Power Games: Bitcoin and Beijing
Examining the historical record of China-domiciled mining operations, international entrants, and the implications of the CCP’s new blockchain policy on Bitcoin’s security model
Introduction: Miners Still Matter More Than Most Think
The mining process for Bitcoin and other cryptocurrencies of similar architecture (grouped together as “Proof of Work” coins) has remained essentially unchanged in principal since the first blocks were generated by Satoshi in 2009. However, the actual hardware that runs the mining node software has improved dramatically in speed and energy efficiency over the last 10 decades, as the increase in the bitcoin price in the intervening period ignited the dynamics of free market competition for the cryptocurrency mining industry, resulting in the current environment where purpose-built ASIC (application-specific integrated circuit) machines must be used in order to have a chance of successfully finding the next block (to append to the blockchain) and earn the associated BTC reward. As a result, profitable mining operations naturally started to scale up in size as the cottage industry developed from hobbyists tinkering around with FPGAs (field programmable gate arrays, an intermediate step before ASICs became the norm) in sheds, to today’s megawatt-plus deployments of the latest hardware from specialty ASIC manufacturers like Bitmain, Canaan, and others. These operations also tend to be at least partially vertically integrated in the sense that there often is meaningful overlap between the 1) manufacturers of ASICs, as well as the 2) mining farm owners who purchase and run the equipment, and 3) mining pool operators which add a layer of abstraction and volatility reduction (in actual block rewards earned, when compared to solo-mining) by coordinating and sharing rewards among voluntary participants in their associated pool (and taking a cut of profits earned).
While there are some that argue the mining and associated hardware manufacturers matter less for the future of Bitcoin than the ongoing software development led by engineers working on the node software (e.g. Blockstream, Chaincode labs), I think understanding not only the short-term business model (profit as dictated by the inputs of hardware and electricity versus the cost of bitcoin in the open market when sold), but divining the longer-term motivations of the global mining community for bitcoin, including the current majority of China-based participants, is a vital part of sizing risk.
To use another analogy, I think of the node software development and soft-fork network upgrade as Bitcoin’s offense, while the mining stack and its participants represent the less exciting but no less important defensive capabilities of the protocol to stay on top of. Given the CCP’s recently renewed focus blockchain technology and thawing of hostilities towards miners of public protocols, I figured it was time to leverage my unique perspective as a Chinese-born American citizen to try and figure out what the future holds. The initial conclusion is somewhat surprising, as I now believe the current China-dominated hashpower is a local maximum and that global incentives and economics likely will result in more diffusion and diversification going forward, lowering some of the current centralization risk that remains a stumbling block for many.
Status Quo: Creeping Centralization, Though Organic
Source: Cambridge JBS 2017 (top left) and 2018 (top right) Global Cryptocurrency Benchmarking. Coinshares 2019 Bitcoin Mining Update (bottom)
Although it fluctuates from year-to-year and sampling methodologies differ depending on the research done, the general trend appears to be that Asia, and China specifically, contributes the majority of the hashrate supporting many PoW networks, including Bitcoin (65% this year per Coinshares estimates). While some conspiracy theorists suggest nefarious reasons why this is, I tend to agree with the commentary from the latest Coinshares mining update which noted that “Occam’s Razor suggests that it is likely an effect of relational and geographic proximity to manufacturers making barriers to business comparatively lower.”
Personally, I specifically recall the amazingly convenient access to spare parts and labor at the large electronics marketplaces (that are typically found in any major city) from my own experience studying abroad in Shanghai, so it’s not a stretch of the imagination to believe that the development of the ASIC and mining industry in China is more organic than the result of a top-down directive, especially given the quick pace of iteration in the product development cycle which is a cultural norm. That combined with the historically cheap coal and hydroelectric power available in rural provinces like Sichuan, Yunnan, and Inner Mongolia completes the puzzle of at least how the hashrate has developed and concentrated thus far. Nevertheless, it would be fair to say that the large miners/pool operators represent some level of centralization risk, per a tweet earlier in December when many representatives from these organizations all assembled at Canaan Mining’s post IPO gala celebrating the listing of the US ADR. While this particular event appeared to be blessed by the local Hangzhou governors, such a concentration in meat-space represents the potential fulcrum by which future political power might be enacted against the Bitcoin network.
