Methods, Trends and Controversies in Contemporary Benefit Transfer

Abstract

Benefit transfer uses research results from pre-existing primary research to predict welfare estimates for other sites of policy significance for which primary valuation estimates are unavailable. Despite the sizable literature and the ubiquity of benefit transfer in policy analysis, the method remains subject to controversy. There is also a divergence between transfer practices recommended by the scholarly literature and those commonly applied within policy analysis. The size, complexity and relative disorganization of the literature may represent an obstacle to the use of updated methods by practitioners. Recognizing the importance of benefit transfer for policymaking and the breadth of associated scholarly work, this paper reviews and synthesizes the benefit transfer literature. It highlights methods, trends and controversies in contemporary research, identifies issues and challenges facing benefit transfer practitioners and summarizes research contributions. Several areas of future research on benefit transfers naturally emerge.

Disentangling the Pure Time Effect From Site and Preference Heterogeneity Effects in Benefit Transfer: An Empirical Investigation of Transferability

“…So far, ambiguous results have been found (Johnston and Rosenberger 2009;Bateman et al 2011). Some find that BT is unreliable (e.g.…”

“…Some find that BT is unreliable (e.g. Rozan 2004), while others conclude that the reliability of BT depends on the similarity of the study and policy sites as well as the type of BT considered (Brouwer and Bateman 2005;Johnston and Rosenberger 2009;Baskaran et al 2010;Bateman et al 2011). Furthermore, the required level of precision can vary depending on the context that the BT is used.…”

“…2 BT is sometimes the most viable option when quick valuation is required for urgent decision-seeking occasions. There could be insufficient time available to do primary valuation or data could be inaccessible (Johnston and Rosenberger 2009). For example, if a primary study is to be done to assess the impacts of an oil spill while oil is spilling, this would lead to substantial decision delay costs.…”

“…Brouwer and Bateman 2005;Zandersen et al 2007). However, the BT studies focusing on temporal reliability, and especially those testing a transfer across more than 2 years (which will often be very relevant in practical BT for policy purposes) are few in number and several aspects have yet to be investigated thoroughly (Johnston and Rosenberger 2009;. Particularly, temporal BT studies that use the open-ended contingent valuation method (CVM) for transfers over a relatively long time interval (e.g.…”

“…2 The main drawback of using BT relates to the fact that transferring values from a study site to a policy site will inevitably introduce an error associated with the transfer. The sources of transfer error (TE) include any errors or mistakes made in the primary valuation survey at the study site from which values are transferred, differences between the study and policy sites not taken into account, choosing the wrong model or method of transfer, as well as mistakes made during the transfer process (Brouwer 2000;Colombo and Hanley 2008;Johnston and Rosenberger 2009;Baskaran et al 2010). Despite this downside, it would seem reasonable to use BT as long as the survey cost savings are greater than the potential negative consequences of the TEs.…”