Due to the level of publicly-available information about Bitcoin, you would think we’d get better attempts at analysis from critics. One of the most widely debunked, yet still widely referenced claims of “academia” is that Bitcoin will single-handedly increase the planet’s temperature by 2 degrees Celsius. By the end of this piece, you’ll see that the opposite is true, with Bitcoin’s emissions likely to have already peaked a few months ago, and that in 10 short years, it’s likely that Bitcoin won’t emit anything at all.
When one understands the basic fundamentals of business, competition and innovation, projecting future energy use of Bitcoin is trivial. Indeed, I and many others have done it with some degree of success in 2014, 2016 and 2018. I’m not able to see into the future this way because I’m necessarily wise or intelligent, I’m able to see into the future because Bitcoin voluntarily shows it to me. I just have to understand where to look. The key to my “predictive successes” over the past seven years all boils down to a strong assertion that Bitcoin mining is the closest thing to a perfectly competitive market that has ever existed in the real world.
In the next sections, I will go through the basics of perfect competition and how miners tick as a result of this. I will then provide five- and 10-year predictions on price, hash rate and technology, and the energy mix of the Bitcoin network. From there, I will conclude with the total energy use and emissions of the Bitcoin network in 2026 and 2031. PERFECT COMPETITION The example of “the hypothetical firm in a perfectly competitive market” is taught in most introductory economics classes.
A literature review of primary academic texts identifies nine conditions that define a perfectly competitive market. In 2014 (page 35), I argued that only four of the conditions had been met. With the benefit of an additional 18 months of lived experience and data, I then argued, perhaps prematurely, that six of the nine conditions had been met (page three). Two and a half years later, in August 2018, we were still stuck at six (pages three to six). Another three years later, today, I’d be happy to say that seven conditions have now been met.
We will go through the nine conditions, very briefly, point by point. For a fully detailed analysis, you can revisit my earlier work linked above. Homogeneous products: Met. Guaranteed property rights: Met. Your keys, your coins. Your node, your rules Non-increasing returns to scale: Met. See GHash.io in 2014. Zero transaction costs: Met. See Lightning Network and Bitcoin Layer 3 as contemporary examples. Perfect factor mobility: Met. See recent seasonal and current China ban miner migrations as contemporary examples. No barriers to entry or exit: Met. Bitcoin is voluntary to enter/exit.
Many buyers and sellers: I wanted to say “not met in the short term” on this one, but the data is suggesting that there are over 70 million Bitcoin users as at December 2020, which misses the past six months of mania that we just witnessed. It is arguable whether 70 million to 100 million users is actually that many, as the market is still illiquid enough to shed 50% of its value in a few weeks. Although there are “many buyers and many sellers,” the “many buyers, few sellers” and “few buyers, many sellers” scenarios still occur too frequently. Perhaps we’ll call this one “halfway met” in the short term, fully met within the next five years.
Perfect information: Not met in the short term. While there are over 70 million Bitcoin users, the extreme volatility leads you to conclude that information is not yet perfect. There are still several insiders that get inside access during extreme market events, while retail investors panic sell in absence of this inside information. As we grow from 70 million users to 700 million users, this will stop being an issue. Indeed, most people still haven’t even heard of Bitcoin, let alone accessed any information about it. Most likely, this needs another five years to resolve, but not longer than 10. No externalities: Not met in the short term.
Only externalities that exist are driven by the grid, not Bitcoin. All ASIC mining equipment is nearly fully recyclable. Further, ASIC equipment is now running for longer than ever due to slow hash rate growth (more on hash rate growth later!). As Bitcoin becomes the world’s flared methane sink, the externalities will be zero or negative. Most likely, this needs another five years to resolve, but not longer than 10. Now that we have demonstrated that the nature of competition in Bitcoin is near perfect, we can discuss the significance of this assumption, and what it allows us to conclude when it comes to making five- and 10-year projections.