Comment by arthurcolle | Hacker News Reader

Comment by arthurcolle | Hacker News Reader

Parent Full thread

Care to share your sources?

Naturally, a summation of the energy expenditure of all of the related components of the existing fiat system must be included, from the totality of all people operating the various functions that facilitate the existing fiat system; i.e. their commutes, aggregate expenditures due to needing to operate physical banks, power usage at these physical banks, etc. are all factored into this calculation... right? All the way to the minting of new currency (probably almost negligible in the aggregate), to all of the security apparatus that ensure that bills are legitimate, to major parts of the US Secret Service that verify that bills are legitimate (including all of their expenses scaled to the proportion of agents dedicated to this task force)... I honestly don't believe that the electricity usage of BTC is higher than fiat currency operation, in the aggregate.

Please share any sources if you have any information that contradicts this. It's an interesting area of open research + I'm curious if I've missed any publications recently.

This is a bit of crypto-delusion I have never understood: why compare Bitcoin with the financial system as a whole to get out of the "it is mostly powered by coal plants in China"-argument?

Does Bitcoin assess credit risks of governments, companies and people and then extend+manage credit?

What does Bitcoin offer in regards to risk management/hedging for operating businesses?

Maybe Bitcoin has some liquidity management feature for real businesses that I am not aware of?

Bitcoin has very low uptake in payments, so trying to compare its energy consumption to all retail branches and the oil burned by the employees on their commute strikes me as misleading and ignorant

Modern banks and financial institutions are much more than a long list of payments. Please learn a little bit about what a company like Bank of America actually does

I worked at Goldman Sachs for 3 years in technology and trading roles so I'm fairly confident that I am largely familiar with how a bulge bracket investment bank as well as how most consumer banking firms operate work.

It's not misleading at all. It's important to consider that a large group of human operators, using a separately large set of semi automated processes, are a large component of the current energy spend of aggregate financial activity - all those busy worker bees are using energy.

Not really interested in doing some super thorough calculation, hopefully this general outline makes a little bit of sense. I'm just asserting that if we are going to criticize one thing's energy spend, you need to actually define what you are measuring and determine a baseline with respect to what the current thing's energy consumption is, ideally as line items and then sources for each input component.

Looking forward to some sources. (If you are referring to the Argentina article, please just identify that as your source). No need to be antagonistic, we're all friends here.

>It's not misleading at all. It's important to consider that a large group of human operators, using a separately large set of semi automated processes, are a large component of the current energy spend of aggregate financial activity - all those busy worker bees are using energy.

A lot of that is compliance though. Some (or maybe even most, I'm not in the industry) of that doesn't need to exist for the system to function - the banking system would work just fine without suspicious activity reports. I don't think it's an apples to apples comparison as these checks don't and can't exist in cryptocurrency.

That's just scratching the tip of the iceberg, as mentioned above the traditional financial system provides lots of services that cryptocurrencies can't provide or don't provide at any meaningful scale.

But institutions like Goldman Sachs aren't going away if Bitcoin suddenly became the world currency, so including them in such calculations is misleading.

Look up e.g. VISA corporate responsibility report for 2019. It’s been some time since I’ve seen that, but iirc Bitcoin hash power takes like 120 TWh, “Big Tech” combined is like 50 TWh, and VISA with all the nuts and bolts (and people) is like 0.2 TWh - with proven consumption of renewable energy.

I've done this before. No, I'm not going to share my sources because a) I haven't saved the exact ones I found before and b) whatever I include, you're always going to criticize me for not including yet more calculations in my side of the ledger. But the answer is going to come out the same anyways: Bitcoin uses orders of magnitude energy more than traditional currency.

Take a simple thought experiment. Someone makes a transaction with VISA. That requires updating the seller's bank's ledger, the buyer's bank's ledger, and VISA's computer. Let's round that up to a dozen computers who need to know about that transaction. Now consider Bitcoin. Those same three computers need to know about that transaction... and so does everybody else who runs a full node. We haven't considered a picojoule of energy consumed by proof-of-work yet, and Bitcoin has already consumed far more energy than VISA.

So how about that proof-of-work? Well, let's compare a bank's energy consumption. Go to their financial statements, and pick out the line item that includes energy costs. If you're not sure which line item that is, add all of the plausible ones together (and round up, because what's the harm in overestimating here?). I don't know how much they pay for energy, so let's say it's 1¢ per kWh because what's the harm in rounding up? Crunch the numbers, and banks still use far less energy than Bitcoin's proof-of-work, even by our wastrel overestimation numbers. Oops, I forgot to account for the different volumes. Doesn't matter--banks are moving more money on less energy than Bitcoin.

Much of the ancillary functions you want to include to make Bitcoin's energy consumption seem less embarrassing would still exist even if the world magically switched from USD to BTC overnight. The loan officer evaluating how much interest to offer on a mortgage to a first-time home buyer would still be doing exactly the same thing. The loan shark driving to bust someone's kneecaps for not paying back loans on time would still be doing exactly the same thing.

> The loan shark driving to bust someone's kneecaps for not paying back loans on time would still be doing exactly the same thing.

Tangential thought...

I might be mistaken, but how can one even collect on defaulted debt denominated in BTC? I thought that the BTC owner still has to transfer BTC themselves, since only they would know the private keys.

Wouldn't the collector literally have to physically threaten the individual to transfer the BTC back, since it's not like they can get a court to order the bank to do it.

I know that ETH supports smart contracts that could in theory automate the terms of the loan, but then what happens if the debtor goes bankrupt?

> Wouldn't the collector literally have to physically threaten the individual to transfer the BTC back, since it's not like they can get a court to order the bank to do it.

What makes you think the court can't order the bank to do it? The court can order you to give some assets up, and if you refuse, they can order your employer to deposit wages directly into creditor's accounts, or the creditor can seize other assets for nonpayment. (For example, some creditors managed to temporarily have an Argentinian Navy ship seized after their default.)

> What makes you think the court can't order the bank to do it?

The bank doesn't have the private keys for the wallet

You make a good point wrt. physical cash (coins and bills) versus electronic currency. I think it's sensible to analyze the two separately.

Also, it's important to consider that commercial banks do much more than just keep track of "who owns how much USD". Banks are primarily middle men in the credit markets, and this is not related to establishing people's balance (which is all the Bitcoin network does). If Bitcoin persists then BTC credit markets will also develop, which would be very different from just keeping track of "who owns how much" (which is all the Bitcoin blockchain does).

I agree, consumer banks (and general financial technologies/innovations offered by private equity firms, private trading companies, financial engineers, etc on the whole) offer a greater breadth of services than crypto-asset financial innovations at present. This is obvious - I'm not sure anyone would argue against the validity of that statement.

The largest single asset within the category is bitcoin + the majority of crypto-assets (in market cap terms, in aggregate) are currently PoW based systems (exception of Eth2 if you're just looking at top 5 but the bridge isn't live yet to my knowledge) so it makes sense to take a step back and really try to understand whether we are measuring the same things across both systems - the terms (in a linguistic sense) need to be defined for any serious discussion/analysis can be undertaken. Basically the Fed Reserve system's budget (some % of their budget) + cost of operating basic banking operations across all banks needs to be part of the calculation, at a minimum. Any other terms (in the equation that represents the relative spend across the two systems) could add greater accuracy, but the record keeping seems like the biggest cost that needs to be factored in to accurately reflect the energy consumption differences between the two systems.

Reply on news.ycombinator.com