I feel like a lot of people go into a PhD program for mediocre reasons. Many of them would probably be better off doing a 2 year Masters program to learn 2 important things:
1) How to do research and run experiments (where you don't already know the answer).
2) How much they actually don't know about a very deep field (and how to study independently).
The extra 4 years of a PhD, as I learned, were to make me an expert in a piece of the field with enough experience writing papers, meeting with colleagues, and designing my own research (not just next-step experiments), and writing funding proposals, that I could teach in the field. Like most others I chose not to become a professor... it's a very competitive and odd position.
Part of the issue is the pressures that have been put on PhD programs. The fight for funding, papers, lab space politics, and relevance seem to have become all consuming as the number of students have grown and positions have remained fixed (filled and tenured). The days when you could follow a reasonable path into academia and Professorship without a lengthy post-doc are gone in most fields at most Universities. Professors have become unhappy managers of giant groups cranking out complex papers based on a huge apparatus (equipment, code, or lab) in order to get tenure and keep their position. CS and Math have a real advantage here as the tools for the job don't usually require as much yearly funding.
Perhaps all this is fine for Particle Physics, but it feels like more and more of the positions at research universities are run this way... and it's not conducive to how many people want to earn their Doctorate... which is often just a badge of honor in the Business world anyway... so maybe we shouldn't try to push so many people through an academic mold so that they can have a different set of letters next to their name.