The effect of the quality of institutions on the relationship between public health expenditure and the infant mortality rate: evidence from the PVAR (X) and PSTR approaches

Infant mortality remains an important indicator of a country's public health insofar as it goes hand in hand with socio-economic conditions and public health practices, the quality and access to medical care, maternal health and many others. The main purpose of this study is to test the effect of public health expenditure (% GDP) on the infant mortality rate, taking into account the role that institutional quality can play. We use two approaches which are the autoregressive vector panel model with exogenous variables (PVAR (X)) and the Panel Smooth threshold regression (PSTR) model on annual data covering the period 2002-2016 and covering 37 African countries. Subsahdarian. Our main results through the PVAR (X) reveal that in the absence of institutional variables, public health expenditure has a negative and significant effect on the infant mortality rate, whereas in the presence of the various institutional variables, this effect is still negative but is no longer significant. Our results show that the presence of institutions reduces the weight of public health expenditure in explaining the infant mortality rate by half. In addition, our results show through the PSTR that there is a certain level of institutional qualities that these countries must achieve for public health expenditure (% GDP) to positively affect infant mortality rates. These threshold levels oscillate for all institutional variables around 7%.