7793 - Laserfiche WebLink

Home Browse Search

INSTREAM FLOW 95 There is an ongoing debate about the validity of existence value estimates. 24 Bishop and Welsh 18 discuss the debate in detail, and assert that part of this contro- versy revolves around the fact that off-site benefits can only be measured by the con- tingent value method. Cummings et al.t3discuss the early tests of the contingent value method's accuracy in predicting real market outcomes. But, Mitchell and Carson25 point out that tests of the ability of contingent value method studies to pre- dict public referenda outcomes or similar nonmarket allocation processes are more relevant. Duffield and Patterson24 recently field tested CVM instream flow existence benefits estimates by soliciting actual contributions to a trust fund. The results were inconclusive. Carson et al.,26 however, provided strong empirical evidence of the operational significance of existence benefits estimates. Carson et al.26 used data from a contingent value method study to estimate the outcome of a referendum that dealt with a market versus nonmarket surface water resource conflict. Their study shows the remarkable accuracy of contingent value method predictions of the ballot results. Moreover, the prediction by Carson et al.26 of the outcome of an actual 1984 California state-wide referendum vote was improved by including existence benefits in their estimates of nonmarket aggregate benefits conferred. However, policy relevant testing of the accuracy of CVM esti- mates along the lines explored by Carson et al.26 is rare. Kahneman and Knetsch27 recently explored a type of CVM bias they call the "embedding phenomenon" in a telephone survey. Kahneman and Knetsch27 show that the estimated aggregate will- ingness-to-pay for the provision of more highly trained emergency and disaster res- cue personnel and equipment varies with the questionnaire format. Kahneman and Knetsch27 asked respondents their willingness-to-pay for improved emergency res- cue services after being queried with regard to their willingness-to-pay for improved environmental amenities, and broad based enhanced disaster preparedness. They also query respondents with regard to their willingness-to-pay for improved emer- gency rescue services in a survey in which there is just one amenity. The mean will- ingness-to-pay for improved emergency rescue services increases ten-fold if the other amenities are omitted from the telephone survey instrument,21 Bishop and Welsh 18 point out that the actual survey questions used by Kahneman and Knetsch27 have basic errors in survey design. The Kahneman-Knetsch questions neither tacitly nor explicitly designate a baseline condition, and the hypothetical improvements are not quantified. These survey design errors are related to various types of estimation bias listed in Cummings et al.13 and Mitchell and Carson.25 Thus Kahneman and Knetsch27 have also demonstrated the fact that CVM survey design flaws can induce "embedding bias." Bishop and Welsht8 correctly note that Kahneman and Knetsch27 do not, however, demonstrate lack of accuracy in non- market amenity value estimates obtained with properly designed CVM survey instruments. THE DOLLAR VALUES AND THEIR POLICY USES The preceding discussion provides useful background information for a review of the range spanned by nonmarket instream flow benefits estimates. Much of the Temarkable variation in estimates of benefits conferred by instream flows stems from an equally remarkable variation in the range of attributes addressed by these studies. 2 The more inclusive the set of amenities provided by the instream flows, the greater the aggregate acre-foot benefits conferred per linear mile of instream flow