Skip to content

Friedman Rule

Set the nominal interest rate to zero — via a steady deflation at the real rate — so that the private opportunity cost of holding money equals its near-zero social cost of production, eliminating the shoe-leather distortion; a benchmark that isolates one welfare cost and prices money at marginal cost.

Core Idea

The Friedman rule is the optimal monetary-policy prescription derived by Milton Friedman in "The Optimum Quantity of Money" (1969): a central bank should set the nominal interest rate to zero — operationally achieved by engineering a rate of deflation equal to the real interest rate — so that the private opportunity cost of holding money equals its social marginal cost of production, which for fiat currency is effectively zero. The welfare case is precise: when the nominal interest rate is positive, holding money rather than an interest-bearing asset carries an opportunity cost equal to the nominal rate; agents respond by under-holding money relative to the social optimum, expending real resources to economize on cash balances — the "shoe-leather cost" of making frequent trips to liquidate assets, holding near-money substitutes, and restructuring portfolios to minimize idle balances. Setting the nominal rate to zero via the Fisher equation (nominal rate = real rate + inflation) requires a steady deflation at the real rate, after which agents are indifferent at the margin between money and bonds and the shoe-leather distortion is eliminated.

The rule occupies a fixed benchmark position in monetary economics: it characterizes what optimal inflation policy would look like if the shoe-leather welfare loss were the only welfare cost in the model. The benchmark is then modified by the additional welfare costs that push the socially optimal inflation rate above the Friedman deflation prescription — sticky-price adjustment costs favoring stable rather than falling prices, fiscal interactions via seigniorage revenue, financial-stability concerns, and the practical difficulties of operating near the zero lower bound. The resulting literature frames the choice of inflation target as a weighted trade-off between the Friedman-rule deflation benchmark and these countervailing considerations, with new-Keynesian welfare analyses generally finding that zero rather than negative inflation dominates when nominal rigidities are significant.

Structural Signature

Sig role-phrases:

  • the central bank — the policy authority that chooses the inflation rate, equivalently the nominal interest rate
  • the cash-holding agents — agents who decide how much money to hold given its opportunity cost relative to interest-bearing assets
  • the zero production cost of fiat money — the social marginal cost of producing money, treated as effectively zero, one side of the wedge
  • the private opportunity cost — the nominal interest rate forgone by holding money rather than bonds, the other side of the wedge
  • the Fisher equation — the identity (nominal rate = real rate + inflation) that converts the zero-nominal-rate target into a deflation prescription
  • the shoe-leather distortion — the welfare loss from agents expending real resources to economize on cash balances when the nominal rate is positive
  • the optimal prescription — set the nominal rate to zero via steady deflation at the real interest rate, eliminating the shoe-leather distortion (marginal-cost pricing applied to money)
  • the deliberate silence — the rule answers only what shoe-leather alone demands; the omitted welfare costs (sticky prices, seigniorage, financial stability, the zero lower bound) are named, signed deviations that push the optimum above the deflation benchmark

What It Is Not

  • Not a real-world policy recommendation to engineer deflation. The rule is a benchmark that isolates one welfare cost — shoe-leather — and derives what that cost alone demands; it is deliberately silent on sticky prices, seigniorage, financial stability, and the zero lower bound. Whether mild deflation should actually be targeted in a given economy is settled by those added deviations, not by the rule, which new-Keynesian analysis typically finds dominated by zero-inflation prescriptions.
  • Not a claim that deflation is benign. The prescription is a specific, mild deflation at exactly the real interest rate, motivated solely by setting money's opportunity cost to its production cost — not an endorsement of falling prices in general, and emphatically not the self-reinforcing debt-deflation spiral, which is a different phenomenon the rule's frictionless welfare model abstracts away.
  • Not a Taylor rule or other feedback rule. The Taylor rule is a state-contingent interest-rate reaction function; the Friedman rule is a fixed normative prescription about the optimal level of inflation. They answer different questions — how to respond to conditions versus what target is optimal — and share only the word "rule."
  • Not a zero-inflation target. The prescription is a zero nominal interest rate, which via the Fisher equation requires deflation at the real rate, not stable prices. Inflation targeting at 2% and the Friedman rule sit at opposite poles of the optimal-inflation debate; conflating "zero rate" with "zero inflation" inverts the result.
  • Not a complete theory of optimal inflation. By construction it answers only "what would optimal policy be if shoe-leather were the sole welfare cost?" and nothing else. Treating it as the full answer ignores the named countervailing costs that push the actual optimum above the deflation benchmark.
  • Not a novel cross-domain mechanism. Strip the monetary apparatus — Fisher equation, fiat money, cash-holding agents — and what remains is marginal_cost_pricing applied to one good (money): set the private price of a costless-to-produce good to its zero social cost. That general result, not "the Friedman rule," is what travels; the rule is its monetary specialisation and adds no structural pattern of its own.

