Nexus between household composition and expenditure in Manyara region: A quantile regression analysis

The purpose of this study is to examine the relationship between household (HH) composition and HH expenditure using quantile regression. Secondary data for the analysis were obtained from the 2017–2018 Household Budget Survey, utilizing a sample of 296 HHs. The analysis reveals significant heterogeneity in expenditure determinants across the distribution of HHs.Theanalysis further revealed that secondary education increases expenditure by 0.53–0.77 log points (p < 0.05) across quantiles, while urban residence shows a consistent positive effect (p < 0.01 in all quantiles). There is a widening expenditure gap between urban and rural HHs across quantiles, with urban HHs spending 25–38% more. The findings also revealed that there is a positive effect of HH size and that expenditure diminishes from 0.13 (20th quantile) to negligible at higher expenditure levels. The findings highlight the need for differentiated policies, with education inter ventions showing greater returns for middle-income HHs (pseudo R2 = 0.14–0.16) and social protection being more crucial for larger, low-expenditure HHs. The analysis also shows that medium and large HH sizes are linked with higher expenditures, especially among middle- and upper-income groups, suggesting growing consumption needs with family expansion. Formal employment emerges as a crucial driver of spending capacity from the 40th percentile upward. The f indings underscore the importance of tailored, income-sensitive policy interventions that address both poverty reduction and sustainable HH development across the expenditure spectrum.