This is meant to serve as a post mortem for the recent expansion from December 8th to December 27th and the following contraction cycle that followed. My hope is to spur some useful discussion so we can continue to grow as a community and improve ESD’s protocol. I encourage you to review this information, study it, come up with your own conclusions/objections and post them here as a record for the future.
The data
Expansion happened between December 8th and December 27th. This graph shows the price during that time period. The red arrow indicates when all coupons were redeemable and the DAO started earning rewards. Notice that the time up until the red arrow shows that the price is stable during expansion but following, the price is much more volatile.
Now look at this graph of the percentage bonded vs. LP vs. liquid supply. The red arrow shows where expansion begins on December 8th. Notice that both the amount of liquid ESD and LP increases in the period after the red arrow. This is because the DAO isn’t paying rewards yet and the only way to earn rewards is either through coupons or LP. The orange arrow indicates when the DAO started paying out. Notice that LP and liquid ESD ratios dropped significantly and the DAO bonding ratio increased.
The black arrow indicates a massive exit from the DAO on January 1st which is exactly 5 days after the TWAP dropped below one dollar on December 27th. This correlates perfectly with the 15 epoch waiting period to move funds from bonded to liquid in the DAO. Also notice that this mass exit from the DAO happens in thin liquidity. In fact, liquidity started thinning immediately after TWAP < 1 on December 27th.
This graph shows DAO bonding behavior vs. the TWAP price during the same time period from December 8th to December 27th. The red arrow shows the time when DAO rewards started becoming available on December 13th where ~30M ESD was bonded to the DAO. Comparing this graph to the one above shows that the drop in LP percent on December 13th directly correlates with the massive demand to bond. The green arrow shows a large unbonding event on December 27th when TWAP < 1. Notice the amount of ESD that’s being moved out of the DAO. It’s about 50M! It’s hard to ignore the fact that the price rises dramatically following the red arrow and drops dramatically following the green arrow. Finally, notice how volatile the price is starting at the red arrow all the way through the green arrow and beyond…
Now let’s look at the price post expansion and see what the outcome was from all of this unbonding and later selling during thin liquidity (as I showed above in the graph showing the percent of LP). As we can see, over this short period the price of ESD was quite volatile. It went as low as $0.83 on December 29th back up to $0.92 on December 31st and then as low as $0.68 a week later!
Now combine all of this data with the recent argument that @sabertooth makes in the article "Uniswap is the reason why algo stablecoins can’t maintain the peg and how to fix it". After reading this article, please refer to the graphs again knowing what you know now. What conclusions can you draw from this information?
Conclusion
There are many factors that led to massive expansion in December including new farming opportunities, new integrations, a hot crypto market where BTC was reaching all-time highs almost everyday and more. The list can probably go on and on and I won’t attempt to analyze all the reasons why our current price point is 0.30 (as of this writing). What the data leads me to believe is that our DAO incentives, DAO locks and our reliance on Uniswap are a significant factor in ESDs volatility. Large demand to bond to the DAO significantly decreased the ratio of LP, leading to more price volatility. The subsequent drop below TWAP led to thinner liquidity and a large movement to unbond from the DAO. Flush with cash and uncertain about the future, many people sold their ESD to make a profit which led to more price volatility and the price we see today. In the end, those that remain invested in ESD have to clean up the mess.
To end this on a positive note, I’m excited about where ESD is going. The community is incredibly active and we have a really smart team of people working to improve the protocol. V2 aims to solve some of the issues I’ve discussed here and I have no doubts that we’ll continue to improve. Please share any ideas/thoughts that you have.
References
https://explore.duneanalytics.com/dashboard/empty-set-dollar https://explore.duneanalytics.com/queries/18283/source#37015