Achim Warner gives a compelling argument that, if done on a large scale, ETF investment could bring about a hard fork.

The main pith of Warner's argument is that corporations ultimately might dismiss mining work as a necessary security feature, and just see it as an unnecessary expense:

Well-written, thought provoking stuff.

Is this likely? Could this be prevented?

Warner's article is printed below.

The ETF approval confers some traditional legitimacy to the asset class, and allows billions of dollars (or maybe $30 trillion according to Grayscale) in traditional investment to flow in from a wider pool. For those interested in Number Go Up this appears to be fantastic news. But if that $30 trillion actually does come in, it will destroy Bitcoin as we know it. It will fundamentally change the economics so that the game theory regime upon which Bitcoin was built will no longer be in effect. We won’t bother here to discuss the 180 Bitcoin’s ethos has done — fighting oppression vs building intergenerational wealth, because this pivot is obvious. Here we discuss the game theory.

Bitcoin takes on two very ambitious and distinct goals. One is to create a widely accepted store of value with a fixed programmatic supply. The other is to create a decentralized cash system. In the evolution of Bitcoin, these two objectives have played mostly hand-in-hand; the accepted viability of Bitcoin as a store of value creates a liquid market, while use of Bitcoin as electronic cash creates demand, which in turn gives Bitcoin value. At smaller scales the incentives are aligned.

Let’s start with the former, the ideal of a global Schelling point for store of value. The existence of the Bitcoin blockchain is quite remarkable. The ledger apportioning bitcoin has slowly invaded all corners of the global economy, representing what has been described as a slow-moving immaculate conception. A centralized actor trying to create this from scratch would undoubtedly fail. The distribution isn’t perfect, but there’s no “fair” way to airdrop a new asset across the planet without doing something odious like eyeball-scanning. A slow distribution over 15 years is the best one could hope for. There seems to be agreement among diverse market participants that this thing is useful and desirable and that the appetite for shares in the ledger will persist in the future.

If we, “we” being everybody and anybody who is interested in such a thing, want there to be a fixed supply digital asset than can be traded as an ETF through our brokers, we now have this thing. It’s a unique occurrence that will not happen organically again. Note that, moving forward, the existence of this thing itself has nothing to do with how it is mined or how shares in the ledger are recorded. This is simply a Schelling point that everyone in the world who values a fixed supply asset can trade.

Given any cryptographic payment system, there are many methods for coming to consensus over the canonical ordering of transactions. The easiest and most efficient way is to choose a single entity for the task. Bitcoin mining is the opposite: clunky, competitive and expensive. The downside is that it’s costly. The upside is that it’s extremely difficult for any one entity to arrogate control over the ledger. Miners, who generally benefit from the popularity, have no desire to damage the ecosystem by colluding with other miners or otherwise trying to game the system with untoward behaviors. This is how the virtuous cycle began.

To bootstrap this virtuous cycle, we had to have Bitcoin’s initial use case: Peer-to-peer electronic cash. Money for those not well-served by the traditional financial system and those who want to be financially self-sovereign. The fact that the payments are censorship resistant is a backbone feature here: This property gives Bitcoin non-zero value, allowing people to begin trading and distributing it not just as something they can use to buy drugs, but soon also something to hold onto, and eventually, a speculative investment. An actual functioning censorship-resistant currency is worth something. This was the bedrock of Bitcoin’s value; if this fails, the Bitcoin experiment has failed.

But today, the necessary supporting role that has been played by censorship resistance has been usurped: this role is played by the linchpin of institutional demand, manifested in the ETF. The former is along for the ride, but only until it can be conveniently jettisoned.

Think of Bitcoin like a papier maché piñata that you build from a balloon, paper and paste. To start the process, you need to inflate a balloon. Carefully, over time, you cover the balloon with paste-drenched strips of paper or string, adding layers, one at a time, pausing after each to let it dry, repeating the process, until the balloon is completely covered by a solid surface. Then, you pop the balloon and extract it from a hole in the bottom, which you can then fill with candy or whatever. In the initial stage, you must be very careful not to pop the balloon (don’t use a hairdryer!) if you do so the whole project will be a soggy mess. In the latter stage, once the papier maché has hardened, the balloon is just some extra thing you need to get rid of.

At this point, perhaps you are mouthing something about the Blocksize War, wondering if perhaps the author of this essay has not read it. Yes, in fact, several times and the author has even spent three weeks discussing it in a university course he instructed. There is no single lesson to be learned from Blocksize War; it was a complicated story, filled with blunders, turnabout, well- and poorly-played hands. The idea that the team that won in 2017 is forever the victor is silly. Just like winning the Super Bowl in 2017 did not confer the New England Patriots with eternal infallibility, the victory by the small-blockers in 2017 does not mean every future battle will have the same outcome. This is one of the most non-sensical tenets of Bitcoin dogma.

