1. A large proportions of the hash being from botnets (where the operators immediately dump), and
2. The relatively high emission rate.
This creates a disproportionate downward pressure on price that differs from other coins, and which appears to be consistently outstripping buy pressure, despite strong interest in the coin. This is not a good thing, and it may drive the price into the ground. Without the expectation that the price will go up, fundamentals don't matter a bit. Miners and investors will lose confidence and move on.
There is a solution to this: write a GPU miner. In its current form as a CPU only coin, if the MRO market doesn't ramp up buy pressure to consistently match/exceed the expected constant dumps from botnets, the coin will fail. And it's easier to scale up botnets than it is to scale up the ordinary userbase, so this disparity may worsen. Your average GPU miner is more likely to hold than your average botnet operator who's looking for immediate turnover and who doesn't care if the coin lives or dies. Perhaps equally importantly, a GPU miner would massively increase adoption among the community, which is another thing currently holding MRO back.
I know people here may be currently enjoying nice profits from CPU mining, and may not appreciate the suggestion of a GPU miner. Personally I'm more concerned about the longer term future of the coin, and I think Monero has a vital role to play within the anonymity niche. Its fundamentals are good and I'd like to see that reflected in its market performance. I'm willing to put a good chunk of MRO towards a bounty for an open source GPU miner if there's enough support for it.
Interesting observation and suggestions. Are GPU's not as easily bot net'd? I haven't heard of that problem on LTC, so wonder... I imagine many GPU miners are on Linux and also mining only rigs, so that might help lower the incidents of bots.
My one real fear with CPU coins is just what you mentioned btw.