Scope of Application

The Friedman rule lives within monetary economics — the subfields that share its apparatus (a central bank with monetary control, fiat money, the Fisher equation, cash-holding agents, a welfare metric over transaction frictions); its reach is bounded by that monetary substrate. There is no off-substrate habitat to enumerate — the apparatus has no referent outside a monetary economy, and the portable optimality logic belongs to marginal_cost_pricing, not to the rule.

  • Monetary economics and central-banking research — the benchmark optimum in welfare analyses of monetary policy, the workhorse comparison point even where judged practically inadvisable.
  • Optimal-taxation theory — a special case of the Ramsey / Diamond–Mirrlees result: the optimal inflation tax is negative (deflation) under the relevant assumptions.
  • New-Keynesian welfare analysis — the benchmark is explicitly retained but typically dominated by zero-inflation prescriptions once sticky-price deviations are added.
  • Zero-lower-bound research and inflation-target debates — invoked as one pole of the optimal-inflation question (Japan's deflation, post-2008 ZLB debates).
  • Macroeconomic modelling pedagogy — the standard optimal-inflation exercise in cash-in-advance and money-in-the-utility-function models.

Clarity

The Friedman rule's clarifying force is that it converts the question "what is the optimal rate of inflation?" — which otherwise invites an unstructured weighing of many considerations at once — into a benchmark-plus-deviations problem. It isolates one welfare cost, the shoe-leather distortion from a positive opportunity cost of holding money, and derives exactly what policy that single cost demands in isolation: a zero nominal rate, hence mild deflation. By fixing that pole, the rule makes the other costs visible as deviations from it. The analyst can now name and price each force that pushes the optimum away from deflation — sticky-price adjustment costs, seigniorage and fiscal interaction, financial-stability concerns, the operational hazard of the zero lower bound — rather than reasoning about inflation policy as an undifferentiated whole. The sharper question the rule enables is not "what inflation rate is best?" but "which welfare costs, beyond shoe-leather, are large enough in this economy to justify moving the target above the deflation benchmark, and by how much?"

It also sharpens a distinction monetary debate can blur: the social marginal cost of producing fiat money is essentially zero, while the private opportunity cost of holding it is the nominal interest rate, and the wedge between the two is the source of the distortion. Recognizing this locates the Friedman rule as the application to money of a familiar optimality logic — set the private price of a costless-to-produce good to its zero social cost — which is why the result feels less like a monetary curiosity than a special case of pricing at marginal cost. And by being explicitly silent on everything except shoe-leather, the rule clarifies its own jurisdiction: it tells the practitioner precisely what it does and does not answer, so that disagreement with its prescription can be located in a specific countervailing cost rather than in the rule itself.

