Saifedean Ammous’s The Bitcoin Standard, the New Standard for Bitcoin Books
With all the hype in the crypto space, it’s gratifying to read a book about Bitcoin which (apart from a brief prologue) contains only incidental mention of Bitcoin for the first two-thirds of the book. Ultimately, as you’ll gather from the title, Bitcoin plays a fairly grand role, supplanting gold as the historical reserve standard, but by building a groundwork understanding of money’s emergence, history, and evolution, Saifedean Ammous makes a sober arrival at his final destination.
Money has traditionally had three facets: medium of exchange, store of value, and unit of account. Sound money, Ammous’s argument goes, also requires salability (“the ease with which a good can be sold on the market whenever its holder desires, with the least loss in its price”) and a high stock-to-flow ratio (the ratio of existing captured supply to what’s added to it over a given time period). Much attention is devoted to the importance of a high stock-to-flow ratio with the conclusion that Bitcoin will soon have the highest stock-to-flow of any money, overtaking gold in the year 2022 and doubling gold’s in 2025.
A high stock-to-flow ratio explains why gold has been a consistent store of value for millennia and why silver has been a second-place store. Demand for money fluctuates in the same way it does for goods on the market. When demand spikes, resources are diverted to producing more of the demanded good — since now the effort will be rewarded with a higher price. The best money maintains a high stock-to-flow ratio even through periods of high money demand. And this has precisely been the case with gold. In the past seven decades, Ammous shows, gold has averaged an annual growth rate of 1.5%, never surpassing 2% even during the highest periods of demand. Ramping up mining operations will of course on average result in some increase in gold production, but this increased supply has to be measured against relatively massive existing stockpiles, which have been accumulating for millennia free of corrosion due to gold’s chemical properties. Silver’s stockpiles, on the other hand, diminish via corrosion and higher rates of industrial consumption. At the same time silver production is more amenable than gold to increased demand, being more abundant and easier to mine and refine. All of this accumulates to an overall lower stock-to-flow ratio.
This is all antithetical to the Keynesian perspective which stipulates that governments control and manage the supply of money, stimulating during downturns when consumption flags. To sum it briefly, Ammous isn’t all that jazzed about this perspective. The book expands quite a bit on this, but the primary critique is that government control lowers money’s stock-to-flow ratio. The Keynesian program isn’t mindless continuous stimulation; at certain junctures it’s time to remove the proverbial punch bowl. Yet when these moments arrive no one seems to have the nerve to do it. Alternatively, finding ourselves in situations like the 2008 scenario where the economy doesn’t make the called for turnaround following extraordinary stimulation, Keynesian economists come to the unfalsifiable conclusion that we simply didn’t stimulate enough. There’s no proven formulation for proper magnitude of stimulation and so no credible restraint mechanism to prevent a free-floating currency from inflating — whether it’s the Keynesian “right time” for this maneuver or not. It turns out that a money which can be “managed” will be managed quite heavily upward; a superabundance of imagination isn’t required to conceive of some incentive to create more money given the ability.
Milton Friedman too is culpable in all this as an advocate of free-floating currencies and for overlooked causative factors in his historical analyses. But criticisms here are limited to Friedman’s professional work, while Keynes’s entire life is open to reproach. Keynes devotees will be unimpressed, but these criticisms although bitter aren’t strictly ad hominem. In Keynes Ammous diagnoses (fairly or unfairly readers can decide) personality traits which inform his entire economic philosophy. A major aspect of the book’s argument for sound money versus easy money relates to the time preferences biased by one compared to the other. A selling point for sound money is its low time preference bias, and Ammous paints Keynes as a high time preference individual who consequently advocated high time preference money. Ammous introduces the concept of time preference with the marshmallow test (by 2018 probably familiar to at least 80% of popular non-fiction readers or TED Talk watchers), whereby the children who eat the first marshmallow right off the bat have high time preference (immediate consumption is much more valuable to them than future consumption), whereas the children who forgo consumption in order to double their reward have a lower time preference (current consumption is only somewhat more valuable to them than future consumption, and so when future consumption entails higher payoff than present consumption, they wait).
Low time preference is more likely to prevail with monies that maintain purchasing power over time, while time preference increases with monies whose purchasing power declines. (Although “nudges” would imply top down control — which Bitcoin intrinsically deactivates — if you like thinking in these terms, sound money nudges toward lower time preference while easy money nudges toward higher time preference.)
