Don’t Fear the Reaper
Concerns about Bitcoin’s long-term supply credibility are overblown — but not for the reasons you might think
There’s a popular view within the non-Bitcoin crypto industry that the more educated Bitcoiners are at best naive, or at worst completely in denial on the topic of Bitcoin’s long-term supply. Put simply, Bitcoin skeptics are fond of claiming that, because in their view it’s unlikely that Bitcoin will muster sufficient fees in the long term to compensate miners, Bitcoin will eventually be forced to add inflation in excess of the rate ordained by Satoshi. According to this argument, it’s therefore misleading to characterise Bitcoin as being genuinely fixed in supply. Naturally, this argument is often intended (whether this is acknowledged or not) to justify the insertion of monetary discretion in other cryptocurrencies. Invariably, the “right” rate of issuance for these projects is understood to be nonzero in the long term.
I believe this argument is plainly wrong. In this post I will explain why I think it fails even if the core premise — that fee-driven security might not be sufficient — is a reasonable concern to hold. Lastly, I will explain why I believe that Bitcoiners have nothing to fear, regardless of whether fees are ultimately deemed to be sufficient or not.
A brief note on supply
First of all, I’ll point out the obvious: a hard cap on supply is not essential for something to be valuable. A commodity with a perpetually growing supply can have value, as long as the world maintains some demand for the asset. Every commodity aside from Bitcoin has this quality, and they clearly have nonzero value. Second, the supply of a commodity, physical or digital, does not have to be known to the tiniest increment for it to possess a nonzero value. Gold is treated as a reliable means of wealth storage but its supply is only vaguely known. Of course, Bitcoin is held to a higher standard of auditability because it is untethered from physical realities, and a bug can inflate supply arbitrarily — so we need to keep a closer eye on it. This piece is not about the merits of a hard cap or its role in maintaining a high unit value. Instead it’s about the relationship between that hard cap and the nature of Bitcoin.
For the sake of this piece, I’m going to leave aside any prognostications as to Bitcoin’s long term security budget. I happen to believe that it’s entirely plausible for Bitcoin to be secure in the long term based solely on fees, but I won’t argue that here. Further, common stipulations that hypothetical future Bitcoin fees be a certain threshold to match current security spend make no sense. There’s nothing special about today’s security spend, which is merely a function of unit price and the issuance rate. Historically, Bitcoin has been secure across a range of security spends, ranging from $0/day to $54m/day.
The inflationist dilemma
Let’s return to the core point. It’s common for critics to point out that Bitcoin must add inflation, and as such its disinflationary current nature is illusory. In the interests of specificity, let’s reduce the argument in question to a syllogism:
- (Premise) There is a possibility that one of Bitcoin’s features (the predetermined issuance schedule) might render it fragile
- (Premise) A specific change (inserting issuance above the predetermined level) will be necessary in order to remediate that fragility
- (Premise) If one of your system features is guaranteed to change in the future, you cannot credibly name that feature as a durable constituent of the system
- (Conclusion) Bitcoin at present includes the future obligation to change, and therefore cannot be described as possessing the desirable feature of capped supply
I acknowledge the possibility of (1), although I believe that these distant outcomes are both unforecastable and not as certain as critics let on. (3) is trivially true. I reject both (2) and (4). It’s not guaranteed that, if long reorgs were to start occuring due to a security shortfall, Bitcoiners would propose reinserting inflation. There are many alternative non-inflationary remedies, ranging from the institutionalization of mining to coordinated counter attacks, to soft forks to invalidate lengthy reorgs. And I also reject (4), because, as I will discuss in this piece, a system which reinserts additional inflation is a fundamentally new system and cannot be considered ‘Bitcoin’ as Satoshi described it and as it’s currently understood.
Bitcoin — the thing we today know as Bitcoin — does not contain any assumption held either by its creator or the community that its supply is bound to deviate from its current trajectory. While an alternative version of Bitcoin may be created which deviates from the ordained supply schedule, this would have as much claim to being ‘Bitcoin’ as other forks which changed critical features, like Bitcoin Cash, which is to say, a questionable one.
So it is false to assert that Bitcoin contains the embedded assumption by users that a change to the rate of issuance is either possible or necessary. While Satoshi’s Bitcoin may yet fail — for any number of reasons —it cannot deviate from its supply schedule, because any deviation would entail the creation of an entirely new and distinct asset. And lastly, if a transition from Satoshi Bitcoin (i.e. the thing we call ‘Bitcoin’ today) to a novel asset were to become necessary, it’s in no way guaranteed that this would be a traumatic event.
The thing that we call ‘Bitcoin’ cannot suffer an alteration to its supply schedule, because the supply schedule is intrinsic to the protocol, asset, and system. It’s one of the very few features that Satoshi clearly stipulated and encoded in the original design with no ambiguity. And in the eyes of the broader community, the 21m cap is perceived as inherent. Ask a Bitcoiner to define the system and “scarcity,” “fixed supply,” and “21 million units” will invariably be mentioned. There is no doubt that the system dubbed “Bitcoin” as baptised by Satoshi contained a deeply ingrained stipulation over the issuance rate; and that this endures today in the popular conception of the asset.
Take this passage from the white paper: the first ever description of the Bitcoin system:
Once a predetermined number of coins have entered circulation, the incentive can transition entirely to transaction fees and be completely inflation free.
Note the presence of ‘predetermined.’ Satoshi had determined the number of coins prior to the protocol going live, and the white paper — the closest thing we have to a constitutional document — clearly stipulates that this could not change post-launch. Satoshi had already accounted for how miners would receive revenue in the post-issuance future.
Now take the original content included on the first version of bitcoin.org to describe Bitcoin.
