Thank you for this write up. I have one follow-up (or perhaps an add-on): why doesn’t anyone talk about the capital costs of the Bitcoin mining network? I have heard the debates about energy input costs, but haven’t heard anyone point out that Bitcoin mining is extremely capital-intensive. My understanding is the special computers needed to mine for it cost about $10k (which is on average good for 0.5 Bitcoin) and have a useful life of about 2.5yrs. The cost to replace that equipment will rise with inflation over time. How can that be considered an inflation hedge? Curious to hear your thoughts. Thanks.
Everyone talks about the energy cost since it’s so crazy, and because there are estimates of it. My guess is that it’s harder to come up with an estimate of the capital cost. However, everyone in gaming is mad about that.
My guess is that the only people who might have a good handle on the costs of mining are the miners themselves. To what extent they are rational, they will be comparing their total cost of production to the Bitcoin price, and decide whether it is worthwhile.
On the inflation hedge side, the idea is that the cost of Bitcoin can’t drop too far below the fees charged and the earned BTC. Otherwise, the miners drop out, lowering complexity and thus the cost of mining. The BTC bug argument is that the fixed supply of BTC is the main inflation hedge.
This is why I'm still having a hard time grasping the "fixed supply protects against inflation" argument. Doesn't Bitcoin's very existence rely on the existence of its miners? And so in order for Bitcoin to be a thing, someone (i.e. the miners) has to incur the capital costs of buying the computers to mine in order to generate any fees and maintain the integrity of the "network." The Bitcoins themselves are basically units/claims on the network. But if maintaining that network becomes more expensive due to rising energy and capital costs, how would a limited supply of Bitcoin protect holders from inflation? How is it any different from a company saying it will only ever issue 21 million shares of stock? The share count quantity does not by itself protect its shareholders from inflation. The quantity of shares has no bearing on the intrinsic economic value of the enterprise. If real costs go up, miners have to start dumping Bitcoins. Seems like a terrible inflation hedge. Or am I missing something?
Another contrast would be comparing Bitcoin to gold. I am far from a goldbug, but at least gold is something that, do to its inherent chemical properties, can be a useful technology for transmitting information, because it is corrosion-resistant, malleable, and hard to replicate due to its relative scarcity, density and distinct yellow luster. And, perhaps most importantly, it exists on its own; it doesn't require a network of computers and miners in order for it to exist. I can theoretically pass down a gold coin as a "family heirloom" and, assuming it stayed within the family, it would still exhibit the exact same properties in 100 years as it does today. That is why, imo, people consider gold as a "store of value." Same thing with precious stones: my in-laws run a small family owned jewelry business, and frequently make new rings to retrofit someone's grandmother's engagement ring stone from 100 years ago. I highly doubt a USB flash drive with Bitcoin information stored on it will be in any way shape or form readable or "valuable" as a family treasure in 100 years.
Apologies for the long post but given you have such a strong grasp on money and economics I'm curious to hear your thoughts.
If the cost of mining is greater than the expected payoff, they shut down mining rigs. Less rigs means lower complexity, which means that the cost falls. Although the accounting depreciation might be fixed, non-operating servers will last longer.
The linkage between production costs and bitcoin prices and transaction fees just tells miners whether mining is profitable *right now*. There’s no long-term relationship between mining costs and bitcoin prices. You also cannot ignore the transaction fees they earn - eventually, there will be almost no new Bitcoin mined.
The Bitcoin backer belief that a fixed supply guards against inflation is just a simplistic Quantity of Money variant. I don’t think the argument is valid, but they do.
Hi Brian,
Thank you for this write up. I have one follow-up (or perhaps an add-on): why doesn’t anyone talk about the capital costs of the Bitcoin mining network? I have heard the debates about energy input costs, but haven’t heard anyone point out that Bitcoin mining is extremely capital-intensive. My understanding is the special computers needed to mine for it cost about $10k (which is on average good for 0.5 Bitcoin) and have a useful life of about 2.5yrs. The cost to replace that equipment will rise with inflation over time. How can that be considered an inflation hedge? Curious to hear your thoughts. Thanks.
