Purpose – The purpose of this paper is to analyze the variety of relations existing in tourism networks, identified as complex and mutable entities, where a vast range of stakeholders coexist. Design/methodology/approach – After a deep review on stakeholder theory, the research applies techniques of network analysis to a case study. Specifically, the analysis focuses on 354 hospitality firms acting in Molise Region (Italy). Each operator was asked to judge the importance to collaborate with other stakeholders to enhance the effectiveness of their management and marketing activities. The answers highlight the degree of preference among stakeholders and the resulting information is the level of confidence in the network. Findings – Results confirm the importance of intensifying relationships between tourism companies themselves and between them and policy makers. It appears that public stakeholders are more important for both management and marketing activities than private sector, since they place a much higher position in the scale of preference. Research limitations/implications – The paper provides a starting-point for further research about non-quantitative destination performance measurement, such as trust and commitment between the stakeholders in tourism destination, and the use of network analysis' techniques. Practical implications – Destination managers and policy-makers may use techniques of network analysis to elaborate useful information for planning and managing the relationships inside the tourism network. Originality/value – The paper offers a novel approach for developing network analysis in tourism network literature. It explores non-quantitative destination performance measurements and uses management and marketing activities to analyze relationships between public and private stakeholders.

We assess the impact on agricultural trade of European Union (EU) trade policies, using a gravity model based on disaggregated trade flows from 161 developing countries (DCs) to 15 EU member countries. We use a sample selection framework to account for potential selection bias of positive trade flows and provide an explicit measure for relative preference margins. From a policy perspective, our results debunk some of the most widespread criticisms of preferential policies: EU preferences matter and have a positive impact on DCs agricultural exports at both the extensive and intensive margins, although with significant differences across sectors.