Mining Equivalence Measure: Valuing Bitcoin


Dec 29

I have been trying to think about the value of Bitcoin and where it comes from.

It certainly doesn’t come from the source code which is open source: though the code is a perquisite without which it couldn’t exist.

One way to think about the question is to rephrase it as what properties might we value in a crypto?

  1. Network effect: Number of users.
  2. Security: Hashing power, algorithms, quantum computing resilience.
  3. Trust: The length of time in existence
  4. Trust: The degree of decentralization. This is in both operating and distribution of initial wealth.
  5. Trust: The anonymity provided. As Edward Snowden points out, while Bitcoin is pseudo anonymous, the lack of better anonymity is both puzzling and glaring.
  6. Power: The number of transactions supported over a period of time, such as a day.
  7. Efficiency: The cost to operate, the greenness, the cost to transact, the speed of transactions.
  8. Resilience: The degree at which it can exist in offline and online states, the difficulty of disruption.
  9. Rate of Appreciation: This is an interesting measure because it largely draws from the other factors and scarcity. Importantly, investors want high rate of appreciation. One might think about PlanB’s stock to flow model.
  10. Other good exchange characteristics: I.e. highly divisible

The Winklevoss twins have probably made the best arguments for why Bitcoin is “digital gold” or “gold 2.0”. I tend to agree they look more right then wrong now. And, the analogy is very apt except Bitcoin does have some superior qualities in terms of transaction.

Curiously, the mining aspect of Bitcoin is required for transaction today. Though eventually one envisions that mining wouldn’t be required. So, you have this mining mechanism that is tied to the transaction mechanism which is somewhat peculiar.

Now one thought might be that once gold is mined it doesn’t cost much to store. And, that’s true with Bitcoin: once mined, there’s no real cost to store it. But, there is a high cost to transact it which is borne largely by the miners and fees. Now, you could argue the same is true with gold as it costs fuel to transfer. The difference is that the cost is not so much obfuscated.

In any regards, one argument goes that because coins are mined each day (every 10 minutes with each block) and the miners incur real-world expenses to mine those coins that there is a pressure to sell them which could depreciate the price. It is also true that as the value of Bitcoin goes up then the difficulty will increase and thus the cost to mine will increase: so one could view it as the cost to operate the network.

The Mining Cost Equivalence Measure

The Winklevoss twins make some compelling arguments why Bitcoin could completely replace gold. They call this “the 10x” factor: if a new technology is 10 times better then a prior technology then the prior technology will be replaced. But, an alternative way to think about gold and bitcoin is they are somewhat equivalent: even though we know they aren’t truly equivalent. Importantly, there is no value judgement required: just to assume that at some point Bitcoin took on enough properties of gold to be the same from an investing standpoint.

There’s a certain amount of gold mined each year. Mining that gold has a certain cost. The cost is real in terms of environmental and energy use. There’s a certain amount of Bitcoin mined each year. Fortunately, Bitcoin being digital has more potential to be green mined then something physical like gold. But, in any regards there’s a certain cost to operate the network.

One way to think about equivalence is that equivalent things should be priced similar. Another way might be to think about it in terms of efficiency if gold and bitcoin are both equivalent then an efficient society should mine whichever is cheaper until the costs rise to the same.

There approximately 900 Bitcoin mined per day at (every 10 minutes) and that yields 328,500 Bitcoin mined per year. At let us just say $25,000 cost then we can imply the network cost will be similar to the mining profit: $25,000 * 328,500= 8,212,500,000. So, it cost 8 billion and 212 million dollars to run the Bitcoin network.

How much does it cost to mine gold per year? The estimate I found was that between 2,000 to 3,000 tons of gold are per year. The cost estimate was as low as $1,000/OZ but one should probably use the market value to be consistent. So, there are 32,150.7 ounces per ton times 2500 tons times $1890 market price yields $151, 912,057,500 (151 billion, 912 million).

If we divide this number by 328,500 Bitcoins then we get an interesting number: $462,441. This number is interesting because it is close to the value of Bitcoin assuming a market cap of gold. The market cap of gold is estimated between 7 to 10 trillion estimated. 9 Trillion divided by 21 million is $428,411. The numbers are curiously close. What makes it curious is I took the cost of mining gold assuming approximate price/cost parity divided by the number of Bitcoin produced each year which yielded a price per Bitcoin that is similar to the price estimate of Bitcoin given a total market cap of gold. Completely different inputs yield a similar output. More over if there is validity in the calculation it might suggest a simple explanation for why Bitcoin has increased in value: simply, at some point Bitcoin became equivalent to gold but cheaper to use, and an intelligent society would adopt the cheaper solution, giving it greater and greater value, until price parity is reached.

Does this analogy or computation even make sense? Or am I off by order of magnitude? Of course, gold still has properties of gold even if mining ceases to exist. Let me know what you think. It also makes one question if the “work” is a “sink” or drain or if it is what gives Bitcoin value. To be honest, this is a outside what I normally do, and so I could be wildly wrong too.


I was thinking about it: this is likely another variation of “stock to flow”. but thinking about it a bit differently. Essentially, it suggests that both Bitcoin and Gold have similar rates of inflation now. To confirm that, we take 151 billion/9 trillion yields .0167*100=1.67% inflation. While, 328,500/21 million = .0156 = 1.56%. Because the Bitcoin inflation is marginally lower in this method, the fair value is slightly higher then the estimated FV when using market cap.

So, what’s different though is that stock to flow model suggest that as the inflation rate drops then the price must go up. While, the line of thinking I have expressed here, suggests that the price will go up but only until parity with gold. The line of reasoning suggest there might be a reason for a break in the stock2flow model. Or perhaps that Bitcoin price can go up but shouldn’t be able to sustain values above the total mining cost of gold. Example, next halving Bitcoin inflation rate is .84%. So you get $151, 912,057,500 / 164,250 yields $924,883 while I believe stock2flow projections $1.2 million.

About the Author

The author is passionate about markets. He has developed top ranked futures strategies. His core focus is (1) applying machine learning and developing systematic strategies, and (2) solving the toughest problems of discretionary trading by applying quantitative tools, machine learning, and performance discipline. You can contact the author at