The amenity value of climate change across different regions in the United States

ABSTRACT

This article estimates the effect of climatic variables on house prices near ski resorts in different regions in the United States. We find that among the climate variables we test, average winter temperature has the most significant and robust effect where an increase in this climate variable increases house prices near ski resorts at a decreasing rate. At the mean average winter temperature levels, an increase in average winter temperature reduces housing prices for all regions except the Northeast. The consumer surplus from projected average winter temperature changes is negative across all regions and the largest negative effects are in the Midwest and Mountain regions.

KEYWORDS:

• Climate change
• house price
• amenity value
• hedonic price
• consumer surplus

JEL CLASSIFICATION:

• Q51
• Q54

Acknowledgments

We thank Jonathan Yoder and Hayley Chouinard for comments on the early versions of this article. We also thank two anonymous reviewers and Mianfeng Liu for comments and help in revising the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ The Box–Cox test favours the double log specification over the linear and semi-log specifications.

² It is unlikely that there is endogeneity of the climate variables based on a bicausal relationship since housing purchases are unlikely to affect climate variables especially in the short run. We also tried to control for many variables in order to reduce the possibility of endogeneity from omitted variable bias. However, we cannot be assured of a causal link, so we can only interpret such results as a correlation.

³ See Appendix 2 for a list of states in each region. We do not include the Southern region because there are not many ski resorts in that location.

⁴ The white book of ski resort is not published every year, so we were only able to use the closest year at the time of analysis which was 2006. Note that some of the ski resort characteristics such as the elevation of the ski resort or the area of the ski resort should be consistent from 2006 to 2010.

⁵ To illustrate the graph, we use the formula, $P_{ci}=\frac{{\beta }_0}{1-\gamma {\sum }_n{\omega }_{ns}}+\frac{{\beta }_i}{1-\gamma {\sum }_n{\omega }_{ns}}{Z}_{ic}^W+\frac{{\beta }_y}{1-\gamma {\sum }_n{\omega }_{ns}}\bar{y}+\frac{{\beta }_n}{1-\gamma {\sum }_n{\omega }_{ns}}\bar{n}$, where $\bar{y}$ and $\bar{n}$ are average income and average population density respectively.

⁶ The owner-occupied housing unit population is 307 276; 434 156; 169 359 and 413 598 in the West, Midwest, Northeast and Mountain region, respectively (U.S. Census tract 2010).