if miners can't make ends meet in case of a price drop then they will shut down their miners. but this is not something that happens for all of them. only some miners are affected and as a result only some of them shut down their miners while others continue mining.
and as this happens, difficulty comes down and things go back to normal.
People are obviously confused about mining costs. They think there is some certain cost which is more or less the same for all miners. And it is not the first time when I see this argument emerging. Indeed, we can calculate an average cost but the variance will be such that it won't have any practical importance and significance as a reference point for making any reliable inferences. It is not even about the difficulty adjusting (though the adjustment will inevitably kick in if a lot of miners jump off the ship) as it is about the variance in mining costs as such which, as I suspect, is higher than bitcoin's volatility itself.
As an aside, it is interesting that most people including me and my grandma are deeply confused about almost everything which is not part of their rather narrow field of interest or professional occupation. We all intuitively draw conclusions about some question or another, and it turns out that we are wrong most of the time.
