RESEARCH DOSSIER

Price Rigidities and Nonclearing Markets

Abstract. This monograph examines the economic foundations, measurement, and consequences of price rigidities and nonclearing markets. It integrates effective-demand analysis, constrained transactions, nominal adjustment costs, asynchronous price setting, state dependence, and information-based credit rationing. Formal illustrations distinguish desired prices from posted terms, intended trade from completed exchange, and equilibrium of strategies from market clearing. Evidence from historical US microprices, harmonized euro-area studies, Japanese institutional experience, wage adjustment, and international import pricing is evaluated with attention to measurement and causal identification. The original analytical synthesis connects repricing incentives, implementation constraints, and realized allocation, explaining why equal observed price-change frequencies can conceal different economic outcomes. The study combines selective documentary research and conditional theoretical reasoning without claiming new empirical estimates or a new named theory.

Keywords: Nominal rigidity; nonclearing markets; quantity rationing; effective demand; menu costs; staggered pricing; state-dependent pricing; credit rationing; wage rigidity; exchange-rate pass-through; microdata; monetary transmission; JEL D45, E12, E24, E31, E52

Introduction and Research Method

A price system performs two tasks that coincide only under particular institutional conditions. It expresses the terms on which exchange may take place and coordinates the quantities that agents choose to acquire or deliver. A market can maintain a publicly available price even when the associated transactions fail to exhaust the plans of buyers or sellers. The corresponding economic question concerns the allocation of scarce opportunities when price adjustment is delayed, constrained, or strategically undesirable.

Price rigidity is especially consequential once transactions create commitments. The proceeds of selling labor finance purchases; a supplier's deliveries permit downstream production; a lending decision changes the resources and risk borne by a borrower. A missed transaction may thus alter demand elsewhere before a quoted price changes. The timing of the response, the distribution of unfulfilled plans, and the organization of exchange are all central to the explanation.

This dossier investigates nominal rigidities, quantity constraints, and nonclearing allocation at the intersection of Economic Theory and Macroeconomics. It compares historical contributions, builds transparent pedagogical examples using standard economic reasoning, and evaluates selected microeconomic findings. Formal exercises establish conditional implications rather than estimated universal laws. The empirical discussion reports documented study results, with their geographic and temporal coverage, rather than estimating new parameters from confidential transaction data. The research design is a structured, selective synthesis of primary theoretical texts and published or institutional research; it is not a preregistered systematic review or meta-analysis.

The principal distinction throughout is between the duration of a price, the degree to which its desired level changes, and the feasibility of completing intended trades. These are separately measurable properties. Confusing them leads to excessive confidence about both the status of a market and the effects of stabilization policy.

Concepts of Rigidity, Clearing, and Equilibrium

Let D(p) denote intended purchases of a homogeneous good at price p and S(p) intended sales, holding income, expectations, product characteristics, and available technology fixed. Market clearing in this simplified setting requires D(p)=S(p). The excess demand schedule is z(p)=D(p)-S(p). A market may clear at an administered price accidentally, at an auction price through bidding, or through an adjustment of willingness to transact. All three are compatible with the same observed transaction volume.

Nominal rigidity concerns infrequent or incomplete revision of a monetary price. Real rigidity concerns reluctance of a desired relative price or markup to change when aggregate conditions change. Downward rigidity is directional: cuts may face different costs than increases. A flexible price with insignificant desired changes can appear unchanged, while a frequently revised price may remain far from its frictionless benchmark after a shock.

Nonclearing means incompatible desired quantities at the prevailing terms under the specified information and constraints. It is essential to identify whose plans are being compared. Desired demand conditional on obtaining a job differs from demand conditional on actual employment. Notional plans use hypothetical unconstrained choices; effective plans incorporate perceived transaction constraints. Realized sales incorporate the allocation rule, so they are a third object.

Ex post purchases always equal ex post sales for completed bilateral transactions. That accounting identity cannot establish clearing. A store may sell every available unit while customers remain unserved. A firm may meet all orders while employees remain involuntarily idle. An equilibrium of strategies can also contain unserved applicants when no feasible unilateral contract change improves a lender's expected return. In that sense, equilibrium refers to consistency of behavior under specified institutions, whereas clearing refers to compatibility of quantity plans at specified terms.

Scientific measurement therefore needs a benchmark, a time interval, and a defined contractual unit. A national inflation rate measures movement in an index, not the share of households that obtained their intended allocation. A labor-market unemployment measure describes labor-force status under its survey definition; it does not itself reveal the flexible-wage counterfactual or prove any particular cost of price adjustment.

Historical Development of the Scientific Problem

Keynes (1936) locates the determination of employment in the intersection of expected aggregate proceeds and the aggregate supply schedule. His account in The Principle of Effective Demand allows an economy to settle at employment below full capacity when planned expenditure does not support additional production. The historical contribution is broader than a modern menu-cost story. It concerns investment, expectations, income determination, and the absence of an automatic assurance that potential output will attract adequate demand.

