What's the big deal with the "Big Deal"?

Right now, in the Canadian academic library world (and beyond to some degree as well), there's an ongoing discussion regarding the value of the Big Deal.  What's the Big Deal?  It's basically bulk purchasing of electronic journals from a specific vendor which saves time, money (per title at least), and workload processing these resources.  The concern is whether this deal is actually such a deal in the end and when does it stop being so.

My opinion about this is that the problem rests not with the vendor or the deal but with both budget stagnation in academic libraries and with our failure to truly evaluate value on resources like ejournal packages.  The former issue seems primarily a problem with the thought on the part of university administration, government agencies and the public that, because "everything is online now", why should we be paying more for information?  We should be paying less, right?  That's not true obviously.  The latter issue comes from the complexity of assigning value to these resources individually by journal title say, or even moreso, as a huge package of thousands or tens of thousands of titles.  There are too many variables (list price, usage, differing usage style by subject discipline, non-transparency of past and future pricing, variable industry pricing, institution size, etc.) to be able to put it all together into one figure that can then be compared to the value figure for other resources/packages or even to be used to signal when (and why) to cancel, acquire, renew or resubscribe later on.