protection, in relative terms (when WTP is expressed as a fraction of the current water cost) their
WTP is smaller.
Returning to the original model, the marginal effect for income is positive but statistically
insignificant. This suggests WTP for the basin preservation is nearly constant across income
classes. A positive and significant relationship between income and WTP has been found in
other WTP studies for water service improvements (Whittington et al., 1990; Griffin and Mjelde,
2000; Montes de Oca et al., 2003). However, prior studies have not identified a statistically
significant effect of income on the WTP for water quality improvements (e.g., Johnson and
Baltodano, 2004).
The marginal effects for the six dummy variable parameters corresponding to dummy
variables included in the regression model to control for individual characteristics are measured
relative to an individual without the characteristics. Dummy variable are used to control for (1)
the sex of the respondent; (2) respondents who think the current cost (tariff) paid for water is not
expensive; (3) households that have access to water for less than 24 hours in a day; (4)
households that are not satisfied with current water quality; (5) households that are not satisfied
with current water quantity; and (6) households that are not satisfied with current system
reliability. Relative to an individual who perceives the current water bill as expensive WTP is
$4.70 per month lower than for an individual that views the current bill as reasonable. Male
respondents have a $3.80 higher WTP than females. Individuals who have access to water 24
hours per day have a WTP that is $4 higher than those that receive the service less than 24 hours
per week.
The gender effect on WTP for water service improvements has not been consistent
among prior studies. A study conducted in rural communities in Nicaragua by Johnson and