Is growth ultimately fully endogenous or semi-endogenous? A quarter-century of theoretical and empirical growth economics has still kept both possibilities open. Consequently, I assume that R&D-driven growth is a general combination of both semi-endogenous and fully endogenous mechanisms. I here prove that if the semi-endogenous growth component is essential to the actual growth mechanism, the long-run growth rate will follow the semi-endogenous growth predictions. On the other hand, if the semi-endogenous is not essential, the fully endogenous growth mechanism may dictate the long run if the world population does not grow too fast. This result holds regardless of whether fully endogenous growth is essential. I also prove that if no other (third) growth mechanism exists, it suffices to prove that less research always leads to fewer innovations to ascertain semi-endogenous growth essentiality.