The other aspect is the miner's inevitable margin compression. From the numbers we have seen (e.g. KnC in 2014), it seems that these companies are already operating on razor thin margins. Whatever money they have made, they needed to re-invest into upgraded hardware in order to stay competitive. Unless there is a meaningful increase in price before the reward halving, a lot of miners will likely run into trouble.
The reward drop will instantly price out of the market a large proportion of the current hash rate.
This will make the next 2016 blocks take substantially longer to find than normal, which will result in a large drop in the difficulty. This, in turn, will price a large chunk of mining back in - enough to mean that the following 2016 blocks take substantially less time to find than normal.
So I think we'll see see-sawing difficulty in alternate 2016-block periods for a while, until a new equilibrium is reached (at a lower difficulty / lower network hash rate than before the reward drop).
I don't think there'll be a significant effect on either the bitcoin price. Transaction fees should rise a bit in the "high difficulty" periods, and drop in the "low difficulty" periods, but once equilibrium is reached they should be much the same as they were before the reward drop.
As long as there isn't sigificantly more transactions than are reasonable to be included in a block there simply won't be too much pressure on the senders side to increase the amount of fees paid.