Economics

Valuing farmland protection: Do empirical results and policy guidance depend on the econometric fine print?

Document Type

Article

Abstract

Economists have generated an abundance of willingness to pay estimates to help inform U.S. farmland protection policies. Stated preference approaches are often used for such estimations, with an increasing reliance placed on discrete choice models such as mixed logit, in which results can be sensitive to minor specification changes. This paper evaluates how and to what extent this sensitivity is policy relevant. Drawing from a case study of agricultural conservation easements in Georgia, we characterize the sensitivity of choice experiment welfare estimates to common variations in mixed logit specification. Results suggest that practitioners should provide greater attention to the potential implications of often unnoticed modeling choices in discrete choice welfare estimation. © The Author(s) 2011. Published by Oxford University Press, on behalf of Agricultural and Applied Economics Association. All rights reserved.

Publication Title

Applied Economic Perspectives and Policy

Publication Date

12-2011

Volume

33

Issue

4

First Page

639

Last Page

660

ISSN

2040-5790

DOI

10.1093/aepp/ppr020

Keywords

farmland protection, mixed logit, sensitivity analysis, stated preference, willingness to pay

Link to Full Text

Share

COinS