1/ Great article by @LaurentMT !
The TL;DR is that energy spent per block contributes not just to UTXOs belonging in that block, but retroactively to all global & past UTXOs.
The often-cited “energy spent per Bitcoin tx” number in many economic papers is flawed for this reason.
The TL;DR is that energy spent per block contributes not just to UTXOs belonging in that block, but retroactively to all global & past UTXOs.
The often-cited “energy spent per Bitcoin tx” number in many economic papers is flawed for this reason.
https://twitter.com/LaurentMT/status/1034046547896492032
2/ Another way to put it is that Bitcoin transactions are secured by high stock-to-flow ratio.
I mentioned this recently in terms of mining hardware stock. But economic history is also another kind of stock.
I mentioned this recently in terms of mining hardware stock. But economic history is also another kind of stock.
3/ So Bitcoin possesses high stock-to-flow ratio in two dimensions:
- Stock #1: ledger history
- Stock #2: mining hardware
- Stock #1: ledger history
- Stock #2: mining hardware
4/ Stock in ledger history is the accumulation of *past* transactions.
Stock in mining hardware is the accumulation of *future* transactions. It is equal to the stream of future rewards, discounted back to the present. Discussed in my article here: medium.com/@hugonguyen/bi…
Stock in mining hardware is the accumulation of *future* transactions. It is equal to the stream of future rewards, discounted back to the present. Discussed in my article here: medium.com/@hugonguyen/bi…
5/ To revert a tx, an attacker needs to overcome both types of stock. He needs to acquire the necessary hardware, AND spend an amount of energy proportional to the distance from the tip of the UTXO he wants to revert.
6/ Of cos, if he wants to revert the most recent transaction (distance=0), then there’s not much stock in the ledger history for that transaction, but there’s still an enormous amount of stock in the hardware dimension.
7/ The longer a transaction has occurred, the more stock it accumulates in the ledger history, and the harder it is for an attacker to revert.
8/ In terms of hashes, stock in ledger history only goes up, never goes down. In terms of actual economic cost (and @LaurentMT ’s “number of Bitcoin.days secured” metric), it could in theory go down, if say, a Quantum computer is invented & mass-produced tomorrow.
9/ In terms of economic cost, stock in mining hardware could also go up & down. A sustained bear market would likely cause the cost of acquiring ASICs to go down, while a bull market would do the opposite.
10/ Mining centralization reduces the security provided by stock in mining hardware. But mining centralization does *not* reduce the security provided by stock in ledger history. A miner with majority hash rate still needs to spend tons of energy to go back far in history.
11/ To sum it up, Bitcoin is protected by high stock-to-flow ratio in 2 dimensions:
(i) Stock in ledger history - security can be weakened with advances in computing
(ii) Stock in mining hardware - security can be weakened with bear markets & a high level of mining centralization
(i) Stock in ledger history - security can be weakened with advances in computing
(ii) Stock in mining hardware - security can be weakened with bear markets & a high level of mining centralization
• • •
Missing some Tweet in this thread? You can try to
force a refresh
