Witness SBD pegging policy update

in #witness-category7 years ago (edited)

I have updated my witness SBD interest rate to 15% from 10%. My feed discount is unchanged at -12% (1-1/1.136).

I have done this because SBD is effectively debt to the Steem blockchain, representing a claim to receive upon demand (with 7 days notice) approximately 1 USD worth of STEEM. Given the decline in STEEM price and market cap, this debt has become distressed. SBD holders may reasonably doubt whether they would actually be able to receive and successfully liquidate 1 USD worth of STEEM as repayment). This is somewhat exacerbated by the 'SBD stability' feature which reduces repayment rates should SBD debt exceed 10% of STEEM market cap. It is currently at about 7%, meaning SBD holders are close to taking a 'haircut'.

The previous 10% interest rate has been in place since SBD creation started on July 4. It was almost never sufficient incentive for investors to want to invest capital in holding SBD on a new and in many ways experimental system. For that reason SBD has consistently traded at a discount (resulting in a higher implied interest rate), even when the debt load was modest. Thus it is reasonable to assume, outside of debt issues, that the interest rate should probably be somewhat higher anyway.

If we do not pay an interest rate commensurate with the risk of holding a token with the properties of distressed debt, then SBD will trade a larger discount. Recently the price of SBD declined quickly from 95 to 87. This will encourage more SBD holders to want to "cash out" by demanding their 1 USD worth of STEEM. While orderly debt reduction is good, collapsing demand to hold SBD is not good.

The cost of an increased interest rate is modest. Currently there is 1.9 million SBD outstanding, a number that continues to steadily decline as debt is reduced. The increase from 10% to 15%, assuming adopted by other witnesses, would cost approximately 260 SBD per day.

Finally, when feasible I believe we should continue to work to keep SBD close to 1 USD in value. Increasing the interest rate will facilitate this, by removing the need for the market to impose a higher implied interest rate with a trading discount. This will encourage continued use of SBD and development of a larger Steem-based economy (which we have seen in its embryonic stages as many initiatives and projects have been funded and operated with SBD as a payment instrument, including Steemfest and the associated travel reimbursement fund). The more participants are concerned about SBD declining in value while they hold it, the less its use as a payment instrument and store of value will flourish.

For all of these reasons I have increased my witness SBD interest rate to 15%.

(payout declined)

Sort:  

I still don't understand the economics of it, but what you stated makes sense in helping to keep the SBD stable. Thanks for your dedication to the platform and the community.

I will add to the above that with SBD being in a distressed state and multiple objectives in play including price stability and wanting controlled debt reduction, further adjustments may be needed. This includes the possibility that SBD could become too strong, providing insufficient incentives to perform conversions and reduce debt. The 'haircut' rule at 10% complicates things considerably. Hopefully we can move a comfortable distance from 10% debt ratio reasonably soon.

Haircut is a euphemism for going bald here. A haircut grows back, but it is hard to see how SBD can recover once the peg is history.

It can recover if held and the STEEM market cap increases and/or if debt is sufficiently reduced such that total SBD outstanding drops below 10%.

The payoff profile for holding SBD while near the 'haircut' limit, ignoring the question of debt reduction, is similar to that of holding a covered call position on STEEM: downside as with STEEM, but trading upside for premium income (interest here).

I think in lieu of some blockchain-based (but reasonable) incentive to maintain the peg, Steemit Inc. could use their STEEM to hold the peg. I speak in ignorance as to what the burn rate would be though. It would have to cost something, and sustained over time it could cost the @steemit account more than they'd like.

But they don't seem to be doing that, so I'm willing to give this experiment a shot.

If the parameters are too far out balance then trying to force the peg by spending money can be very costly, including by potentially ending up accumulating a lot of SBD that needs to be held without being adequately compensated for the risk it could fail. It also could tie up a lot of their money indefinitely that they want to use for other development purposes.

Where direct trading can work well is once the parameters are close to a good overall balance, keeping the price stable between 99 and 101 instead of what might be natural fluctuations of 97-103 or even wider.

The balance between SBD and STEEM isn't understood by most Steemit members. I know if I had if unstood this earlier I would have held in to more SBD as a STEEM price deflation risk insurance. I guess eventually the STEEM price has fallen t a level where the insurance company is going to go broke paying the risk claim.

I can't agree more and people should do that too because I think this is the best for the community. People should remember Steemit is still in Beta and we all should see the bigger picture (it will be better for the future of this community). People should pay attention to​ what a witness says because we choose them and we have faith they will do the best for all!

Economic principles always perplex me. Wouldn't increased interest rate on SBD further encourage users to hold it (for stable rewards, rather than decreasing reward value from holding steem/vests) resulting in further reduced demand for the $1 worth of steem per SBD? You will have to excuse my ignorance in these areas, as this question is likely one which appears dumb to those with a better command of these principles.

The use cases for a stable value token and a highly speculative token such as STEEM are going to be very different. This is a bit like asking whether a lower price for bicycles would reduce demand for airplanes. In some extreme theoretical sense, it might, but not very meaningfully.

I may still have misunderstood your question in which case please let me know. I'm happy to try to explain things to the best of my ability.

Maybe I have misunderstood the post from the outset, your intention is to further incentivise people to hold SBD by increasing interest rewards from 10 to 15% correct (when I originally read it I thought it was the opposite, which is why I was confused)?

Assuming that is what you want then increasing the interest for holding makes sense. However by what mechanism other than an increase in steems market cap can the burden of SBD debt be reduced, and is a further increase in interest reward sustainable? Also why is an interest reward necessary, should not protection from asset volatility be reason enough for people to want to hold SBD, especially in a down market?