Putting Beijing’s Recent Moves in Perspective
As a quick summary of recent developments, President Xi’s recent top-down directives at the end of October saw a doubling down of the “blockchain, not bitcoin” thesis, likely driven by the ongoing development of China’s DCEP on the monetary tool front, and the understanding from the CCP that elements of the blockchain can be levered to better ensure compliance with social cohesion goals without risking the effectiveness of the panopticon-style command and control mechanisms in place. That’s why I believe there’s greater risk of interference at either the provincial or national level in the mining industry for public PoW chains, a risk that I believe is not taken seriously yet but is a meaningful overhang, though not without its own mitigating factors.
For now, I am guessing the CCP intends to boil the frog slowly and let the mining industry continue to develop, but I also believe the potential exists for crackdowns to re-emerge, probably most likely in a scenario where Bitcoin and other open protocols demonstrate greater traction than the DCEP product (thus, low probability especially in the short-term), since that represents a case where the political risk of losing access to the financial data and accounts of its citizens outweighs any economic benefit of supporting the mining and ASIC-manufacturing industry.
Zooming out, I also think there’s potential risk of eventual nationalization in 10–20 years if cryptocurrencies really follow the same adoption curve as we long-term bulls expect, since decentralized/uncensorable public protocols are almost total anathema to the harmonious society narrative that the CCP works hard to maintain. In such a scenario, whichever mining / operators are around will likely be forced to merge into a full state-owned / controlled entity, similar to Norinco today, which is a Chinese defense industry corporation that not only manufactures munitions for the PLA, but also generates 10’s of billions of $ each year in export contracts to other national governments despite China being one of the most restrictive nations in the world with respect to allowing its own citizens to possess personal firearms.
Motivation: The Two-Front War on Dollar Hegemony
Given the above, you might wonder why China even tolerates the mining industry today — given the potential for additional headaches in the future. I believe the answer is fairly simple, namely that the near-term focus of de-dollarizing the world is important enough that China’s willing to take a two-prong tack, both of boosting the adoption of the RMB through the belt-and-road initiative, but also tolerating the growth of an industry that supports yet another potential driver of a weaker dollar (e.g. more Bitcoin bulls from the US will sell dollars to own BTC), while at the same time effectively keeping their finger on the panic button in the sense that if Bitcoin becomes too successful too quickly, they are in a position to pressure miners to act in ways that are disruptive (though I don’t believe necessarily fatal) to the network.
To clarify, I’m guessing the CCP doesn’t care if Americans sell their dollars and buy BTC in large numbers, that’s probably a success scenario for them, so long as their own citizens are less aware or are unable to do the same. It feels to me like the combination of 1) relaxing controls on mining, while also 2) inviting closer relationships with selective on-shore exchanges that are then effectively “blessed” (such as Okex and Huobi) likely gives the government enough comfort that they’ll be able to manage or mitigate demand for public/permissionless protocols in China (or worst case, fork / force a more CCP-friendly Bitcoin, similar to what they’ve done for the internet). I don’t know how the rest of the world would react to something like that, but it will most likely depend on the availability of non-China-based hashpower, as well as how the social consensus has evolved between now, and even how the codebase and feature-set for Bitcoin looks at that time. For example, if in 5–10 years China only has ~30–40% of the total SHA-256 hashpower, then the ability for shenanigans decreases dramatically, particularly if additional American semiconductor companies like Intel, AMD, or Nvidia can at that point (if crypto has scaled) justify getting into the ASIC business, which would further shore up supply-chain-level exposure to China given a lot of the fabrication process actually takes place in non-PRC locales like Taiwan.
Implications for the Present Day
The near-term implications are somewhat troubling and require a guess on how miners are ultimately thinking about these scenarios. You’re already seeing players like Bitmain plan for larger deployments outside of China, suggesting they may be seeing the chips start to fall in an unfavorable way in the future, and are looking to diversify quickly enough to have some leverage, but not quickly enough to alarm Beijing or given the impression of escape. It’s a delicate dance, and one I don’t envy having to participate in, given the downside for these actors is perilous, with even China’s one-time richest man, Wang Jianlin of Dalian Wanda having been pressured to sell off his international assets and reinvest more capital in domestic businesses.