Manages Complexity

The optimal-inflation question is, taken whole, an unstructured weighing of heterogeneous and incommensurable forces: shoe-leather distortions from the opportunity cost of holding cash, the menu and adjustment costs of moving sticky prices, the fiscal value of seigniorage revenue, financial-stability considerations, the operational hazards of the zero lower bound — each pulling the optimum in a different direction, with no common scale on which to add them. The Friedman rule tames that sprawl not by resolving every force but by fixing a single sharp pole and recasting the whole problem as benchmark-plus-deviations. It isolates exactly one welfare cost — the shoe-leather distortion arising when a positive nominal rate makes money costly to hold relative to bonds — and derives precisely what that one cost demands in isolation: a zero nominal rate, achieved via the Fisher equation by a steady deflation at the real interest rate. With that pole nailed down, the analyst no longer confronts a many-bodied optimization but a base case to which each remaining force is added as a named, signed deviation: sticky-price costs push the optimum up toward stable or mildly positive inflation, seigniorage needs push it up, financial-stability and zero-lower-bound concerns push it up, and the magnitude of the move is set by how large each of those is in the particular economy. The high-dimensional question "what inflation rate is best, all things considered?" collapses to a short, tractable checklist — start at the Friedman deflation benchmark, then ask which countervailing costs are large enough here to justify shifting the target above it, and by how much — so the qualitative answer reads off a handful of force magnitudes rather than a holistic intuition. The compression has a second payoff in jurisdiction: by being explicitly silent on everything but shoe-leather, the rule fixes exactly what it does and does not settle, so any disagreement with mild deflation must be located in a specific, nameable countervailing cost rather than in a vague rejection of the prescription — which is why the new-Keynesian finding that zero rather than negative inflation dominates under significant nominal rigidities can be stated cleanly as "the sticky-price deviation outweighs the shoe-leather benchmark here," a single comparison rather than a re-litigation of the whole question.

Abstract Reasoning

The rule's signature move is benchmark-and-deviation reasoning, which structures how a monetary economist attacks the optimal-inflation question. Rather than weigh all welfare costs at once, the analyst first holds every consideration but shoe-leather fixed at zero and derives what that single cost demands in isolation — a zero nominal rate, hence steady deflation at the real interest rate via the Fisher equation. That derived pole then becomes the reference point against which every other force is read as a signed deviation. The reasoning runs FROM "this economy has significant nominal price rigidities" TO "the sticky-price cost pushes the optimum up, away from deflation toward stable or mildly positive inflation"; FROM "this fiscal authority needs seigniorage" TO "the revenue motive pushes the target up"; FROM "the zero lower bound is an operational hazard here" TO "a positive inflation buffer is warranted." The characteristic prediction is comparative and directional: name a countervailing cost, sign its push relative to the deflation benchmark, and the optimum moves that way by an amount set by the cost's magnitude.

A diagnostic-attribution move is the converse. Confronted with a recommendation that departs from mild deflation — say, a new-Keynesian result favoring zero rather than negative inflation — the analyst does not treat it as a rejection of the rule but localizes the disagreement to a specific term: the result holds because the sticky-price deviation outweighs the shoe-leather benchmark in an economy with significant nominal rigidities. Any quarrel with the prescription must therefore be stated as "which countervailing cost is large enough here, and by how much?" rather than as a vague preference for higher or lower inflation. This converts an open-ended dispute into a single, locatable comparison between two named magnitudes.

The interventionist move is the rule's operational core. To set the private opportunity cost of holding money to its essentially-zero social cost of production, drive the nominal interest rate to zero; because the nominal rate is real rate plus inflation, the only way to hold it at zero when the real rate is positive is to engineer deflation at exactly the real rate. The predicted effect is precise: at a zero nominal rate, agents are indifferent at the margin between money and bonds, they stop expending real resources to economize on cash balances, and the shoe-leather distortion is eliminated. The instrument (the inflation rate the central bank targets) and its mechanism (the wedge between the cost of producing money and the cost of holding it) are both fixed, so the move predicts not just a direction but the specific deflation rate required.

The rule's boundary-drawing is unusually explicit, and is itself a reasoning aid. By being deliberately silent on everything except shoe-leather, it fixes its own jurisdiction: it answers what optimal policy would be if the cash-holding distortion were the sole welfare cost, and it answers nothing else. The inference this licenses is one of scope — whether the deflation prescription should actually be followed in a given economy is not settled by the rule but by the added deviations, so the rule tells the practitioner precisely the question it leaves open (which other costs dominate here) and the question it closes (what shoe-leather alone demands). Recognizing that the result is the application to money of marginal-cost pricing — set the price of a costless-to-produce good to its zero social cost — further bounds it: where that pricing logic is offset by frictions the simple model omits, the benchmark is a starting point to be adjusted, not a standalone policy.

