Self publishing is only really viable the last 2 or so years, so there's going to be ZERO stats on the subject, but I wonder if that's actually true at all?
Actually, unlike self-publishing music (which used to require significant investment in instruments and recording equipment or studio rental and now requires little more than a computer), affordably self-publishing your books has been possible for a long, long time. What's more, an occasional smash hit has in fact risen from that level of publishing. But those are the anecdotes--the stories told by struggling amateurs while the vast majority of successful authors warn them that the odds are against it.
It's true that distribution and marketing logistics are dramatically simplified by the internet, and that is obviously worth further examination. But I'd submit that without the editorial process, something very important is still missing. Let's take your experiment:
Take 100 starting out authors today (50 self publishing, 50 submitting drafts to big publishing houses). Check their average earnings from books in 2, 5, and 10 years.
You really think the ones just starting submitting, as opposed to publishing NOW (and for cheap), will wind up better on average?
Do you actually think that a single author out of those 50 just starting submitting manuscripts will make it at all in the standard writing game? (by make it I mean earn a living family wage from books alone)
I wouldn't be surprised if 1 in 50 really did make a living wage ten years out (assuming they stuck with it), though others have told me that the number is more like 1 in 1000--it may be that I have just met too many successful authors to make a reasonable guess.
However, something your thought-experiment is missing is process--editorial, market, and (perhaps most importantly of all) writing.
Editorial: How much would the authors in each group improve? Which authors would be more likely to write something notable? Assuming I'm wrong and epubbing became the norm, what impact would that have on our literary sophistication as a community?
Market: What would be their chances at stardom? Let's say a published author becomes so successful, she hires a personal assistant. She becomes so prolific, her friends can't keep up, and she contracts out her editing. She becomes so popular in the ebook realm that Barnes & Noble wants her, so she she goes and pays for a printer. Pretty soon she's going to realize that her profit margin is shrinking (though her income continues to rise) and so she forms a co-op with other authors to diffuse costs like paying a publicist or a lawyer to examine those self-pub contracts she so gleefully inked.
See where this is going?
While e-pubbing has definitely loosened the control of the big publishers a little, it hasn't been the disruption everyone keeps saying it's going to be, any day now. I don't know what the true, final implications will be--if dead-tree publishing will ever go away, etc. However, I have one guess that no one ever wants to talk about:
Writing: The digital world moves fast. Writing is often slow.
My wife writes about 3 books every two years. Some of her colleagues put out 3 a year by themselves. Some put out more by hiring assistant authors.
And some of them write one book ever 5 to 10 years.
These books are different. I don't know that they're better, but they do tend to appeal to a different kind of reader. The only way to make money on these books is to make a lot of money, to last you from one book to the next.
Most of the semi-successful e-published authors I've seen, by comparison, churn out books the way lawyers churn out paperwork. They write, and write, and write. And it is often painfully bad stuff, because even the best of them have the book edited by their friend who is an English professor or something. And wide distribution combined with high profit margins means that if they can get a couple hundred sales by pricing their book so low that statistically, that many people will just throw money at it to see what's new...
Yeah, they'll make some cash.
But you've got to realize how quickly that market is going to become glutted.
If you really want to understand where e-publishing is headed, check out XBox Live Marketplace, or even the Apple App Store. Everything is flash-in-the-pan. Companies rise and fall over the course of weeks, sometimes days. Most of the stuff they're churning out is like appetizers compared to the AAA titles that still sell in Wal*Mart. And the folks in control--the hardware companies--are already involved in quality control, and looking for ways to do more there rather than less of that.
As epublishing grows increasingly mainstream, the demand for intervention--from traditional publishers, from "booksellers" (the people who decide whether B&N will carry your title), from someone--is only going to increase. And you might be inclined to suggest user moderation a la reviews, but this strikes me as even more likely to create a handful of superstars and a whole lot of losers than the current state of affairs.
So I guess to the extent that Konrath is at all correct, it is--at best--a limited-time offer. We're not going back to the patron system, where a greater number of artists can make a living by fragmenting the market and doing away with "superstars" like J.K. Rowling or John Grisham. Which may be a shame--I think it would be nice if more people could make a living doing art.
But then again... quality stuff is already often hard to come by. I don't think opening the floodgates and amping the signal-to-noise ratio ever toward infinity is going to improve that. I'm not a big fan of corporations generally, and I have certainly been known to rail against "the man" keeping "art" and "authors" down in my day. But viewing the process from the other side, I have to admit--it's more complicated than I was led to believe.