Perhaps the most striking example of the halo effect that I have noticed is the weight given to very high-impact journals (say Nature). Most scientists I know would love to publish in such journals, because of the visibility and prestige, and because they know it helps their career (for example to get grants). But the same scientists generally also agree that papers in those journals are not necessarily better than those published in good journals in their field, and everyone has examples in mind about obviously flawed papers in their field in such journals. There is in fact little objective reason that those journals may publish better science. The reviewing process is essentially the same as in other journals (2-3 reviewers writing a report), except faster, and most of the selection is done by professional editors (i.e., whose full-time job is to evaluate papers), not by people in the field, on the basis of “significance”. Significance is obviously a highly subjective notion, especially when assessed by non-specialists. In any case, publishing in Nature should not make you a better scientist than publishing in, say, Journal of Neuroscience.
Some might have a different opinion, but it is not my point here. The point is that committee members also often share the same opinion, and yet it is almost impossible for them to resist the appeal of a Nature paper on a CV. Publishing in such journals is so desirable that one cannot help but think, consciously or not, that a scientist who got a paper accepted in such journals must be some sort of genius, forgetting their opinion that the competition in those journals is not so much about scientific quality. There clearly seems to be a halo effect with prestigious journals.
Taken together, all the cognitive biases I have described make a rather gloomy picture of academic hiring committees. To be clear, my intention was certainly not to imply that committee members are stupid or crazy. They are not at all. But they are humans, and humans suffer from well-documented cognitive biases in difficult decision processes. The solution to these issues is therefore not to appoint different humans. It is to change the processes so as reduce these issues. Even though selecting a permanent faculty is a very difficult task, it is not the worst hiring scenarios. Compared to other professions, the work of scientists is public and there might be therefore more useful information available for the hiring committee than in other professions, where only a CV and an interview can be used.
For example, one simple recommendation would be that documents produced by the hiring process (reports etc) be made public after the decision. This would: 1) increase feedback, 2) suppress some irrelevant criteria (quotas etc).
I would be interested in hearing about how academic hiring processes work in other countries.