Postwar disequilibrium analysis sharpened the distinction between intended and feasible transactions. Barro and Grossman's (1971) General Disequilibrium Model belongs to this research program. The bibliographic record establishes its publication; the full article was inaccessible during this review, so the numerical constructions in this dossier are independently derived illustrations, not claims about the original article's detailed equations.

A different qualification appears in Friedman's (1968) Role of Monetary Policy: monetary action can influence activity during adjustment without providing a permanently exploitable employment-inflation trade-off. A policy rule must consider expectations and the distinction between temporary responses and long-run attainable real outcomes. That historical argument does not imply instant clearing of every individual market.

Calvo (1983) introduced a convenient formal representation of asynchronous nominal price revision. Mankiw (1985) examined the distinction between private adjustment incentives and social consequences. Later state-dependent pricing research connected the decision to revise prices to deviations from desired pricing, and microdata studies examined whether those models reproduced the observed patterns. Information-based rationing supplies a separate explanation of constrained trade. The sequence is analytically cumulative: each contribution changes the economic conditions under which price, output, and welfare can adjust.

Fixed-Price Allocation and Rationing

Consider a one-good market with D(p)=120-2p and S(p)=20+3p for prices in a range where both quantities are nonnegative. The clearing benchmark solves 120-2p=20+3p, so p*=20 and Q*=80. At p=16, demand is 88 and supply is 68. Under a short-side transaction rule, Q=min(D,S)=68 and the unmet demand is 20. At p=24, demand is 72 and supply is 92. Actual transactions are 72 and desired sales exceed transactions by 20.

This numerical example assumes immediate matching of feasible trades, no inventories entering the market, and a common quality and price. The short-side rule gives the aggregate volume, not the identity of buyers or sellers. If supplies are allocated proportionately, a buyer requesting 11 units at p=16 obtains 11 times 68/88, or 8.5 units. If units are indivisible, implementable allocation instead requires probabilities, lotteries, priorities, or rounding. If a firm chooses trusted repeat customers, two buyers with the same willingness to pay can face different access.

The ratio of completed demand to demand in the shortage example equals 68/88, approximately 77.27 percent. The remaining 22.73 percent is unfulfilled. Neither percentage estimates a real observed market. It reveals that a fixed price alone leaves an incomplete allocation problem.

Queues can partly perform the work of prices by imposing waiting costs. If a unit's monetary price is p and obtaining it requires a waiting time t valued at v per hour, a buyer's generalized cost is p+vt. Different valuations of time then change access. Cash prices can remain stable while generalized costs diverge. Resale can transfer units toward buyers with higher willingness to pay, but may introduce transaction costs, uncertainty, or legal restrictions.

Inventory adds another margin. Let I(t+1)=I(t)+production(t)-sales(t), with nonnegative inventory and no spoilage for simplicity. After a demand contraction, stocks can rise even while the posted price stays fixed. A supplier may initially fulfill additional demand out of stock after an expansion. Nonclearing at a posted price therefore need not immediately appear in production or measured unemployment. Backorders, emergency procurement, cancellations, and changes in lead times are direct evidence of adjustment along the quantity and service margins.

Cross-Market Constraints and Income Feedback

A one-good rationing diagram leaves the incomes of constrained traders unchanged. In an economy with multiple markets, the failure to realize intended sales may reduce the ability to purchase elsewhere. A household that expected ten paid hours may obtain six; its effective budget shrinks. A firm that fails to sell its intended output may revise labor demand even when the wage is unchanged.

An elementary closed-economy illustration uses real planned expenditure E=A+cY, where Y is real income, A autonomous expenditure and c the marginal propensity to spend with 0<c<1. Assume fixed prices, idle capacity, no government budget response and no induced import leakage. Goods-market expenditure consistency is Y=E, giving Y=A/(1-c). With c=0.75 and A=25, income is 100. If autonomous expenditure falls to 20, income becomes 80. Potential output could remain 120 in both cases; the difference is feasible demand under the illustrative closure.

This calculation is a familiar multiplier mechanism. It does not identify an empirical fiscal multiplier. The assumed consumption coefficient is constant, supply can satisfy demand up to capacity, and financial settlement creates no new constraint. A change in interest rates, expectations, taxes, imports or capacity invalidates the simple numerical response.

To specify labor consequences, let production be Y=aL where a=2 units of output per labor-hour. Employment in the two states is 50 and 40 hours. If 60 hours are offered at the unchanged wage, unfulfilled labor supply rises from 10 to 20 hours. The accounting is exact under the assumed technology and the chosen wage. A wage cut might permit additional employment if effective demand expands or unit cost reductions are passed on, but the effect depends on spending, debt, relative prices and firms' ability to sell.

These equations distinguish a goods market that clears at effective demand from an economy in which labor plans remain incompatible. Calling both 'equilibrium' without specifying the set of constraints obscures this distinction. Likewise, reducing one constrained trader's income can influence other traders who would otherwise have been unconstrained. The policy-relevant issue is the direction and strength of those feedbacks, not merely the initial excess-demand sign.

