Premium
Valuing Farmland Protection: Do Empirical Results and Policy Guidance Depend on the Econometric Fine Print?
Author(s) -
Johnston Robert J.,
Bergstrom John C.
Publication year - 2011
Publication title -
applied economic perspectives and policy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.4
H-Index - 49
eISSN - 2040-5804
pISSN - 2040-5790
DOI - 10.1093/aepp/ppr020
Subject(s) - mixed logit , discrete choice , easement , economics , logit , estimation , willingness to pay , welfare , econometrics , preference , public economics , econometric model , revealed preference , logistic regression , microeconomics , statistics , mathematics , market economy , management , political science , law
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.