There is a gold bar sitting in the Fort Knox bullion depository in Kentucky. It weighs about 400 troy ounces. If you melted it and sold the metal for industrial use, in electronics, jewelry, dental work, it would trade at market price. That price is partly driven by industrial demand and partly by something else: the persistent, centuries-old belief that gold is a store of value. Remove the belief, and you’re left with a heavy yellow metal whose industrial applications are real but modest.
There is no gold in the US Federal Reserve backing your bank account. The $100 bill in your wallet is not exchangeable for anything other than other money, at conversion rates set by markets, which are also sets of beliefs. Modern economies left the gold standard behind not as a failure but as a feature. The gold standard was a way of anchoring monetary value to a physical substance. Fiat currency is a way of anchoring monetary value to institutional credibility and collective agreement. It turned out the latter was more scalable.
Money is a story. The gold bar is a story. The dollar is a story. The question worth spending time on is: what kind of stories make good money, and why do some of these stories collapse while others persist for centuries?
The Problem of Coordination at Scale
To understand what money is, you need to start with the problem it solves.
Human beings are capable of direct reciprocal cooperation in small groups without any monetary system. In bands of 150 or fewer people, direct observation and reputation are sufficient to coordinate economic exchange. You help me now because you know me, you know I’ll need help in the future, and everyone around us knows both things. The informal credit system of a village is robust because everyone is watching.
The problem is that this system doesn’t scale. Once your community exceeds the size where direct personal knowledge is possible, you can’t cooperate through reputation alone. You need a different mechanism.
Every monetary innovation in human history is an attempt to solve this problem. Cowrie shells, which have no functional utility beyond their monetary use, worked as currency across parts of Africa, South Asia, and East Asia for thousands of years because they were difficult to counterfeit, easy to count, and, crucially, widely accepted within the relevant communities. The acceptance was the function. Nothing else mattered.
Minted coins added two innovations: they tied the monetary medium to state authority (increasing the credibility of the system through the threat of enforcement) and they standardized the unit of account (reducing the cost of every transaction that used them). The coin isn’t valuable because it’s made of silver. It’s valuable because the ruler who minted it says it is, and has sufficient military and administrative capacity to enforce that declaration.
Paper money removed the physical tether entirely. The note was always a claim, initially a claim on a specific quantity of metal, then a claim on the credit of the issuing government. Modern fiat currency is a claim on nothing physical at all. It is a bare coordination mechanism: a token that everyone agrees to accept in exchange for goods and labor, because everyone else agrees to accept it.
This is circularity in the precise sense. The reason to accept money is that others will accept it. The reason others accept it is that you will. The system works because the belief in it is self-fulfilling, maintained by the network effects of universal acceptance and the institutional infrastructure that enforces that acceptance.
What Makes a Monetary Story Durable
If money is fundamentally a story, the natural question is: what makes some monetary stories more durable than others?
The answer has several components, none of which involves the physical properties of the monetary medium.
Institutional track record. The most durable monetary systems are embedded in institutions with long, consistent histories of maintaining purchasing power and honoring obligations. The dollar’s reserve currency status isn’t a natural property. It’s an artifact of the US Federal Reserve’s decades of credible inflation management, the depth and liquidity of US financial markets, and the legal infrastructure that makes dollar-denominated contracts enforceable globally. These properties took a century to accumulate and would take a long time to replicate.
Network effects. The more widely a monetary unit is accepted, the more useful it is to any individual user, and the more motivated each individual is to accept it. This creates strong incumbency advantages for established currencies. Abandoning the dollar as a reserve currency would require every country that holds dollar reserves to simultaneously find a substitute, coordinate on what that substitute would be, and absorb the transition costs, while the dollar-based system continues to function during the transition. The network effect is a form of coordination value that compounds over time.
Narrative coherence. Even purely fiat currencies need a story that explains why they should be trusted. For the dollar, that story involves American economic and military dominance, the petrodollar system that required oil transactions to be denominated in dollars, and decades of inflation management. When that story becomes incoherent, as when inflation accelerates significantly, the currency’s credibility erodes even if nothing physical has changed.
Scarcity design. One of the innovations of cryptocurrency was to encode scarcity directly into the protocol. Bitcoin’s supply is capped at 21 million coins, and the schedule of new coin issuance is mathematically predetermined. This is an attempt to substitute algorithmic credibility for institutional credibility, to make the monetary story unfalsifiable by making the rules of the system immutable. Whether this design is superior to institutional credibility is an open question. It has different failure modes (code bugs, governance failures within the developer community, loss of key holders) rather than fewer failure modes.
