I agree with everything Peter Dorman has to say in this post at EconoSpeak. A couple key points:
First, economic theory, taken as a whole, is culpable. The core problem is that each theoretical departure, whether it is a knotty agency problem or a behavioral kink, is inserted into an otherwise pristine general equilibrium framework. The only way you can get an article published in a mainstream economics journal (and therefore reproduce the conditions of your existence as an academic economist) is to present your departure piecemeal, showing that it exerts its effects even in an otherwise pristine universe. According to the standards that rule the profession, a GE model with one twist is theory’s version of the controlled experiment.
Second, the shift toward empiricism is not an unalloyed gain. Yes, much of this work is refreshingly open-minded, allowing the data to lead. An honest tally of the published literature, however, particularly in the top journals, will show that a majority of quantitative articles are concerned to calibrate existing theoretical models.
…economics in both its theoretical and empirical modes is implicated in the current debacle. Making the funding of economists more transparent would help, and attention should be given to the institutional structures that favor conformism within the profession, but economics itself needs to be reformed.
Bingo. And this is why a provost or dean who simply quotes top-twenty journal publishings as a criteria for “serious” research is not taking his/her own job seriously.