>But making it profitable to be the one with stock on hand in an emergency would encourage holding bigger buffers during the good times. This strategy only wor

>But making it profitable to be the one with stock on hand in an emergency would encourage holding bigger buffers during the good times. This strategy only works if companies can accurately predict the likelihood of future supply/demand mismatches. This isn't possible for most industries, and it's not sane to hold out for hundred year events. There's also something like a reverse prisoner's dilemma: if all companies in an industry keep a buffer, they're all worse off (no price spikes = less profit, years of unused capacity = less profit). Only small players can really win from spare capacity, and only if they have a huge excess, and only if their timing is right. They lose if they guess wrong and go broke before the price spike, which is made even more likely by the inefficiency of running with tons of spare capacity.