Varietal seed technology and household income of maize farmers: An application of the doubly robust model

Abstract

The study uses a sample of 962 maize-farming households to examine the effects of a varietal maize seed technology on annual household incomes of rural poor farming households in the Upper West region of Ghana. An approach that combines the regression adjustment method with weighting to estimate the treatment parameters was employed for the study. The approach is a generalization of the doubly robust estimation methods for different treatment parameters, which are defined in a multivalued treatment effect estimation framework. Doubly robust methods combine weighting and regression methods to produce consistent estimates of the treatment parameters even if one of the models is misspecified. Average gains in annual household income are estimated for the entire population, as well as conditional on having a specific varietal seed technology (i.e., pannar and/or Obatanpa). Results from the study show that the pannar varietal maize seed technology significantly increased the annual household incomes of the selected maize-farming households. Thus, the study recommends that safety net policies should generally pay more attention to the factors that allow increased maize production technology adoption by maize-farming households, since they could then build on existing pattern of increasing maize productivity whilst alleviating rural poverty.