Online access to papers + university libraries - info wanted

Now that we live in the Information Age, where I am reliably told that Information Wants to be Free, I'm confused by a trend that is coming in terms of how university researchers access electronic versions of journals.  (For those of you under 30, there was once a time when journal articles were published in an arcane format that predates pdf called "paper".)   The electronic availability of journals, including historical archives, has largely been an enormous boon to scientific progress.  It is far easier and faster now than ever before to do proper literature research when writing a paper or a proposal.  If I'm using google scholar or Web of Knowledge or Scopus or any other reference crawling aid, I can now find and (and if my institution subscribes or the content is available free) download copies of relevant references very quickly and efficiently.  If anything, the technology to provide this content is continually becoming cheaper and faster, since providing print content has far lower bandwidth requirements than the streaming video demands that are really driving innovation.

That is why I am concerned and confused by a trend popping up in the perpetually-financially-stressed university libraries around the country (and the world, presumably).  We all know that commercial publishers have been cranking up prices and applying annoying/evil tactics like bundling one high impact title with a dozen expensive, low-impact journals in forced package deals.  (Wiley, Elsevier, Taylor and Francis, that's you.)  Now, though, there is this idea being pushed that it would somehow be cheaper for university libraries to actually drop their subscriptions (!!) and instead use Get It Now, a product of the Copyright Clearance Center (those people you have to contact if you want permission to use a figure in a review article).  The problem is, Get It Now is misnamed; really it's Get It In Seven Minutes.  Needless to say, if you are trying to trace references and write a paper or proposal, having to wait seven minutes for every article you want to examine (which could easily number in the dozens while proposal writing) would be a major mess.   

Given that the publishers have the capability to provide content essentially instantly, and that the infrastructure to support that capability is steadily getting cheaper, and that the publishers could quite readily track download statistics (and could charge per download if they really wanted to), I don't understand how Get It Now is a positive step.  Surely if per-article billing was an economically viable approach, the publishers would do it themselves, right?  The publishes are going to recoup their costs somehow, passing them along to CCC, and CCC will pass those along to the universities, so it's hard for me to see how interposing a middleman like CCC can really do anything except slow down researchers and make money for CCC.  This idea seems to go directly against the trend of open access, public archives, etc. 

Do any of my readers work at institutions that use this service?   How does it work for you?  Is it as annoying as it sounds?  Does it actually enable your university to save money (that is, provide more or better content for the same actual cost) relative to the old approach?  A major challenge faced by universities in budgeting is that libraries don't sound as exciting as new buildings or major initiatives, and yet libraries and their services are essential to the scholarly mission of the institution.