I think, for how multi-purpose D&D tries to be, two books makes sense, accounting for many pre-made monsters and traps and such. The DMG can really be folded into the other two books. D&D generally claims to handle "any fantasy and sometimes non-fantasy" to limited success without heavy modification. A more specified setting or view of character capabilities could probably manage it all in one book.
Simulation rules also devour pages. Though their inclusion is a matter of taste, game design should closely examine what rules are needed. Ultimately, it is a game, and that should be clear. The basic structure of a game can take only a few dozen pages, honestly, if built well.
Consolidating multiple rules and removing extraneous and convoluted rules is a great starting place. With abstract HP, attack rolls determine "success" rather than "hits," and then fit nicely with the fluff of magic -- removing the sarcasm from a remark I stumbled upon at 339, there's an underlying wisdom: "Sure, you can make a fireball with the spell, but a poorly made fireball won't burn anyone" (Tiballagher).
Reducing rules speeds up learning, preparation, and play, and leaves more room for abilities or reduces pages needed, saving paper costs on wide distribution. Then consider a 'feature' video game costs $60. A well-built, open-ended, highly replayable and adaptable game should easily be able to command, say, a $40 price tag. But that's the whole game not just the basics. For instance, when 4e hit, the PHB1 was pretty lackluster: playable, but with strikingly few options, especially for feats -- especially for wizards! Very specific expansions - campaign settings, adventure modules, etc. - can merit their own sales, but the game itself should be one package.
I won't advocate piracy, but if a game doesn't look like it's worth your dollar, skip it. Find a company making a quality game, or keep playing whatever you play now. D&D is not the only fantasy TRPG; and people should not let companies yank them around as if they hold a monopoly.
Game companies have a lot of opportunity right now to change. There's a great interview with Peter Adkison (CEO of WotC at the time TSR was purchased) where he talks about thinking at a time they could cut the D&D (game) staff down to about half a dozen people. Of course, that's not what happened, but having a few people with clear roles - content generation, art direction, etc. - saves costs and the team can often work much more quickly. WotC's MtG line works similarly with the Design, Development, and Creative teams for each set.
