utilized because 15 percent of the surveyed respondents stated they were not willing to pay
anything (i.e., their WTP was $0) for the basin protection plan, therefore the household WTP
variable is left censored. We hypothesized that household WTP for the basin protection and
maintenance may be affected by household socio-economic characteristics, perceptions about the
water supply service quality, quantity and price, and perceptions regarding the impact that
human activity in the basin watershed is having on water quality.
Because the estimated Tobit model coefficients do not measure marginal effects, the
marginal effect of each estimated coefficient were calculated and are reported in table 2 (see
Greene, 2003 p. 766 for the marginal formulas). The asymptotic covariance matrix of both the
coefficient estimates and the marginal effects was approximated using a non-parametric
bootstrapping procedure as outlined by Wooldridge (2002, p.379). A total of 999 replications
were used to generate the standard errors.
Results
The average age of the surveyed respondents is 40 years. Fifty two percent of the surveyed
individuals were male and 76 percent reported having some college education (individuals with
college degrees or at least some college education) and is consistent with the 2001 National
Census that reports Loja has one of the highest levels of education attainment in Ecuador.
Average household income and household size are $789.80 per month and 4.5 members,
respectively.
Survey respondents pay an average of $19.60 per month for water services. Most
respondents (65 percent) perceive their water bill cost as normal. However, many individuals
believe the water service is poor. Seventy eight percent of respondents stated that the quantity of