Validation of Benefit-Transfer FunctionsEPA Grant Number: R833459
Title: Validation of Benefit-Transfer Functions
Investigators: Boyle, Kevin , Kuminoff, Nicolai , Parmeter, Christopher , Pope, Jaren
Institution: Virginia Polytechnic Institute and State University
EPA Project Officer: Wheeler, William
Project Period: January 1, 2007 through June 30, 2008
Project Amount: $194,374
RFA: Methodological Advances in Benefit Transfer Methods (2006) RFA Text | Recipients Lists
Research Category: Economics and Decision Sciences
1. Investigate criterion validity of benefit-transfer equations using simulated data. 2. Investigate convergent validity of benefit-transfer functions using “off-the-shelf” valuation data. 3. Investigate the sensitivity of the validity findings to applications of in situ versus marginal-effect use values. 4. Investigate the robustness of the convergent validity findings to freshwater fishing and coastal beach use applications. 5. Investigate whether supplementing existing valuation study data with GIS data enhances benefit-transfer function validity. Hypotheses tested will include the null hypotheses that preference parameters recovered from the estimated benefit-transfer functions are not significantly different from those defined for the simulated policy site (Objective 1) and from those estimated in original studies at policy sites (Objective 2). The null hypotheses tested will also include that there is no difference between benefit transfer predictions and known/estimated values at the policy sites. These tests will be replicated for Objectives (3), (4) and (5) to test the sensitivity of the validity investigations to situations typically faced by policy makers.
An overarching conceptual framework will be developed to guide the estimation of benefit-transfer functions. Benefit-transfer functions will be estimated using the study-site data to predict value estimates for the policy sites. The criterion-validity tests (Objective 1) will be undertaken through the use of simulated data where the “true” value is known for both the study and policy sites. The convergent-validity tests (Objective 2) will be undertaken through the use of actual data from study sites to estimate the benefit-transfer functions. The benefit-transfer functions will include traditional Meta analyses, calibration functions, Bayesian models and two new benefit-transfer function approaches introduced in this research, and two new nonparametric approaches that provide estimates of upper and lower bounds of transfer estimates. The policy sites will be freshwater (Maine) and trout (Delaware) fishing, and beach use (mid-Atlantic and Texas coast); estimated random-utility, travel-cost models exist for each site.
1. Identification of benefit-transfer functions that are the most credible. 2. Identification of benefit-transfer issues that are related to transfer method and those related to data limitations. 3. Clarification of issues that need further research to enhance the validity of benefit-transfer functions. 4. Identification of where original research on the values of freshwater fishing and coastal beach is needed to support benefit transfers. 5. More accurate measurement of benefits and costs by EPA and other federal and state agencies.