[For Discussion] Limiting Access to Automated Vault

129

No that is completely incorrect. Based on what Samsara said, you would need 20000xAlpaca to deposit 10000 $ into a single vault. You would eb able to depsoit 10k $ into each vault.

One full AV is 25 million. When looking at 8x AVs, this means its 25/8=3,125 million $ to be doposited + 21,875 million $ borrowed. This means that you would need 3.125*2x=6.25 million xAlpaca to fill one 25mil automated vault. As we have currently approximately 40million xAlpaca locked in governance vault, this should not create a huge buy alpaca to put into governance vault trend.

This is just an extra utility that is provided to xAlpaca holders and might only toss in a small extra percentage to increase Alpaca locked at current 2xalpaca = 1$ ratio.

1 Like
130

Thanks for your reply. You are correct that Sharpe ratio measures return per unit of volatility. Increasing the fee doesn’t change return’s volatility.
However, increasing the fee will reduce the expected return of the vault. As in the formular you attatched, increasing the fee will reduce the Rp, it will not impact the Rf and the volatility, so the Sharpe Ratio will drop. That’s my point.
image

131

Thanks for sharing this. My view is that such calculations are likely to systematically underestimate the capacity of AV that can be opened for mainly 3 reasons. And because those two parameters (APR current and Target APR in the table) are very sensitive. A change of one percent can bring about tens of millions of TVL capacity changes. So I suggest that the team should further scrutinize this capacity estimation.

  1. In the estimation, the target APR is calculated based on the 3x pool, which will result in a small calculated capacity. Regardless of 3x, 5x, 8x, investors basically only look at the final APY. So if you calculate with 8x as the starting point, that is, if most of the pools are 8x, the capacity will be greatly increased.

  2. The calculation assumes that the reward of each pool is fixed and will not change as the capacity of the pool increases. In fact, the total amount of Farming reward (the number of cakes in each pool) is indeed constant, but the trading fee will increase with the increase of capacity. Because DEXs are passive pools, the larger the pool, the larger the transaction volume is required to move the the price of the coin (BNB, ETH, etc) to the market level. Therefore, the reward of the pool is underestimated, especially for those pools with a relatively high trading fee portion.

  3. The borrowing cost is overestimated in the calculation, which leads to the overestimation of Target APR. Stable coins like BUSD or USDT may need an interest rate of 10%~12% to attract depositors, but another part of the borrowed assets are non-stable coins, such as BNB or ETH. And the higher the leverage multiple, the higher the demand for the unstable portion. For example, 8x AV, non-stable coins will be almost half of TVL. For non-stable coins, we don’t need a interest rate of 10% to 12%, 6 to 8 may attract enough deposits. Therefore, the target APR does not need to be so high. If it is lowered a little, the return to the investor will not be less, but the capacity of the pool will be greatly increased.

1 Like
132

Ok, so if xAlpaca requirements are based on equity deposit and not on the TVL, it means a target of around 23/24M xAlpaca to sold out the BNB-USDT vault PCS if 50% of the estimated max additional TVL go to the AV x8, like before.

$188M estimated max additional TVL for the BNB-USDT - PCS vault.
= $94M additional TVL for the AV x8, if half of the additional TVL go to it like before.
This TVL is composed of around $11,8M equity deposits and around 82,2M borrowed leverage.

And, with 2 xAlpaca requirement for each $ of equity deposit allocation in the 8x vaults, it means between 23/24M xAlpaca to sold out.

Well…

There is actually 48M of xAlpaca and counting.
if AV x8 sold out with 24M xAlpaca, every xAlpaca holder won’t be served…

And it bring me to my principal concern :
Regardless of the assumptions about the number or xAlpaca required for each $ of allocation, this new rule doesn’t benefit to all xAlpaca holders in the same way.

If you have a lot of xAlpaca but you don’t have stable coins to put in the AV 8x, you won’t benefit from this restriction.

It’s really different from the others sources of revenue for xAlpaca holders : FAQ - Alpaca Finance

Each of precedent rules was benefiting to each xAlpaca holder in the same way, burn, redistribution of fees, there is no difference of treatment between holders.

And I think, this restriction doesn’t maximise the global revenues neither :

If there is to much demand for AV vaults,
It’s because you restrict the offer by choosing to lock the APY at the expense of vault capacity.

Please don’t answer to a problem of centrally fixed price by quotas…

In place you should let the market decide which capacity he wants by letting the the APY fluctuate freely.

Maybe the APY of the AV would go down progressively to 15 - 20% and you will capt billions of TVL in place of two hundred millions.
Anchor has 15 Billions of tvl with an APY of 19,5%
And Luna holders have been happy :slight_smile:

In my opinion, letting APY and Capacity fluctuate freely would be fair and efficient. The market would be way bigger, and every xAlpaca holder would benefit it, not only the ones in the AV.
And there would be no moral hasard with the first depositors in the AV. Being first they benefited of high APY which is fair, but there is no free lunch nor special rules for the chosen ones.

And if the lending market isn’t big enough, it’s not a problem, let fluctuate the borrowing rates when usage is too high.
BNB lenders will come.

