I saw a blog where someone reverse-engineered it and created something as a test. So it's possible.
I see 4 difference approaches Apple could take:
1) Come out with an "iWallpaper" app for $2.99 that lets you make your own. They'd sell a LOT of that but it would have to be somewhat limited in its abilities to make it usable by the public. (Remember, they're not videos so you can't just import camera footage.) This would give Apple lots of up-front profit but I imagine interest would fade when it turns out you can only do so much. (Imagine the Mac's "Flurry" screen saver controls as your inspiration for this. Fun at first, but it gets old.)
2) Start making more wallpapers of their own and either release them in an update or sell them on the store like they sell ringtones. This would probably be a good way to go, but there's only so much they can do. You're not gonna find the color theme to your favorite college team with a system like this.
3) Release an API that allows developers to make their own wallpaper-making apps. Not much up-front profit for Apple since most will probably be fremium apps, but I see this path increasing the value of iOS the most of the 3 plans. You'll see generic apps that are like the "iWallpaper" idea. You'll also see really specific ones (like one that only does wallpapers to match college sports teams, for example). And you may see super creative ones like, maybe, upload a picture of your face and the app analyzes it and reverse-rasterizes it to give you a cartoon of you that runs as a wallpaper. Who knows what people will come up with.
The difference between this and an Apple-made app is that they could be very specific and certain apps could have a steeper learning curve than others. So you'd see a wide range of amateur-to-pro apps that Apple wouldn't be able to do with just one app.
I think #3 is the best choice, but, of course, there's always option #4: Do nothing and forget about it.