The value of scientific information on climate change: a choice experiment on Rokua esker, Finland

Abstract

This article presents an application of the choice experiment method in order to provide estimates of economic values generated by water quantity improvements in the environment. More importantly, this is the first choice experiment study valuing scientific information and in particular scientific information on climate change. The case study of interest is Rokua in Northern Finland, a groundwater dependent ecosystem very sensitive to climate change and natural variability. The study deals with the uncertainty about the actual dynamics of the system and the effect of future climate change by exploring whether the public values sustained provision of resources for scientific research to better understand long-term environmental changes in Rokua. Data are analysed using a nested multinomial logit and an error component model. Evidence from this study suggests that individuals are willing to pay in order to assure scientific research so as to better understand long-term environmental changes. As a result, policy should consider investing in and supporting related research. Other aspects of water management policy valued by the public are water quantity, recreation, and total land income.

Keywords:

• choice experiment method
• nested logit model
• error component model
• willingness to pay
• improved scientific information
• water management practices

Acknowledgements

We gratefully acknowledge the financial support from the European Union via the 7th Framework Program GENESIS: Groundwater and dependent ecosystems: New Scientific basis on climate change and land-use impact for the update of the EU Groundwater Directive; WP-6 Groundwater systems management: scenarios, risk assessment, cost-efficient measures and legal aspects. We finally thank two anonymous referees for constructive and insightful comments.

Notes

1. According to that property, the ratio of choice probabilities between two alternatives in a choice set remains unaffected by the introduction or removal of other “irrelevant” alternatives.

2. A violation of the assumption occurs whenever the Hausman-McFadden IIA test value is strictly higher than the critical value for the χ ² statistic which in our case was 16.87. Hence, acceptance of IIA was firmly rejected with the Hausman statistic being large and statistically significant at the 5% level.