Hayek and Keynes: Two Theories of Why Money Matters
The two most influential 20th-century theories of money offer usefully different framings.
Friedrich Hayek’s view was that the price system is primarily an information technology. Prices communicate dispersed knowledge about relative scarcities and preferences in a way that no central planner could replicate. A change in the price of copper tells the copper market something specific about supply and demand conditions, concentrating information from thousands of actors into a single number that allows everyone in the market to adjust behavior accordingly.
In Hayek’s framework, money’s value is its function as a unit of account and medium of exchange in this information system. The story that money tells is the story of relative scarcities, and that story is valuable because it allows coordination without centralized information gathering.
Keynes had a different emphasis. He was more interested in the psychological dynamics of monetary systems: how expectations, animal spirits, and liquidity preference shape economic behavior. In Keynes’s framework, money demand is itself partly a psychological phenomenon: people hold money not just for transactions but as a hedge against uncertainty, and this demand fluctuates with confidence about the future.
The “Keynesian beauty contest” is his most memorable illustration of the psychology underlying financial markets. He compared stock investing to a newspaper contest where you had to pick which face other readers would find most beautiful: the winning strategy is not to pick the face you find most beautiful, but to predict what the majority will predict that the majority will prefer. Asset prices, in this view, are set not by fundamental value but by collective psychology about collective psychology. The story about the story.
Both framings are useful, and they’re not as contradictory as they’re often presented. Hayek captures how monetary systems transmit information when they work. Keynes captures how they can fail when collective psychology overwhelms the information-transmission function.
A useful synthesis: monetary systems are information technologies that work as long as the stories embedded in them remain coherent. When the stories fracture, as in a hyperinflation, the information technology fails, and coordination collapses. The physical economy continues to exist, but the coordination layer that made complex exchange possible at scale no longer functions.
The Comparative Table of Monetary Objects
Consider what the following things have in common: gold, cowrie shells, fiat currency, and Bitcoin.
None of them have significant functional utility as consumption goods. You can’t eat any of them, wear any of them, or use them directly for physical survival. Their value in each case derives entirely from collective agreement about their monetary function.
Gold’s monetary value exceeds its industrial value by an amount that depends on belief in its store-of-value properties, a belief that has persisted for thousands of years across cultures. Remove that belief, and the market price of gold would fall substantially.
Cowrie shells had no industrial utility whatsoever. Their monetary value was entirely a function of the social agreements that recognized them as currency within specific cultural contexts. When those contexts changed (contact with European traders who had more abundant supplies of cowries, or transition to different monetary systems), their value collapsed.
Fiat currency’s entire value is social agreement, maintained by institutional infrastructure and network effects. A Venezuelan bolivar held most of its value for decades, then lost it catastrophically when the institutional story behind it became incoherent.
Bitcoin’s value is a social agreement with algorithmic properties designed to make it resistant to certain kinds of institutional failure. Its value has fluctuated enormously, reflecting the fact that the social agreement around it is still being formed: who accepts it, what it’s used for, what institutional frameworks recognize it. The story is still being written.
The comparison is instructive not because any of these things is better money than the others in all circumstances, but because it makes visible what money actually is: a socially agreed coordination mechanism whose effectiveness is a function of the quality, durability, and coherence of the story it embodies.
What Inflation Is, At a Deeper Level
Economists define inflation as a general rise in prices. That’s accurate as far as it goes. But the deeper description is: inflation is what happens when the story behind the money becomes less credible.
When people believe the purchasing power of money will decline, they try to spend it sooner and hold it less, which causes prices to rise, which confirms the expectation, which accelerates the process. The inflationary dynamic is, at its core, a narrative crisis. The story that money stores value reliably becomes untrue, and everyone acting on that truth makes it more untrue.
This is why hyperinflations are so fast relative to the underlying economic conditions. Germany in 1923, Zimbabwe in the 2000s, Lebanon in the 2020s: in each case, the underlying economic dysfunction had been developing for years before the hyperinflationary episode. The crisis happened when the monetary story broke. Once the break was sufficiently visible, the self-fulfilling dynamic took over.
Central bank credibility is, in this frame, exactly what it sounds like: the credibility of the story that the monetary authority will maintain purchasing power. When central banks raise interest rates to fight inflation, the primary mechanism is partly psychological: demonstrating willingness to accept economic pain to maintain the monetary story, which reinforces belief in the story, which reduces the expectation of future inflation, which itself reduces inflation.
