Evaluacion De Impacto

MP A R
Munich Personal RePEc Archive

Introduction to contingent valuation using Stata
Lopez-Feldman, Alejandro Centro de Investigacion y Docencia Economicas (Cide) ´ ´

2012

Online at http://mpra.ub.uni-muenchen.de/41018/ MPRA Paper No. 41018, posted 04. September 2012 / 18:44

Introduction to Contingent Valuation using Stata⇤
Alejandro L´pez-Feldman o Centro de Investigaci´n yDocencia Econ´micas, CIDE o o alejandro.lopez@cide.edu

1

Introduction

Cost benefit analysis is a key input for the ex-ante evaluation of public projects and policies. An ideal cost-benefit analysis incorporates all the social costs and benefits of a project for all members of a society. Boardman et al. (2006). One of the big challenges to achieve this is the need to place monetary values onnon-marketed goods and services. The objective of this chapter is to provide the reader with the basic tools to obtain estimates of these values using the contingent valuation method. A basic introduction to the method is provided but the focus of the chapter is the use of Stata to do basic empirical analysis. The use of the doubleb command is illustrated with an example.

2

Contingent ValuationValuation methods for non-marketed goods can be divided in direct and indirect methods. With indirect methods the estimations are based on the observed behaviour of individuals in the market of a good or service related to the one of interest. An example of this is the estimation of the use value of the services provided by a national park using information on the costs that travellers mustincur in order to get to that place. Among indirect methods we have the travel cost method, hedonic pricing and averting behaviour and defensive expenditures. Direct methods, on the other hand, try to elicit information about the value of the non-marketed good or service directly
Taken from the book ”Aplicaciones en Econom´ y Ciencias Sociales con Stata” (forthıa coming) and translated withpermission of the publisher, Stata Press (College Station, TX: StataCorp).
⇤

1

from the individual. Among these methods we have contingent valuation and choice modelling.1 Contingent valuation implies asking to a sample of the population about their willingness to pay (WTP) for the provision of a given good or service.2 The name of the method comes from the fact that the elicited values arecontingent to the hypothetical scenario that is presented to those being interviewed Portney (1994). This is a very flexible method since one can obtain estimations for public policies or projects that have not been implemented. Furthermore, with this method it is possible to obtains estimates of non-use values.3 Originally contingent valuation was basically an intellectual exercise with limitedpractical relevance. That changed when, by request of the State of Alaska,Carson et al. (1992) did a contingent valuation study to get an estimate of the non-use value loss associated with the Exxon Valdez oil spill. Not long before that, the US government had approved a federal act in which contingent valuation was accepted as a valid method to measure the loss associated with environmental disasters.After the study of Carson et al. (1992) Exxon and other companies sponsored a series of research e↵orts looking to discredit the method in order to avoid its future use. As a response to this the National Oceanic and Atmospheric Administration (NOAA) commissioned a group of experts to do a report in the validity of contingent valuation. The panel concluded that contingent valuation was capable ofgenerating estimates reliable enough to be used in court and made some recommendations and guidelines for the application of contingent valuation studies (Arrow et al., 1993).

2.1

Contingent valuation with dichotomous choice questions

The focus of the rest of this chapter is on how to econometrically analyse the data obtained from a contingent valuation survey. Nonetheless, it is
1 For...