Everyone talks about the energy cost since it’s so crazy, and because there are estimates of it. My guess is that it’s harder to come up with an estimate of the capital cost. However, everyone in gaming is mad about that.
My guess is that the only people who might have a good handle on the costs of mining are the miners themselves. To what extent they are rational, they will be comparing their total cost of production to the Bitcoin price, and decide whether it is worthwhile.
On the inflation hedge side, the idea is that the cost of Bitcoin can’t drop too far below the fees charged and the earned BTC. Otherwise, the miners drop out, lowering complexity and thus the cost of mining. The BTC bug argument is that the fixed supply of BTC is the main inflation hedge.
Thanks Brian. I found one write-up online (https://www.toptal.com/finance/blockchain/what-is-bitcoin-mining#:~:text=Bitcoin%20Mining%20Economics&text=These%20tables%20represent%20typical%20costs,on%20values%20from%20December%202021.&text=Capital%20expenses%20(Capex)%3A,~5.22%20years%20%3D%20~0.48%20BTC) that estimates total CAPEX of $22.6k per Bitcoin (jives with my original statement of $10k for 0.5 Bitcoin so $20k for 1). According to this guy, total b/e cost for miners is around $32k. These computers, which represent 2/3 of total mining costs, use up real resources and depreciate rapidly, requiring constant reinvestment. Businesses that require rapid rates of reinvestment generally do not fair well when inflation is high, because an increasing portion of the profits get chewed up by growing capital costs. Recently, Bitcoin miners have been dumping Bitcoin in order to raise cash to cover their growing fixed costs (https://www.coindesk.com/business/2022/07/06/riot-blockchain-starts-move-away-from-new-york-hosting-site/#:~:text=Miners%20have%20been%20selling%20their,price%20of%20bitcoin%20and%20ether.).
This is why I'm still having a hard time grasping the "fixed supply protects against inflation" argument. Doesn't Bitcoin's very existence rely on the existence of its miners? And so in order for Bitcoin to be a thing, someone (i.e. the miners) has to incur the capital costs of buying the computers to mine in order to generate any fees and maintain the integrity of the "network." The Bitcoins themselves are basically units/claims on the network. But if maintaining that network becomes more expensive due to rising energy and capital costs, how would a limited supply of Bitcoin protect holders from inflation? How is it any different from a company saying it will only ever issue 21 million shares of stock? The share count quantity does not by itself protect its shareholders from inflation. The quantity of shares has no bearing on the intrinsic economic value of the enterprise. If real costs go up, miners have to start dumping Bitcoins. Seems like a terrible inflation hedge. Or am I missing something?
Another contrast would be comparing Bitcoin to gold. I am far from a goldbug, but at least gold is something that, do to its inherent chemical properties, can be a useful technology for transmitting information, because it is corrosion-resistant, malleable, and hard to replicate due to its relative scarcity, density and distinct yellow luster. And, perhaps most importantly, it exists on its own; it doesn't require a network of computers and miners in order for it to exist. I can theoretically pass down a gold coin as a "family heirloom" and, assuming it stayed within the family, it would still exhibit the exact same properties in 100 years as it does today. That is why, imo, people consider gold as a "store of value." Same thing with precious stones: my in-laws run a small family owned jewelry business, and frequently make new rings to retrofit someone's grandmother's engagement ring stone from 100 years ago. I highly doubt a USB flash drive with Bitcoin information stored on it will be in any way shape or form readable or "valuable" as a family treasure in 100 years.
Apologies for the long post but given you have such a strong grasp on money and economics I'm curious to hear your thoughts.
If the cost of mining is greater than the expected payoff, they shut down mining rigs. Less rigs means lower complexity, which means that the cost falls. Although the accounting depreciation might be fixed, non-operating servers will last longer.
The linkage between production costs and bitcoin prices and transaction fees just tells miners whether mining is profitable *right now*. There’s no long-term relationship between mining costs and bitcoin prices. You also cannot ignore the transaction fees they earn - eventually, there will be almost no new Bitcoin mined.
The Bitcoin backer belief that a fixed supply guards against inflation is just a simplistic Quantity of Money variant. I don’t think the argument is valid, but they do.
Thanks Brian
One of the best takedowns of the crypto craze that I have read so far!
Thanks!