The drive to increase renewable electricity production in many parts of Europe has led to an increasing concentration of new wind energy sites at sea. This results in a range of environmental impacts which should be taken into account in a benefit-cost analysis of such proposals.

The authors use choice modeling to investigate the relative gains and losses from siting new wind farms off the coast of Estonia, relative to the option of creating a new marine protected area. They find that, while respondents are generally opposed to converting marine shoals to conventional wind farms and prefer the establishment of marine protected areas instead, benefits from constructing ‘environmentally-friendly’ wind farms – an alternative program which is also considered by the government – are not statistically different with respect to consumers’ welfare to those associated with creating a new marine protected area.

Methodologically, the paper makes a contribution by showing the ability of the latent class mixed logit model to represent both within- and between-class preference heterogeneity, and thus its power to provide a more sophisticated representation of preference heterogeneity than stand-alone latent class or mixed logit approaches. The paper also presents the first use of the latent class mixed logit model in willingness-to-pay space for environmental goods.