Notes from the seminar by NCourtois and Robert Sams. 12 June 2014, revised on 13 June ==================================================== About the speaker: Robert Sams is a former interest rate derivatives trader and central bank specialist. Currently entreprenuer, co-founder of company that provides institutional liquidty services for cryptocurrency and blockhain 2.0 solutions for financial settlements. BTC monetary supply: a "BUG NOT A FEATURE" STUPID INELASTIC MONEY SUPPLY built in BTC Not like gold, rather like rare collectibles (stamps). MORE IMPORTANTLY: even gold has a supply function that responds to its price. (elasticity) Why DigiCash failed? They had the right investors, 1/2 billion offer from Microsoft, however negotiations broke down... (maybe key people did not believe in it enought to carry on?) The speaker is working on one of the most important open problem of bitcoin design: "adaptive money supply", could save bitcoin from beign inadequate/obsolete etc... Thought experiment: NEGATIVE INTEREST RATE: central bank, -10% Simple: invalidate some bills with randomly chosen numbers? the acutal supply of dollars, most in 100 dollar bills, was increasing recently... (it is like moslty criminals neede dmore liquidity...?) Trying to understand if fixed mon. supply is good or bad in BTC All boils down to "The Money Demand Function" a curve x=quantity y=price TWO schools of thaught on modelling the money demand 1. classical economy: depends on tx volume and velocity 2. Keynes: E of delta price affects the money demand function IRONIC people in bitcoin believe in 1 but act in a way that confirms 2. He assumes ONE DOMINATING CRYPTO CURRENCY, however you always have at least BTC and fiat, 2 currencies UOA= Unit Of Account deflation is bad - usually pb with UOA side, almost the wrong word to use Deflation is bad namely because debt and employment contracts are written in nominal terms, so an unexpected decline in the price level has effects on the real economy. Don't think anyone in crypto would ever denominated their contracts in BTC prices, practice is for contracts to be written in fiat prices settled in BTC at fx rate prevailing at time of payment. Key remark, bad for bitcoin: Volatility INCREASES transaction costs by A LOT! ======================================================== STABILIZATION with elastic money supply! no crypto currencies for now have it... no paper, blogged about it... not aware of other people proposing such things (except Fernando Hayek Money paper) Vitalik thought about it but is not a fan. Fernando has written about it in summary form on cryptonomics.org. If others have, not aware of it. HOW CAN WE REDUCE MONEY SUPPLY also for existing coins?????? NECESSARY if the demand goes down!!!! How to make the protocol aware of the outside world? through price of electricity?????? These are the two main problems in implementing an elastic money supply rule in crypto. Mining: should be in equilibrium, nobody making money f(Price, difficulty, GH/kW) METHOD 1 just find a way to incentivize miners to communicate the price of electricity to the network this does NOT involve trust... Economics of mining: marginal costs equal marginal value (long run equilibrium condition) MC = MV MC = f(Price, difficulty, GH/kW) MV = Price * (block award + tx fees) ======================================================== CRITICISM by Courtois: maybe we will discover that the electricity is free, and the only cost is to transport it? Speaker: Interesting point. Another reason to disparage hash-based PoW. IF BITCOIN forced people to compute SHA256 with an abacus... it will also work... METHOD 2 "Hayek Money" to make network aware of the outside world in a trustless way Thomas Schelling = game theorist 1960s people send bets along with 1 BTC, people in outer quartiles have their money confiscated? INFORMATION MARKET! (Courtois says it is super dangeorus and can deviate from providing accurate and honest information for profit or hedging reasons) I initially suggested this idea in early february here https://forum.ethereum.org/discussion/31/what-would-a-trusted-data-feed-look-like/p1 Vatalik blogged about it in March http://blog.ethereum.org/2014/03/28/schellingcoin-a-minimal-trust-universal-data-feed/ The most well thought-out scheme I'm aware of is here: https://github.com/psztorc/Truthcoin/tree/master/docs And it's a key feature in Ferdinando's design. Paper "Hayek Money" Bitcoin: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2425270 Proposes to give 10% more bitcoins if you have already bitcoins... or decimate their balances Technically quite easy to implement!! ("indexation" as a opposed to a peg: basket of commodities... consumer prices...) Judgment of Sams: rather bug, not a feature... does nothing to solve "the speculative balances problem being a source of volatility" => Namely, it's the distribution of increases/decreases in money supply that is a bug, IMHO. Nando sees it as a feature. To me it does nothing to counter speculative money demand. How to make monetary supply decline with like a central bank prop it up? Main problem: volatility => never a good adoption as MOE even much worse for UOA Cicious circle!!! chicken and egg... Elco Jol says: bitcoin volatility is decreasing. Sams says it is bogus data... no matter of liquidity can make the volatility go away. Sams: I think I dealt with this claim definatively back when it was popular http://cryptonomics.org/2014/01/22/is-bitcoin-volatility-really-in-decline/ It's a simple empirical conjecture and the data are unambiguous: volatility is NOT declining!!! (Elco: Argentinian currency is also highly volatile and yet widely used).