Knowledge Transfer

Within monetary economics the Friedman rule transfers as mechanism across the subfields that share its apparatus — a central bank with monetary control, fiat money, the Fisher equation, cash-holding agents, and a welfare metric over transaction frictions. The benchmark-and-deviation reasoning, the marginal-cost-pricing-of-money derivation, the deflation-at-the-real-rate instrument, and the explicit jurisdiction (shoe-leather only) all carry intact as the concept moves across its home subfields: from Friedman's original cash-in-advance argument into optimal-taxation theory (where it is a special case of the Ramsey / Diamond–Mirrlees result — the optimal inflation tax is negative under the relevant assumptions), into new-Keynesian welfare analysis (where the benchmark is explicitly retained but typically dominated by zero-inflation prescriptions once sticky-price deviations are added), into central-banking research on the zero lower bound, and into the standard macro pedagogy of money-in-the-utility-function and cash-in-advance models. Across these the rule is not re-applied by analogy; it is the same prescription with the same welfare derivation, with each subfield adding its own named deviation to the same deflation pole. The transfer is gated on the monetary substrate being present, and within it the benchmark logic travels exactly — which is precisely why the rule functions as the field's workhorse comparison point even where it is judged practically inadvisable.

Beyond monetary economics the honest report is that the named rule does not travel at all, while the optimality logic it instantiates does — as the parent, not as the Friedman rule (case B). The rule presupposes a monetary economy with a central bank, fiat money, a Fisher equation, and cash-holding agents; none of these preconditions exists outside that substrate, so there is no Friedman rule for ant colonies, ecosystem nutrient flows, or distributed-systems coordination protocols, and there is no honest metaphorical extension to mark — the apparatus simply has no referent off-substrate. What is portable is the thinner structural insight the rule embodies: set the private price (opportunity cost) of a good that is costless to produce equal to its zero social marginal cost, by adjusting the relevant price. That is recognizable not as a monetary curiosity but as the application to money of marginal-cost pricing, the general optimality result familiar from microeconomics and public finance — and it is that general result, already carried at prime level by marginal_cost_pricing (with pigovian_tax and externality as the surrounding wedge-correction machinery), that genuinely recurs across domains: price any costless-to-provide good at marginal cost, drive any distortionary wedge between private and social cost to zero. The home-bound cargo the Friedman rule leaves behind is everything that makes it the Friedman rule: the nominal-versus-real-rate distinction and the Fisher equation that converts the zero-nominal-rate target into a deflation prescription, the shoe-leather welfare cost, the seigniorage and sticky-price deviations, and the whole inflation-target debate. So the correct cross-domain lesson carries marginal_cost_pricing (the private price of a costless good should equal its zero social cost; any wedge is a distortion to be removed), not "the Friedman rule," which is the monetary-economics specialization of that result and adds no structural pattern of its own to the catalogue. Strip the monetary vocabulary and what remains is exactly marginal-cost pricing applied to a particular good (money), which is why the Friedman rule is a domain-specific abstraction whose portable core belongs to its parent rather than a prime (see Structural Core vs. Domain Accent).

Examples

Canonical

Milton Friedman derived the rule in "The Optimum Quantity of Money" (1969). Suppose the real interest rate is 3%. By the Fisher equation, the nominal interest rate equals the real rate plus inflation, so to drive the nominal rate to zero the central bank must engineer inflation of −3% — a steady deflation of 3% per year. Why aim for zero? Producing an extra unit of fiat money costs society essentially nothing, yet whenever the nominal rate is positive an agent who holds money forgoes that rate and so burns real resources ("shoe-leather") economizing on cash. At a zero nominal rate, money and a bond earn the same real return — the bond's 0% nominal plus 3% deflation, and cash's 3% gain in purchasing power, both come to +3% — so agents are indifferent at the margin and the distortion vanishes.

Mapped back: The policy authority is the central bank; savers choosing money-versus-bonds are the cash-holding agents. That an extra unit of money is nearly free to make is the zero production cost of fiat money, and the forgone nominal rate is the private opportunity cost. Using nominal = real + inflation to convert a zero-rate target into 3% deflation is the Fisher equation, and eliminating the wasted-resource wedge is removing the shoe-leather distortion — the optimal prescription.

