I would be more interested in such a model, since it was not so purposefully advantageous to early 'miners'.
In a system without a dramatic fall in return-of-investment on the mining you'd end up with one or two situations. Either it's profitable to mine for bitcoins, in which case you'll get more and more computer power assigned
to mining for bitcoins and thus runaway inflation, or it will not be profitable, in which hardly any mining will be done, and you'll get deflation (particularly as the system will "bleed" bitcoins in hard-drive crashes, people dying or forgetting the password to their bitcoin wallets.)
It's perhaps
theoretically possible to hit a sweet spot where the roi of bitcoin-mining is
just right for the amount of bitcoins mined to correspond to, say, 2% inflation. This isn't a stable system though, not even assuming an even, steady rate of increase in computing power, so there's no way this could ever occur without regulatory oversight -- which would rather defeat the point of bitcoins in the first place.
Then there's the issue of sudden breakthroughs in computer technology or mathematics or for that matter the discovery of a flaw in the bitcoins algorithm or implementation (not exactly unheard of in the field of cryptology.)
Then there is the issue of computer power actually being very unevenly distributed. A few large actors such as IBM, Google, the US government, IBM own the vast majority of the world's CPU power and there are huge thresholds of entry on computer power at those scales. (And if we look at it from the point of view of macro economics, all that computer power is located in a handful of countries. How would Nigeria or Pakistan fare in a bitcoin economy?)
All of those are minor concerns compared to the elephant in the room, though. Without backing -- whether in a commodity such as gold or the power and credibility of an authority -- the value of a "currency" is solely speculative.