This paper focuses on the analysis of the relationship between inflation and economic growth, using a sample of 32 sub-Saharan African (SSA) countries, over the period 2002-2021. Beyond a linear association widely discussed in the literature, we are particularly interested in testing a non-linear relationship, using the democracy level as the moderating variable. Empirically, we use a renovated framework of Gonz & aacute;lez et al. (2005) by using a panel smooth transition regression (PSTR) model. Our strategy validates the non-linearity between inflation and economic growth, with the presence of a single transition function, reflecting two inflation regimes, namely a low-inflation regime in which economic growth is an increasing function of inflation, and a high-inflation regime in which the inverse relationship is observed. On this basis, we establish two main findings: (1) the optimal inflation threshold in SSA is estimated at 9.27%. Such a value stipulates globally that any inflation below this threshold is pro-growth, while inflation above this threshold is anti-growth; (2) political regime determines the estimate of the optimal inflation threshold in SSA. The results show that, unlike democratic regimes, autocratic regimes have a higher optimal inflation threshold. Thus, depending on the level of inflation, these results point to more opportunities and options in central banks' inflation targeting policies, in formalizing economies to reduce the inflationary tax, and in building more resilient and inclusive institutions.