I still think there are some fundamental flaws in my understanding of the economics of all of this. No worries if you've no time to answer my questions. I understand you are busy! Thanks for getting me to think more about this. I will likely spend some more time reading and thinking about it in the coming weeks.

I'm not really trying to force any particular behavior here, I'm more trying to keep SBD from becoming (more than it already is) a hot potato. As distressed debt that carries the real possibility of principal losses, a significantly higher interest rate is necessary to make it a viable holding. 15% may well not be high enough, but I'm taking a somewhat gradual approach. If I still see signs that it is being sold for lower and lower prices, I will propose further increases.

As for the mechanism for reducing debt, that comes (and is already coming) from conversions, where SBD holders request to be paid in STEEM, which destroys the SBD, retiring the debt. This happens for several reasons:

  1. Those with SBD rewards or payments who want to power up may use the conversion function to request STEEM instead of trading it and then power up the resulting STEEM.
  2. Traders may engage in a form of statistical arbitrage where they buy SBD, convert it, and then sell the resulting STEEM. This is profitable if the value of STEEM that is received and liquidated is higher than the cost of SBD (plus a cost of capital covering time and risk).
  3. Some whales and other community members have been steadily converting SBD without too much regard for profitability. Of course, no one wants to take huge losses either.

All of these are supported by witnesses including a discount in their price feeds, which gives more slightly more STEEM for each SBD converted, as well as the price of SBD not being too high. Recently the discount factor has been about 9% and prior to recent weakness, price had been in the range of 92-96. Personally I feel that a discount factor of about 12-14% with a price of 95-101 is preferable (while offering the same conversion gross margin), but some witnesses have wanted to be more cautious about the size of the discount.

Interesting, thank you for taking the time to explain your thoughts to me! I appreciate it smooth. I think I understand this a bit better now, or at least I feel a bit more comfortable.

This is somewhat exacerbated by the 'SBD stability' feature which reduces repayment rates should SBD debt exceed 10% of STEEM market cap. It is currently at about 7%, meaning SBD holders are close to taking a 'haircut'.

I get 5.7% at the moment.

5.7% is by 7-day median. If you check latest price, it's about 7%.

The total supply and sbd supply will have changed by the time the price feed catches up to the current price. I would expect it to be somewhere between 6% and 6.5% in 3.5 days when the feed catched up to the current spot price.

I do not understand what the code is doing but by my estimate it is already 6.6% right now:

Feed price is 0.145 USD/STEEM * total supply 198618992.113 STEEM = market cap 28799753.856385 USD.

SBD supply 1893566.413 / market cap = 6.6%

Perhaps my formula is incorrect.

The latest feed update is 0.119 so there is an 18% drop in market cap left to reach the current price. The supply increase is slightly below 1% per day (1.75 million / 199 million) and that would only matter if the price doesn't drop by at least 1%/day. The pace of debt reduction is likely <1%/day (but jumpy as it depends when large conversions occur).

I expect to reach 7% quite soon and something close to 8% appears quite likely within a few days. Reaching 10% or higher is not unlikely at all.

Currently,

  "virtual_supply": "212716237.198 STEEM",
  "current_supply": "199335134.831 STEEM",
  "current_sbd_supply": "1886630.621 SBD",
  "median_sbd_price": {
    "base": "0.162 SBD",
    "quote": "1.149 STEEM"
  },

So the percentage = 1 - 199335134.831 / 212716237.198 = 6.3%
Actually, 1886630.621 * (1.149 / 0.162) + 199335134.831 = 212716237.198
If calculate with the price I fed last time

  "sbd_exchange_rate": {
    "base": "0.133 SBD",
    "quote": "1.136 STEEM"
  },

then virtual supply should be 1886630.621 * (1.136 / 0.133) + 199335134.831 = 215,449,513.669 ,
the percentage = 1 - 199335134.831 / 215,449,513.669 = 7.48%

I thought 10% interest was the maximum. Is there a maximum?

I don't know if there is a maximum. I have mine set to 15% and another witnesses has it at 20% so whatever it is, we haven't reached it yet.

That's good to know!

This seems somewhat like a chicken and egg problem. We want people to convert to destroy debt. We also want people to hold SBD instead of dumping on the market, but if the incentive for that involves increased debt via increased interest, isn't that bad? 260 a day compared to nearly 2M seems small, so maybe it's not a concern compared to the value of strengthening the incentives for the peg. It's complicated stuff for sure.

The reason I went ahead and calculated the daily cost of the interest increase was to make sure that it wouldn't add too much to the debt and be counterproductive. Debt reduction was running about 5% ($100K) per week, though in the past 1-2 weeks it has slowed down somewhat. Still, $260/day is not a major problem relative to continuing to neglect properly pegging SBD and undermining what ultimately could be one of the most useful features of the platform.

It is for sure complicated as you said. Only after having spent a lot of time thinking about these issues over the past few weeks have I reached my current (certainly incomplete) level of understanding. I may write a bit more soon on the tools available to witnesses to maintain the peg and some of the ways those tools can and should be used.

It's easy for me to get frustrated with all of this complexity and while seeing my investment dwindle in value. I just have to remind myself we are early pioneers, exploring and testing new economic systems in real time. That's pretty exciting and hopefully worth a little sacrifice along the way.

Thank you for being a consistently beneficial actor for the network here. So many of us benefit from you understanding. Please keep it up.

upped and resteemed

full $teem ahead!

This post has been linked to from another place on Steem.

Learn more about and upvote to support linkback bot v0.5. Flag this comment if you don't want the bot to continue posting linkbacks for your posts.

Built by @ontofractal