This is untrue, of course. 51% attack resistance requires that mining be resource-intensive and have significant opportunity costs. Otherwise - if mining is no

This is untrue, of course. 51% attack resistance requires that mining be resource-intensive and have significant opportunity costs. Otherwise - if mining is not costly and wasteful- it's not costly for an attacker. If electricity prices went down to 1% of their current value tomorrow, miners would need to increase their electricity consumption by approximately 100x, or network security would suffer. To the extent that energy or resources used for mining are free, or surplus (would otherwise be wasted)... that mining activity provides no security to the network. This is the same reason advances in miner performance (e.g. better ASICs) don't provide more network security. Miners mine faster, but any attacker would have access to the same technology. The only thing that can actually make the network more secure is using more scarce resources - whether directly, .eg. via energy usage, or indirectly, by manufacturing ASICs, etc. Bitcoin mining is not resource-intensive in the sense that, say, aluminum smelting is. The thing about aluminum smelting is that if there's efficiency improvements, that's good! That means more aluminum from the same limited supply of free/low-cost/surplus energy. I like aluminum! But when somebody finds a way to, say, double the hashrate-per-watt you can get mining Bitcoin... the hashrate increases, but the network-security-per-watt stays the same. It has to. It's not resource-intensive in the sense that automobile transportation is. If cars get more gas-efficient (or energy-efficient in general), awesome! Everybody's life gets better. If Bitcoin mining gets more efficient... miners just mine more. No extra security. No actual reduction in resource usage. Assuming that miners mine when it's profitable (+EV) and don't mine when it's not profitable, Bitcoin's difficulty adjustment system implies the following: "The sum of all transaction fees and the coinbase reward for each block equals, on average, the cost of the resources used by all miners during that block period plus a small profit." In other words, with a 6.25BTC block reward at current prices, every Bitcoin block requires burning about $250,000 in resources- or about $1.5 million an hour. Now, that doesn't have to be electricity. If we discover free energy tomorrow and electricity prices go to zero, the resources used would likely switch to be those used in miner ASIC manufacturing- since every ASIC made would be a literal free money printer. But you'd still need to (amortized across the ASIC's life, of course) use resources worth about $1.5 million an hour on network security - or accept that network security is weaker. And to be clear, that's "worth" including things like transportation costs- natural gas may be worth $X per BTU, but natural gas out in the middle of nowhere where it'd otherwise be flared is worth far, far less, and using it thus provides far less security. So, Bitcoin is either insecure or wasteful on an almost unimaginable scale - and probably isn't even wasteful enough today. (And no, that's not an XOR.) The only way I can see not to wind up at this conclusion is to argue that markets aren't even weakly efficient.