Microeconomic Foundations of Price Inertia

Suppose a differentiated product has demand q=A p^(-epsilon), with epsilon>1 and constant marginal cost m>0. In the static flexible-price problem the firm maximizes (p-m)A p^(-epsilon). Differentiation yields the interior optimum p*=epsilon*m/(epsilon-1). This markup result follows from the stated isoelastic demand and cost assumptions. A change in A alone changes sales and profits but leaves the optimal price fixed in this special model. The example demonstrates why observed price stability does not by itself imply nominal adjustment frictions.

A firm may face a fixed resource cost K>0 of changing prices. Compare the profit obtained after optimally repricing, net of K, with the profit from retaining the old price. In the one-period approximation, adjustment is worthwhile when the gross gain exceeds K. Near a smooth interior optimum, a Taylor expansion gives the profit gap approximately as H(p-p*)^2/2, with H=-profit''(p*)>0. The inaction region is therefore approximately |p-p*|<=sqrt(2K/H). If K=2 currency units and H=4 profit units per squared price unit, the illustrative threshold is one price unit. A desired change of 0.5 yields an approximate gain of 0.5 and is deferred; a deviation of 1.5 yields 4.5 and is worth correcting.

The quadratic approximation concerns small deviations and static profits. A dynamic firm compares discounted expected gains across future states, including the option value of waiting. Menu costs can include actual repricing labor, contract administration, software changes, communication or anticipated customer responses, but an estimated residual adjustment cost cannot uniquely identify any one physical expense.

Mankiw's (1985) Small Menu Costs and Large Business Cycles examines how the private adjustment incentive in a monopoly setting can differ from welfare consequences. A price-setting firm's lost profit near its optimum can be second order in the deviation, while associated changes in consumer surplus or the effects of other distortions can be substantially larger. This is a model-dependent welfare comparison rather than a general assertion that every unchanged price is inefficient.

Coordination can generate real rigidity: if competitors hold prices steady, a firm considering an increase may expect market-share losses; if suppliers' costs and competitors' prices move together, the preferred price responds differently. Stable relative-price targets can amplify the aggregate consequences of small nominal frictions. In empirical work, this mechanism is distinguished from a simple fixed cost by observing responses to competitors, demand, input costs, and anticipated future adjustment.

Time-Dependent Adjustment and Aggregate Dynamics

The original Staggered Prices article by Calvo (1983) represents asynchronous revision using random opportunities and continuous time. A standard discrete-time pedagogical counterpart assumes that a firm cannot reset with probability theta per period, independently across firms and time. A fraction 1-theta can reset each period. The probability of surviving k opportunities without adjustment is theta^k, and the expected number of periods until a reset is 1/(1-theta), counting the successful period.

For theta=0.8, the expected duration is five periods. The probability of at least one opportunity within three periods is 1-0.8^3=0.488. These values are illustrative and are not empirical estimates. If the probability of a reset changes with the distance from the desired price, the constant-hazard expressions cease to be valid.

With a continuum of symmetric differentiated products, elasticity epsilon>1, and all resetting firms choosing the same price P*(t), a CES price index satisfies P(t)^(1-epsilon)=theta P(t-1)^(1-epsilon)+(1-theta)P*(t)^(1-epsilon), assuming random nonadjustment and a stationary weighting structure. Near a zero-inflation steady state with equal relative prices, its log-linear approximation is p(t)=theta p(t-1)+(1-theta)p*(t). An unexpected rise in reset prices changes the aggregate price level by only a fraction of the reset-price movement in that period.

In a benchmark New Keynesian environment with optimizing households, monopolistic competition, Calvo adjustment, no indexation, a simple cost structure and a zero-trend-inflation linearization, inflation can be written as pi(t)=beta E(t)[pi(t+1)]+kappa mc(t), where beta is the discount factor, mc(t) a real marginal-cost deviation, and kappa=(1-theta)(1-beta theta)/theta under the simplest corresponding cost specification. More elaborate real rigidities alter the mapping from output or marginal cost to inflation; the equation is not the original continuous-time Calvo system reproduced verbatim.

For beta=0.99 and theta=0.75, kappa=0.0858333 under this specification. Lower reset probabilities reduce the immediate response to marginal cost. The forward-looking term matters because today's resetting firm anticipates conditions while its price may remain fixed. One cannot infer the slope of the output-gap Phillips curve from theta alone unless the relationship between output, marginal cost, labor supply, and technology is specified.

Importantly, a demand-constrained monopolistic firm can fulfill every order at its posted price even if the price is not its currently desirable flexible-price choice. Standard staggered-price models can therefore feature nominal rigidity and equilibrium goods transactions without explicit queues. The presence of price dispersion or a sticky-price output gap is not logically sufficient to prove literal quantity rationing at every product market.

State Dependence, Selection, and Counterexamples

State-dependent adjustment conditions the probability of revision on the value of doing so. Firms facing large deviations from their desired prices can be disproportionately represented among those that reprice. A monetary shock therefore changes not only the price set by adjusters but also the identity of adjusters and potentially the number of revisions.

Golosov and Lucas (2007), in Menu Costs and Phillips Curves, formulate a monetary economy with idiosyncratic productivity shocks and a real cost of changing prices. Their calibration uses US individual-price evidence. The mechanism is useful because it ties firm-level decisions to a distribution of price gaps. The model's quantitative consequences depend on that distribution and calibration; its results should not be treated as independent estimates of all countries' monetary transmission.

