Haircut is around the corner unless STEEM price is above $0.245

Repository

https://github.com/steemit/steem

Update: As forecasted, the haircut had started, and as you may see, SBD print rate is still 1% (maybe you remember this from the last haircut). This is due to a bug that I analyzed in details before: Why SBD print rate is still 1% despite the haircut? Bug report, explanation, and suggestions, and as far as I know, @timcliff fixed this (thanks!), and that fix will be merged in HF21.

How to know if we are experiencing a haircut? https://api.steemjs.com/get_feed_history If current_median_history does not show the quote of 1.000 STEEM you can assume that we're under the haircut. In this case: quote is steem supply, and base is the haircut sbd supply.

Introduction

SBD print rate decrease starts when debt ratio is highter than 9% and completely stops (no SBD) at debt ratio is 10%, and then the haircut begins.

What is the haircut?
The haircut is the situation where 1 SBD no longer worths $1 USD of STEEM. For instance, Splinterlands' cards are quoted in USD not SBD, so you need to use more SBD to buy the card with the same quote price.

About 8 months ago, we experienced the haircut, and the second haircut is just around the corner within 36 hours unless STEEM price quickly recovers above $0.245.

Note that the feed price is the median of 3.5 days.

Scope

Data: since 2019-07-01

• Steem/SBD supply
• Steem feed price
• Steem conversion request

By analyzing the above data, I also provide some rough forecast when the haircut will begin.

Results

The feed price when the haircut begins without conversion effort

Current supply information can be fetched with API (dynamic_global_properties) or just visiting steemd.com

Here is the data as of 2019-07-17 UTC 1AM (approx)

Current supply and the feed price where the haircut starts

Based on the current STEEM/SBD supply,the haircut may begins at the feed price of $0.250.

However, when the situation is getting serious, normally conversion effort comes in.

Since the conversion (SBD->STEEM on chain and then SBD goes to null) reduces the supply of SBD, it also reduces the debt ratio.

Recent trend of conversion requests

The recent conversion requests since 2019-07-01

The above graph shows the conversion request amount since 2019-07-01.

As you can see, the conversion requests increased tremendously on 2019-07-14, which clearly shows the effort to prevent the haircut.

By whom?

Here it is.

• blocktrades, alpha, smooth-a, fermion massively requested conversions. Thank you all!

Note that the conversion takes 3.5 days.

So, the effort on 2019-07-14 will help the situation on 2019-07-17. For simplicity, let's say it's 300K SBD (and STEEM/SBD supply increases due to other reasons except for conversion are negligible, which is okay based on my test). Taking this into account, the feed price where the haircut starts $0.241 which is slightly lower than $0.250 without any conversion.

The feed price where the haircut starts with 300K SBD conversion

Note: Initially, I didn't take the converted STEEM into account in the calculation. That is, the conversion not only decreases SBD supply (which I included) but also increases STEEM supply (which I didn't include). Considering it, the approximate feed price of the haircut start should be about 0.241 not 0.243. I'll not draw the plot again, since the only red line matters, so please just read the red line as 0.241. But the time when the haircut may occur is quite similar, just plus 1-2 hours.

When will the haircut begin?

Based on the previous data, we can forecast when the haircut will begin.

First, say STEEM price stays the same (~$0.224)

In the following graphs,

• black line $0.250: price at which the haircut begins if there is no conversion.
• red line $0.243 (but should be 0.241, as in the above note): price at which the haircut begins with 300K SBD conversion.

• Due to the conversion effort, haircut may start within 33 hours, about 10 hours extended (without conversion).

Even if STEEM price increases around $0.240, still it's around within 34 hours.

• The reason why there is no big difference is because the feed price is 3.5 day median and we have already been in the very low price for past a couple of days.

However, if STEEM price jumps above $0.245, the haircut may not happen.

So if you don't wanna the haircut, buy more STEEM or convert SBD on chain (not on exchanges).

Conclusion

Based on the recent history of feed price and conversion requests, we can forecast the haircut roughly. Since I already explained the details, I'd rather like to express my personal opinion why Steem is in this situation. (The previous haircut happened when the bitcoin price is 3~4K, and now the bitcoin price is still 9~10K) The rank of STEEM marketcap is continously falling. I believe the main reason is Proof-of-Brain isn't working. It's just dead. Very sad.

I personally believe that the only solution is ad-revenue + centralized curation. (See this: Why is ad-revenue sharing important? Steemit's marketcap is 0.215% of Facebook's ad revenue , which I wrote 3 weeks ago, and now steem marketcap is 0.15% of FB's ad-revenue.) Individual seeks their profit. What's wrong? That's the very basic assumption in economics. Nothing wrong if they do whatever they want to maximize their profit if it is allowed on the platform. That means, decentralized PoB is impossible from the beginning. It just looked possible only due to the crypto bubble.

Fortunately, Steemit Inc is getting some ad-revenue. When the ad-revenue surpasses all the costs, Steemit should make some efforts on some curation problem, including Utopian, of course. Moreover, the ad-revenue should be shared with authors.

Tools and Scripts

• steemsql for getting conversion history

• https://api.steemjs.com/get_feed_history for the recent feed history

• Python script can be found on https://github.com/economicstudio/steemanalysis

Relevant Links and Resources

Here are some of my utopian analysis articles

• 100% SP vs 50:50 Which one is better? Does the haircut matter?
• Effect of haircut, early voting, beneficiary on dust payout
• Why SBD print rate is still 1% despite the haircut? Bug report, explanation, and suggestions by @blockchainstudio
• Witness of the people, by the people, for the people
• Drugwars - revenue and transaction analysis