I wonder if this amounts to optimization of an easy-to-measure and reasonably-looking but incorrect metric, much like the previous "common sense" wave of "fat makes you fat" that led several generations into a dead end.
You assume above that glycemic spikes are unhealthy which I think was never proven for the general population. Eating too much sugar or eating too much in general is bad, but I'm not aware of evidence that croissants kill you if you eat reasonably.
Another thing to note: one of the potential suspects in the obesity epidemic is HFCS, and fructose doesn't actually cause glycemic spikes. If this turns out to be true, then parallels with the "fat makes you fat" theory become uncannily strong.
I tend to think this could be the case because it matches my personal observations. I moved to France a few years ago and the amount of croissants I and people around me consume is at the stereotypes level. My weight is stable but each time I go to US for a couple of weeks I bring back 3-5 extra kilos. Something is seriously different between the food here and there, and I don't think it's glycemic spikes per se.