The Problem:
Liquidity incentives are being gamed by liquidity "providers" buying and selling SD back and forth among themselves by using two accounts that they control and a large war chest (a large amount of STEEM or SD). There are many Steemit posts that already cover this problem, so I will not go too far into it. [1]
This results in 1200 STEEM being printed every hour and paid gifted to people who aren't actually providing useful liquidity to STEEM users (which is the intended purpose of liquidity incentives.) At the current market value, STEEM is paying $3,756 USD per hour, $90,144 USD per day, $631,008 per week, $2,208,528 per month or $26,502,336 per year for a broken feature. That money is mostly going to people that are trading among themselves, gaming the intended purpose of liquidity incentives, and not actually providing enough useful liquidity for the amount they are paid.
Obviously liquidity incentives are broken, and there has been word that the developers intend to fix the incentive system somehow (I have not seen any published details.) However I posit that no matter what reward-based incentive system is conceived, or what parameters are tweaked, that someone will eventually figure out a way to game it. Liquidity incentives result in a perpetual "cat and mouse" game, with the system constantly being tweaked, ending up being eventually gamed, then tweaked again (and so on and so forth.)
The Solution:
Instead of providing game-able incentives for market makers to provide liquidity to the SD market, I propose eliminating the incentives altogether and pivoting to an alternative way to provide liquidity in the SD market.
How It Works:
Basic Idea:
I am proposing that we revert the value of 1200 STEEM printed per hour for liquidity incentives to a decentralized and autonomously controlled fund that provides actual and useful liquidity to Steemit users. The fund would autonomously setup buy orders on the SD market an arbitrary percentage away from the median price feed and sell orders an arbitrary percentage above the the median price feed.
Balancing The Books:
In certain market conditions, it is plausible that there may be one side of the books whose value heavily outweighs the other. For instance, in a bear market the SD sell side may be much thinner than the buy side, or in a bull market the SD buy side may be much thinner than the sell side. This results in the larger side sitting there mostly useless on the books. There are two ways that the autonomous liquidity fund can offset this. Changing the asset that is printed for liquidity (SD or STEEM), or changing the percentage the liquidity is sold from the median price feed. By using both tools at Steem's disposal, eventually equilibrium will be reached.
Unlike the currently implemented liquidity incentives in which 1200 STEEM is printed a hour, STEEM can autonomously decide whether the same value (1200 worth of STEEM) should be printed in the form of SD or STEEM. The funds printed for the purpose of liquidity should be printed in the form of whichever side of the books is thin. If the SD sell side is thin then the funds should be printed in SD, or if the SD buy side is thin then the funds should be printed in STEEM. Alternatively, both SD and STEEM could be printed proportionately to the weight on each side.
Alternatively, if one side of the books ends up being disproportionate to the other, then the arbitrary percentage the SD/STEEM is being sold away from the median price feed should be reduced on the side that has amassed the greater amount of value. In theory, the closer to the median price feed the SD/STEEM is being sold at, the more likely it is to be sold to third part market makers operating closer to the median price feed than the autonomous liquidity fund.
Extreme Market Conditions:
If the books are not automatically balanced by slowly inching the sell orders towards the median price feed, or printing 1200 worth of STEEM in either STEEM or SD, then extreme measures can be utilized to balance the books. Extreme measures such as selling at (or slightly below) the median price feed can be taken if extreme market conditions occur. Selling at the median price feed until the books are balanced may suffice, but I posit the losses from selling slightly lower than the price feed (to even the books in extreme scenarios) would still be much less costly compared to the value that SP/STEEM stakeholders are losing via the currently implemented game-able liquidity incentive program.
Combating "Whale Gaming":
If the autonomous liquidity fund is being sold too close to the median price feed, then it is possible for a whale to buy it up and then proceed to sell it at a higher percentage away from the median price feed. If a whale had enough ammunition in their war chest (a large amount of STEEM or SD), then this could prove to be a lucrative opportunity for the whale.
