The publishers see it your way, too. And if that was all there was to it, the developer got to keep the I.P. rights, etc., then I'd probably agree with you. In fact, in my ideal world, that's exactly how it would work - the publishers would be commoditized, and would really just be a one-stop shop for developers seeking financing, marketing, and distribution for their games.
But that's not how it works anymore (or almost never how it works), and it's the developers who have gotten commoditized.
Scott Miller explains it at great length with a lot more detail than I could, here. He explains why paying for the advance out of royalties is effectively "double dipping" on the part of the publisher. I'll let you read the article and see if you are persuaded:
Scott Miller: Royal Tease
Thanks for posting that, it was an interesting read. I'm a business person (investments) so I probably see things differently than people working for developers. It seems like the writer of the article is arguing for a system that is more of a partnership than that of an investor/proprietor relationship.
He wants a system where all costs are covered before ANYONE sees a dime of profit, and I can understand why. I think though that he discounts the cost of capital and required return on investment that any investor sees.
Let's consider three different scenarios and how they affect each party under the current system and his proposed:
1) The game is a flop
2) The game breaks even on actual cost
3) The game is a hit
Under the current system, in each case, the employees of the developer (at least theoretically) have all been paid for the work they put into the game, paid out of the advance. The only risk in scenario 1 or 2 to the developer is that they may not have jobs anymore as I imagine it will be hard to get future games funded and published. The employees are not going to have their wages garnished or their savings raided cover the loss (in scenario 1) that the publisher has incurred.
However, the publisher in both scenarios has lost. In scenario 1, it has actually lost capital that it has no way to recover. So while the developer (and its employees) are still whole, the publisher is not. In scenario two, the publisher has recouped its capital, but it has opportunity lost. Say development was 3 years and the risk free rate of investment (IE US Treasuries in my case) is 3%, that means they've lost out on at least 9.2% total return (and possible more had they put the money into a better game). The point is that even though the costs of production were recouped, the developer has still lost out.
Now obviously in scenario 3, everybody wins, but the publisher wins more.
In investments, this is considered a perfect example of the risk/reward ratio. It's why US Treasuries yield say 3%, while IBM bonds yield 6% and Joe Blow Inc. yield 20%. As the risk to capital increases, the required return to make the investment increases. If all three yielded 3%, why would anyone buy IBM or Joe Blow Inc. bonds when they can get treasuries, with theoretically no risk, at the same return?
Now in the proposed system of treating it like an equal partnership, where all costs are pooled before anyone sees any profit, in scenario 1, both parties lose, same as before, and just as before, the publisher isn't able to recoup any of the lost capital from the developer. Scenario two is the same as well, except that it continues to a higher sales point.
Scenario three is different however. The developer gets a larger share of the profit than before because the advance is not charged against the royalties. So the publisher could actually still lose money in terms of opportunity cost (IE the return on capital is less than what their other options are as I described above), and always gets less than with the current setup. So the publisher has now taken increased risk with less return.
While you may say "so what, they're still making money" realize that owners of capital will (mostly) take their capital to the venue that offers the best risk/reward ratio. By compressing that ratio, you end up actually decreasing the amount of capital available to make new games because other avenues of investment become more attractive, relatively speaking, which means less developers in existence (though perhaps better for the developers that stick around).
I'm actually having a similar discussion with my brother. He's a home builder who wants to go out on his own (flips more then new builds though), but he doesn't have the capital. I have a decent amount of capital (mainly in my retirement assets, but enough for him to get started), but due to my position, not many places I can invest it. So we're trying to make a deal, but he doesn't understand why I want a percentage so much higher than he pays for his mortgage. I've had to explain to him that risk/reward relationship. If he fails, he can just go get another job and worst case he makes less over that time period than he would otherwise. My money is gone though.
Now the other issues he brings up in the article (lack of competition among publishers, depressed royalty rates, etc.) I do find very persuasive.