Applied / In Practice

The rule functions in practice as the benchmark in central-bank optimal-inflation research. Schmitt-Grohé and Uribe, computing the optimal long-run inflation rate in calibrated new-Keynesian models, take the Friedman prescription of mild deflation as the frictionless starting point and then add the welfare cost of adjusting sticky prices, of seigniorage, and of the zero lower bound. They find that once sticky-price costs enter, the optimum sits close to zero — with other frictions nudging it slightly positive — well above the Friedman deflation. This is a large part of why real central banks target low positive inflation (around 2%) rather than the deflation the bare rule implies: the analysis begins at the Friedman pole and reads each real-world cost as a named, signed deviation from it.

Mapped back: Starting from mild deflation and adding sticky-price, seigniorage, and ZLB costs is exactly the benchmark-and-deviation use of the optimal prescription. The finding that sticky prices push the optimum up to near zero is the deliberate silence made operational — the rule settling only what shoe-leather demands, leaving the omitted costs to move the target above the deflation benchmark.

Structural Tensions

T1: Benchmark versus prescription (the isolation that gives analytical power disqualifies it as policy). The Friedman rule's value is as a fixed pole from which every other welfare cost is read as a signed deviation, and that value comes entirely from its willingness to isolate one cost — shoe-leather — and answer only what that cost demands. But the same isolation that makes it a clean reference makes it a poor recommendation: the omitted costs (sticky prices, seigniorage, the zero lower bound) typically dominate, so new-Keynesian analysis finds the actual optimum well above the deflation the bare rule implies. The tension is that the rule is most useful precisely where it is least followed — indispensable as the comparison point in optimal-inflation research, yet targeted by no central bank. Treating its analytical cleanliness as policy advice inverts its purpose; treating its practical inadvisability as a defect misses that a benchmark is supposed to abstract. Diagnostic: Is the rule being used as the reference pole against which real costs are signed (its proper role), or as a standalone deflation recommendation (a misread of a deliberately partial benchmark)?

T2: Welfare-optimal zero nominal rate versus the hazards of the deflation instrument. The rule's welfare target is a zero nominal rate, but the only way to hold the nominal rate at zero while the real rate is positive is, via the Fisher equation, to engineer steady deflation at exactly the real rate. That instrument is where the frictionless welfare model and the rest of macroeconomics collide: the model abstracts away precisely the phenomena that make deflation dangerous — the zero-lower-bound trap that disables conventional stabilization, and the self-reinforcing debt-deflation spiral. The tension is that removing the shoe-leather distortion requires deploying deflation, and deflation carries its own distortions the derivation cannot see, so the cure for one welfare cost is the trigger for others outside the model's frame. The prescription is optimal in a world with no nominal rigidities and no debt overhang, and hazardous in the world that has them. Diagnostic: Does the setting actually lack the frictions the derivation abstracts (deflation is safe here), or are ZLB and debt-deflation hazards live (the instrument reintroduces distortions the welfare target ignored)?

T3: Zero private-social wedge versus the revenue role of the inflation tax. The rule drives the wedge between money's zero production cost and its positive holding cost to zero — marginal-cost pricing applied to money. But the nominal rate that constitutes that wedge is also the inflation tax, a source of seigniorage revenue, and optimal-taxation theory (Ramsey/Diamond–Mirrlees) delivers the Friedman result of a zero inflation tax only under specific assumptions; where other taxes are distortionary and revenue is scarce, a positive inflation tax can be part of the optimum. The tension is that setting the wedge to zero for allocative efficiency forfeits a fiscal instrument, and the two objectives — eliminate the shoe-leather distortion versus raise revenue at least total distortion — pull in opposite directions. The rule prices money efficiently precisely by declining to use money as a tax base. Diagnostic: Are the fiscal preconditions for a zero inflation tax met (lump-sum taxes available, no binding revenue need), or does the economy's revenue requirement make a positive inflation-tax wedge part of the true optimum?

