bAsset Usage
Terra Bounty 65 - What is the number of users who are using Anchor to both borrow against their bAssets and place those borrows directly in Earn? What is the average leverage LTV risk they are taking in doing this?
Anchor Protocol has two sides of passive income at the current moment.
- EARN - One can either deposit UST and earn a stable 20 percent yield
- BORROW - One can bond a collateral like LUNA or ETH, then borrow to capture yield in the form of ANC governance tokens.
As both sides and good yield farming opportunities, a saavy investor would attempt to play both sides of the Anchor protocol to maximize their yield. So he would deposit a collateral into the Borrow side and then borrow UST stable coins, which are then deposited into Earn side of the Anchor Protocol.
This has a couple of advantages :
- By locking a value-appreciating collateral and then borrowing against it to take advantage of the EARN side, the user ensures they don't miss out on the gains of the said asset. The alternative to this is to liquidate the asset and then deposit the UST.By playing both sides, the investor is able to capture price action gain, while also using the borrowed stable coins to have a stable investment.
- Anchor BORROW pays out consistently above 10 percent APR on the borrowed amount. This combined with 20 percent APY stable interest on the deposited amount. Say on average an investor maintains 30 percent LTV, this means, he is able to get 30 percent return on 30 percent of his locked Collateral. This means he manages to get atleast 9 percent APY. When the BORROW side is stressed due to less amount locked, its APRs hit 200, giving the investor a much better alternative to delegating LUNA for around 5 percent APR return.
The Anchor protocol defines a money market between a lender, looking to earn stable yields on their stablecoins, and a borrower, looking to borrow stablecoins on stakeable assets. To borrow stablecoins, the borrower locks up Bonded Assets (bAssets) as collateral, and borrows stablecoins below the protocol-defined LTV ratio. The diversified stream of staking rewards accruing to the global pool of collateral then gets converted to stablecoin, and then conferred to the lender in the form of a stable yield.
In order to keep it simple, we will make a few assumptions and rubrics to define a certain LOCK COLLATERAL - BORROW - DEPOSIT event as farming.
- All events must take place within a time gap of 30 Mins atmost. Any trio of events that are separated by more than 30 mins between the two , will not be included. This will
- Borrow and Deposit value must be comparable (within 20 percent of each other in this analysis). This means when an user, for example borrows 100 UST, he must deposit back atleast 80 UST or atmost 120 UST
- LTV must be below 60. This ensures that borrow is purely from a freshly Locked Collateral, instead of previously Locked Collateral. Though this is detrimental to the analysis, as it eliminates users who keep assets locked all around and borrow and deposit occasionally to farm, it allows the easy capture of LTV.
- The collateral will be bLUNA alone, as bETH data is relatively small.
All data for this analysis is taken from FlipsideCrypto's Terra tables.
- For Locking Collateral transactions, the
terra.msgs
table is used to find interactions with theANCHOR OVERSEER
contract -terra1tmnqgvg567ypvsvk6rwsga3srp7e3lg6u0elp8
. - For Borrow and Deposit transactions, the 'terra.msgs' table is used to find the interactions with the
ANCHOR MARKET
contract -terra1sepfj7s0aeg5967uxnfk4thzlerrsktkpelm5s
Thus from the FlipsideCrypto's Terra tables, we can retrieve all events that are either Locking Collateral, Borrowing UST or Depositing UST. Then we can join them by their comparing the timestamps, and users (a.k.a. senders)
Comparing the total volume of these transactions on the daily basis, a similar volatile ranging oscillating value is seen. One must remember that these graphs are on log-scale, hence the actual volatility is under-visualized.
Beginnning of September has seen an increase in the volume these leveraged farming methods.
Though we did ignore a decent chunk of farming activity by enforcing a LOCK-BORROW-DEPOSIT rubric for the farming, we seem to gathered a good chunk of activity.
All the plots indicate a general stability inspire of short term volatility in the values until last few months.
The stability could be attributed to smart farmers keeping their farming positions healthy, while the uptrend in the last few week could probably attributed to inflow of new users in the Terra Ecosystem as the entire Blockchain space gears up for the next leg up the bull market.
First we will compare the Average USD value of Locked Collateral with Borrow and Deposit, per transaction.
Evidently, everyday on average 1000 USD per transaction is being put into Leveraged Farming in Anchor. This average is volatile, but steady generally. Only very recently we are noticing a surge in average USD value per transaction.
Comparing the number of these farming events with the price of LUNA, we can see a higher number of farming events during the dips. This is especially because, during LUNA dips, Liquidations and Low Staking yields implies there is a strong pressure on the BORROW side of Anchor to keep up with the EARN side. This results in increase in ANC emission, which increases the incentives for borrowing. So generally higher number of farming events occur during these low-value times.
Also, in the graph above, we can notice a sharp decrease in these Farming events since the start of the protocol. This is due to the third assumption we made during the selecting of Farming events. We miss out on potential farming activities, in order to calculate the proper LTV of Farming events.
This however suggests that there are a good chunk of farmers who Lock collateral all around to ensure they take advantage of such Farming opportunities.
Comparing the LTV on average farming event to the LUNA price is interesting. A stable LTV of above 30 percent is consistently maintained. This is true even before the MAY crash and liquidation spiral, when a farmer can borrow upto 60 percent LTV. This could mean one of the two :
-
most farmers seem to understand the protocol and the possible down side to over-leveraged positions, thus keeping their positions healthy.
-
or these are smash-grab farmers, who don't farm for long periods.
However, once again we only see a rise in these values during late September.
We can see the top 20 farmers, have an almost similar average USD value invested in these farming practices. On average these elite farmers have deposited collateral, taken borrows and deposited UST of value upwards of 5 Million USD. These are rich farmers indeed.
Once against inspite of the log-scale graph, we can notice that these farmers have total invested north of 20 Million USD into farming activities.
Doing some simple math ( 9 APR ~ 0.025 DPR(daily)) , they must have atleast gained 2500 UST via farming activities, if they had farmed for a single day.
This value is the bare minimum with a 10 percent APR on borrow. Imagine the yields post the May crash when the BORROW APR was north of 150. Roughly 10,000 UST in farming would have yielded around 100 UST daily in ANC
In general the top 20 farmers seem to have very few farming events. This implies almost 10-20 Million worth of collateral was deposited in a couple of events per user. This implies that these farmers were taking advantage of BORROW APR.
In the below plot, we will select top 20 users based on number of farming events. Comparing the graphs above and below, we see that high volume farmers seem to have fewer interactions that others.
We will also do a quick top 20 user metric. The top 20 users are selected according to the total value of USD employed in farming.