The intent of SBD is not that it be 'open to speculation' other than speculation within a narrow range and relatively short time period due to liquidity considerations (with enough volume, a speculator can make a large amount of money betting that SBD at $0.99 will rise to $1.01 or vice versa). Instead of a speculatively determined price, increased demand for SBD should result in more SBD being circulated and decreased demand for SBD should result in less SBD being circulated, in both cases pushing the price back toward $1.

The current mechanism doesn't fully implement this but that is the intent, and the proposed change gets a bit closer while removing an odd quirk how the thresholds are currently set: between 5% and 10% there is literally _nothing_ being done by the system to control or even influence either the supply or value of SBD (none of it being created nor destroyed, backing ratio not affected, etc.). No one really intended to create a market cap ratio 'window' for fully-free-floating SBD with no controls at all; it seems to have been an accident.

EDIT: @timcliff in his reply also noted some other perverse effects of the mechanism of shifting payouts from SBD to STEEM, which are good points too. By reducing size of the range where this occurs those effects are reduced, which is good. (I, and numerous other stakeholders and witnesses, would like to see more done, but this is a great start and kudos to @timcliff for getting it implemented and scheduled for the hard fork.)