Academic journal article Agricultural and Resource Economics Review

Willingness to Pay for Water Quality Improvements in the United States and Canada: Considering Possibilities for International Meta-Analysis and Benefit Transfer

Academic journal article Agricultural and Resource Economics Review

Willingness to Pay for Water Quality Improvements in the United States and Canada: Considering Possibilities for International Meta-Analysis and Benefit Transfer

Article excerpt

This paper presents a multinational meta-analysis estimated to identify systematic components of willingness to pay for surface water quality improvements, developed to support benefit transfer for Canadian policy development.Metadata are drawn from stated preference studies that estimate WTP for water quality changes affecting aquatic life habitats-a type of study with few Canadian examples. The goals of this paper are to assess the properties of a multinational (United States/Canada) meta-analysis compared to a single-country (U.S.) analog; illustrate the potential information that may be derived as well as the analytical challenges; and assess the performance of resulting meta-functions for benefit transfer.

Key Words: meta-regression analysis, stated preference, water quality, willingness to pay, benefit cost analysis

Policy analysts frequently rely on benefit function transfer (BFT) as a source of nonmarket welfare estimates, with benefit functions drawn from individual primary studies in the literature. The availability of applicable site-specific benefit functions, however, is limited, particularly those drawn from study sites of sufficient similarity to the policy sites in question. In cases for which a closely matching study site function is not available, a parameterized meta-analytic benefit function, estimated using a meta-regression model (MRM) of many primary study results, can provide the analyst with a viable alternative (Bergstrom and Taylor 2006, Rosenberger and Loomis 2000). Results of prior MRMs have been promising with regard to the ability of such models to synthesize information regarding willingness to pay (WTP) and reveal systematic patterns unapparent from individual primary studies (Rosenberger and Johnston 2009).

BFT applications of meta-analysis (MA) typically involve the use of MRMs in which the dependent variable in a Bayesian or classical regression model is a summary welfare statistic (e.g., WTP estimate) drawn from comparable primary studies. Independent or moderator variables in the regression equation characterize corresponding attributes of the resource, policy, site, and affected population that are hypothesized to explain observed variation in the dependent variable across primary study observations (Nelson and Kennedy 2009). The estimated regression equation characterizes a value surface that reflects multidimensional patterns in estimatedWTP across the empirical observations that comprise the metadata. The use of the estimated regression equation for BFT requires the analyst then to assign values (i.e., choose variable levels) for independent variables reflecting conditions or attributes at an unstudied policy site, so that a transferable WTP prediction can be obtained.

Much of the literature on BFT is derived from work in countries such as the United States with extensive valuation literatures. Other nations, however, have less extensive collections of nonmarket valuation studies. In such cases, options for within-country benefit transfer are limited, causing some to explore options for international benefit transfer (Ready et al. 2004, Ready and Navrud 2006). Options include standard site-tosite function transfers adjusted for exchange rates and other cross-border differences (Brouwer and Bateman 2005, Kristófersson and Navrud 2007, Ready et al. 2004, Ready and Navrud 2006), as well MRMs estimated with multiple country data (Brander, van Beukering, and Cesar 2007, Brouwer et al. 1999, Lindhjem and Navrud 2008, Santos 1998). Multinational MRMs may be particularly tempting in countries for which a relatively small number of primary studies are available, both reducing the likelihood of a good site match for site-to-site BFT and the number of in-country studies suitable for MA. In such cases, multinational metadata might combine incountry studies with out-of-country information to obtain sufficient degrees of freedom for MRMs. A potential benefit of MA in such cases is the ability to identify systematic value surface differences between nations, thereby promoting more informedWTP adjustments. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.