nexis206
commented
2 years ago signmessage 8Yv2DLTc8Yu8VV3ypC58V26wmRnJ7D6YuB "DFIP-2211-F-yes" H/deiyMjQN8lbOzrerM2j6+YQ7aW2/fPqiNfTWqKoJadVu/ExeAPgEk+2EPa55bM5kKKNw0fvLBAaeuVvZFS9Fo= signmessage 8S8PY5NFD1BsEwhJg1ddxNqptCJ68BgDKN "DFIP-2211-F-yes" IB04eFm0DS0y/WbimUHqe+7o5mcw+zRMrraK8u35YLnYHGwsJiBbUl0GfvG7tnmtVUSTWqYO6eMwLCoj0IhJMm4= signmessage 8Pw9mWvXJYsVZ92M1wNkTKP93SFbJP39PD "DFIP-2211-F-yes" HwxeQmsMMgQDENZ0cE/G9S5fQO5C5qnukBGx56dtcRdEL3lEpul2UdhO8JlteIRnRt8iKSnYWE7O8mUtHhRGedg= signmessage 8GaAZK5JUsBgANLfphU1gC5BEheU6r8t7X "DFIP-2211-F-yes" IH+vAvLgsU0oIdQoviWgg9+0kimfGuS+wme7o9CJAeKpaYgIgfHhKhz3LI2wEMcw/rX1XbCiGeyT+JEcGGNtD78= signmessage 8SixmmDzZqo4vzor1tg8btYLdXyqbJKuYT "DFIP-2211-F-yes" H9SWB5MlpfPiED44HP4SxF6eHFEXqA2j2KbZU+PKh3iHAqHrI6Ir7JLhuuHlMeh4IeeNO6xesgiNQxnJRJbu++U= signmessage 8bRXKNLWs9YRP32DD7hAbwUegoZRkN23mF "DFIP-2211-F-yes" IOPCxQnD0WUjNAbtLA6Q3+N+8fhkaPdjL78m6czquhRJdSn4OBNUINETlv7cMgXunoJozy51PqjnHp6wYJ0ZOtk=
carboncloudcomputing
defiaal
SilverSurfer2000
ghost
AESILDV
dfip-voting
wetsnakes
BleepMaster
ojbcrypto
mrgrauel
benameless
Belroyce
naiirad
Niecoo
WestendWood
RackiTacki
KennyKentucky
CHS-Tallamore
Sunnyville1234
Oldman12345
defichain-mn-8P24-krmb
Kaplan111
wmbst
seb18781
Anaconda7777
Ppavlov0405
pieceofcrypto
OKnr66
Buschfunker
Cockerooo
htimm1963
andreasideen
defidfi
gradionxxx
doc1970
Boulemaster
cosmox1797
slikskils
svenol
thephilippk
maxiimal
DrDrCat
NapierSG
keenmaker
Merc87
rmcookie
grmpfx74
CiTSLA
DFIP Overview
The history of DUSD showed that unlimited creation of "unbacked" tokens can lead to problems down the road. Currently the FutureSwap is unlimited which might lead to such problems. That's why I am proposing to limit the FutureSwap to a total volume of 10% of the average liquidity of the last 28 days in the corresponding liquidity pair.
The idea behind FutureSwaps
The FutureSwap was introduced to ensure a loose binding of dToken prices on the DEX to the corresponding oracle prices. With this mechanism users can either swap dToken for DUSD or DUSD for dToken, each with a penalty of 5% against the oracle price (and only once per week).
This does not create any immediate attack vector on the dToken system, but excessive usage of the FutureSwap might put the system into an imbalance which we do not want. In general the FutureSwap is designed to stabilize the system. It should only be used if the DEX price is in a strong premium or discount (>5%) for a long period.
Normally creation (and removal) of dTokens should work via loans. So a premium in the dToken should incentivize the creation of loans, while a discount should incentivize paying them back. Only if this mechanism fails, the FutureSwaps should be used.
The risk
Let's consider a pool like GME-DUSD with 2 mio in liquidity. So 1 mio $ worth of dGME and 1 mio DUSD. Now if someone would swap 100 mio DUSD into GME, they would not move the market in any way, but this user now has a 100 mio long position "against" the chain. This was clearly not to stabilize the DEX, but to prevent slipage and get a great entry price. Now GME pumps "only" 100% for any reason ("worst case" they push up the real GME price themself with enough funds) and they swap the GME back to 200 mio DUSD. So they made 100 mio DUSD profit, created out of the chain. This also creates a massive amount of algo DUSD which again hurts the dToken system as we are seeing right now.
The solution
If we limit the usage of FutureSwap to 10% of the average liquidity of this token in the corresponding DEX pool, this would be far less of a problem. In the example above they could only swap 100k DUSD per week, and since they would block the FutureSwap with that, the dGME might even go into a premium. And that's what a market should do when there is such a massive demand.
I also often hear the fear that algo-dToken might flood into algo-DUSD. With this cap we would limit this risk to a maximum of < 5% new algo-DUSD per week, because the max swap is limited to 10% of the total liquidity in the dToken pools (which is the same amount in DUSD). We usually have the same amount of DUSD also in the "gateway" pools, so DUSD in dToken pools are less than 50% of all DUSD, and that is not even counting the DUSD in vaults.
On the other hand a 10% cap per week would be enough to move the DEX by 21%. So more than enough to deal with any overboarding discount/premium coming from normal trading and inflow/outflows. Even a higher premium/discount would only take an additional week to be resolved. During this time the high premium/discount would be a strong incentive for vault owners to mint tokens and benefit from it. Which is a great benefit for the ecosystem again.
How does this DFIP benefit the DeFiChain community?
Limiting the usage of the FutureSwaps will prevent future exploits and unhealthy shocks to the system while still allowing the FutureSwaps to make the dToken system efficient. This will increase confidence and trust into the dTokens and therefore drive adoption and marketshare.
Non-obligation
I understand that vote of confidence for DFIP carries no obligations by any developers to implement the proposals. DeFiChain is a community projects. Pull requests can be submitted by community and reserved to be evaluated for safety and general community acceptance.