Interest groups are a major channel through which citizens can express their opinions to decision-makers. Their participation in policymaking may improve decision-making processes by supporting policies that are in line with citizen preferences and blocking policies that solely reflect the interests of the governing elite. At the same time, however, intense interest group pressures may make it difficult for policy-makers to implement the most efficient policies since such policies often impose costs on parts of the public. Competition among interest groups over the distribution of economic gains may also slow down the rate of economic growth (Olson 1982). Finally, if some groups constantly win, interest group politics may undermine the legitimacy of electorally accountable decision making in a democracy.

This article looks at the impact of multilevel governance structures on state-society relations. It contends that rather than focusing on relations between state actors and societal actors, it is more useful to look at shifts between competing advocacy coalitions in a given issue area. It argues that the impact of multilevel governance structures on domestic advocacy coalitions depends on the political opportunity structure provided at the international level, the types of policy outputs international institutions can deliver, and the extent to which members of an advocacy coalition have the organizational capacities to be active at the international level. These factors are explored in two cases of public health policy: anti-smoking policy and alcoholism policy. Moreover, both cases show that multilevel governance structures offer better opportunities for challengers than for defenders of the domestic status quo.

Scholars have avoided studying interest group influence because of the difficulty operationalising the concept. The research presented here introduces a new way of measuring lobbying success and lays out a model of its determinants. To understand why interest groups sometimes succeed and at other times fail, we must consider the institutional structure of the political system within which the advocates are operating; the characteristics of the issue at hand; and finally the characteristics of the interest group itself and their lobbying strategy. I test these factors with original data based on interviews with 149 advocates in Washington D.C. and Brussels active on a random sample of 47 policy issues. From the results, issue context emerges as a much more important determinant of lobbying success than institutional differences. The institutional differences that do emerge suggest that direct elections coupled with private campaign finance lead to winner-take-all outcomes biased in favour of wealthier business interests, while the lack of these institutional characteristics leads to more balanced policy compromises with more advocates achieving some of their policy goals.

Studies of lobbying try to determine the influence and power of non-governmental actors on public policy. Although influence is very difficult to measure empirically, many continue to push for better research design to solve the problem. Through case studies of business-government relations in the United States and the European Union, this article argues that the difficulties with power and influence concern not only their operationalisation, but they also reflect conceptual confusions. Trying to determine the ‘winners’ and ‘losers’ of a policy issue can be misleading, since power also structures apparently harmonious exchange relationships. The perceived success of business lobbying in the cases studied depended on the governments' receptiveness to their demands, which in turn depended on strategic advantages they saw for themselves in international negotiations. Even when business appears to lead the dance, it is more promising to look at resource distribution and the interdependence of both sides, instead of assuming the domination of business power over policy outcomes.

The actions of nongovernmental organisations (NGOs) concerned with such issues as development, human rights, and the protection of the environment voicing concerns to public authorities raises the questions: Do these newly mobilised societal actors influence EU trade policy outcomes? We answer in the negative, arguing that such groups, which have diffuse costs and benefits from trade policies, do not dispose of resources with which they can threaten or enhance political actors' chances of re-election or re-appointment. A survey of NGOs and business groups as well as two in-depth case studies on the negotiations concerning the EU's Economic Partnership Agreements and the EU's policy on access to medicines in developing countries support our reasoning. The analysis shows that although NGOs have gained access to policy-makers, they have largely failed to shift policy outcomes in their favour.