If society were a little girl in that marshmallow experiment Keynesian economics seeks to alter the experiment so that waiting would punish the girl by giving her half a marshmallow instead of two, making the entire concept of self-control and low time preference appear counterproductive. Indulging immediate pleasures is the more likely course of action economically, and that will then reflect on culture and society at large. The Austrian school, on the other hand, by preaching sound money, recognizes the reality of the trade-off that nature provides humans, and that if the child waits, there will be more reward for her, making her happier in the long run, encouraging her to defer her gratification to increase it.
The U.S. Federal Reserve targets a 2% inflation rate. This continuous devaluation of stored wealth is intended as a mild economic stimulant. As Ammous has it, knowledge that your wealth’s purchasing power isn’t preserved while stored in USD incents levels of risk-taking beyond what people would otherwise be disposed to if they knew their wealth was safe. The risky investments that become necessary for individuals to compensate for the drain on their wealth aggregate to systemic overcompensation. Indeed, this stimulates the economy, but the planned injection synchronizes local ups and downs into systemic oscillations of boom and recession.
When, on the other hand, wealth is preserved, people remain free to take risk but without being goaded into otherwise intolerable levels. They’re staking real wealth for the prospect of real return rather than lobbing around financial hot potatoes. Under this more Austrian specification we don’t get the systemic cyclicality of top-down injection. Planned inflation also amounts to an opaque tax — a tax that a population might be less willing to abide, for activities they might not approve of, were all this publically tabulated for clearer inspection. Although this isn’t necessarily a small government argument; funds will be allocated most optimally when all costs are most accurately accounted for whatever the size of the governing body in question. (Much more on all of this as it relates to fractional reserve banking, interest rates, and Austrian business cycle theory is outlined in the book.)
Nakamoto’s design choice of a 21 million supply cap for Bitcoin turns out to be more of a psychological anchoring point (some number had to be settled on, and scarcity but not a level of scarcity intimidating to newcomers would be a good range to shoot for with a sound money experiment) than an economic consideration. The network would function just as well with only one bitcoin so long as it were divisible to the full extent of users’ needs. (And here’s where Bitcoin starts to show possible superiority over gold: subdividing a block of gold is much more cumbersome than doing the same to a bitcoin.) While money supply itself is in fact inconsequential, adjusting it increases unpredictability, destroying information signals, creating distortions and bubbles, and harming what should really be protected: purchasing power. Tying back in with time preference, the idea is that if you spend your money in ten years it should purchase just as much or more as it would spending it today.
If money supply itself doesn’t matter (given adequate divisibility), resources diverted to increasing the money supply could be put to more productive use instead. The easier a money is to create, the more capital that will be unproductively diverted into increasing its supply. The sign of a good money is when intensifying efforts to create more of it is a bad investment, and Bitcoin’s difficulty adjustment ensures that more effort will not result overall in a greater number of newly minted coins, preserving holders’ purchasing power.
As Ammous shows, the ability to manipulate money supply is perpetually abused. It’s a lure that — whether stemming from tyrannical greed, strategic desire for obfuscation, or ill-guided good intentions — doesn’t go unheeded for long. There’s the example of coin clipping where metal clipped from around the edges of officially minted coins allowed Sovereigns to decrease the official weights of their coins in order to mint more for themselves. There’re also examples of isolated preindustrial societies whose formerly stable currencies (such as rai stones used by Yap Islanders and the glass beads at one time prevalent in west Africa) lost all value when foreign industrial-scale reproductions flooded local markets. The devastation to local wealth and purchasing power was so great in Africa that they resorted to selling their own people as slaves.
Although these industrial-preindustrial asymmetric interchanges offer disturbingly vivid accounts of monetary wreckage, they highlight the dangers of currencies that aren’t robust against supply inflation. Interestingly though, these anecdotes paint money as technology, perhaps not the first designation everybody would consider. The best form of money so far — by sound money accounts — has been gold, which certainly doesn’t seem technological. But exactly this kind of fruitful thinking is how Bitcoin — not without its own IT ancestry — arose.