T4: Fixed normative pole versus state-contingent implementation. The rule is presented as a fixed prescription about the optimal level of policy — a zero nominal rate, hence mild deflation — and the entry sharply distinguishes it from the Taylor rule, a state-contingent reaction function. But the implementation is not as fixed as the statement: because the required deflation equals the real interest rate, and the real rate varies over time and is not directly observable, holding the nominal rate at zero demands tracking a moving, estimated quantity. The tension is that the clean normative pole ("zero nominal rate") conceals an operationally state-contingent target (deflate at whatever the current real rate is), which blurs, in practice, the very level-versus-reaction-function line that separates it from feedback rules. The benchmark is fixed in the space of nominal rates but a moving target in the space of inflation rates the bank must actually engineer. Diagnostic: Is "the Friedman deflation" being treated as a single fixed number, or as the real-rate-indexed, time-varying deflation the Fisher-equation mechanism actually requires?

T5: Autonomy versus reduction (a monetary specialization or an instance of marginal-cost pricing). "Friedman rule" is a canonically studied monetary-policy result with its own machinery — the nominal-versus-real-rate distinction, the Fisher equation converting a zero-rate target into deflation, the shoe-leather welfare cost, the seigniorage and sticky-price deviations, the inflation-target debate — earning its own study, and within monetary economics that apparatus travels intact. But its portable core is not proprietary and, unusually, the named rule does not travel off-substrate at all: set the private price of a costless-to-produce good equal to its zero social marginal cost by adjusting the relevant price is marginal_cost_pricing (with pigovian_tax and externality as the wedge-correction relatives), and that general result is what recurs across microeconomics and public finance. There is no Friedman rule for an ant colony or a distributed protocol, because the apparatus has no referent there; strip the monetary vocabulary and only marginal-cost pricing of one good remains. It is also not a Taylor rule (a reaction function) nor a zero-inflation target (it prescribes deflation). Diagnostic: Resolve toward the parent (marginal_cost_pricing) whenever the good is not money; toward the named rule only when the Fisher equation, shoe-leather cost, and monetary apparatus are doing the work.

Structural–Framed Character

The Friedman rule sits at the framed-leaning position on the structural–framed spectrum: it is a clean optimality derivation with real mathematical content, but its every operative term is bound to a human monetary institution rather than to a natural system, which holds it well short of the structural end even though it convicts no one. On evaluative_weight it carries a mild framed tint: the rule is a normative prescription — it names what policy is optimal — so it is not the evaluatively-neutral description of a mechanism that "feedback" or "diffusion" gives, yet the "optimality" is a welfare verdict relative to a model, not a judgment rendered on any person or piece of reasoning, so the tint is faint. On human_practice_bound it is emphatically framed, and this is the decisive criterion: the concept is constituted by a monetary economy — a central bank with monetary control, fiat money, cash-holding agents, an inflation tax — and the entry is explicit that "the apparatus has no referent outside a monetary economy" and "there is no Friedman rule for ant colonies, ecosystem nutrient flows, or distributed-systems coordination protocols." Nothing here runs observer-free in nature; remove the human institution of money and there is no rule left. Institutional_origin is pronounced: the result is an artifact of a specific scholarly lineage (Friedman, "The Optimum Quantity of Money," 1969), derived inside monetary economics as a special case of the Ramsey / Diamond–Mirrlees optimal-taxation apparatus — a proposition of a theory, not a fact the world discloses. On vocab_travels it scores firmly framed: the nominal-versus-real-rate distinction, the Fisher equation, the shoe-leather welfare cost, seigniorage, the zero-lower-bound deviation are all pinned to the monetary substrate. On import_vs_recognize it patterns even more strongly toward the home domain than a typical framed entry: within monetary economics the same prescription with the same welfare derivation is recognized intact across subfields, but beyond it the named rule does not travel even by analogy — as the entry stresses, "the named rule does not travel at all, while the optimality logic it instantiates does — as the parent, not as the Friedman rule."

