Another baseline for burn proposal #117 [burnsteem100]

It occurred to me last week that the market activities from proposal #117 could be offset, to some extent, if the proposal were to drive an increase in SBD Conversion activity. As background, here's what I believe are the first-order effects of the various things that can affect the virtual supply size and the haircut price (this is a draft, and hasn't been reviewed by anyone else, so be suitably skeptical. I welcome feedback.):

As a second-order effect, anywhere that the virtual supply increases/decreases, the activity should have the same effect on inflation (in terms of STEEM per day). The top visual shows conversion requests, and the bottom visual shows conversion fulfillments, so the spikes on the bottom are about 3 ¹/₂ days behind the ones on the top.

Right now - ignoring the SBD burning from the proposal, we're operating below the haircut threshold, so proposal 117 is increasing the haircut threshold by burning STEEM, and it's also decreasing the virtual supply. But, what happens if someone buys those SBDs on the internal market and then converts them to STEEM?

1. Proposal 117 buys STEEM and burns it, which raises the haircut price and lowers the virtual supply.
2. A Buyer buys SBDs and converts them. This comprises two relevant activities:
   2.1 Burning SBDs decreases the haircut price without changing the virtual supply.
   2.2 Creating STEEM decreases the haircut price and increases the virtual supply.

I haven't run the numbers, but I assume that Steps 1 & 2 are basically in balance. As long as we're burning more STEEM than conversion is creating, the proposal will reduce inflation - as expected. If conversion activity creates as much or more STEEM as what we're burning, then we might not see the anticipated inflation reductions. If it triggers more conversion than SBDs withdrawn from the SPS, then we would expect the haircut price to decline faster than if we had burned SBDs directly (increasing the value of the SBDs that remain), but we might not see a reduction in inflation.

So, I think it makes sense to pay attention to conversion activity, as well as activity on the internal market. Unfortunately, the SteemDB link was stale, so I created a script to collect the information. Here's my first baseline, running from October through today (again, this has not been validated, so mistakes are possible).

Points of note:

1. We have now seen 7 consecutive days with higher-than-normal conversion activity (more than 1000 SBDs per day), so it's possible that the proposal has influenced conversion activities, but it's probably too soon to know for sure.
2. We can see the increase in value of SBDs during burn proposal #116 from Nov. 7 through Dec. 7. During that time, SBDS increased from ~6.5 STEEM to ~7.3 STEEM in value.
3. Based on requests submitted, but not yet fulfilled, we can expect that inflation reductions from burn proposal 117 will be damped until, at least, January 5.

It will be interesting to see whether conversion levels drop back to lower levels or if they remain elevated during the next few days.

This conversion activity, I believe, is how the sliding peg works for the SBD when it's below the haircut price. The price is set by the current STEEM price (3 ¹/₂) median, divided by the haircut price. If the price drifts too far below that value, it triggers conversion activity that will bring it back up again. It's the same mechanism that was intended to peg it at a dollar when haircut pricing was not in effect.

So, if the STEEM price rises quickly (or arbitragers believe it's about to rise quickly), I suppose we would expect an increase in conversions.

Reminder: this page updates daily with supply information in order to observe effects from the proposal. Unfortunately, it doesn't have the conversion activity, at least for now.

Update: I realized after posting that January 3 was missing from the visual. It has now been added, and that makes it 7 consecutive days with more than 1,000 SBDs converted.