Even with its rarity and chemical stability gold has its attack vectors. It’s not easily divisible for everyday small transactions (the mythic cup of coffee). (Substituting silver or another lesser metal for everyday transacting solves this problem but opens the door to exchange swings between the two poles of a bimetallic system.) Meanwhile, gold is relatively transportable (it’s a denser store of value than, say, cattle — another historical currency from the book), but it’s really not that convenient to carry around with you. For this reason people began storing gold in banks. The banks then began issuing paper notes to be traded in place of gold while the gold itself remained in the banks’ vaults. While paper notes solved the divisibility problem, the fact that the gold now remained in the banks was a new point of centralization — one that didn’t go unnoticed by governments. It’s laid out in nice detail in the book but over the course of a few decades of the 20th century gold made its way from these dispersed local banks to the official central banks of governments, accompanying our transition from a gold standard (paper notes backed by gold) to a system of free-floating exchange. Aside from destroying people’s saved wealth this caused havoc for international trade:
It is an astonishing fact of modern life that an entrepreneur in the year 1900 could make global economic plans and calculations all denominated in any international currency, with no thought whatsoever given to exchange rate fluctuations. A century later, the equivalent entrepreneur trying to make an economic plan across borders faces an array of highly volatile exchange rates that might make him think he has walked into a Salvador Dali painting.
But dreams of a return to a halcyon gold standard, of gold as a world reserve currency, are misguided — and not merely due to the technological limitations. Monetary and fiscal policy are firmly established tools of government — with plenty of additional beneficiaries. It’s difficult to see how any entity would willingly relinquish this kind of control. Friedrich Hayek voiced some awareness of this in a 1984 interview quoted by Ammous: “I don’t believe we shall ever have a good money again before we take the thing out of the hands of government, that is, we can’t take it violently out of the hands of government, all we can do is by some sly roundabout way introduce something that they can’t stop.”
The Bitcoin project may still fail, but if it can’t be Hayek’s sly roundabout introduction of something governments can’t stop it shouldn’t have had much value in the first place. While we’re still waiting to see if Nakamoto was sly enough to pull it off, with each passing day the network becomes more robust. Ammous invokes Nassim Taleb’s antifragility concept (and Nassim Nicholas Taleb followers will be interested to know that he wrote The Bitcoin Standard’s forward), citing how attacks on the network have only so far strengthened it and inoculated it through code patches against repeat attacks on any given vulnerability.
If capacity for inflation will never remain latent in a money for long, the technological aspect of money is precisely the capacity to resist this. Bitcoin removes the ability to modify supply altogether. Critically, Bitcoin’s difficulty adjustment makes it so that more resources put into mining bitcoins will not result in more bitcoins being created than planned for by the code, but instead result in making the network itself more secure and harder to attack. Greater security increases its attractiveness as a store of value, which raises the price, which increases attractiveness to mine bitcoin, which increases network security, and so on.
The protocol’s predictability promises to preserve as much of the information signal that travels through the monetary medium as possible. Bitcoin’s robustness, it’s resistance to modification, is achieved via mass distribution of its ledger on nodes across the world. But such extreme redundancy makes for a functionally rather inefficient system. If this is the case, why do we keep hearing so much about blockchain technology these days? Ammous paints the blockchain-not-Bitcoin rhetoric that’s popped up over the last few years as so much hype for doomed endeavors to wrest Bitcoin’s market and mind share over to projects which have no intention of ceding control or centralization and therefore have no use for the core value proposition of the blockchain (with its inherent inefficiency) in the first place:
‘Blockchain technology,’ to the extent that such a thing exists, is not an efficient or cheap or fast way of transacting online. It is actually immensely inefficient and slow compared to centralized solutions. The only advantage that it offers is eliminating the need to trust in third-party intermediation. The only possible uses of this technology are in avenues where removing third-party intermediation is of such paramount value to end users that it justifies the increased cost and lost efficiency. And the only process for which it actually can succeed in eliminating third-party intermediation is the process of moving the native token of the network itself, as the code of the blockchain has no integrated control over anything taking place outside it.
Similarly, the book has little positive to say with regards to altcoins, making the case that most projects have zero grounds for their trumpeted decentralization claims:
[V]irtually all altcoins have a team in charge; they began the project, marketed it, designed the marketing material, and plugged press releases into the press as if they were news items, while also having the advantage of mining a large number of coins early before anybody had heard of the coins. These teams are publicly known individuals, and no matter how hard they might try, they cannot demonstrate credibly that they have no control over the direction of the currency, which undermines any claims other currencies might have to being a form of digital cash that cannot be edited or controlled by any third party.