This is monetary policy as narrative management. Which sounds dismissive. But if the economy runs on a collectively maintained story, then narrative management is exactly what monetary policy is and should be.
When the Story Can’t Be Told Anymore
There are failure modes for monetary stories that go beyond inflation.
The more interesting ones involve the displacement of one monetary story by another. The dollar displaced the British pound as the primary global reserve currency over the course of the 20th century, not through any single crisis but through a gradual shift in which story was more credible, backed by which economic and military power.
Crypto advocates argue that Bitcoin represents a different kind of displacement: not one government-backed story replacing another, but a mathematically constrained story replacing institutional ones. Whether that argument succeeds depends on whether the Bitcoin story can achieve the network effects, institutional recognition, and narrative coherence that make a monetary system durable at scale. That’s still genuinely uncertain.
What seems more likely in the near term is a proliferation of competing monetary stories, each with different design properties and different kinds of credibility. Central bank digital currencies, stablecoins backed by various assets, Bitcoin, and traditional fiat: each represents a different answer to the question of what the monetary story should be grounded in, and each will attract different users based on which answer they find most compelling.
This is uncomfortable for monetary theorists who prefer elegant unified accounts. But it’s probably a more accurate picture of what money actually is: a contested, evolving set of social agreements, rather than a fixed property of specific physical objects.
What This Means for Everything Else
If money is a story, then institutions more broadly are stories. The corporation that owns nothing but shares in other corporations is a legal fiction that generates real effects because enough parties agree to treat it as real. The contract is a story about what happens if certain conditions are met, backed by a legal system that is itself a story about the legitimate use of force.
The implication is that a significant portion of “the economy” is not, in the first instance, about physical goods and services. It’s about maintaining, revising, and competing over the stories through which economic coordination happens. Who owns what, under what conditions, with what rights and obligations: these are questions of social agreement, not physical fact.
Property rights don’t inhere in physical objects. They exist in the collectively maintained story that certain parties have certain enforceable claims over certain objects. When that story is recognized by sufficient institutional authority, property rights function as effectively as if they were physical properties. When the authority collapses, the rights collapse with it, regardless of what’s physically true.
This isn’t nihilism. The stories have real consequences. A strong property rights system generates different economic outcomes than a weak one. Credible monetary systems generate different behavior than incredible ones. The stories matter enormously, which is why people fight about them.
But understanding them as stories, rather than as natural facts, changes how you think about what it means to maintain or change them. The dollar doesn’t have reserve currency status because of an objective property of the dollar. It has it because of a complex of historical relationships, institutional arrangements, and network effects that would need to be systematically unwound to change. That’s a political and institutional challenge, not a physical one.
A Final Thought on What “Real” Money Is
The question “what is money really worth?” is, when you push it, meaningless in isolation. Money is worth what it can be exchanged for, which is determined by what people believe it can be exchanged for, which is determined by institutional track record and network effects, which is determined by history and collective agreement.
There is no money that exists independent of social agreement. Even gold’s monetary value, which has persisted for 5,000 years, is a collective agreement that has lasted an unusually long time. The persistence is remarkable and probably tells you something about gold’s properties (scarce, durable, divisible, portable, beautiful) as inputs to a monetary story. But the value is still the agreement, not the metal.
What this suggests is that the question worth asking about any monetary system is not “what does it really represent?” but “how robust is the story it tells, and under what conditions does that story break down?”
Applied to dollars: the story is robust because of deep institutional infrastructure, global network effects, and a track record of reasonable inflation management. It could break down under sustained institutional dysfunction, a credible competing story achieving sufficient network effects, or a crisis of confidence that becomes self-fulfilling.
Applied to crypto: the story is still being formed. The algorithmic scarcity is a feature with real design value. The institutional recognition and network effects are still limited relative to traditional monetary systems. The story has enough adherents to maintain a market, but not yet enough to provide the coordination function that makes money genuinely useful at scale for ordinary transactions.
Applied to historical monetary systems: every one of them, without exception, has eventually been superseded. The cowries gave way to coins. The coins gave way to paper. The paper gave way to fiat. Fiat will eventually give way to something.
What persists across all of them is the underlying function: a coordination mechanism that allows strangers to exchange without direct trust, complex supply chains to function without central planning, and economic cooperation to scale beyond the limits of personal relationships.
The medium changes. The function is ancient. And it’s entirely social, all the way down.