I don't think the question is well-posed. Why would the market rate for electricity affect the price at all if stolen computing time were a reasonable componen

I don't think the question is well-posed. Why would the market rate for electricity affect the price at all if stolen computing time were a reasonable component of the current hash power? If bitcoin mining were competing for the market for stolen computing time, then the "pwner" of each zombie would make an economic decision as to whether to use the machine for other tasks (spamming, password cracking, etc.) based on the relative value of those tasks in bitcoin vs. the amount of bitcoin produced by the machine. As the price/hashrate ratio increases, more zombies would switch to mining; as it decreased, the flow would go in the other direction. The price of electricity is irrelevant to the "pwner" of the machine as regards this decision; for them electricity is free. In terms of how this were to affect price, that's a more complex relationship, even in theory. As the difficulty increases, it gets more expensive to attack the network (and thus artificially expand supply), so the price will presumably face upward pressure to keep the relationship of difficulty to inverse price constant (assuming price/hashrate as the function is disingenuous, but is probably reasonable for local price and difficulty changes). That said, it is unlikely that the antecedent holds -- it is extremely doubtful that stolen computing time (or even total CPU + GPU mining) is a reasonable component of the current hash power.