An analytical counterexample clarifies the selection problem. Imagine 100 equally weighted firms, 20 of which adjust in each of two hypothetical economies. In the first, 20 firms are randomly chosen and the average increase among them is 2 percent. In the second, 20 firms are selected because their price gaps are large and their average increase is 10 percent. Under a small-change arithmetic approximation, aggregate price changes are 0.4 percent and 2 percent, respectively. Both adjustment frequencies are 20 percent. Different conditional changes produce different aggregate inflation.

A related counterexample separates the frequency of revisions from monetary nonneutrality. If an economy with many idiosyncratic changes also has selection that rapidly accommodates aggregate nominal shocks, prices may appear sticky in individual spells while the aggregate responds substantially. Conversely, frequent temporary sales with unchanged underlying regular prices can coexist with relatively sluggish adjustment to persistent costs. These logical possibilities do not establish the empirical prevalence of either regime.

Price-setting models should consequently be assessed against several dimensions simultaneously: spell lengths, distributions of price-change size, the inflation response of upward and downward revisions, product turnover, and the timing of adjustment after plausibly identified shocks. Matching one unconditional moment can leave the economically central dynamic mechanism unidentified.

Information, Credit, and Equilibrium Nonclearing

The credit market reveals a form of nonclearing that survives freely chosen nominal terms. A bank sells a contingent claim to future repayment. The quoted interest rate changes the payoff if repaid and potentially the identity or behavior of those who borrow. Competition does not guarantee that raising the rate above its expected-return-maximizing level is advantageous.

Stiglitz and Weiss (1981) establish this possibility in Credit Rationing in Markets with Imperfect Information. Their central mechanisms are adverse selection and incentives: the credit price itself can affect portfolio risk. This provides a distinct explanation of rationing even when banks can alter interest rates.

An elementary numerical illustration assumes a unit loan, repayment of 1+r in the success state, zero recovery upon default, and funding cost of one. Let repayment probability of the relevant loan pool be s(r). Expected gross repayment is s(r)(1+r). At r=0.10 with s=0.95, the value is 1.045; at r=0.30 with s=0.75, it is 0.975. Raising the interest rate lowers expected receipts in this hypothetical comparison. The numbers are not estimates or a reconstruction of Stiglitz and Weiss's model.

If lenders possess scarce lending resources and several applicants offer the lower rate while a higher rate worsens expected risk, rationing may persist. Establishing a full equilibrium requires specifying borrower choices, lender entry, capital and funding, observability, the contract space, and whether collateral or screening can profitably separate risks. Two hypothetical observations alone do not prove an interior optimum over the entire rate schedule.

Distinguish these cases from an externally imposed interest-rate ceiling. A ceiling might prevent otherwise profitable repricing. Information-related rationing may exist in the absence of a ceiling. A bank may also reject applicants because of its regulatory capital position, expected default costs, or unavailable collateral. Identifying a specific mechanism requires evidence beyond low credit volume.

Credit constraints propagate into goods and labor markets through working capital and household budgets. A supplier denied financing may fail to produce goods despite strong order books; a household denied a loan may curtail expenditure despite willingness to pay the quoted loan rate. A general demand stimulus can increase applications without relaxing information or balance-sheet constraints. The appropriate comparative statics therefore depend on the contractual and financial foundations of observed nonclearing.

Labor Markets, Nominal Wages, and Asymmetry

Labor contracts include a wage but also hours, benefits, performance requirements, job security, recruitment costs and usually continuing relationships. Market clearing at one cash wage is therefore an incomplete description of employment. Adjustments can occur through hiring, quits, layoffs, shifts in hours, bonus changes, or the composition of jobs.

Downward nominal rigidity poses an asymmetry in wage adjustment. Consider a fixed nominal wage of 100 monetary units and a price index initially equal to 1. If prices rise to 1.04 while the wage remains 100, its purchasing power becomes 100/1.04=96.1538 original-price units, a decline of approximately 3.846 percent. If prices fall to 0.96, the unchanged wage becomes 104.1667 original-price units. Changes in the price level can thus modify real labor costs even when money wages do not change.

Daly and Hobijn (2014) investigate Downward Nominal Wage Rigidities in a model compared with US wage patterns. The strength and shape of the Phillips-curve relationship depend on prevailing inflation and the degree of resistance to nominal wage cuts. This makes the inflation regime relevant to the feasible adjustment of real wages. The study is not evidence for a mechanically fixed economy-wide wage floor.

Worker-level data are needed to distinguish wage freezes from changes in worker composition. Average earnings may increase in a recession if low-paid jobs disappear, despite unchanged wages among continuously employed workers. A wage histogram with a spike at zero can reflect contracts, measurement rounding, or infrequent observation. A credible study should compare intended with actual adjustments and observe changes in hours, bonuses, hiring wages and mobility.

