Applying rational expectations to the pilot supply model has one interesting short coming: Pilots actually didn't know the model and therefore their assumptions were not correct.
What I am implying is that there is a great deal of leverage where supply of pilots exceeds demand.
This has generational consequences as companies could push down terms and conditions. They did this for decades. Aided by an adversarial IR posture this was a foregone outcome with ample supply ensuring more candidates than needed.
Moral hazard fell on the pilot. The company no longer bore responsibility for training or endorsements, contracts didn't compensate when a commercial cancellation occurred or sick leave provisions were less than an office worker. Pilots were paid only to fly, sitting around was at their expense.
From General Aviation to Airlines this became the new model.
When pilots adjusted to the reality of the model, it was oftentimes too late (age and financial commitment)...
With real wage growth stagnant and other opportunities available, people do make rational choices: they find ways to be compensated for their time, they do not spend $150k learning to fly. The statistics although time lagged and opaque are bearing this reality: There is inadequate supply.
The great weakness of the model that IR/HR created, aided by pilot representatives who knowing there was adequate supply continued to give things up to 'secure the flying'
This loop has returned and choked off a lot of supply and with a demographic surge in retirements it is going to be interesting to see how it plays out.