To combat any such "whale gaming", I suggest that the arbitrary percentage the SD/STEEM is sold away from the median price feed start large and slowly creep toward the price feed once an arbitrary (but sufficient) amount of value is built up. For a maximum amount of flexibility, that arbitrary percentage for both sides of the books should operate independently from the other. Eventually, the more SD/STEEM that is built up in the autonomous liquidity fund (it is receiving 1200 STEEM per hour,) the closer the autonomous liquidity sell orders can slowly inch closer toward the peg. The spread will therefore be lessened over time as funds are built up further away from the median price feed.
By selling the initial autonomous liquidity a larger arbitrary percentage away from the peg, this eliminates the profitability of a whale buying it up and selling at a higher spread. For instance, say that arbitrary percentage is 15% away from the median price feed and a whale buys it up. To make any profit from this trade, the whale must then sell the funds at over 15% away from the median price feed. At a certain percentage away from the median price feed, no reasonable person would buy or sell STEEM/SD that far away from the peg as they would be losing too much money in the trade. This leaves the whale, whose intent was to game the autonomous liquidity fund, stuck with SD/STEEM they paid too much for and cannot sale. Their only way out of their position is to sell it at a loss.
Eventually, as the value/amount of SD/STEEM grows at the highest determined arbitrary percentage away from the peg, future STEEM printed for autonomous liquidity purposes can be inched closer and closer to the peg. As the value/amount of both sides of the book builds up, the amount that is on the books will be too much for any whale to profitably "whale game", even as the spread is lessened. Over time as the autonomous liquidity funds build up, the market spread will be less and less.
References:
1. Steemit posts covering the issue with liquidity incentives being gamed:
If you like this proposal, or want me to continue brainstorming and posting solutions to Steem-related problems, then please consider reading and voting for my other proposals/blogs:
Formulating The Optimum Cryptocurrency Speculation Theory:
STEEM PROPOSAL: Offsetting Steem Power Debasement Through Unobtrusive Advertisements:
STEEM PROPOSAL: Implementing Third Party Certificate Authorities To Combat Sybil:
#proposal #proposals #steemit-ideas #steemit #liquidity #liquidityincentives
I just thought of another benefit of autonomous liquidity as compared to liquidity incentives.
It is seemingly obvious, and I am not sure why I did not think of it before. I can no longer edit the original post. The HUGE benefit is the following:
SP/STEEM stakeholders are not debased (diluted) with my solution as compared to any liquidity incentive scheme you could come up with. The STEEM/SD printed for autonomous market making liquidity stays on the on-chain books for eternity, for the purpose of market making. It never makes its way into the STEEM off ramps as downward pressure on the market.
I have no idea why your post or proposal didn't garner more attention. It should have. This liquidity reward thing needs to be fix as soon as possible. It's ridiculous. Steem is such great. This liquidity reward shouldn't be complicated to fix compare to everything Steem has accomplish.
Thanks. I think I have brought more attention to the issue, and hopefully my solution can be of some use to the developers. I am sure that it will be fixed.
The autonomous trader you propose is similar to the liquidity bot I developed and ran for a while on btc-e and coinbase. I was going to implement it on STEEM, but realized that the economics of the current liquidity rewards mean that the inflation there is going to outstrip any gains I'd make. In my opinion it's not worth trading right now, and I pulled all my money back out.
Very well thought out, I can see you put a lot of time into this and it sounds sound to a laymen like myself. I think it should have gotten more attention and discussion, maybe repost it?
Thank you. Yeah, I am not sure why it didn't receive much attention. I posted it as a Github issue last night to make sure the developers are aware of it. Hopefully if they find it valuable they will just up vote this one. It is kind of shocking that the people that are paying for this broken feature don't care more (SP shareholders... judging from upvotes).
keep posting...
Thanks for your encouragement on all my blogs. I really appreciate it!!
Holysh! Great post. This all really is like a complex math equation, made more complex by the unpredictability of people.
It is kind of complicated, yet fairly simple at the same time once you can grasp all of the moving pieces. Thanks for the encouragement!