The one portable structural skeleton is marginal-cost pricing — set the private price (opportunity cost) of a good that is costless to produce equal to its zero social marginal cost, driving the wedge between private and social cost to zero. That skeleton travels as mechanism across microeconomics and public finance, tempting a structural reading, but it is what the Friedman rule instantiates from its umbrella prime (marginal_cost_pricing, with pigovian_tax and externality as the surrounding wedge-correction machinery), not what makes "the Friedman rule" travel: the cross-domain reach belongs to that parent result, while the monetary-accented specifics — the Fisher-equation conversion of a zero-rate target into deflation, the shoe-leather cost, the seigniorage and sticky-price deviations, the whole inflation-target debate — stay home. Its character: an evaluatively mild, monetary-institution-bound optimality result, structural only in the marginal-cost-pricing skeleton it borrows from its public-finance umbrella but framed by the central-bank vocabulary that pins it to the theory of a monetary economy.

Structural Core vs. Domain Accent

This section decides why the Friedman rule is a domain-specific abstraction and not a prime — an unusually clean case, because the named rule does not travel off-substrate at all, and its entire portable core is a single parent prime applied to one good.

What is skeletal (could lift toward a cross-domain prime). Strip the monetary apparatus and a thin optimality result survives: set the private price (opportunity cost) of a good that is costless to produce equal to its zero social marginal cost, driving the wedge between private and social cost to zero. The portable pieces are abstract — a good with negligible marginal production cost, a positive private price that opens a distortionary wedge, and the prescription to close that wedge by adjusting the relevant price. That result is genuinely substrate-portable, which is why it is already carried at prime level by marginal_cost_pricing (with pigovian_tax and externality as the surrounding wedge-correction machinery). It is the core the Friedman rule shares with every other marginal-cost-pricing application; it is not what makes the Friedman rule distinctive.

What is domain-bound. Everything that makes it the Friedman rule in particular is monetary-economics furniture and none of it survives extraction: the central bank choosing the inflation rate; the cash-holding agents; the zero production cost of fiat money set against the private opportunity cost of holding it; the Fisher equation that converts a zero-nominal-rate target into a deflation prescription; the shoe-leather welfare cost; and the named, signed deviations (sticky prices, seigniorage, financial stability, the zero lower bound) that push the optimum above the deflation benchmark. The decisive test is stronger than for most DS entries: the apparatus has no referent outside a monetary economy, so there is no Friedman rule for an ant colony, an ecosystem nutrient flow, or a distributed-systems protocol — not even a metaphorical extension to mark. Remove the human institution of money and there is no rule left to state. The result is constituted by the very monetary institution the prime bar asks it to shed.

Why this does not clear the prime bar. A prime's vocabulary travels and its transfer is recognition of the same mechanism, not analogy. The Friedman rule's transfer is bimodal, but with an unusually sharp off-substrate cliff. Within monetary economics it travels as full mechanism — the benchmark-and-deviation reasoning, the marginal-cost-pricing-of-money derivation, the deflation-at-the-real-rate instrument, and the explicit shoe-leather-only jurisdiction all carry intact across central-banking research, optimal-taxation theory (where it is the Ramsey/Diamond–Mirrlees result that the optimal inflation tax is negative), new-Keynesian welfare analysis, and macro pedagogy; it is the same prescription with the same welfare derivation, not an analogy, which is why it is the field's workhorse comparison point even where judged inadvisable. That is recognition. Beyond the monetary substrate the named rule does not travel at all — the apparatus is simply silent. And when the portable optimality lesson is wanted elsewhere, it is already carried, in fully general form, by marginal_cost_pricing: price any costless-to-provide good at marginal cost; drive any distortionary wedge between private and social cost to zero. The cross-domain reach belongs entirely to that parent (with pigovian_tax/externality); the named entry is its monetary specialization and adds no structural pattern of its own. It earns its own study in situ, but strip the monetary vocabulary and only marginal-cost pricing of one good remains — which is exactly what keeps it below the prime bar.

