You see this in tons of domains, not just chess. My personal hypothesis is that it has become so much easier to study and improve yourself based on the past insights of others, and this network creates ever accelerating standards.

For example, I recently (as a middle aged adult) took up ice skating. It probably won't be too long before I'll be at "early 1900s Olympic champion" level. Reason being that those early Olympic champions basically had to figure out how to do everything by trial and error. These days I can watch countless videos on YouTube explaining exactly how to practice certain elements.

So yes, the inequality factor is certainly there, but with even homeless people getting access to a base level smartphone these days I don't necessarily think it's insurmountable.

> I wonder if this cohort effect is exacerbating inequality.

It seems to me that increased capability would be more strongly correlated with productivity than with inequality. Certainly there's some correlation between productivity and inequality, but I think it's more complicated than a simple relationship. For example, the richest people are all past what would be considered their peak capability.