I've seen quite a few instances of folks going to the trouble of answering a question without up-voting the question itself. Surprisingly, the answer often continues to receive further up-votes, and the question still receives no up-votes.
I suppose I am a little confused about this. If we expect the OP to show gratitude by up-voting and (hopefully) accepting our answer, then shouldn't we at least up-vote the question in an act of mutual respect?
While I can see the merit of Alan's answer, I'd like to present the opposite view:
The upvote button's hover text says: "This question shows research effort; it is useful and clear [...]". There are questions that are answerable, often because missing information can be guessed or inferred, but that don't show any effort of the OP to find the answer themselves, e.g. How to use font in code format?, or don't provide a MWE even though it would be possible and helpful. While egreg's answer to that question is worth an upvote because it explains the matter to an appropriate degree of detail, is correct and useful, the question itself doesn't deserve an upvote imho.
There are issues with this point of view as well: Is it sufficient just to write "Googling didn't give me the answer"? And what about provocatively basal questions, to which the asker may even know a (partial) answer? You could well argue that my own question bibtex vs. biber and biblatex vs. natbib doesn't show any research effort, nonetheless it has proven quite popular. Probably many people upvoted it because of the great answer it triggered. Does this make it a good question? Looks like it. Then again some questions seem to pile up upvotes just because they're popular and enjoyable -- How do I make my document look like it was written by a Cthulhu-worshipping madman? ... :) (To be fair, the question is quite specific in its description, but then again, many of the things it lists are well-Googleable.)
Coming back to Alan's statement: If a question is worth your while to answer, then it's worth an upvote as well. This would be true if people only answered question because they think "Oh that poor soul doesn't know any further, I'll help them". I think, however, this is only one motivation among others like "Can I do this?", "Let's provide a really comprehensive answer to this simple task", "I want rep", "I want to propagate XYZTeX by showing that this can be done using it", or simply "I enjoy *TeXing, I don't even care what the task may be". So the incentive to answer a question isn't directly related to the quality of the question, which is the gauge of upvoting it, however.
Institutionalized mutual upvoting, as cmhughes suggests in the question, would be a dull tradition of politeness, threatening to make upvoting as a tool of quality assurance quite blunt. In turn, I don't upvote every single answer a question of mine gets. I upvote most, but not those that don't help me solve my problem or at least make me understand it better, e.g. I didn't upvote https://tex.stackexchange.com/a/18749/4012, even though Tobi put some honest effort in it -- it did not get me any further with my problem.
I think the general answer to this question is Yes: if a question is worth your while to answer, then it's worth an upvote as well.