It is entirely worth looking closer at the Blocksize War and asking what we did in fact observe and what sort of realities do persist. The dominant takeaway seems to be that users — Bitcoin’s grassroots plebs, ultimately had the stronger hand. They flexed this muscle with the UASF, and eventually most of the corporations and miners on the large-block side capitulated.
There’s a crucial piece of game theory here. If large blockers had elected to call the bluff on the UASF, it probably would have been catastrophic for Bitcoin. The miners had capital investment they couldn’t stand to lose money on, so they had no choice but to swerve in this game of chicken. In 2017, without the support of the grassroots small-blockers and their decentralized ideals, Bitcoin would have faced a possibly disastrous constitutional crisis. In this sense, the users did have the upper hand.
Part of the genius of Bitcoin is that it was designed with a failure as a possibility. To take the failure mode away fundamentally changes the game theory.

What happens today? The locus of economic nodes has shifted drastically away from raspberry pi hobbyists and towards large financial institutions who are doing Bitcoin for Wall Street. When retail consumers, pension funds, sovereign wealth funds, etc., are buying shares in the fixed supply currency through traditional finance channels, they are cementing a demand for share in this ledger that doesn’t depend on its decentralized maintenance. If the ETF invites an influx of such investment, any cartel of corporations who would attempt a takeover can now safely ignore the threats of Bitcoin purists; these early adopters can no longer claim the be the crucial pillars propping up the system. If there is another game of chicken, Wall Street can stay the course. Would Michael Saylor sabotage his investment by dumping his stash because he doesn’t like a miner-backed OFAC enforcement? Wall Street is smart to call the bluff.

If Bitcoin becomes no more than an agreement by major financial players and their clients that tradeable entries in this ledger are valuable, it becomes less important how we keep track of who has what share, provided the process is efficient, predictable, and legible to regulators. The current Bitcoin mining process was designed to be the opposite. Basic forces of economics suggest this inefficiency eventually will be shed.
Every round in which institutional investors ape into the ETF is like another layer on the papier maché balloon. Enough of these and the balloon is no longer necessary.

Why not have 8 MB blocks? Or 256 MB blocks? The retail ETF investor buys shares because they believe that someone in the future will want to buy their share, presumably because these future investors in turn believe that a future demand exists for their share. This may very well be a good bet, but none of this self-fulfilling dynamic relies on the minutia of how the transactions are ordered. When registered market participants are trading shares through registered brokers, uninterested in double-spending or shielding activity from authorities, the mining process becomes an expensive and obsolete Rube Goldberg machine, at least as it applies to an economic preponderance of activity. The costs associated to mining leak billions of dollars out of the Bitcoin ecosystem. Without a clear reason for this expensive process to exist, the financial sector will seek to end it. A thriving ETF market, especially when celebrated by Bitcoin holders, provides a sturdy structure and leverage for this attack.

To put it simply, there are now corporations with billions of dollars of investment in not just the asset, but the ecosystem. Mining is a huge economic drain on the system, and all it accomplishes is censorship resistance. If censorship resistance isn’t worth tens or hundred of billions or dollar per year to the corporations; they will find a way to end it. When they do try to end it, they can point to the ETFs enduring appeal and ignore any constitutional brinkmanship from OG Bitcoiners who threaten to reject the new rules or sell their stacks.

Mining itself represents a huge arbitrage opportunity available to someone who can leverage capital or organize a cartel. Take Cathie Wood’s $1.5 M in 2030 case. This is over $100 billion per year vaporized in order to keep the network censorship-resistant, just counting the block subsidy. These same transactions could be processed for a microscopic fraction. If some entity can come in and monopolize the process and lower the difficulty without destroying the game, this is free money.

It’s a valid question to ask, how? You can’t just change the code or fire the miners, right?

First note that there will be two sets of conversations going forward. One is the usual one, on Twitter spaces or other open venues, where we argue about the relative merits of OPCAT vs CTV as means to enable protocols that will make Bitcoin a self-sovereign option to billions of people. But there’s another set of conversations, which may have already begun, happening in private board rooms. These conversations we will not be shared with the rest of us. Think Dragon’s Den, but more like Vulture’s Roost. The folks in the Vulture’s Roost also want to make Bitcoin an option to billions of people, just without the self-sovereign part. Rent-seekers gonna rent-seek, and this isn’t simply about holding the asset and watching the number go up. It’s about re-intermediating the financial sector.
So for the next part of the exercise, put yourself in the Vulture’s Roost. What are your incentives? What tools do you have available to you? What risks are you trying to minimize? Who else is on your team? In the Vulture’s Roost, you are free to say without judgment things like “will noone bankrupt these turbulent miners?”

The goal from the Vulture’s Roost will be to destroy mining and replace it with an inexpensive cartelized process that routes all transactions fees directly to a small group of insiders (instead of towards miners’ energy bills) and puts insiders in positions to do insider things, like profit from rich data streams or fancy financial products.