Total circulation will be 21,000,000 coins. It’ll be distributed to network nodes when they make blocks, with the amount cut in half every 4 years.
first 4 years: 10,500,000 coins
next 4 years: 5,250,000 coins
next 4 years: 2,625,000 coins
next 4 years: 1,312,500 coins
On that page, Satoshi also describes Bitcoin’s ‘main properties,’ one of which is No mint or other trusted parties. For supply to remain untrusted and for monetary discretion to remain absent in perpetuity, a predetermined issuance schedule would have to be adhered to. Picking an inflation rate which could respond to changing circumstances requires a trusted entity to make that choice. Satoshi was clearly not prepared to accept that.
Satoshi reiterates their commitment to the hard cap in subsequent posts on BitcoinTalk.
Otherwise we couldn’t have a finite limit of 21 million coins, because there would always need to be some minimum reward for generating. In a few decades when the reward gets too small, the transaction fee will become the main compensation for nodes.
There’s simply no ambiguity on the question of whether the supply is essential to Bitcoin. Bitcoin’s supply schedule cannot change, because Bitcoin is the supply schedule. Any alteration produces something that is decidedly non-Bitcoin.
Many Bitcoiners will disagree with me on this point, arguing that we should adopt a more moderate stance and make our conception of ‘Bitcoin’ malleable to account for potential contingencies like a change to the supply schedule. I disagree. I believe that Bitcoin is far more concrete and easier to reason about if we agree on its most essential features. It’s this intransigence which has thus far warded off capture and perversion from projects which assumed the name yet which failed to instantiate Bitcoin’s vital qualities. The biggest threat to Bitcoin isn’t its outright failure — it’s far too important now for the community to roll over and accept defeat — but rather its conceptual dilution. By committing to a firm vision of what Bitcoin is, we have a concrete notion to build towards and rally around.
Bitcoin entitles you to its successor
That said, this does not mean that Bitcoiners are fated to fritter away on a doomed chain if Bitcoin should suffer some fatal flaw. While this is hardly a new point, it’s worth reiterating. If Satoshi Bitcoin indeed falls out of favor, any worthy successor will be one which is allocated to all Bitcoin holders. This maneuver is not unlike Ethereum’s impending transition. Ethereum 2.0 will have little in common with Ethereum 1.0 aside from the name and the fact that owning units on 1.0 entitles to you to a pro-rata share of 2.0. While Bitcoin Cash and other such forks failed, they were on the right track from a distribution perspective. Any worthy Bitcoin successor would have to be built atop the pillar of investment and thermodynamic security accumulated within Bitcoin which gave it such vibrance and dispersion. A fair launch is scarcely possible today — the stakes are too high, and large investment funds will scoop up all available supply before the public, rendering it hardly different from a premine. The only way to imitate Bitcoin’s truly fair distribution is to inherit it.
So in the final calculus, if Satoshi’s Bitcoin fails; so be it. It will have failed gloriously. A worthy successor will definitionally be one which incorporates the $19B worth of accumulated miner work and resumes atop Bitcoin’s fairly-distributed UTXO set. The fair launch is just as essential a design characteristic as the supply schedule is. Thus, Bitcoiners need not worry. They own a claim on current Bitcoin, but also all of its most viable descendants if, for whatever reason, v1 doesn’t pan out.
I’ll leave you with this. There is no optimal nonzero rate of monetary issuance. The issuance or dilution rate is a political question, not an engineering one. Always, there are groups that will stand to benefit by increasing the flow of new units — groups that are protocol-proximate and can potentially exploit the new flow: Cantillon insiders. And conversely there will be groups who will push back against new issuance because they already have a share of supply and don’t want more dilution. All monetary debates boil down to this: a conflict between those who benefit from a loose money regime and those who suffer from it. The only way to arrest these political squabbles is to take monetary discretion off the table from day 0, as Satoshi did. Any positive rate of issuance is completely arbitrary and the pro-inflationists will always be able to argue that a small increase will bring a large gain (for instance, to finance some new infrastructure project). Thus the anti-inflationists who occupy a nonzero issuance regime will forever be open to attacks from the pro-inflationists, and the rate of issuance will always be a political pawn. In short: you cannot commit to a stable nonzero monetary rule, because there is no stable nonzero equilibrium.
Bitcoin breaks the cycle, entirely removing any discussion of monetary rules and instead ordaining an initial release (or vesting) schedule. The rules of the game were clear, even before the first block was mined. No one can complain that they were hard done by or that the issuance rate took them by surprise. Thanks to the beauty of PoW, issuance is a competitive free market process, with no special privileges to protocol insiders. So Bitcoin is genuinely new in that it entirely removes monetary discretion from the equation.
No other monetary system has done this before, and as far as I can tell, none has done it since. I have yet to come across a Bitcoin alternative that can credibly allege to be truly nondiscretionary in its monetary features. This uniqueness is important and worth preserving. Instituting monetary discretion is completely mundane and exactly how every government and central bank already operates. Non-Bitcoin crypto elites would just have us substitute central banks for protocol architects. But that’s an old story, and it’s not one I’m particularly interested in.
So Bitcoin, Satoshi-Bitcoin, Bitcoin-with-21-million-units — that’s a grand, bold experiment in eliminating monetary discretion. It’s the only remotely new monetary project at play here. Everything else is just a high-tech reinstitution of what we already have: monetary arbitrariness which allows elites to plunder a torrent of new issuance for their own gain. If it fails, so be it. We will pick up the pieces and move on, and Bitcoin will spawn a worthy successor — which will have to be called something else. But it’s worth putting a marker down and believing in something. Otherwise we have nothing.