Does development assistance for health buy better results in maternal health in Tanzania? evidence from Autoregressive Distributed Lag (ARDL) model

Abstract

This paper establishes whether development assistance for health buy better results in maternal health in Tanzania using annual time series data for the period between 1995 and 2014 based on Autoregressive Distributed Lags (ARDL) and Error Correction Model. A long run cointegration relationship exists between GDP per capita, maternal mortality, unemployment and development assistance for health over examined period of time. The results show that in the long run Development Assistance for Health (DAH) and Economic growth (measured by real GDP per capita) was significant in reducing maternal mortality in Tanzania. In the short run, unemployment was statistically significant on increasing maternal mortality in Tanzania between 1995 and 2014. Furthermore, the short run results show that both DAH and real GDP per capita reduces maternal mortality between 1995 and 2014. The results imply that Development Assistance for Health (DAH) channeled to the health sector is an important component in improvements of maternal health in Tanzania. The findings are robust to sensitivity analyses and estimation methods.

Keywords: Maternal Mortality, DAH, Unemployment, Autoregressive Distributed Lags, ARDL