No, because the authors don't understand how miners gain revenue. For every dollar of miners revenues, currently $0.02 come from tx fees and $0.98 come from the block reward (12.5 BTC today.) Therefore with everything remaining equal (same average fee per tx, which is the authors' assumption), if the tx rate grew 50x it would increase miners revenues to $1.98, hence only doubling revenues and power consumption.
However, looking at historical data the current fees are well below the cap. https://bitinfocharts.com/comparison/bitcoin-transactionfees...
Now, currently Bitcoin can’t scale transactions so extrapolation is kind of up in the air. But, in practice if it’s going to survive it needs to create value to offset the cost of mining.
True, we will never be able to extrapolate future fees. I'm merely pointing out the authors did it in a way that is completely bogus.
That’s a reasonable approximation. Over time mining is paid for by transaction fees in a competitive market. If you get 2x the fees and the cost to mine stays the same, you get 2x as much mining.
Granted that assumes transaction fees are ~1:1 with transaction count. However, protection from a 51% attack eventually requires 1:1 increase in mining as the value of a 51% attack increase, so the lower bound is there.