Predicting impact of price increases on demand for reproductive health services: Can it be done well?

Abstract

To assess criterion validity of a survey that uses contingent valuation to elicit estimates of client willingness-to-pay (WTP) higher prices for family planning and reproductive health services in three developing countries. Criterion validity was assessed at the individual client level and at the aggregate service level. Individual-level validity was assessed using a longitudinal approach in which we compared what women said they would do with their actual utilization behavior following a price increase. Aggregate-level validity was assessed using predictions derived from cross-sectional surveys and comparing these with actual utilization data. Phi coefficients and correlation statistics were calculated for individual and aggregate-level analyses, respectively. None of the three individual-level cohorts exhibited statistically significant relationships between predicted and actual WTP. Approximately 70% of clients returned for follow-up care after the price increase, regardless of their responses on the WTP survey. For the aggregate analysis the correlation coefficient between predicted and actual percentage change in demand was not significant. Many clinics experienced higher demand after prices increased, suggestive of shifting demand curves. A validated technique for predicting utilization subsequent to a price increase would be highly useful for program managers. Our individual and aggregate-level results cast doubt on the usefulness of WTP surveys for this purpose.