Last point, the leverage impact differently the market if it’s x3, x8 or anything between.
With high leverage people hurt more the lending market and APY of the vault with less capital.
You could put different fees structure based on the amount of leverage.

All in all, on reflection, I am against this restriction to xAlpaca holders.

I strongly prefer that the maximum capacity of AV be set by the market letting rates fluctuate freely and capacity uncapped.
I think this is what would benefit all xAlpacas holders the most.
Your automated vault is a wonder, let it run at full capacity, the market will tell you when you have reached it.
Billions in TVL, not millions, it’s all I wish to Alpaca :heart:

2 Likes
134

Am I the only one that though “what if Alpaca price shot up because of demand?”
Would it still be 2 xAlpaca per dolar? It isn’t really efficient.
What if the price of Alpaca goes to 5 usd? We need to buy 1000 Alpaca (5k dolars) to lock and can only put 500 usd into AV?
I think it should be volatile with the price of the Alpaca token. - Maybe something like a moving average so it does not spike too much and become able to be manipulated

135

Adding to that we only emit 100m Alpacas till the last batch and we also burn some, so the maximum AV usage would be 100m/year (not considering burnt tokens)? Not really efficient aswell, what if the protocol really expands because of this move?

136

And this would actually make Alpaca be REALLY volatile because pretty much ALL Alpaca tokens will be LOCKED and not in a LP or in CEX to make the price fluctuate a little less.

I think a different tier would be better because it would not strech Alpaca tokens too far:

  • Blank Tier: no xALPACA - can use x3 vaults
  • Green Tier: at least 1,000 xALPACA get to use x4 vaults
  • Bronze Tier: at least 25,000 xALPACA get to use x5 vaults
  • Silver Tier: at least 50,000 xALPACA get to use x6 vaults
  • Gold Tier: at least 75,000 xALPACA get to use x7 vaults
  • Platinum Tier: at least 100,000 xALPACA get to use x8 Vaults
2 Likes
137

I still think that this lock up Alpaca in order to use high leverage 8x AV is something that will benefit only actual xAlpaca holders and not new ones.

Because if you are in the game for stable coin farming, you really want the closes to zero risk as possible, so you will going to use 3x for sure, because for the 8x you need to take a huge risk and lock up x amount of a token that we all know that can pump and dump in hours like every other token in defi plus this crazy market conditions.
So the risk for new people in order to use the 8x will be a lot higher (low risk in the 8xAV + high risk of Alpaca price in this crazy market conditions (imagine we enter in a bear market with your token locked, you are pretty much FU and need to wait months/years to price recovery).

In the case of new investor that looks to farm stables 3x will be the way with this proposal, 8x will be only for the people that already hold xAlpaca, new ones will need to take a huge risk if they want 8x, so i bet that not much new people will go 8x way.

If you really want to push the idea of hold xAlpaca in order to use 8xAV, Why not implement a privilege period for xAlpaca holders, every time a new AV is live, they will have 24hrs to fill it, after that is open for all…and please consider that pancake can increase the rewards for their LP, so limit the amount of AV because of the rewards split should not be relevant if is well managed, lets push pcs to do that :slight_smile:

Hope this comment gets well receibed, thanks.

1 Like
138

If you deposited 10,000 USDT it will count as 80,000 in the vault space. But the xAlpaca requirements will count on your initial 10,000 so 20,000 xAlpaca (for simpler understanding) is needed

139

That is still incorrect.

The denominator is standard deviation of the portfoilio’s excess return. So as you pointed out, When Rp reduces, denominator would also decreases proportionally.

140

Thanks for the detailed outline.

As mentioned, this is a high-level estimation to get a sense of magnitude of capacity the vaults can handle. In reality, once we have added additional $200 million in capacity, if there are still room for more addition, there would be no reasons not to add more.

4 Likes
141

Vault participation based on xALPACA weight would be on a per-vault basis. There is no hard limit to AV usage other than the lending pool capacity and perhaps the desire to keep a somewhat attractive APY.

142

I don’t think you are right.

The standard deviation is a measure of the volatility of the return.

When Rp reduces, because it’s a consistent reduction, so the volatiltiy of the return (the denominator) will not change.

Just like you move a time series down along the Y-axis will only reduce the average (mean) of it, but not the volatiltiy. You still have the same ups and downs of the time series.

143

This is a great proposal.
Given the limited amounts that can be allocated to 8x pools, the higher demand and the revenue not being affected it absolutely makes sense to give xAlpaca another awesome utility.

I don’t understand why anyone in the governance would be against this.

A whale scooping up Alpaca tokens to allocate more on AV should be welcomed by the governance.
Why would you be in favor of whales allocating in public open AV if the pools are projected to be full anyway?

Public pool money might have no interest in Alpaca whatsoever! And so seem some of the posts in this Thread.

I am a big Alpaca believer and super excited about what AV will bring to Alpaca price itself.
This is also the main point here.

If such proposal brings value to the token without affecting revenue. It’s a clear winner.

