Now for my two cents on the debate in the blogs over what college is for. Is there significant value-added that justifies the cost? Some bloggers (and especially the commenters) have suggested that it is largely signaling your ability and/or your motivation. Others claim that there is revealed value deriving from the large and widening differential in lifetime earnings between high school and college graduates. An interesting discussion and summary can be found here, where Steve Postrel argues that the inflated demand for college education by students and by employers boils down to the fact that college is getting dumbed down, but high school is getting dumbed down by a proportionally greater margin. This implies that the increase in demand (and tuition) is fundamentals-based.
I found the argument about the dual decline in high school and college quality compelling at first, but I always have to remind myself to be skeptical about the perceptions of academics about the quality of the average student "back in my day." For example, Mr. Postrel notes that, " It used to be that everybody could read and understand something like Orwell’s Animal Farm, but the typical college graduates could also understand Milton or Spencer." But I am a bit skeptical. It may have used to be that more graduates survived Milton or Spencer, but I'm not sure that the typical graduate really grasped deep literature or philosophy as deeply as Postrel suggests. Much of his "evidence" is based on perceptions by professors. This is likely to be biased by the fact that professors tend to be right-tailers.
I'm inclined to believe that at most levels depth of understanding has been sacrificed for breadth of coverage, and perhaps we need to rethink where we are on the margin. Judging from economics curricula, I would say that this is the case. More topics are covered in, say, intermediate microeconomics, but with less mathematical rigor and depth than when I was taking it almost 20 years ago. Additionally, grade inflation has made completing the degree (and doing so with a reasonably high GPA) somewhat less informative than it perhaps once was).
Which brings me to "signaling". Let's suppose it's all about signaling. Many have used this to suggest that college doesn't "do" anything that job market experience or military service or whatever can do just as well, which in turn leads some (like Mark Cuban) to conclude that college education is a bubble. Why? If the signal is valuable for both applicants and firms, then why does this not have value. Which brings me back to Milton. Did kids who learned Milton ever use that knowledge (or the literary analysis they applied in their class) on the job? So, as my students frequently wonder "why did I have to learn that stuff?" Because, for employers, it is useful to know that someone wants to learn new things, or that they can learn new things.
Ninety percent of the folks in my intermediate microeconomics class in college will never solve a Lagrangian optimization problem again, and probably have forgotten how. But having been able to learn the necessary calculus to do it is worthwhile, and signalling that ability is not trivial.