With so much debate over blockchain project governance these days (be it on- or off-chain schemes), Ammous doesn’t seem to find much use in any of it. Bitcoin’s core value proposition is again ultimately it’s resistance to modification. It’d only undermine confidence in the currency as a store of value if it could be modified willy-nilly. Ammous posits it as a virtue that the upgrade process is slow and that persuading various factions of the merit of network upgrades isn’t something that happens overnight. It’s a strength of the network that there’s debate at every step of the process. So in this model, while Bitcoin isn’t likely yet in its final form, it’s not far from it. There’ll be some efficiency upgrades, but so resistant to modification is the protocol that we’re not likely to see any wild boosts to on-chain capacity (this dovetailing additionally with the assessment that protocol-level scaling increases centralization pressures).
And so getting back to the Bitcoin standard of the title, Ammous envisions (unlike some of Bitcoin’s earliest adopter who foresee Wall Street in nothing but rubble in Bitcoin’s wake) a return to a system of banking similar in style to that of the gold standard heydays but with Bitcoin taking over as reserve currency. In a system of sound money banking,
bankers perform two highly pivotal functions for economic prosperity: the safekeeping of assets as deposits, and the matching of maturity and risk tolerance between investors and investment opportunities. Bankers make their money by taking a cut from the profits if they succeed in their job, but make no profit if they fail. Only the successful bankers and banks stay in their job, as those that fail are weeded out. In a society of sound money, there are no liquidity concerns over the failure of a bank, as all banks hold all their deposits on hand, and have investments of matched maturity. In other words, there is no distinction between illiquidity and insolvency, and there is no systemic risk that could make any bank ‘too big to fail.’ A bank that fails is the problem of its shareholders and lenders, and nobody else.
Where transaction scaling is concerned, Ammous sees banks minting their own cryptographic tokens pegged full-reserve to their bitcoin stores so as to avoid using the scarce Bitcoin block space. The blockchain itself will function as an ultimate settlement layer in the way that gold was used to settle balances between banks in the pre-central bank days. This vision of Bitcoin’s future is far from uncontroversial within the Bitcoin community, but the argument is well laid out and the book should be informative to whatever side the debate insiders fall on.
If curious about how banks clearing in bitcoin instead of gold will avoid the centralization pressures gold eventually caved to, it seems that one of the main problems with gold is its cumbersome clearance process — transporting it. Gold centralized because gold clearance centralized. Moving data around the Bitcoin network won’t have gold’s same transportability burdens and risks.
Although the vaunted lightning network is mentioned, little is said about its potential role. Likewise, Bitcoin’s privacy/fungibility issues are completely unaddressed except maybe implicitly in the fact that Ammous counts major protocol changes as especially unlikely going forward. You might also implicitly conclude from the book’s argumentation that protocol-level privacy features would diminish the transparency necessary for Bitcoin to function as final settlement. Furthermore, banks can fractional reserve on a gold base because gold reserves are difficult to audit; if blockchain activity were more obfuscated banks would have a more difficult time proving their bitcoin reserves, opening doors to fractional reserve. Leaving some of these issues unaddressed may seem like an oversight, but by not weighing in on each and every brewing debate and controversy the book better meshes Bitcoin into the historical framework it begins with and preserves the impression that a decade or two from now it’ll still be a deeply relevant read.
Devoted altcoiners are unlikely to ultimately be won over by the book’s arguments but will nonetheless find some value in the read and may perhaps gain a new appreciation (from a Bitcoiner’s eyes) of how Bitcoin’s value proposition differs from other projects in the space within this techno-historical framework. And with this historical emphasis it’ll be a great read for curious newcomers to the cryptocurrency scene and economics alike provided they have the background knowledge of a couple of YouTube Bitcoin explainer videos — The Bitcoin Standard doesn’t delve to any extremes into the protocol’s technicals, but it at least assumes you have the What’s a blockchain? type of questions behind you.
Readers of economics who’ve already absorbed and concluded against Austrian economic accounts will be the least likely of the reader base to be converted to Ammous’s side, but in the Bitcoin project at least there’s now an alternative, sound, non-state currency, and if it doesn’t fail for technological reasons we’ll all be able to observe over the next few years whether it’ll fail or succeed on economic grounds. And all sides should at least be enthusiastic about that.