From the perspective of allocation, real-wage flexibility may relieve one margin of adjustment while household cash-flow losses or lower spending worsen another. The causal sign of a generalized wage cut consequently depends on the joint response of product demand, production costs, creditor claims and expectations. A local labor-demand elasticity measured at fixed product prices is insufficient to answer the aggregate question.

Microdata Measurement and United States Evidence

A price-change frequency is a weighted or unweighted fraction of comparable observations whose prices differ from the preceding observation. The definition of a comparable product matters: model replacement, package size, quality revision, store switches and stockouts can create apparent changes or erase genuine ones. A price spell starts and ends according to the observer's product and sampling rules. Left- and right-censored spells require distinct treatment.

Bils and Klenow (2004), in Some Evidence on the Importance of Sticky Prices, study US BLS information for 1995–97 across 350 spending categories, representing about 70 percent of consumer expenditure. Their reported price durations are substantially shorter than many earlier estimates, with half below 4.3 months. The evidence is historical and sample-defined. It does not imply an unchanged present-day duration or a common value for every category.

Nakamura and Steinsson (2008) show in Five Facts about Prices that sale treatment changes the measurement considerably. For identical consumer items they report median monthly frequencies of 9–12 percent for nonsale revisions and 19–20 percent when sales are included. When substitutions are counted, the reported ranges are 11–13 and 21–22 percent. Their documented seasonality and lack of an upward-sloping individual hazard in the data pose additional restrictions on benchmark models.

The two studies do not present a simple contradiction. They use different observation windows and measurement treatments. In an application concerned with changes in persistent nominal contracts, a return from a promotion to an unchanged regular price differs economically from a firm permanently resetting its baseline. In an analysis of expenditure actually paid by consumers, temporary sales matter directly. Selecting one price measure without reference to the question is an avoidable identification error.

A common transformation of a monthly reset frequency f into a mean spell length is 1/f months. It is justified for homogeneous, memoryless, uncensored spells with one relevant opportunity per period and complete observation. The median duration is instead the smallest integer m for which (1-f)^m<=0.5. Mixtures with heterogeneous hazards do not generally obey a single geometric distribution. The reciprocal of a weighted mean frequency is not the weighted mean of individual durations.

Price changes and availability must also be observed together. Missing prices during outages or stockouts may be treated as missing data by the statistical system while precisely documenting an economically relevant quantity constraint. A data set of only surviving products may underrepresent firms unable to maintain supply. Researchers should report coverage, imputation, turnover, sale identification and weights before interpreting recorded zero changes as behavioral rigidity.

International Comparisons and the Inflation Regime

The harmonized European evidence makes the distinction between product composition and institutional differences particularly important. Gautier and coauthors (2022) use detailed CPI observations spanning 2010–19 in 11 euro-area economies. Their New Facts on Consumer Price Rigidity study covers roughly 60 percent of the relevant consumption basket. Reported average monthly price-change frequencies are 12.3 percent in the euro area and 19.3 percent for the US comparison; excluding sales gives 8.5 and 10 percent, respectively.

The differences are seven percentage points with promotions and 1.5 percentage points without them. These arithmetic comparisons describe the reported samples. They are not estimates of the causal effect of European institutions or monetary policy. Sectoral composition, the definition of a sale, sampling frequency and product turnover can influence the measured gap. The study reports substantial variation across sectors relative to country differences, reinforcing the need for disaggregated comparisons.

For a purely illustrative memoryless model, 12.3 percent corresponds to an expected 8.13 months until adjustment, and 8.5 percent corresponds to 11.76 months. Those are reciprocals of reported frequencies, not measured average spell lengths. The exercise shows why a mechanically calibrated representative hazard is sensitive to promotional treatment and why average frequency cannot simply be compared with measured median duration.

A low-inflation sample gives limited information about pricing behavior during unusually large aggregate disturbances. As persistent nominal or cost pressures accumulate, more firms may find it profitable to revise. Adjustments may become synchronized by common supplier contracts, energy shocks or widespread changes in inflation expectations. Models calibrated only to tranquil regimes need out-of-sample validation before their implied policy responses are extrapolated.

Japan supplies an institutional perspective that differs from treating low aggregate inflation as a sufficient statistic. In his September 2025 Transformation of the Global Economy speech, Bank of Japan board member Asahi Noguchi discusses the long-lived zero norm of individual prices and wages, its concentration in services and rents, and the shift in price-change distributions after the pandemic. This is a dated interpretation by a policymaker; it is evidence about the issues and descriptive patterns emphasized by the institution, not an independently identified causal estimate.

An economy with roughly equal numbers of large price increases and decreases can have near-zero inflation while considerable relative-price reallocation occurs. Another with almost all prices fixed may have the same near-zero rate. Comparing the distributions, labor arrangements and shock histories is more informative than comparing mean inflation. The Japanese example also warns against assuming that an estimated parameter is invariant to expectations, contract institutions or monetary regimes.

The international evidence here is selective: United States, euro area and Japan are documented cases, supplemented by general international trade reasoning. Comparable evidence for low-income and high-inflation economies, informal markets and countries with managed exchange rates remains a research priority. The analytical scope is international, but the dossier does not claim geographic representativeness of its numerical comparisons.

