A friend just emailed me on this thread that's going on at Slashdot on whether or not it's advisable to get a PhD in Computer Engineering. Really, this could be applied to a PhD in any number of fields, and as one who is dead set on doing this, it's a great thread to read because there are many PhDs writing in, and many hiring managers writing about hiring said people, so the conversation is pretty lively and insightful!

In my opinion, and the only valid reason for pursuing a PhD is because you really want to, and have a passion (yes, passion!) for the subject matter. Forget career - that should have little to do with your decision. Doing a PhD is hard work, and you will almost certainly go through times when you wish you'd never started and wonder if you should just cut your losses. On the other hand, it can be immensely rewarding, and will teach you a whole new way of thinking.

As for jobs afterwards, outside academia at least, it's a crapshoot. Some companies value them, others don't. So that shouldn't really affect your decision. We have a PhD on staff and he's an extremely valuable team member because we need someone with his research background and wealth of experience to lead that area of the company. I don't think, however, that we could fully utilize PhDs in many other parts of the company.

Give the Slasdot thread a read if you're at all interested in the PhD or further education generally.