You are viewing a single comment's thread from:
RE: STEEM Dollar Purge Update : The Bonfire has been lit!
Since 26-Sep-2018 (HF20) we have printed approx 870,000 fresh SBDs and we purged about 200,000 SBDs in Purge 3 - the only one post HF20 before this latest carnage began.
So considering the purges were already underway I think it is a stretch to claim that we didn't make things worse (more unstable) with HF20.
I agree though that most of the damage was done during the unchecked SBD Pumps and there were plenty of senior witnesses and whales who didn't want to do anything about it. They still don't.
Somewhat true but in fact the change in HF20 was an incremental measure to help reduce these pumps, supported by witnesses and whales so it is false to say that they don't want to do anything about it. It is true that, especially at that time (late 2017-early 2018), there was a great deal of resistance as many people were enjoying a lot of 'free' money. That probably included resistance from witnesses and whales and are part of the reason nothing at all was done until HF20.
Reverse conversions might be better, but what happens when you reverse convert right up to whatever limit is set (say 5% or 10%) and SBD is still pumped?
Really? What would the SBD ratio be right now without the added 870K (we can even ignore the 300K that were converted down, although I don't think we can categorically say that HF20 didn't play a positive role in those happening)? It would still be at or near 10% and I argue this is not significantly worse. It may be marginally worse, but its really ultimately the same thing. Now consider the effect of a supply shut off during a pump because of low soft limit (say 2-5%) being hit, meaning there is nothing to push the price back down and reduce supply. No, I don't agree that overall a supply shut off makes things better. At least it is not clear at all that it does.
Yes, it is only marginally worse - but it's still worse and its possibly only marginal because this market dump occurred less than 2 months since the change. We could have been printing for a lot longer and made things a lot worse before judgement day arrived.
Here are some of my current calculations :-
Expected Feed Price : 0.40533
Current STEEM Supply : 283092063.286
Current SBD Supply : 13929289.792
Debt Ratio : 10.825%
Expected SBD Conversion Rate : 0.915
Now I know this is oversimplifying things a bit but if you knock 800,000 SBD off that Supply then you get a Debt Ratio of 10.267% with a SBD Conversion Rate of 0.971
I don't think thats insignificant. The difference between 0.915 and 0.971 is massive for maintaining the credibility of a Stable Coin.
How much would the STEEM price have to decline to go back form 0.971 to 0.915? I'm pretty sure the answer is a tiny, almost insignificant amount (relative to typical volatility). (EDIT: I guess the answer is pretty obvious since SBD is effectively pegged to STEEM once the ratio goes above 10%: about 6%, well within the range of one day volatility.)
Ultimately it is price changes which account for the vast majority of all of these considerations.
And because of that, and the inherent volatility of the underlying asset, I don't believe it is possible to avoid all sorts of possible 'failures' of a stable value coin, including having the price go up (potentially a lot) because you cut off supply. The tradeoffs that Steem makes are pretty reasonable IMO.
Indeed, I would simply redefine failures as 'residual volatility' of any given design, and the look to minimize rather than eliminate volatility (as well as preferring smooth/gradual to catastrophic failures, which the current 10% rule does quite well). From what I have seen of Steem, big pumps are more of a concern to me than some excursions above 10% where the peg target drops below $1 (including, for the reason that the big pumps probably make the >10% excursions more likely). When SBD is $10 that means you can (and likely will) lose 90% of you money. It would be hard (though not impossible) to lose 90% from the haircut. It would require STEEM to decline another 90% plus not having conversions along the way to reduce SBD supply.
If you are holding SBD that you bought or received when valued at $1 then certainly 91 cents on the dollar sucks, and ideally $1 will be recovered relatively soon. But it is still a hell of a lot more stable (and will continue to be so even if it declines some more) than STEEM or most other cryptos. We need to define success (or at least progress) in relative terms.
As I noted elsewhere the print rate is simply tiny. Whether you print at 100% or not only makes about 2.9% per year difference. That's about a quarter of one day's price volatility!
There is simply no hope for the decision whether to print or not to have much effect at all. That will never work. IMO you might as well just leave the printer on all the time (which probably helps in some sense by removing a variable from market expectations) and let conversions adjust for the difference between that and what is needed. The latter can possibly adjust fast enough to cope with volatility and is really the only hope. (I'm also open to reverse conversion models.)
You have to remember that the print rate limit was not part of the original design and was thrown in by the devs with little real review or analysis. Nobody considered these sorts of numerical factors in whether it was a useful approach, nor that stopping printing of SBD doesn't actually mean stopping printing, it means printing more STEEM, which likely has an immediate negative effect on the STEEM price/market (and therefore possibly the SBD ratio). Some of the same devs who originally implemented the 2-5% range not only agreed with the 9-10% model but actually proposed a 9.9%-10% model. In fact, from a numerical perspective none of this likely matters.