If you want to know which assumptions they made, why not just read the studies and see for yourself? To be sure, predicting the future is super difficult. If we

If you want to know which assumptions they made, why not just read the studies and see for yourself? To be sure, predicting the future is super difficult. If we have a global thermonuclear war in 2020, we will probably not need any tons of lithium at all, what with everyone being dead. Same thing if we get hit by a really large asteroid, etc. Also if someone ever develops man-portable fusion reactors we might not need batteries at all. :) Your point about the self fulfilling aspect of basing policy on research is well taken, though in this case the researchers used multiple scenarios for policy development, including one where lithium demand is heavily stimulated by eg subsidies on EVs and the like. That is the upper part of the range of 12-20 million tons of demand, which is still well below the lower bound of estimated reserves.