The Architecture of Agency › Volume 6 › What Is Money?

What Is Money?

The common denominator — and the fallacy it kills

This chapter is a review — it is readable but still changing.

Alice has a bicycle and wants apples. Bob grows apples and wants a bicycle. Carol has bananas and wants apples too. On paper Alice and Bob look like a match, but try to close the deal: how many apples is a bicycle worth? Bob thinks fifty; Alice thinks four hundred; and even if they settle on two hundred, Alice does not want two hundred apples rotting in her kitchen — she wants a dozen now and more later, from someone who has no later use for a second bicycle. Carol, meanwhile, is locked out entirely unless Bob happens to want bananas. Every party holds real value and has a real desire, and almost none of the possible trades can happen.

Now introduce a currency. Alice sells her bicycle for $100. Bob sells apples at 50¢ each. Carol sells bananas at 25¢ each. Suddenly everything is commensurable:

Alice buys her dozen apples and keeps $94 of open-ended claim on everything else in the economy. Carol sells bananas to whoever wants them and buys apples with the proceeds, without Bob ever needing to want a single banana. The deadlock dissolves — not because anyone’s valuations changed, but because they can finally be compared.

This is the whole definition. Currency is the common denominator of market exchange — literally, not metaphorically. When you compare fractions you rewrite them over a shared denominator so the numerators become comparable; that is exactly the operation money performs on valuations. Every buyer and seller values goods according to their own needs, preferences, and circumstances, which makes all valuations inherently subjective — the position established in the discipline of value. Those subjective valuations are mutually incompatible: there is no fact of the matter about how many apples a bicycle is “really” worth. Currency supplies the scalar unit that lets each participant express their valuations numerically, and once everyone’s valuations are expressed over the same denominator, prices emerge as the ratios between them. A price is not a measurement of intrinsic worth; it is the exchange rate between one person’s subjective scale and everyone else’s — a point with consequences that the price illusion takes up.

Credit Before Barter

The textbook story says money evolved out of barter: first people swapped goods directly, then some convenient commodity emerged as the medium. Anthropological and historical work complicates that sequence. In many documented communities, debt, credit, and remembered obligations organized exchange without a prior spot-barter economy; barter also appears at boundaries, among strangers, and during monetary disruption. There need not be one origin path from barter to commodity to coin.

That complication matters because it reveals one function beneath very different implementations. Money can operate as a record — a way to quantify and communicate claims so that exchange can happen where direct swaps are impractical or impossible. In some histories bookkeeping came before coin; in others commodity, state, temple, tax, and trade institutions interacted. Money is often bookkeeping even when the record travels as a coin.

What the Denominator Requires

Whether the currency is dollars, cigarettes, gold, or Bitcoin, the same three properties do all the work:

Fungibility. Each unit must be identical and interchangeable with every other, so a price means the same thing no matter which particular units change hands. A denominator that varies from instance to instance is not a denominator.

Wide enough acceptance. The unit must be honored across the relevant network, so it can mediate exchanges between strangers who share little but the currency itself. Carol’s ability to route around Bob’s indifference to bananas depends on it.

Easy quantifiability. The unit must divide, add, and count cleanly, so ratios stay transparent and transactions stay simple. You cannot express a 1 : 200 ratio in a medium that comes in indivisible lumps.

From these three properties flow the two powers that make money useful: salability — the ease and speed with which the currency can be converted into a wide range of goods — and comparability — the ability to weigh many priced opportunities on a common scale. Notice what is absent from this list: intrinsic value. Nothing in the definition requires the medium itself to be good for anything apart from the functions its users value. Gold’s physical properties and Bitcoin’s cryptography can support scarcity, transfer, fungibility, or acceptance; none supplies agent-independent worth. Currency is a shared translator: it maps some agent-relative preferences into a common numerical unit without making every value commensurable or reconciling every conflict. Value remains where it always was — in the judgments of the agents doing the valuing.

Money, in short, is a coordination technology. In the layer stack this volume builds in the coat and the ticket — value, wealth, capital, money, currency, credit — this chapter is about the money and currency layers, and its one-sentence summary is that those layers are mathematical, not material. The ticket is not the coat. A cloakroom ticket coordinates the retrieval of coats; printing more tickets does not weave more coats.

That last sentence sounds obvious, yet schools of macroeconomics disagree about how nominal claims mobilize idle real resources and how a currency issuer is constrained.

The Dispute: Modern Monetary Theory

Modern Monetary Theory emphasizes that a monetary sovereign is not financially constrained like a household, that public spending creates deposits before taxes remove them, and that real resources and inflation are the binding limits. “Print your way to prosperity” is a hostile caricature, not the theory’s stated conclusion. The serious dispute concerns how reliably fiscal institutions can diagnose slack, inflation, distribution, currency demand, and political constraints—and whether the framework understates the information and governance failures involved in steering aggregate demand.

Where does monetary demand come from? MMT emphasizes the state’s ability to impose tax liabilities in its unit of account. Taxes can create baseline demand, while legal tender, payment networks, wages, contracts, banking, habit, productive capacity, and expectations also sustain acceptance. The mistake would be to treat either the decree or the voluntary network as sufficient by itself. A currency is simultaneously a social network and, in fiat systems, an institution backed by public authority.

Can institutions observe the real constraint in time? “A sovereign issuer can always make a nominal payment in its own currency” is conditional on continued monetary institutions and says nothing about the real goods the payment commands. MMT explicitly names inflation and resources as limits. The practical objection is that capacity, expectations, distributional bottlenecks, exchange rates, and policy lags are hard to observe and politically hard to correct before issuance consumes trust.

It treats taxation as a behavioral lever. In MMT, taxes do not fund spending; they exist to create demand for the currency and to regulate the behavior of the population — pull money out here, push it in there, and steer the society toward engineered outcomes. Set aside the presumption of governmental benevolence and competence this requires. The deeper problem is what it does to the thing money is for. Currency exists to let subjective valuations find free expression and voluntary reconciliation; a tax deployed as a steering mechanism imposes valuations by force instead — it is coercion wearing the uniform of bookkeeping. A medium built to facilitate voluntary coordination gets repurposed as an instrument for overriding it.

The knowledge problem remains. Prices aggregate some dispersed information, while fiscal and monetary authorities act through coarse data, forecasts, mandates, and political institutions. Decentralized markets also contain sticky prices, credit cycles, externalities, and unequal bargaining. The comparative question is which intervention responds to which failure with what lag and error—not whether one side possesses complete information.

The Ledger

Hyperinflations demonstrate that nominal issuance cannot substitute for productive capacity, fiscal credibility, political stability, and currency demand. Weimar Germany, Zimbabwe, Venezuela, and Argentina are not controlled tests of one theory; their causes include war or political crisis, collapsing tax bases and production, foreign-denominated obligations, exchange-rate pressure, and institutional breakdown alongside monetary finance. They still warn that a state able to issue nominal claims cannot guarantee the real resources or trust those claims require.

Because MMT itself names inflation and real capacity as constraints, a hyperinflation is not by itself a refutation. The open question is whether its policy program supplies institutions capable of respecting those constraints before expectations and politics outrun correction. That is a demanding empirical and governance burden, not a category mistake solved by definition.

Money is a shared claim and accounting system that makes unlike exchanges comparable. It can mobilize idle capacity; it cannot create unlimited real capacity by changing nominal entries. The coat-and-ticket distinction therefore constrains every monetary theory, including MMT, without settling the macroeconomic dispute by definition.