Search

Billions of people globally, living with various degrees of water insecurity, obtain their household and drinking water from diverse sources that can absorb a disproportionate amount of a household's income. In theory, there are income and expenditure thresholds associated with effective mitigation of household water insecurity, but there is little empirical research about these mechanisms and thresholds in low- and middle-income settings. This study used data from 3655 households from 23 water-insecure sites in 20 countries to explore the relationship between cash water expenditures (measured as a Z-score, percent of income, and Z-score of percent of income) and a household water insecurity score, and whether income moderated that relationship. We also assessed whether water expenditures moderated the relationships between water insecurity and both food insecurity and perceived stress. Using tobit mixed effects regression models, we observed a positive association between multiple measures of water expenditures and a household water insecurity score, controlling for demographic characteristics and accounting for clustering within neighborhoods and study sites. The positive relationships between water expenditures and water insecurity persisted even when adjusted for income, while income was independently negatively associated with water insecurity. Water expenditures were also positively associated with food insecurity and perceived stress. These results underscore the complex relationships between water insecurity, food insecurity, and perceived stress and suggest that water infrastructure interventions that increase water costs to households without anti-poverty and income generation interventions will likely exacerbate experiences of household water insecurity, especially for the lowest-income households.