The way to accomplish this is to signal that you are going to do it, then start doing it. Those who want to be part of your coalition are in. Those who want to fight it will eventually fall away to attrition. In this psychological game, you will slowly convince investors that not only is centralization inevitable, but it’s actually not-so-bad, after all.
We can assume that there are high fees on L1. If not, miners will eventually start to suffer (Number Geometrically Must Increase) and you can just buy them up as they go under.

In a high fee regime, your first goal will be to route as many transactions away from the decentralized mempool as possible. If you can just buy up 33% of the hashrate and start selfish mining that will be too easy, so we’ll assume a more difficult base scenario in which you can only start with 5% to 10%. This is still a block every couple of hours. You can contact any major institutional user of the blockchain, and broker out-of-band deals for blockspace. You can undercut the mempool by offering frequent users not only a discount, but an opportunity to shed fee volatility. You can even undercut the mempool openly with transaction accelerators priced below market rate. If you undercut the mempool market enough, you can begin to force many of the other miners out of profitability. You do this openly and brazenly: The goal is that any would-be home miner considering spending $20,000 on the latest ASIC will read the writing on the wall and think twice.
Here it’s convenient that many of the transactions are brokered by corporations, not random self-sovereign people from around the world. It will be easier to capture a large of the transaction volume if most transaction volume is handled by major corporations or generally KYC’d high wealth individuals. But this isn’t a stretch; if fees are high, L1 Bitcoin won’t be an accessible means of peer-to-peer payment for most people.

Most of the hyperbitcoinization scenarios assume that major institutions are willing to pay high fees to etch transactions into the blockchain.
At the same time, you can take advantage of the high-fee regime to build a side-chain network or other alternative means of Bitcoin accounting. All you really need to do this is put together a network with critical economic mass, backed by small amount of trust, but mostly backed by Delaware’s Court of Chancery. For example, a cartel of corporations can create a multisig UTXOs on the mainchain, and then create their own blockchain. This could be something as simple as Fedimint or Liquid, or mildly decentralized and creative like Ark, based on fancy things like timeout trees. You can create an entire permission-mined blockchain downstream of these UTXOs with 3 second blocktimes and 32 MB blocks. As long as nobody thinks the corporate block validators are going to rugpull (and get sued in a meatspace court), people can treat this like actual Bitcoin UTXOs. Anyone who wants to transact in a near zero fee regime can simply slide through certain access points over to this sidechain, saving perhaps millions of dollars in fees. For institutions who want to cut costs, this is a no-brainer. It’s orders of magnitude better than Lightning, and plus, if it goes wrong, you have someone to sue. The financial sector is too smart to sit around and pay billions of dollars of year in transaction fees.
So while we still may have a fixed supply of coins asymptotic to 21 million, the practical ways in which transactions are recorded will mushroom. The ETF is just the beginning of this process. Layer 1 fees will trend towards negligible, except for the few people who really need to fully self-custody and are willing to pay market rate for that.

Once you have moved a large volume of transactions away from L1 and towards the traditional financial sector, you can go after mining itself. Having starved the decentralized miner of much of their fee revenue, you next want to scare up enough hash power to force them out for good. At this point, you’ve been open about your intentions. Miners can see where this is headed and when you offer to buy them out, they say “yes,” or at least the rational ones do. After your cartel gets to 33% you can begin selfish mining, commencing the snowball phase towards 51%. After 51% is achieved you now have full control and can ease down the hashrate, turning it back on if necessary to break the will of any stubborn decentralized miners. Soon, hashrate is minimal but your cartel is still receiving the full block reward.

But wait! Won’t everybody abandon Bitcoin when you start selfish mining and everyone sees that you’re selfish mining? That’s exactly the point here. Once the retail and institutional ETF investors are propping up the price with their 1–5% IRA allocations, nobody cares.

But also wait! Won’t all of the miners fight back? The dynamics of wars of attrition, dollar auctions, and coalitional game theory come into play. The decentralized miners have no way to do selfish mining without coordinating somehow. If some subset of miners tries to selfish mine in retaliation, it will push all the anonymous miners out for good, making your job easier. Your job is to send a strong enough signal that you have centralized resources behind your back, and so will win a war of attrition. Once you send this signal, it’s a prisoner’s dilemma among the remaining miners: Walk away with the full value of your ASICs today, or get nothing in 6 months.

You haven’t broken any rules yet, nor pushed any hardforks. But after arrogating full control, you are free to hardfork at will. At this point it will be a formality. After years of telling people what you were going to do, and then doing it over the objections of the purists, The Hard Fork is just the final step. Not that you have to worry about blowback; the final step to permissioned 10 GB blocks will be widely celebrated as the next step in bringing sound money to everybody.