Valuing Complex Goods: or, Can You Get Anything Out of Experts Other Than a Decision?

• DOI: https://doi.org/10.1016/S0193-5895(07)23013-1
• Published date: 16 October 2007
• Date: 16 October 2007
• Pages: 301-335
• Author: Ben Groom,Andreas Kontoleon,Tim Swanson

VALUING COMPLEX GOODS:

OR, CAN YOU GET ANYTHING

OUT OF EXPERTS OTHER THAN

A DECISION?

Ben Groom, Andreas Kontoleon and Tim Swanson

ABSTRACT

An experiment is undertaken to assess how the level of information

provided to survey groups impacts upon the decisions they make. In this

experiment, a group of experts is surveyed first to determine both the

forms and levels of information important to them regarding an obscure

environmental resource (remote mountain lakes), as well as their ranking

of particular examples of these resources in accordance with their own

criteria. Then three different groups of respondents are given different

levels of this information to assess how their WTP for the resources

responds to varying levels of this information, and how their rankings of

the different goods alters with the information provided. The study reports

evidence that generally increased levels of information provide significant

quantitative changes in aggregate WTP (the enhancement effect), as well

as a credible impact on their ranking of the various goods. On closer

examination, much of the enhancement effect appears to be attributable

to the changes in ranking, and to changes in the WTP for a single lake at

each level of information. In addition the ranking does not respond in any

consistent or coherent fashion during the experiment until the information

Research in Law and Economics, Volume 23, 301–335

Copyright r2007 by Elsevier Ltd.

All rights of reproduction in any form reserved

ISSN: 0193-5895/doi:10.1016/S0193-5895(07)23013-1

301

provided is complete, including a ranking of subjectively reported

importance by the expert group, and then the survey group converges

upon the expert’s group rankings. In sum, the experiment generates

evidence that is both consistent with the anticipated effects of increased

levels of information but also consistent with the communication of

information-embedded preferences of the expert group. It may not be

possible to communicate expert-provided information to survey groups

without simultaneously communicating their preferences.

1. INTRODUCTION

One of the more interesting trade-offs in environmental policy-making

concerns the use of experts’ information and experts themselves in the

decision-making process. On the one hand, many environmental problems

and resources are complex, and it requires a very substantial amount of

prior knowledge and precise information to be able to understand the

policy-making context. On the other hand, any group of persons which

understands and appreciates such an unusual resource or problem is not

highly representative of the general tax-paying public, and so any decisions

they make regarding that resource are representative of a self-selected and

non-representative sample of that society.

Anyone who has sat on a policy panel assessing an environmental

problem together with a group of natural scientists will appreciate the

nature of the problem. On the one hand, it is difficult to appreciate the

scientific nature of the problem, for example, trace pollutants in high Alpine

environments without a thorough appreciation of the nutrient chains and

ecology of those places. On the other hand, those persons who have

dedicated 20 or 30 years of their lives digging around in the mud on the

bottom of high mountain lakes represent a highly skewed set of preferences

regarding these environments. If you ask them for the policies they would

recommend for the resources in which they specialise, then you are certain

to get policies supported by relatively extreme preferences. The simple fact

that the experts have invested their lives in acquiring the unique information

they possess demonstrates the eccentricity of their preferences in this regard.

Then the problem faced by the policy maker is how to divest them of their

unique knowledge, and present it to a more representative sample of the

citizenry for decision-making purposes, without simultaneously commu-

nicating the unusual preferences that generated it.

BEN GROOM ET AL.302

We examine this issue and others regarding the impacts of complex

information – in the context of a valuation experiment using both an experts

group and a set of more representative citizenry. In this chapter we follow

National Oceanic and Atmospheric Administration (NOAA) Panel

guidelines and undertake two tests of the impacts of information on CV

responses. The tests are explored in a case study that estimates the non-use

values for a highly unfamiliar environmental resource, namely remote

mountain lakes (RMLs). These ecosystems constitute a perfect example of

a complex environmental good with which respondents are unfamiliar

and for which non-use values are most likely the predominant element

of total economic value. In the first test, the effects of varying levels of

information on stated Willingness to Pay (WTP) are examined. This is

a theoretical validity test in that it examines whether information provided

to Contingent Valuation (CV) participants affects their WTP in a manner

consistent with economic theory. Such theoretical validity tests have

been undertaken with varying results by numerous researchers.

1

The

experiment presented here explores the information effects on individual

preferences and WTP for a highly unfamiliar and obscure environmental

resource for which individuals have only very indirect use or non-use value,

namely RMLs.

Another important issue regarding information concerns its impact, if

any, on preferences. Does the additional information act to clarify and make

more coherent the preferences of the public regarding environmental

resources? The hoped-for effect of providing information to the public in a

policy-making exercise should be to aid in clarifying public priorities and

objectives. This would imply that additional information should play a part

in rendering these priorities more apparent. This issue is related to the

matter of the ‘optimal amount of information’ in a Contingent Valuation

Method (CVM) study. The issue of optimality is also a validity test, and is

usually conducted by reference to some external benchmark or baseline

against which to judge the outcome of the CV study (Mitchell & Carson,

1989).

2

We thus follow Boyle, Welsh, Bishop, and Baumgartner (1995) and

interpret the NOAA guidelines as prescribing a form of external validity test

for the reasonableness or optimality of the information provided.

3

This

external validity test is undertaken through a comparison of the CV results

with those obtained from a Delphi survey. This comparison allows us to

assess the credibility and validity of using CV as an input into policy

decisions compared to relying on expert-based appraisal methods. We

consider a situation that is commonly encountered by policy makers: the

selection for conservation purposes of a small number of sites from a larger

Valuing Complex Goods 303