There is a group of publishers referred to as the "Big 6". Each of the Big 6 is made up of a lot of different imprints, many of which used to be independent companies. Thus, all of the members of the Big 6 are rollups or conglomerates. Authors who decided not to also be their own publisher signed contracts with people to publish and distribute their work. The contents of those contracts have some stability but are customizable and the stability has changed over time. That is, at any one point in time, virtually all contracts will specify compensation, which rights (hardcover, paperback, audiobook, digital, etc.) are included and so forth, and those terms will tend to be the same for everyone signing contracts that month/year. An individual author can get changes made, if they think to do so and are able to convince the publisher to agree to them.
A fairly typical part of the contract describes what happens if the publisher chooses not to keep (say) a book in print. It is common (but by no means universal) for rights to revert to the author if the publisher does not keep a book in print for a certain period of time. This is all subject to some technical definitions that really, really, really piss people off on occasion.
Digital rights, or rights to ebook publication and, more recently, multimedia/interactive book rights are a relatively recent addition to standard contracts. Authors whose contracts have been in place for a decade, say, may well own the digital rights to their entire backlist. The Big 6 is worried about this and have tried a number of things to make sure they control those digital rights: litigation (so far wildly unsuccessful), applying pressure to authors to sign a new contract for the backlist digital rights in order to continue with the publisher, etc. They are unlikely to be successful in gaining access to all of their backlist titles.
Prior to Amazon (and I don't mean kindle, I mean Amazon), the publishing world was rapidly heading towards twin destinations: bestsellers and niche markets. Bestsellers sold to everyone; niche markets each presented a unique but solvable marketing problem. The midlist -- books without an obsessive, identifiable fan base, but which people bought and enjoyed just not in the hordes that would make them bestsellers -- and the backlist -- books which _had_ sold well, but were no longer new -- were dying a horrible and not particularly slow death. Amazon changed that, because suddenly, shelf space was infinite. The midlist revived, the backlist thrived and there were no returns. Sweet, sweet profitability for publishers.
The successful ereader/ebook revolution presents publishers with a massive risk and, simultaneously, an unbelievable golden ring. If they cannot lock down digital rights to any of their backlist, and new authors switch to selfpublishing, they are goners. As long as they can slow the transition down, they can guarantee that when the year comes that most people buy their books in e-form rather than p-form, they'll have years of recently published, mildly to wildly popular titles that they own the digital rights to.
They've been pretty good at this so far.
To the extent they can, by hook or by crook, get even more digital rights, that's gravey.
Konrath and others have been quite emphatic about several things. (1) Selfpublishing has a big component of luck to it. (2) It helps to write a good book and present it well. (3) It really helps to have lots of other books for sale, so that when someone _likes_ one of your books, they can run right out and buy all the rest at the same time.
There is no way that any one self pubber, small group of self pubbers, or even all the self pubbers, add up to as big a backlist as what even 1 of the big 6 have access to.
The Big 6 aren't going away.
That said, the Big 6 don't own all of "legacy publishing". There's always been a fair amount of churn. Small publishers sprout up and die or sell out to some other publisher -- or start rolling up other publishers. That's where the Big 6 came from, after all. Konrath is already feeling the pain of managing his backlist and explains how he hires services to help with this and talks about paying someone 15% to just take over the job as a whole. He calls it e-stribution and believes agents will be doing it.
I think the sad part of this whole story is that we've confused a pretty normal part of book publishing (new publishers showing up on the scene whenever there's a change in the business or cultural environment; old publishers selling out to more successful ones, cashing out based on their base of backlist contracts) with a pretty normal part of post-Reagan manufacturing woes in the US. There are real people losing real jobs, many of them jobs that paid enough and had benefits appropriate to raising a family. Those are _manufacturing_ jobs, the exact same people that have been laid off repeatedly around the country as newspaper after newspaper closes down or reduces circulation or contracts out its printing or whatever. The people in the New York City offices that are the target of rage in comments threads about selfpublishing -- those people are probably going to be just fine. (Well, to the extent that all those young women with advanced degrees working for incredibly crappy pay can be described as "just fine". They don't call those high prestige/low pay jobs for no reason at all.)
Konrath's talk about e-stribution and his threat to publish his backlist that he doesn't have rights to as multimedia ebooks (rights _not_ covered in his contract) and his discussion of what precisely is and is not provided to him by publishers and whether it's worth the rate they are offering to pay him -- that's all contract negotiation. Some of the people involved in this discussion appear to recognize this. (Eisler and Konrath make jokes about it.) Others do not, and while I think it's just fine for people to make either of a couple choices, I find it a little sad when people shut down a choice without realizing they're doing so.
Authors make money by selling their work. Someone has to want to buy it. If you're irritating enough, it'll tend to discourage sales. While the gatekeeping function of the Big 6 and publishing houses in general has been weaker recently than it has been in a long while, it is not nothing, and there are some indications that gatekeeping function is rebounding from its nadir. If you are writing for a niche market, you can flip off the Big Boys all you like; you'll sell to who you sell to and no one is going to handle the distribution better anyway. But if you want to reach a national audience, now is not the time to be ranting about the evils of the Big 6 and how Self Publishing Will Destroy Them. Once you become Nora Roberts or Danielle Steele or Mary Higgins Clark or Stephen King or whoever, you can be a jackass and be certain someone will still sign a contract with you (altho you might, like them, decide that being super polite gets you even further).
This may also go a long ways towards explaining why Amanda Hocking signed a big contract recently, and why Zoe Winters has recently become a lot more discreet about plugging selfpublishing.
I am in _no way_ arguing against an individual decision to selfpublish and I think for many authors, that is the right way to start out and/or continue and/or to switch to after starting with a "legacy publisher".
I think any author who has benefited from rights reversion and/or no digital rights in the contract on a book that did pretty well should be putting that book out on Smashwords and Kindle. I would point to Julie Ortolon's Perfect Trilogy on the kindle bestseller list as what I believe to be an example of that working out Really, Really Well. There are a lot of other books currently in a similar situation, but going forward, "legacy publishers" who intend to thrive in the new era are likely to write their contracts to guarantee they benefit from long legs in e-space instead.