International Prices and Exchange-Rate Transmission

For a foreign good with exporter price P* denominated in foreign currency and an exchange rate e measured as domestic currency per foreign unit, the domestic converted price is eP*, before distribution costs and tariffs. If e rises by 10 percent with P* fixed, the converted price rises by 10 percent. If a contract instead fixes the domestic-currency invoice, its domestic price does not change immediately; the foreign-currency value of receipts adjusts. Currency denomination identifies a specific nominal commitment.

Actual pass-through depends on the currency of invoicing, the response of marginal costs and markups, the distribution chain, and whether a revision occurs. Using US import data, Gopinath, Itskhoki and Rigobon (2010) report in Currency Choice and Exchange Rate Pass-Through conditional pass-through of 25 percent for the average dollar-priced good versus 95 percent for nondollar pricing. The comparison is conditional on price changes and concerns import prices, not national consumer-price inflation.

In a log-linear thought experiment, a 10 percent exchange-rate movement combined with an assumed constant elasticity of 0.25 predicts approximately 2.5 percent import-price change; elasticity 0.95 predicts 9.5 percent. The calculations are illustrative approximations using the published contrast, not predictions for any individual shipment. Exact finite-change arithmetic and actual elasticities may differ.

Currency selection is endogenous in the study. Firms may choose a currency precisely because their expected pass-through differs, so treating contract currency as randomly assigned is inappropriate. A comparison of local- and foreign-currency prices can be informative only with controls for product, supplier, destination, cost exposure and selection into invoicing.

Quantity restrictions can coexist with incomplete pass-through. A dealer may preserve a posted domestic price but delay purchases of imported inventory after depreciation. Consumers encounter scarcity before the retail price reflects replacement costs. An importer that reprices promptly may still be unable to obtain foreign exchange or physical supplies. The exchange rate, the invoice price, the landed cost and the consumer's final transaction are distinct objects in an international allocation analysis.

Empirical Identification and Falsifiable Implications

The primary identification problem is that both desired price movements and observed adjustment decisions respond to economic conditions. A high revision rate can signal small costs of repricing, large underlying shocks, or a combination. A low revision rate can reflect stable desired prices. Separating the mechanisms requires variation in predicted desired prices independently of adjustment opportunities.

A useful observational design joins product-level prices and sales with inventories, delivery dates, input costs, customer orders, and information on contract renewals. For a documented cost shock that differentially affects products by predetermined input shares, estimate the probability of revision and the distribution of changes by exposure. Compare changes before the shock to test anticipatory trends, and account for common demand effects and correlated cost exposure. Exposure is not automatically exogenous; sourcing choices, strategic inventory and expected demand may determine it.

A time-dependent hypothesis with a fixed exogenous reset opportunity predicts that the arrival of the opportunity is independent of a firm's price gap, although observed nonzero changes may depend on that gap. A state-dependent model predicts that sufficiently large expected benefits increase the willingness to pay the adjustment cost. A test must therefore distinguish review opportunities from completed price changes. Administrative dates or contract review calendars may identify opportunities, whereas price observations alone rarely do.

For nonclearing, define fulfillment as actual completed units divided by valid requested units under a fixed order protocol. Analyze rejected orders, backorders, waiting times and cancellations conditional on price and product characteristics. A rise in canceled orders without a price change can indicate rationing; alternatively, customers may have responded to expected quality deterioration, entry of substitutes or delivery risk. Both the demand denominator and the reason for cancellation require observation.

To investigate monetary transmission, high-frequency policy surprises, narrative shocks, or other defensible identification strategies can provide candidate instruments. Financial prices may react to central-bank information about the outlook as well as to unexpected policy; controlling or separating those components is essential. Local projections can describe dynamic responses but cannot independently fix a defective identification assumption. Monetary shocks may also affect firms differently by debt exposure and contract structure, which should be considered before attributing responses solely to nominal rigidity.

A practical rejection criterion is mechanism-specific. The simplest fixed-hazard specification is strained by systematic variation in true opportunities with price gaps. A simple short-side rationing model is strained if observed intended trade cannot be measured or if actual sales exceed contemporaneous intended supply because inventories were omitted. An information-based credit explanation is strained when complete observable contracting alternatives eliminate the supposed risk-selection effect. Each criterion relates a claim to an observable margin rather than treating all output fluctuations as confirming evidence.

Welfare, Distribution, and Policy

When prices do not clear markets, allocation rules affect both efficiency and distribution. For a scarce homogeneous good, efficient assignment under a conventional surplus criterion gives units to users with the highest willingness to pay relative to opportunity cost. Proportional rationing, queues and incumbent priority can yield different allocations even at the same number of transactions. The ethical objective may differ from willingness to pay if income or access is unequal; the welfare criterion must be stated instead of smuggling distributional assumptions into an efficiency calculation.

In the earlier linear example, inverse demand is 60-Q/2 and inverse supply is (Q-20)/3. At Q=68 the marginal willingness to pay is 26 and marginal supply cost is 16. At Q=80 they coincide at 20. Under efficient assignment of the 68 units, the lost surplus relative to Q=80 is the integral of their difference from 68 to 80. The linear wedge narrows from 10 to zero, so the lost surplus is 60 currency units. A queue or mismatched priority can introduce additional costs, and transfers between buyers and sellers do not themselves constitute resource losses under equal welfare weights. This calculation assumes the schedules are valid welfare measures with no externalities.