3 Likes
144

People who don’t feel comfortable locking and supporting Alpaca token for long term would still be able to allocate to 3x pools.

The 8x would be filled anyway.
Don’t you see it does makes sense to then give another utility to the token itself?
This will feed back to the price of Alpaca. It will also rewards fixed long term holder for the “risk” you said they are taking.

To me the real risk here is for people to not invest in Alpaca itself

1 Like
145

Another remark,

The APY in the AV is the same for everyone because the borrowing rate is fixed at around 12,5%. Correct ?

What happen if utilization go above 85% ?
Borrowing rate go up ?
APY go down ?
And it impact everyone the same way in the AV ?

In consequence, new BNB lenders will come (or AV depositors will start to withdraw).
And utilization will go back to normal, borrowing rates too and APY will rise back to precedent level.
That’s the plan ?

Or you will stop adding capacity to not cross 85% utilization ?
But how to expend the lending market if you don’t let the AV capacity raise and the demand pay for it ?

146

I’m all in for this proposal. But I would go for a Linear allocation. I think only people who don’t have or don’t want to buy $ALPACA have a problem with limiting access. So in that sense who cares…

1 Like
147

I’m a big fan of Alpaca, I own a substantial amount of Alpaca, and I’m against this restriction, let me explain you why :

Problem : Demand for AV >> offer

Why ?
Because capacity is capped to not let the APY go down.

It’s a classical economic problem : Fixing prices creates shortages

Wrong solution : imposing xAlpaca quotas, it’s the Venezuela way…

Good solution : uncap the capacity and let APY fluctuate and the market inform Alpaca what it wants.

Presently, Alpaca don’t know what max capacity the market wants because they don’t let APY fluctuate. It could be billions !
For exemple Anchor raised 15 billions of TVL with an APY of 19,5% !! And Luna holders have won amazing amount from this.

Protecting a ridiculous high APY (>110% on x8) by limiting the capacity of AV, benefit to the depositors for sure, but will make Alpaca miss billions of TVL that would have benefited ALL xAlpaca holders, not just those who deposited in the vault.
Anchor did not put any capacity limit to their vault nor impose any Luna holding to access.
They attracted billions of TVL, Look at the Luna price action…
And they will let the APY fluctuate.

This is a much better strategy in my opinion.

New problem :
BNB lending market may not be big enough to new borrowing demand if AV capacity is uncapped

Solution :
let raising the borrowing rate when utilization is above 85%. With better APY, lenders will come.

Last point :
I don’t see the utility of the AV x3 nor providing anything less optimal than Alpaca best automated vaults aka x8…
The risk is the same, it require more capital, it provides lesser returns…

It’s not a problem for Alpaca to let people borrowing money on the lending market.
It’s the main purpose of leverage yield farming !

Depositors on AV x8 will be willing to pay really good rates to the lenders, let them pay high rates and attract more lenders on Alpaca.
Other LYF strategies won’t be competitive anymore.
But it’s not a problem if you let the capacity of AV x8 uncapped, they will switch their TVL to it until the APY reach an equilibrium with the others farms.

Morality :

Please, don’t limit Alpaca market with fixed prices and quotas.
I think Alpaca has the best automated pseudo delta neutral strategy in the world !!
It’s a blessing ! Please let the market use it at full capacity and use this big demand to expand your TVL and your utilization metrics. In one word, let your best strategy expand Alpaca market as far as it can be.
Doing that you will provide best APY on the lending market, best APY on pseudo delta neutral strategies, you will attract billions, not millions and Alpaca will worth billions not millions !
:llama: :heart:

3 Likes
149

The key word here is standard deviation of excess return

150

I dont think you realize how AV works and what its limitations are. If they wouldnt control the cap on them, then there might be potential need for liquidation down the road, and they wouldnt be “automated” anymore as people should close them before hitting red numbers fast. You should realize that current 1xx APYs are not real APYs those vaults will achieve. The APY you see is calculated based on last 7 days from trading fees, so when there is huge trading in the pair you get higher APY and temporarily higher rewards, which lasts usually for few days, then there might be days where you get 5% from fees cause nobody traded. In 8x AV you should expect realisticly approximately around 35-75% APY based on the market conditions and BSC attractivity in time. If you choose to medle with the numbers the way you suggested you could reach less (even much much less) than 20% which is not desired from such a strategy and could endandger the AVs itself. SO where would be the benefit? Temporary increase in revenue than big loss? not sure how that would be worth.

On the note of comparing to luna and UST, you should realize its two completely different models and Terra is not utilizing anything and their yields are based on completely different model of borrowing and holding yield bearing assets which offset the high yield they provide, while still not generating any revenue for the platform from itself. They are leveriging their basic principle of minting UST from Luna which is increasing due to popularity on price, and once minted it has the algo-peg. They are replensihing the reserves from such minting with xx millions so they dont run out, and can actually afford the 19%. Here in Alpaca it is selfsustaining solution which doesnt need to be saved with some additional money. I am not saying terra model is bad or that it cant work, but it cannot be applied here under any circumstance.