This paper tries to examine and analyze households' willingness to pay for improved waterrnsupply in Comboldw town, Soutlt Wallo Etltiopia. It also explores determinants ofrnwillingness to pay and examine the applicability of cost recovery strategy.rnA contingent valuation method based on a single bounded dichotomous choice formatrnfollowed by open ended question was employed. A total of 247 saml)le Iwuseholds ran(lomlyrnselected from all kebeles of Combolcha town were covered during the survey.rnResponses to tlte valuation questions revealed thot 98.6% of the respondents expressed theirrnwillingness to have private connection to the new improved water and are willing to pay forrnthe new service with mean willingness to pay of 15.70 cents. This implies that, on average,rnthe respondents are WiLling to pay birr 7.85 for one m' of improved water sewice which isrngreater than even twice of tlte town's highest consumption block.rnResults of the probit model also show tlwt income, initial bid, and attitude of tlte respondentrnfor tlte responsibility of town water provision Iwve positive effects and age of tit< head of thernhousehold Iws a negative effect on the Iwuselwlcls' choice for improved water services. Thisrnimplies that Iwuselwlds with high incont< are more likely to choose improved services.rnHowever, there is a tendency for aged Iwuseholds not to choose the improved water sewice.rnMean willingness to pay for the probit model is 26.10 cents per 20 liters water, whichrnimplies tlwt rIte mean willingness to pay for the new improved water supply ranges betweenrn15.70 to 26.10 cents.