Gauntlet IST Nov 2nd 2023 Non-parity Event Analysis

Inter Protocol
1

Summary:

IST experienced a non-parity event on November 2, 2023. We hypothesize that the non-parity event was likely due to the increase in trading after TIA airdrop

Here is the IST price on November 2 from the Osmosis DEX price source. The price went above 1.01 at 10:31 AM UTC, peaked at 1.11 at 11:15 AM, and returned to below 1.01 at 12:09 PM.

Untitled-6
Untitled-62028×1198 155 KB

Here is the IST price on November 2 from CoinGecko.

Untitled-7
Untitled-71616×1022 77.8 KB

Here is the IST price on November 2 from CoinMarketCap.

Untitled-8
Untitled-81248×852 35.3 KB

As shown, the non-parity of IST was different between different sites (Osmosis DEX, CoinGecko and CoinMarketCap). Specifically, the Osmosis DEX price peaked at 1.11, whereas the CoinGecko price was the highest at around 1.03, and the CoinMarketCap price was the highest at 1.15.

The prices among the sites are different because Coingecko and CoinMarketCap may pull IST prices from different protocol pools. For example, Coingecko pulls the prices from Osmosis, Shade, Crescent, and Astrovault. Also, These sites may use different block timestamps for these transactions.

Deep dive into the IST Non-Parity Event

There are a few factors that may lead to a non-parity event for IST

  1. Market Conditions: If there are sudden changes in the demand for a stablecoin, it can affect its peg to the USD. For example, if there is a sudden surge in the buying of the stablecoin and there isn’t enough liquidity to maintain its price, then a non-parity event would occur.
  2. Liquidation: If there are a lot of liquidations in the vault on ATOM, it might increase the demand for IST.

We suspect some unusual trade activities on IST on Nov 2nd. However, getting an accurate picture of the IST trading data can be challenging.

From flipsidecrypto.xyz shows that the total swap in dollar value by date for IST spiked up after Oct 31st with an average swap in dollars of $11,000 for IST between 10/31 to 11/03. This spike corresponds to the TIA token airdrop in COSMOS, which took place on 10/31.

	Date.        Total_swap_in_IST
2023-10-30    $2039
2023-10-31    $12327
2023-11-01    $11824
2023-11-02    $6128
2023-11-03    $13499
2023-11-04    $2479

Untitled-9

We suspect that people were trading in IST from TIA. A closer look at the data from Oct-31 to Nov 2nd data shows a large increase in the IST from the following tokens: USDC, USDC.axl, USDC.grv, and JKL based on flipsidecrypto’s data.

Untitled-10
Untitled-101189×590 73.6 KB

However, we also found that the swap in dollars for IST was relatively lower on Nov 2nd (compared to the number on Oct 31st, Nov 1st, and Nov 3rd), around ~$6000.

From flipsidecrypto.xyz data, we found that the net swaps into IST on November 2 were ~$6k and the net swaps out of IST were ~$27k. Swaps into IST were also in mostly small chunks, less than $200 per transaction. During the non-parity event (from 10:31 AM to 12:10 PM), a total of $310 was swapped into IST. The following chart is the IST hourly swap in USD volume by the hours. We did not see much trading activity before or after the price spike. The chart below shows the swap in USD throughout the day on 11-02.

Untitled-11
Untitled-111005×585 47.3 KB

We also examined the swap data on Osmosis DEX. We have observed that there are a lot of intermediate trading for IST that is not captured in Flipdata. In the example below, we can see a trade where the swap-in and swap-out currencies are both OSMO. However, it went through several routes - OSMO/TIA, TIA/USDC, USDC/IST, IST/OSMO. In this case, the money went through the USDC/IST and IST/OSMO pools.

An example of a potential arbitrage transaction of OSMO/OSMO.

Untitled-12
Untitled-121768×488 79.7 KB

Based on the 11-13-2023 data from Osmosis swaps, we found that the total money that touched IST is $19k. However, the flipdata only shows a swap-in of IST at $3900. There is a big discrepancy between the swap data from Osmosis and Flipdata.

Based on the data from Flipdata and Osmosis swaps, we learned that:

  1. There is a huge increase in IST swaps from Oct 31 to Nov 3rd (5x higher than other days). The average size of the swap-in is around $200 and there was $310 in total during the non-parity event (from 10:31 AM to 12:10 PM). This is based on Flipdata.
  2. We believe that Flipdata is underestimating the demand for IST. From the swap data on Osmosis, we found out that the demand for IST is much higher as many routes will take IST in the transactions. For example, the swap value for IST was $19k on Osmosis vs. $3900 on FlipData.

Based on these two learnings, we hypothesize that the non-parity event was likely due to the increase in trading after TIA airdrop. There was a surge in IST demand from Oct 31st to Nov 3rd and the non-parity took place on Nov 2nd. There were a lot more tradings involved in IST in the pools which resulted in the increase of demand for IST.

Temporary price fluctuations for stablecoins due to low liquidity are not uncommon. The protocol currently has $367k of total collateral value locked. In our internal data, the 2% open liquidity for IST was around $2500 on Nov 2nd. If more than $2500 was traded in or out of IST, it could have led to a peg deviation.

There is no action required at this time, but we will continue to monitor the price of IST.

Improving Gauntlet’s Internal Risk Monitoring and Alerts System

At the moment, we have access to daily price data from Agoric. To further enhance our internal risk monitoring and alerts system, we’ll need to create a real-time data feed that can monitor such discrepancies. We have requested a list of real-time data API endpoints from the Agoric team.

2 Likes