Slightly off topic, but you seem to be advocating cross-subsidisation here. Why is it right that the success of one artist should be used to pay for the failure of another? If the major labels didn't produce so much crap, then the music that people would want to listen to would be cheaper as they don't have to pay for the costs of producing rubbish that people don't want - and the company would still be profitable.
The first thing to keep in mind is that if consumers weren't buying crap companies wouldn't be selling it. The second thing to keep in mind is that having more losers than winners is the nature of the beast whether you are talking about music, books, or TV shows. There are too many variables and it's too subjective a purchase to be able to accurately say how these things will do before they hit the market place. It's a crap shoot. How many people predicted the cult success of "Clerks" or the absolute failure that was Daikatana? It's not like people are walking around going, "Gee, where can I find something that's gonna really tank and be a detriment to my company?" If you can accurately and consistently predict how well a musician, movie, book, video game or TV show will do out in the market place based on only preliminary data then get yer butt to LA or NY 'cause you'll be rich in no time flat.
Or to put is another way, the company takes a risk - why should we insure their risk by paying more for the music we want?
Paying more for the music who wants? Do I want the same music as you? Do you want the same music as CanadaRam? Do we really want to be limited to only mainstream acts that only appeal to broad demographics? Personally, if the success of B. Spears means that a few more unknown artists get their shot at the big time then I'm all for it (Or for a slightly different example, the success of a big budget, cookie cutter action film means Studio XYZ has enough money in the coffers to buy and distribute an Indie flick that you absolutely love, but never would have heard of if it wasn't for the cliched action film that you hate). Until the label signs the artist, spends the money and gets a CD on store shelves they don't know if they have a dud, stud or something in between. Just like in pro sports. Some guy make look awesome on paper and may have torn it up in college, but until he suits up and plays some pro games you don't know if he's a star or a very expensive bench warmer.
To varying degrees this is common throughout business. I mean, Apple uses the profits of it hardware sales on software development. iTMS is basically a loss leader designed to sell iPods. Hell, the success of the iPod has been a windfall for Apple. Should Apple only take that iPod generated revenue and spend it only on iPod R&D? Or should Apple take some of that iPod generated revenue and spend it on R&D for other Apple products as well?
Oh, and a third thing to keep in mind is that just because you think it's good doesn't mean it will sell and just because you think it's bad doesn't mean it won't. And when it comes down to brass tacks a business w/o sales isn't a business for long.
Lethal
EDIT: I don't think it's really that off topic as the thread was moving this way anyway. I'm very intrigued by the new, successful business models that are slowly appearing (especially since I'm working on an independent documentary part time), but there is a reason the why the "tried and true" methods are tried and true and why large companies are cautious to move away from them. I'm not meaning to be argumentative (although I'm playing a bit of devils advocate), but I would like the discussion to get deeper than the cliched "they are greedy dinosaurs that can't get w/the times."