It's the opposite advice that makes up common wisdom, which is to take risks when you are young, inexperienced and without responsibilities, and then settle into the comfortable long-term job with small raises once you get past that phase. I'm not surprised by this; if technical skills compound at all, it would seem that a developer's value would trend exponential and not logarithmic. That is, so long as they don't weary and stagnate.
But then, there is also value for the inexperienced in a startup, in that you'll learn a ton, and be given a lot of responsibility and autonomy. It might be that common wisdom is only half-broken.
Granted, there are opportunity costs involved. You might be in a position where you have to trade capitalistic striving for a paycheck, as a ribbonfarm article posted here a while back put it. It's as much being aware of those opportunity costs, though, as it is knowing whether or not you are in a position to need to accept them.