A demand-related recession with spare capacity and slowly changing prices may respond to monetary or fiscal expansion by increasing sales and employment. A severe supply restriction can instead convert additional nominal expenditure into bidding for scarce resources, longer waits, or inflation. A credit constraint caused by capitalization or screening may respond more directly to repairs in credit intermediation. The instrument should correspond to the constraint, including the fiscal and distributional consequences of the instrument.

An administrative price ceiling can protect some existing buyers and simultaneously suppress the monetary signal of shortage. Its consequences depend on supply response, compensation, access rules, enforcement, duration and complementary provision. A temporary ration book with credible procurement is economically different from relying indefinitely on a ceiling without expanding supply. Price liberalization can improve supply incentives while increasing near-term expenditure burdens for constrained households. Targeted transfers may address purchasing power but cannot instantly create physically unavailable output.

Monetary stabilization faces a related timing problem. Expanding spending before all prices reset can raise real demand; persistent unexpected inflation is not a sustainable way to exploit that margin because expectations, contracts and behaviors adapt. Conversely, tightening to reduce inflation may impose disproportionate losses on sectors with slower wage or contract adjustment. These observations establish the relevant policy trade-offs without selecting a universal optimal instrument or numerical inflation target.

Welfare assessment must include the cost of actual price revision, dispersion in relative prices, involuntary idleness, queue time, lost matching opportunities and informational losses. Reducing any one component may raise others. A policy evaluated only by the speed of CPI adjustment can therefore miss the welfare benefits or costs borne through transactions and access.

Original Analytical Synthesis

The research supports a separation of three empirical questions within a common account of market coordination. The first concerns desired-price adjustment: how does the optimizing or institutionally determined target respond to changed costs, demand, expectations and competitors? The second concerns implementation: when and at what cost can the quoted terms move toward that target? The third concerns realized exchange: given prevailing terms and institutional allocation rules, which planned transactions take place?

The distinction has substantive consequences. An unchanged price with unchanged desired price and full fulfillment is compatible with a flexible market. An unchanged price with a changed target and full fulfillment indicates an implementation friction without necessarily creating product rationing. An unchanged price with persistent unfulfilled orders can involve both frictions and an allocation constraint. A freely chosen interest rate with unserved borrowers can reflect an information-dependent feasible contract set even if the price is revised when profitable. These cases have different scientific explanations despite superficially similar price observations.

Consider two sectors with identical observed price-change frequencies and total sales. One clears orders by drawing down inventories, while the other rations scarce output through waiting lists. The same price statistic and sales total describe different dynamics and welfare. In the first, sustainability depends on replenishment; in the second, the distribution of users, time costs and forgone consumption already matters. A follow-up observation of inventories and order fulfillment distinguishes the cases. This is a direct analytical implication of the measurement definitions, not a newly named theoretical framework.

The international comparison strengthens the argument: identical monetary values can mean different commitments depending on contract currency and indexation. A wage freeze during inflation changes workers' purchasing power; an import-price freeze in local currency changes the foreign supplier's conversion revenue. Contract structure identifies which agent initially absorbs the shock. The subsequent change in sales, hours, credit or inventories determines whether the burden remains with that agent or propagates.

A central methodological conclusion follows. Rigidity is estimated relative to a specified counterfactual and time scale, while nonclearing is diagnosed relative to intended trades and the actual settlement process. Analysts must connect these observations before attributing a macroeconomic response to sticky prices. This connection also limits welfare claims: absent evidence about feasible transactions and the allocation of scarce units, price-change frequency alone is not an adequate measure of economic loss.

The synthesis integrates established contributions, independently worked examples, and interpretation of documented empirical studies. It asserts no priority claim, new theorem, named model, doctrine, or universal empirical parameter. Its contribution is the systematic reconciliation of different mechanisms and the explicit consequences for identification and welfare.

Limitations and Open Scientific Questions

The scientific conclusions are conditional on the concepts and assumptions stated. The numerical examples are constructed rather than estimated. The literature selection is purposive; it prioritizes canonical arguments and microeconomic studies directly relevant to the subject. Some publications are consulted through primary reproductions or publisher and repository abstracts, and Barro and Grossman through metadata only. No confidential CPI microdata, matched transaction panel, or monetary shock series was obtained for independent estimation.

Future work should test the simultaneous evolution of regular prices, actual transaction prices, and measures of product availability through high- and low-inflation intervals. Linking stockouts and order-level data to producer and retail prices could improve the identification of quantity constraints. Researchers should distinguish missing observations generated by market exit from changes caused by a sampling or statistical protocol.

A second question concerns the aggregation of firms that face different repricing opportunities, size distributions and customer contracts. The macroeconomic response depends on expenditure weights, sectoral networks and selection among firms. Estimating an unconditional average hazard does not establish an invariant distribution of price gaps.

