The latest version of the Nature license has generated another round of discussion about the pros and cons of confidentiality clauses (see Section 8.5 of the Terms for the offending passage). The topic pops up from time to time on liblicense-l, and the positions of the disputants are pretty well staked out. What caught my attention in particular this time was Sally Morris's comment that "making the price actually negotiated would be most unfair to the vendor, wouldn't it?"
I'm generally sceptical when people start talking about what's fair or not. Librarians do this all the time, complaining that it is unfair for Elsevier (for example) to charge the prices that they do. I've never been able to figure out how "fairness" plays into this. Mostly it seems to me that when librarians complain about "unfair pricing" they mean "prices that I don't want to pay." I'm sympathetic to the frustration of the librarians, but I don't see what is unfair about the pricing.
I feel similarly about Sally's comment -- I can understand why publishers/vendors think it may be to their advantage to keep the terms of negotiated agreements confidential. They would like all of their customers to believe that they're getting the best possible deal, and if I find out that the university down the road appears to have gotten better pricing, that's going to make me push harder on my end (or make me resentful if I've already sealed my deal). So it could be that more transparency about pricing would make negotiations more difficult for publishers in some circumstances. But why is that unfair?
In the liblicense discussions, several issues seem to get muddled together -- differential pricing, negotiated pricing, and confidentiality. Some of the proponents of confidentiality argue that without it, differential and negotiated pricing would not be possible (to the detriment of the publisher), and yet that's clearly not the case. Many publishers use tiered pricing based on fte or type of institution and publish those prices.
It may be more of an issue with negotiated pricing, but again, I don't see this as a matter of "fairness". It would require publishers to be more consistent, perhaps, with the elements that they use in determining a negotiated price. It could make them less willing, on occasion, to drop the price as far as they might for a particular customer because they're leery of setting a precedent. It might require them to spend more time explaining why one customer got a deal that they're not willing to give you. It could lead some publishers to give up negotiated pricing altogether and rely on various differential formulas.
I don't expect publishers who rely on negotiated pricing to bend much on confidentiality as long as they believe it is in their best interest. But it's a business decision. It's got nothing to do with fairness.