The idea is that the price gets set by the publisher (and, presumably, violators are somehow punished, altho I have no idea what their plan is for that -- and they'd better have one, because I know a thing or two about retail history and what tends to happen to cartels), who collects 70% of it for themselves and everyone else and the retailer gets 30%. Unclear who pays for moving physical product around between the publisher and the retailer, but let's just assume that doesn't matter.
Macmillan has been taking pains to say they would set the ebook price somewhere between the hardcover list currently and the 9.99 they are so unhappy about -- initially. They would have a dynamic pricing model (think about the way airline seats get priced) that would reduce that under some circumstances, presumably based in part on what the thing is available for in physical formats. It is completely opaque to me whether this agency model thing is limited to ebooks, or if Macmillan really does plan to put a stop to 40% bestseller discounting. Macmillan (and other agency model supports) emphasize that in many cases, this would mean the ebook would become cheaper than any physical format. Eventually, or under some circumstances. Given Macmillan's behavior with backlist titles, and how much Barnes & Noble charges for non-bestselling hardcover titles not yet out in paper, I wouldn't hold my breath on anything.
I'll imagine I'm a book purchaser in this universe. I've got some choices -- just like I have choice of airline when I fly, including the choice to not fly at all. In this universe, I might hear about a new book by someone like Gawande. I might enjoy it -- I might not. I liked one of his books and I didn't care for the other; I find the man personally appealing when interviewed and the topic is part of a major reading interest. I know from past experience that if I go to buy it online, and see that it is $25 new, I'm not going to buy -- I wouldn't pay that in person, either (even to support an independent bookstore -- sorry Gawande!). If I see it is not available as an e-book, I'm not going to buy it (that just happened), altho I'll at least toy with the idea of buying it for $15 and change and immediately donating it after reading it once -- but after I check with the librarian to make sure she wants it for the collection, which means I may forget, or she might have already ordered it or someone else may have already donated it. At $13 and change, I'd probably buy it, but I'd note the price in the review (that has happened with Jayne Ann Krentz novels). And at $9.99, I'd buy it, read it, review it, and forget about the whole thing in painless short order.
Let's say I can't buy it online right away in that $10-$13 range. If it drops AND I happen to think of it, I might buy it later. But by that time, there are going to be a lot more customer reviews on Amazon. If it turns out like _Better_, I'm never going to buy it at any price, and will probably skip checking it out at the library. There's a risk here, also a risk that I may never think of Gawande again until I see it at the library, and then no one makes any money off me.
Let's say it turns out that none of the big 6 are putting out $9.99 titles when I think to buy them. I don't wait for paperback anymore, but I might pick up a used copy if it showed up in a timely fashion and, with the $4 shipping, was still $10 or less. I've done a lot of that, too; those books get donated or TitleTrader'd when I'm done with them. Publisher gets nothing out of those, either.
Would I have trouble finding stuff to read? That's my big question. Half my purchases are indie/academic anyway -- and I'm completely accustomed to spending sometimes way more than $9.99 for those as ebooks on the rare occasions they are ebooks, and putting up with physical formats that are often over $25 as well (for a paperback, even). Harlequin has its own pricing strategy, and they're plenty cheap enough for me -- I don't need to give them up, but if I did, I'd just go figure out how to navigate Ellora's Cave and other online romantica publishers anyway. I may do this, regardless; there's reason to suspect I'd be happier if I did this.
I think Amazon will do much better with a slightly higher price under the agency model -- but I also think I'll buy fewer big 6 published books. I'll buy my series (and I'll occasionally add new series when something catches my eye), but big 6 books often have a disposable feel to them that the indies and academics don't have. I don't think I'm the only one who feels this way, and altho I'm prepared to believe it was not always this way, I think the trend is towards greater disposability.
The next question is, are there cheaper options out there that can be found as reliably as big 6 choices (which, after all, have a lot of TV and radio outlets doing interviews with authors to plug them) that would be as satisfying for what they are? I think the answer to that right now is no. You can find free stuff and really cheap stuff, but the big 6 do a decent job on branding. This is really what they offer. If you want what you want, and they are offering it for sale, as long as the price isn't too annoying, you're going to buy. Me and series books. I'll be shelling out for Sookie and Rachel Morgan and Harry Dresden and Kris Longknife and etc. for the foreseeable future.
And the last question is, are there cheaper options out there that are worth wading through to find something you don't mind reading, even if you're not sure quite when you're going to find the nugget that justifies the dross? The answer to that is emphatically YES. Over time, Amazon's Vine program, or a bunch of bloggers, or any number of other solutions could speed up that find-the-good-stuff process. That's what the big 6 really have to fear. Conversely, that's the big opportunity for the young-and-hungry.
Price is only part of it, but in that middle ground, it does matter. Push the price too high, and you move out of the middle. And if the big 6 don't supply stuff that people want, the price can't go low enough to attract people.