A third question concerns market institutions and informal exchange outside the documented samples. Comparable evidence about bargaining, mobile pricing, regulated utility tariffs, provisions of foreign currency, and rural product markets would extend the international scope. Studies should make local contract definitions compatible where possible and explicitly account for differences that cannot be harmonized.

Finally, welfare effects require evidence about workers, households and firms that fail to complete desired trades. Observed transactions record a selected subset of economic plans. Surveys and administrative records of unfulfilled demand, constrained supply and rationing criteria can help recover that omitted information without treating hypothetical responses as observed choices. No single data set is likely to identify every margin.

Conclusion

Price rigidities affect the speed and distribution of economic adjustment because contracts, expectations and the organization of price setting govern when nominal terms change. Nonclearing markets arise when prevailing terms and feasible allocation rules leave desired trades unrealized. The two phenomena intersect frequently, and they can also occur independently. A rigorous analysis states the relevant benchmark, the constraint, the adjustment mechanism and the trading rule.

The short-side allocation example shows how fixed prices transfer adjustment to quantities and account for unfulfilled demand or supply. Income feedback extends that mechanism between product and labor markets. The menu-cost and staggered-price examples show why multiple microeconomic mechanisms can generate slow nominal adjustment. Credit rationing illustrates how information can make nonclearing consistent with optimizing contract choices.

Microdata results establish the empirical importance of price-change definitions, temporary promotions, heterogeneity and institutional context. International pricing and wage evidence identifies additional contractual margins. The appropriate interpretation of market outcomes requires a joint account of price changes, conditional changes in preferred prices, orders and transactions, and changes in the resources available to buyers and sellers.

Scientific and policy evaluation consequently benefits from explicit identification of which constraint binds, who carries the adjustment cost, how long the constraint persists, and what alternative allocation is feasible. The economic significance of rigidity resides in the path of real decisions and exchanges that takes place before and during price adjustment. This is the foundation on which competing explanations of nonclearing markets can be compared and tested.

References

  • Barro, Robert J., and Herschel I. Grossman (1971). A General Disequilibrium Model of Income and Employment. American Economic Review 61(1): 82–93. Research text or record

  • Bils, Mark, and Peter J. Klenow (2004). Some Evidence on the Importance of Sticky Prices. Journal of Political Economy 112(5): 947–985. DOI: 10.1086/422559. Research text or record

  • Calvo, Guillermo A. (1983). Staggered Prices in a Utility-Maximizing Framework. Journal of Monetary Economics 12(3): 383–398. DOI: 10.1016/0304-3932(83)90060-0. Research text or record

  • Daly, Mary C., and Bart Hobijn (2014). Downward Nominal Wage Rigidities Bend the Phillips Curve. Federal Reserve Bank of San Francisco Working Paper 2013-08, January 2014 version; subsequently Journal of Money, Credit and Banking 46(S2): 51–93. Research text or record

  • Friedman, Milton (1968). The Role of Monetary Policy. American Economic Review 58(1): 1–17. FRASER, Federal Reserve Bank of St. Louis. Research text or record

  • Gautier, Erwan, Cristina Conflitti, Riemer P. Faber, Brian Fabo, Ludmila Fadejeva, Valentin Jouvanceau, Jan-Oliver Menz, Teresa Messner, Pavlos Petroulas, Pau Roldan-Blanco, Fabio Rumler, Sergio Santoro, Elisabeth Wieland, and Hélène Zimmer (2022). New Facts on Consumer Price Rigidity in the Euro Area. Working Paper 2225, Banco de España; also ECB Working Paper 2669. Research text or record

  • Golosov, Mikhail, and Robert E. Lucas Jr. (2007). Menu Costs and Phillips Curves. Journal of Political Economy 115(2): 171–199. DOI: 10.1086/512625. Research text or record

  • Gopinath, Gita, Oleg Itskhoki, and Roberto Rigobon (2010). Currency Choice and Exchange Rate Pass-Through. American Economic Review 100(1): 304–336. DOI: 10.1257/aer.100.1.304. Research text or record

  • Keynes, John Maynard (1936). The General Theory of Employment, Interest and Money. London: Macmillan. Chapter 3, The Principle of Effective Demand; electronic transcription hosted by Duke University. Research text or record

  • Mankiw, N. Gregory (1985). Small Menu Costs and Large Business Cycles: A Macroeconomic Model of Monopoly. Quarterly Journal of Economics 100(2): 529–538. DOI: 10.2307/1885395. Research text or record

  • Nakamura, Emi, and Jón Steinsson (2008). Five Facts about Prices: A Reevaluation of Menu Cost Models. Quarterly Journal of Economics 123(4): 1415–1464. DOI: 10.1162/qjec.2008.123.4.1415. Research text or record

  • Noguchi, Asahi (2025). Transformation of the Global Economy and Developments in Monetary Policy. Speech at a meeting with business leaders in Sapporo, September 29. Bank of Japan. Research text or record

  • Stiglitz, Joseph E., and Andrew Weiss (1981). Credit Rationing in Markets with Imperfect Information. American Economic Review 71(3): 393–410. University-hosted JSTOR reproduction. Research text or record