Given estimates that 70% of the hashrate deployed in 2019 thus far (a fair amount that has outstripped even BTC’s price performance) has been in China, it’s clear that at least the near-term decision for miners that are able to profitable mine is to double-down and reinvest ahead of the upcoming halving, thereby committing their hardware for a likely 3-year period. I’m guessing that they’ll be paying close attention to the CCP’s messaging and level of interference, and as a result, we should be paying close attention to what they do, particularly in 2 years as they weigh decisions on how much of their capital base will continue to stay in China versus be deployed elsewhere external to China. No decisions are ever made in a vacuum, and so I believe watching the overt and covert back-and-forth between the mining industry and government officials will be extremely important to the actual and perceived stability of the Bitcoin protocol over the next few years.
Non-Chinese Mining: How Can Hashrate Catch Up?
Although the tone for this article hews more towards a cautious wait-and-see approach, there are nevertheless developments which are positive for Bitcoin to offset the confusing Sword of Damocles situation of Chinese hashrate going forward. Specifically, the steady drumbeat of smart operators and investors such as the ones involved with Crusoe Energy Systems sourcing near-free and environmentally friendly sources of energy to apply towards bitcoin mining should help diversify the hashrate away from China at the same time that Chinese operators are seeking to do the same. Doing the quick math on this, if we take the Cambridge University power model which estimates the current Bitcoin network consumes 8.5 GW of power, this implies Chinese hashpower is 5.525 GW vs. non-China at 2.975 GW, or that 2.55GW of shortfall needs to be added to ensure the majority of hash is outside of the potential realm of interference of the CCP. Applying the Crusoe press release numbers to this mental model is straightforward, as the 70 units said to be coming online next year directly match up to $70M in funding and each unit adds 1MW. At this simplified valuation (caveating that of course Western mining deployments will vary in energy density, cost etc. based on source of power, cooling and other factors), this implies that only a $2.55B investment is needed over the next few years (assuming no new Chinese investment, which can be debated since they appear to have pulled forward their capex), at a time when private equity dry powder is at record levels ($900B as of May), a minor drop in the bucket comparatively.
In addition to the announced deals, I’m personally also hearing of more interest from traditional investors who are still hesitant to buy BTC, but appear to be extremely excited about the infrastructure investment play given the ability to hedge BTC exposure and earn fiat-denominated yields that nevertheless offer extremely favorable payback periods / IRRs versus more traditional private equity assets. I expect this trend to be a hidden tailwind over the next couple of years that seems to be flying under most people’s radar, probably because it’s not as fun to talk about as an IEO or main net launch.
US Response: Jobs Creation and Rural Investment Narrative
If we believe that there’s an intentionality in the shift to tacitly encourage the continued growth and deployment of hash rate for the mining industry in China as part an effort to open up a second, more surreptitious front in the war on global dollar dominance, this leaves the US with a few options in response. One option is to do nothing and leave the free market to develop on its own naturally, which is basically feasible given the relatively pragmatic regulatory approach at least to the mining industry given the CFTC’s classification of bitcoin as a commodity which has stood for years and is unlikely to change. However, I think we could see a step-up in state and local-level government incentives to attract mining investment, especially since the geographies most amenable to mining tend to have experienced tax base deterioration from rural residents moving to cities. It’s notable that the announcement report for the Bitmain facility cited the winning of tax abatements from the county government. I expect this to continue in full force and even accelerate once more government officials realize that such actions serve to counterbalance the 1) Chinese offensive against the dollar, 2) revitalize American regions that need it the most, while also 3) potentially serving to support the environmentalism narrative to the extent sources of power are renewable or would otherwise have been simply wasted (in natural gas flaring), while also being more and more politically palatable over time.
Conclusion: China-Based Hash % Likely Decreases in Future
The current mining dynamic reminds me of the classic fable of the tortoise and the hare. Given China’s proximity to semiconductor foundries and the earlier focus of the CCP compared to the rest of global governments (the sovereign currency project now known as DCEP was announced as focus as early as January 2016), it makes sense that the manufacture and distribution of ASICs has so far concentrated in the country, especially given the relatively low cost power available. However, the factors I’ve walked through suggest the rest of the world will catch up eventually and that we’ll see China’s contribution to the SHA256 hashrate dip back below 50% over the next 5 years, with timing subject to the incentive structures offered by non-Chinese regional and potentially national governments. Specifically, I believe that the combination of 1) Chinese miners being incented to diversify geographically as a personal/political hedge, 2) increased diffusion of mining equipment + longer hardware cycles (enabling larger capital investment projects), and 3) greater appetite for mining operations as infrastructure investment plays outside of China will naturally result in diversification of hash in the base case, though I’ve highlighted some of the obstacles and sources of variance that bear watching (and potentially adjusting risk appetite as a result of).