Relationships to Other Abstractions

Local relationship map for Friedman RuleParents appear above the current abstraction, mutual partners to the right, and children below. Node labels state whether each abstraction is prime or domain-specific; colors identify relation types.Friedman RuleDOMAINDomain-specific abstraction: Deflation — is part ofDeflationDOMAINDomain-specific abstraction: Interest Rate — is part ofInterest RateDOMAINPrime abstraction: Opportunity Cost — is part ofOpportunity CostPRIMEPrime abstraction: Frictionless Benchmark Reasoning — is a decomposition ofFrictionless Be…PRIME

Current abstraction Friedman Rule Domain-specific

Parents (4) — more general patterns this builds on

  • Friedman Rule is part of Deflation Domain-specific

    The live Friedman rule contains a steady deflation at the real rate as the price-level path that implements its zero nominal-rate target.

  • Friedman Rule is part of Interest Rate Domain-specific

    The Friedman rule contains the nominal interest rate as the private holding cost and policy target that the rule sets to zero.

  • Friedman Rule is part of Opportunity Cost Prime

    The rule contains the return forgone by holding money rather than the best available interest-bearing asset as its private marginal price.

  • Friedman Rule is a decomposition of Frictionless Benchmark Reasoning Prime

    Stripped of money, the rule is a sharp ideal optimum used as a coordinate origin for a separately named catalog of real-world deviations.

Hierarchy paths (13) — routes to 8 parentless roots

Not to Be Confused With

  • The quantity theory / k-percent money-growth rule. Friedman's other famous monetary prescription: hold money-supply growth at a constant rate, resting on the quantity theory (money growth drives inflation). This is a positive/operational rule about the money stock, distinct from the Friedman rule's welfare-optimal prescription about the nominal interest rate. Both bear Friedman's name and concern money, which is exactly the trap. Tell: is it a constant-money-growth prescription tied to the quantity theory (k-percent rule), or the optimal-deflation result setting the nominal rate to zero (the Friedman rule)?

  • The Taylor rule. A state-contingent interest-rate reaction function — how the central bank should adjust rates in response to inflation and output gaps. The Friedman rule is a fixed normative claim about the optimal level of inflation, not a response schedule; they share only the word "rule." Tell: is it a formula for reacting to current conditions (Taylor rule), or a statement of what long-run target is optimal (Friedman rule)?

  • Inflation targeting / a zero-inflation target. Targeting stable (or mildly positive, e.g. 2%) prices. The Friedman rule prescribes a zero nominal interest rate, which via the Fisher equation requires deflation at the real rate — the opposite pole of the optimal-inflation debate. Conflating "zero rate" with "zero inflation" inverts the result. Tell: is the target stable or mildly positive prices (inflation targeting), or a zero nominal rate implying steady deflation (Friedman rule)?

  • The debt-deflation spiral. The self-reinforcing contraction in which falling prices raise the real burden of debt, depressing demand and prices further. The Friedman rule prescribes a specific, mild deflation at exactly the real rate and abstracts this hazard away; it is emphatically not an endorsement of falling prices in general. Tell: is the phenomenon a destabilizing price-debt spiral (debt-deflation), or a benchmark mild deflation that prices money at marginal cost in a frictionless model (Friedman rule)?

  • The optimal inflation tax (Ramsey/Diamond–Mirrlees). The public-finance question of how much revenue to raise via inflation, weighing seigniorage against distortion. The Friedman rule is the special case — a zero inflation tax — that holds only under specific assumptions (e.g. lump-sum taxes available); where revenue is scarce and other taxes distortionary, a positive inflation tax can be optimal. Part-to-whole relation. Tell: does the analysis weigh seigniorage revenue against distortion across taxes (optimal inflation tax), or isolate shoe-leather alone and set the inflation tax to zero (Friedman rule)?

  • marginal_cost_pricing (the parent). The general optimality result the rule instantiates — set the private price of a costless-to-produce good equal to its zero social marginal cost, closing the private–social wedge (with pigovian_tax and externality as the wedge-correction relatives). This is the portable core; the Friedman rule is its monetary specialization and adds no structural pattern of its own. Tell: strip the Fisher equation, fiat money, and cash-holding agents and what remains — price a costless good at its zero social cost — is carried by this parent, not by "the Friedman rule." (Treated fully in a later section.)

Neighborhood in Abstraction Space

Friedman Rule sits in a crowded region of the domain-specific corpus (10th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Monetary Policy & Financial Fragility (15 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-07-12