Over at OrgTheory.net, he looks at the evidence -- a couple of bullet points each -- for and against prevalence of meritocracy in academia really is (and the post has attracted some fantastic comments). Here's the evidence against:
There is evidence of un-meritocratic components of the academic labor market:
In many studies, there is a correlation of labor market outcomes and gender, even when controlling for # of articles and other relevant performance measures. It may be the case that there is outright prejudice. It may also be the case that gender is correlated with other behaviors that are judged differently by the labor market. Erin Leahey has a series of papers, for example, arguing that gender is correlated with specialization in research, which is correlated with labor market outcomes. Meritocratic? It’s up for debate.
Studies like Burris (2004) that show that elite programs dominate the market and people are often judged by status of the PhD program, even when there is evidence of publication.