Pedro Ojeda Profile picture
May 11 24 tweets 9 min read
Update on Luna Tokenomics
How to predict when depeg will end.

Bringing back my “death spiraling Luna” model. Turns out it was accurate. All of those “Didn’t age well” jabs at it Didn’t Age Well.

Prerequisites: you must understand how Luna rewards work and have a basic understanding of the death spiral model (links in next tweet).

Or just read this thread and then go back to these later.
Crash courses:
[1] how luna accrues and distributes fees.
[2] death spiraling luna
[3] counter cyclical fees + real FUD for Terra.


Lets begin! —>
Here is where we are at. Luna and UST have both expanded and contracted ~7% respectively.

Let’s discuss the value of Luna.

The sheer speed of this contraction has left thin order books and hurt investors. Perfect recipe for Arbs!
Terra has a virtual AMM which wants to hold Luna and Stablecoins in perfect balance.

When a sellers burns UST for Luna, in unbalances the pool.

This makes the next trade have a less favorable UST to Luna price. Same as any AMM like Uniswap.
The uniqueness of Terra’s AMM is this imbalance decays over time.

The variable is called TerraPoolDelta and it drops by 63% every 36 blocks -> asymptotically to zero.

Right now delta is at 5401034127083 which translates to 18% UST -> Luna fees.
A quick check of UST price reveals that UST trades at $1 miss the swap fee (1-2% error). Why would Arbs not let the Fee drop?

Because to them these two are equivalent:
- Buying UST at 0.80 from you and getting 0.82
- Buying UST at 0.97 from you and getting 0.99

They get 0.02
So the true determinant of these swap fees and hence the price of UST is how quickly participants want out of UST.

The average swap fee has been 13.27%. This is charged in Luna. Where does this Luna go? Well it gets distributed to stakers over time. I can tell you exactly —>
The 7.1M Added to the Oracle pool looks like this.

You can see the wallet that stores these fees here:…
Let’s appreciate this graph.

Dotted lines = expected Luna or UST left to distribute IF 0 fees were collected since Jan/22.

Note how actual UST (Red solid) is >>> expected UST (Red dotted)?

That is because SO MUCH UST was being minted which gave extra UST in fees!
Now note how Luna actual and Luna expected (yellow lines) were exactly parallel?

That is because no Luna fees are accrued during expansions!

But now we have a lot of contraction and high swap fees. This is juicy for the value of Luna.

How much? Let’s find out —>
Here we see the MINIMUM APR from staking Luna over the next two years (based on current Luna staked, 286M).

Unlike Ethereum or Polkadot, these are fees already captured by Terra, actual revenue, not some target inflation.
Here is the same Data in graph form. See how it curves up to the left?

That is because there are stablecoins (1.2B UST) being distributed to stakers.

So as Luna price drops, APR goes up because it has some non-reflexive rewards.

VS Ethereal staking is 4% regardless of price.
A valid counter here, is getting stable coins is useless if UST is worth $0. True, but the argument also necessitates the total failure of the blockchain which is a different argument. IF it does not fail this is the APR.

Either way, 0.05 Luna per Luna is 5% yield natively.

1 Luna staked will give you:

0.05 Luna + 1.8 UST per year. MINIMUM. Regardless of price.

So as Luna price drops that $1.8 becomes a bigger percentage vs Luna price!
Sanity check. Model says 15% minimum, actual APR is 20%. Why? Because MORE fees are coming in per minute. Hence the “minimum”.
Phew almost Done!

Next how long will it depeg?!

1.4B has exited over 33hrs or 1.3 days. Let’s check that with my death spiral prediction in my old post…
Based on actual average swap fee of 13.27% and the liquidity of the virtual AMM (the TerraDeltaPool) we should be able to burn $1.2B UST per 1 day.

We have burned 1.3B in 1.3 days.

This is good news because since model worked so far, it might also work to predict the future…
Quit rest-stop at the rest of the death spiral model.

Average price of Luna mint so far was $24 which predicted a 7.1M added to the oracle pool. Actual value was 7.1M.
So far so good!

Expected staking APR at $27 was 11%, but actual price at $18 so we get 20%.
Predicted = 1.2B UST per 1 day.

Actual = 1.3B in 1.3 days.

So our burn / day prediction is close enough so far! 20% off.

The ONLY question is how much UST needs to burn?

The following is an untested assumption and I need community feedback…
UST issuance is currently 17.14B. But the 24hr average price is $0.85.

I say 17.14B * ($1 - $0.85) needs to burn to return to $1.

That is 2.57B.

The fee tolerance for this is 15%.

