If the described effects do exist, I think it would be a much more surprising finding if they didn’t disproportionately affect cohorts that tend to have weaker

If the described effects do exist, I think it would be a much more surprising finding if they didn’t disproportionately affect cohorts that tend to have weaker labor-rate bargaining positions than others. Do you see problems with the core argument of the paper?