate that the summits do accomplish something, but perhaps not as much as we would like them to.

[Document Contents]

G8 Centre
This Information System is provided by the University of Toronto Library and the G8 Research Group at the University of Toronto.
Please send comments to: g8@utoronto.ca
This page was last updated .

All contents copyright © 1995-99. University of Toronto unless otherwise stated. All rights reserved.