Which gives us a re-peg date of 1.79 days.
What will Luna look like in such a scenario. Again it depends on the AVERAGE mint price of Luna. Here is the model for the total 3.871B UST we need to burn.

REMEMBER final Luna price is not necessarily = to average luna price. So APR might be higher or lower.
Length is totally depended on swap fee required and amount of UST that wants out. Here are 4 more scenarios.
This gives you a foundation. But it's only a small part of the story. Sentiment, meme, market feeling will all come into play.

The argument that Luna gets more valuable and that it hasnt broken yet Vs the market sentiment of 17b ust. Truly anyone's guess.

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Pedro Ojeda

Pedro Ojeda Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @pedroexplore1

May 12
I’ve always tried phrasing my threads as educational know hows, this one is going to have a lot of opinions.

1. Update on Luna/UST coins.
2. Update on Terra Chain
3. “Wasn’t this obvious all along?”
4. It is not all about the Tokenomics™

So much to say, such little sleep.
1) Luna/UST supply changes.

In terms of tokenomics, the game theory incentive came out. So much sell pressure and speed of contraction made Luna “a valuable asset” (as I erroneously called this scenario before).

Owning Luna now yields +999.99% APR (I think the dashboard maxed) ImageImage
We will table (1 for now). 2) The Terra Blockchain.

A backstop to UST contraction giving Luna sell pressure, is Luna inherent value, as any blockchain token, for the demand of its blockspace.

As long as Validators continue validating, Luna would not go to zero. Image
Read 29 tweets
May 10
46M terra have been minted, supply is same as a month ago.

1.2b UST have been burned.

Ddos like congestion for 2 days.

Nothing has survived such pressure!

But here we are, no death spiral, one block every 6 seconds, Staking high from actual fees.
Ust will trade bellow $1 for as long as the flow outward is bigger that the burnable amount.

The less patient people are, the more fees to stakers.

I'm sorry for those that wanted to buy the dip. I hope we have deeper liquidity in the future.
UST is trading exactly at $1 minus the current swap fee.

Arbs are making money.
Read 5 tweets
May 10
This should end the fud. Anchor update.

First bug reported, not a bug.

Second bug could be real.

Don't quote me. Usually I explain after official announcement as to not spread misinformation but since I missed 2nd one in previous tweet I need to lyk.

Deposits are safe.
Clarity lenses.

This bug only happens for the vast minority of liquidations when:
A) There aren't enough bids. Which there currently are.
B) There are only high discount bids >18%. Which is not the case, see Pic. $80M @ < 18%

Bug is not likely to be encountered soon. Image
Basically as found by @skgBanga

If collateral_value_liquidatable < amount_required_to_pay_loan

Then Fail liquidation.

Which happens when not enough bids or in previously mentioned liquidations.
Read 6 tweets
May 10
.@anchor_protocol has some insolvent loans.
1. Not a bug.
2. Anchor is not insolvent and deposits are safe for now.
3. The docs need some serious updating.

To preempt le-fud and for the sake of my DM’s here the explainer —>


Anchor loans are collateralized. When collateral falls in value, Anchor sells collateral to Liquidators.

Anchor allows Liquidators to buy the assets at a 1-30% discount in order to:
- incentivize Liquidators to buy this collateral quickly.
- Keep deposits safe.

Sales occur in an order book fashion from lowest discount to 30%.

Most of the time, competitions to buy these discounted assets makes the discount small, average 5-8%.

This is good because it means the proceeds from liquidations can fully cover the Anchor Deposits.
Read 24 tweets
May 10
Juicy fees accumulated in the Astro Maker contract. $350k market buy on ASTRO executed.

This transaction got through in 6 seconds after sending (one block). So I'm assuming all issues are with front-end syncers like terrascope, anchor and wallets being behind.
FYI Terra uses account sequences. If you broadcast a new transaction before the old one gets in a block, it gets rejected since your address' account sequence is not updated yet.
Read 4 tweets
May 9
There are two things that matter:
- The peg.
- The intrinsic value.

Everyone is confounding the two...the peg is fine, and the intrinsic value is in demand shock. 🧵

During market collapses, everyone wants stablecoins.
"INTRINSIC VALUE" or the current value of the stablecoins becomes >$1. No "de-peg".

In cases when everyone wants to buy QUICKLY: They market buy with their stables. These stables will have a <$1 intrinsic value. No "de-peg".
Why isn't this a de-peg?

Because the CURRENT intrinsic value ONLY measures two things:

- The effectiveness of arbitrageurs aka Market Makers.
- The depth of liquidity with respect to demand shock.
Read 20 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!


0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy


3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us on Twitter!