The growth has to stop very soon - exponential functions can go on only so long. And when it stops the volatility will not stop - this is a prediction of course - but I am pretty sure that it will be this way.
But why would it stop sooner than a $200 billion market capitalization? Or $700 billion for that matter?
When Facebook was raising money at $10 billion valuations would it be wise to predict that it had to "stop very soon"?
I don't disagree that Bitcoin will at some point stop revaluing upwards by orders of magnitude, and I don't pretend to know what the market cap will be in a few years time. But I don't think it would be all that surprising if Bitcoin were to be more valuable than say, What's App. And for that to happen we're going to likely see another order of magnitude adjustment.
But that's my point. Arbritrarily declarations that exponential growth must stop soon are not meaningful in the positive or negative.
There are reasonable methods, however, to estimate likely outcomes of successful growth. I think its reasonable to predict that Bitcoin is likely to reach valuations of large publicly traded corporations. There are those that estimate it reaching valuations similar to gold, or the GDPs of industrialized nations.
But saying that the growth must stop because growth must stop isn't really saying much at all.