In studies of the effects of various factors on development, in order to control for the potential endogeneity of these factors (i.e., the possibility that growth and development are affecting the supposedly "causal" variable, rather than the other way around), instrumental variables are often used. Consider the case of settler mortality as an instrumental variable in analyzing the effect of institutions on growth. If settler mortality is to be a VALID instrumental variable, which of the following MUST hold (select all that apply)? Group of answer choices Settler mortality must be correlated with long-term growth. Settler mortality in the colonial period must logically influence variation in current institutional quality, and not the reverse. Settler mortality should affect growth and development directly in the short term but not in the long term. Settler mortality must vary across countries less than institutional quality. Settler mortality should only weakly influence growth through other channels. Settler mortality must not affect growth and development directly, but only through institutions