Cost of Production
Meter uses dynamic block rewards to achieve stable cost of production.
Moore’s Law says the amount of energy needed for the same amount of hash power drops by half every 18 months (= 540 days). Assuming a gradual upgrade on mining equipment, the future efficiency of sha256 mining is calculated as
Efficiency(Day) = StartingEfficiency * 0.5 ^ (Day / 540)
Whatsminer M20S, a sha256 mining equipment currently used in mainstream, operates at an efficiency of 49J/Th requiring 3360W power to perform computations at 68Th/s when run on full capacity.
Assuming this is the industry average for efficiency today, let’s plug in numbers to our formula (Day=0, StartingEfficiency =49 J/Th) to determine the amount of computation needed for 1 MTR today.
To consume 10 kwh electricity one module of M20S needs to run for (=10 kwh / 3.36 kw) 2.97 hrs at full capacity performing (=2.97 hr * 68 Th/s * 60 s/min * 60 min/hr) 728,570 Th of computation.
Today (day=0, efficiency=49j/th) the amount of computation needed to produce 1 MTR is 728,570 Th. Unlike BTC, the amount of computation needed for 1 MTR is fixed regardless of the amount of hash power in the network.
However based on the formula above, tomorrow (day=1) the computation required will become 728,570 / ( 0.5 ^ (1/540) )= 729,505 Th.
The increase in computation needed is expected to be offset by improvements in sha256 mining efficiency resulting in a constant cost of production around 10 kwh electricity
A price of MTR less than 10 kwh electricity expense implies that it’s cheaper to buy MTR on exchange than to produce it. It’s not hard to see production to cease below this price and supply growth to become zero.
However before naively assuming that production directly starts at any price above 10 kwh electricity expense we need to also take into consideration the opportunity cost of not mining other sha256 coins (mostly BTC). I would then assume production to start upon following condition
MTR price — Cost to produce — Opportunity Cost of MTR mining > 0
Opportunity cost is a subjective value. Depending on their risk appetite, size of pocket and perception of future value of BTC, some miners will continue to mine BTC for extended amounts of time even when losing money in hopes to capitualize on their holdings in future when BTC price rises. However besides extreme bull cycles (in late 2017, many manufacturers didn’t have enough inventory to sell to the market during the bull market) there will always be some miners remaining out of play because of their inability to afford running costs as a result of ever increasing dilution in hash power and/or inefficient operation (= zero opp cost)
Below table demonstrates all in electricity costs ($/kWh) of BTC miners
Assuming Meter will appeal first to the segment with highest electricity cost (segment with little to zero opp cost), we reach to a safe low end of production cost at (=0.07 usd/kwh * 10 kwh) 0.7 USD. It’s very unlikely for new MTR to be produced when price of MTR reads below this figure.
Stability
In Meter design there is a correlation between hashpower and the need of new currencies in the system. In that sense, instead of an exchange rate targeting used by many stablecoin projects or inflation targeting common in nation based currencies, the stability target in Meter is a monetary targeting; that is reaching a consensus on how much money supply is needed to achieve certain price level assuming predictable velocity of money
1st layer stability mechanism
Fundamentally, all stabilization techniques are based on the elementary economic model of supply and demand. The price of a currency can be modeled as the level at which its supply and demand meet each other on the market.
Following this logic, it’s not hard to foresee how things will evolve during periods of high demand; Price of MTR increases -> miners profit increases -> demand for hash rate in the network increases -> total hash rate increases -> supply growth increases -> supply increases. Eventually supply outperforms demand getting price back to equilibrium
On the contrary when demand decreases -> price decreases -> miner profit decreases -> demand for hash rate in the network decreases -> total hash rate decreases -> supply growth decreases. Eventually supply growth hits zero (expected to be at price level 0.7 usd as calculated above) at which point 1st layer mechanism reaches it’s limit and total supply remains constant.
In such a situation, in order for price to recover and get back to equilibrium, total supply must start to decrease. Moreover, the decrease rate of supply must be higher than that of demand.
2nd layer stability mechanism
Before diving deep here lets make a short recap on Meter’s two token hybrid PoW/PoS design. Unlike in Bitcoin, in Meter PoW doesn’t involve txns and block ordering. It’s merely consensus algorithm to determine coin creation. Txn validation is done by PoS validators who stake MTRG (governance token of Meter) and in return recieve txn fees and auction proceeds (to be explained) in MTR. MTRG is a finite supply token with a smooth reduction curve and around 12% inflation in first years.
2nd layer stability mechanims in Meter consists of an on chain auction which removes MTR out of circulation. A fixed amount of MTRG gets auctioned off roughly every 24 hours in an on chain dutch auction. The bidding happens with no specific price and bids are placed in MTR only. At the end of auction all participants receive MTRG at the same settlement price calculated as
Auction Settlement Price = MTR bidded / MTRG Auctionned
Majority (initially %60) of the MTR tokens from the auction proceeds goes to Reserve, getting out of circulation. Reserve is collectivelly owned by MTRG hodlers and it’s intended to be used as collateral in defi apps in future. The rest goes to PoS validators as block rewards
The amount of MTR bidded is completely decided by the market. As you may infer, if participation is low, bidders are able to buy MTRG at a dicount. If however the auction oversubscribed all bidders will buy MTRG at a premium. The characteristics of auction (many buyers, freedom to transact, voluntary participation, good information symmetry) make it an ideal market. In such environment, there will be full participation and daily auction settlement price will match with daily average CEX & DEX price with little deviation in the long run.
Open Issues/Challenges
Certain parameters of auction can be played with via governance which include
- Reserve allocation ratio -> initially %60. Higher ratio would mean more MTR removed
- MTRG offered in auction-> this amount is capped by smooth reduction supply curve and is determined by weighted average of the settlement price for the past 30 on-chain auctions (mtr bidded/mtrg offered). If price is higher more MTRG will be offered and vice versa.
Obivously these parameters influence the rate at which MTR can get removed.
It’s important to transparently forsee the ability of auction to remove MTR out of circulation in order to make an assement on the band of behavior from which MTR price can recover during periods of low demand. At the end of the day no price will hold without confidence of market participants.
The exact figures are still being worked on by the team as they stress test the stability of the system. I’d consider an ability to remove %50–60 supply within 5–6 months as a strong sign for stability. We will see how things will play out.
Finally, the use cases around and capabilities of the Reserve is still unclear to me after reading through whitepaper and tokenomics paper(meter link shared below). If MTR in Reserve can enter back in to the system via governance decision, a conflict of interest between pow miners and pos validators might rise assuming they are different parties. Also such a case could auction to become a temporary supply reduction mechanism without much influence on supply in the long term, potentially putting the system in “growth collateralized stablecoins” danger zone.
Conclusion
Meter is a hybrid Pow/Pos chain with two tokens. It uses dynamic block rewards coupled with a demand creating dutch auction mechanisms to create a completely decentralized, layer 1 low volatile currency around a free floating 10 kwh electricity price rate. It’s unique design removes the need of oracles or collateral making it’s stablecurrency highly scalable. The parameters around auction and reserve can be calibrated via governance resembling to open market operations performed by central banks. It aims to be the first completely crypto native unit of account.
Taxonomy of Meter Token
https://www.blockwaresolutions.com/research-and-publications/2020-halving-analysis
https://www.aniccaresearch.tech/blog/bitcoin-minings-three-body-problem
