The contingent valuation method for estimating the existence value of
natural resources is examined for psychophysical robustness, statistic
al reliability, and economic sensibility. Extensions of standard model
s for willingness-to-pay, and suitable econometric techniques for anal
yzing these models, are developed. The analysis is applied to a series
of experiments on the value of preserving wilderness areas in the wes
tern United States. The results call into question the reliability of
the CV method for estimating existence values.