Expectations And Inflation

Inflation is the rate at which the general level of prices rises.

10 frameworks, from 1900 to today · Macroeconomics · Economics

still a working view ended · each leads to its entry below

Overview

Inflation is the rate at which the general level of prices rises1. Expectations are what people believe inflation will be1. The two are bound together because almost every important economic decision—setting a wage, posting a price, lending money, buying a house—depends not on today's prices but on what prices are expected to be tomorrow. A worker who expects 10 percent inflation will demand a raise to match; a firm that expects its costs to climb will raise prices preemptively. This means inflation can become self-fulfilling: beliefs about future inflation help create present inflation1. The subfield of expectations and inflation is the body of theory and evidence that tries to explain this feedback loop—how expectations are formed, how they feed into wages and prices, and how central banks can manage them. Its history is a sequence of models that each tried to fix what the previous one got wrong, and each fix changed what economists thought monetary policy could and should do.

The Quantity Theory and the Classical Dichotomy

Before expectations were a named problem, classical economists treated money as a veil2. The quantity theory of money held that the general price level is proportional to the amount of money in circulation: double the money supply and, other things equal, prices double3. The mechanism was mechanical, not psychological. In its classic form, associated with Irving Fisher's The Purchasing Power of Money (1911), the theory was expressed as an equation of exchange: the money supply times its velocity of circulation equals the price level times real output. If velocity and output are stable, changes in money translate directly into changes in prices3.

This framework implied that people did not need to forecast inflation in any sophisticated way; prices adjusted to money, and real decisions were unaffected. The classical dichotomy—the separation of nominal variables (prices, wages, money) from real ones (output, employment)—rested on this2. The Great Depression shattered the confidence behind it1. Prices fell, money contracted, and output collapsed in ways the quantity theory could not easily explain, and John Maynard Keynes's General Theory (1936) argued that output and employment could be stuck below capacity because wages and prices were rigid, not flexible. In the Keynesian world, expectations were present but informal—Keynes wrote about "animal spirits" and the "state of confidence"—and were not modeled as a distinct variable that could be measured or manipulated.

The Phillips Curve: A Stable Trade-off?

The next turning point came from empirical observation. In 1958, the New Zealand economist A. W. Phillips published a study of nearly a century of British data showing an inverse relationship between wage inflation and unemployment: when unemployment was low, wages rose faster. Paul Samuelson and Robert Solow extended this to price inflation in 1960, and the Phillips curve became the workhorse of the neoclassical synthesis. It appeared to offer policymakers a menu: accept a little more inflation to get a little less unemployment, or vice versa1. The trade-off looked stable and exploitable.

The problem was that the curve was a purely statistical relationship with no theory of why it should hold4. It described correlation, not causation, and it said nothing about what would happen if governments deliberately tried to use it. In the late 1960s, as the United States pursued expansionary policy, inflation began to rise without the expected fall in unemployment—the curve seemed to shift.

The Natural Rate and the Expectations-Augmented Phillips Curve

Milton Friedman and Edmund Phelps, working independently in 1967–68, supplied the missing theory. Their key move was to distinguish between nominal and real wages. Workers care about what their wages can buy, not the number on the paycheck. If prices rise by 5 percent and wages rise by 5 percent, the worker's real wage is unchanged and there is no reason for employment to rise. The apparent trade-off in the data existed only because workers were temporarily fooled—they confused nominal wage increases with real ones.

Friedman introduced the natural rate of unemployment: the level consistent with the economy's structural features, independent of inflation. Any attempt to hold unemployment below the natural rate would require ever-accelerating inflation, because workers would eventually catch on and demand compensating raises1. The expectations-augmented Phillips curve formalized this: inflation depends on expected inflation plus the gap between actual and natural unemployment. In the long run, when expectations adjust to reality, the curve is vertical—there is no trade-off at all1.

The concrete case that vindicated this view was the stagflation of the 1970s1. Inflation and unemployment rose together, which the original Phillips curve said was impossible14. The Friedman–Phelps framework explained it: expectations had caught up with the inflation policymakers had been generating, and the short-run gains had evaporated1. The lesson—that a central bank cannot permanently buy lower unemployment with higher inflation—became a foundation of modern central banking.

Rational Expectations and the New Classical Revolution

Friedman and Phelps assumed that people learn from experience, but slowly and with systematic errors. In the early 1970s, Robert Lucas, Thomas Sargent, and Robert Barro pushed further with rational expectations: the assumption that people form forecasts using all available information and the correct model of the economy, so their errors are random and unpredictable rather than systematic. This was not a claim that people are omniscient; it was a claim that they do not leave predictable money on the table, and that policy cannot systematically exploit their ignorance5.

The implication was radical. If workers and firms understand the policy rule, they will anticipate its effects1. A central bank that announces an expansionary policy will see wages and prices adjust immediately, leaving real output unchanged5. This was the policy ineffectiveness proposition (Sargent and Wallace, 1975): systematic monetary policy cannot affect real variables. Only surprise policy can, and surprises cannot be repeated6. Lucas's 1976 critique of econometric policy evaluation drove the point home: models estimated on past data would break down when policy changed, because the expectations embedded in the data would shift.

The new classical macroeconomics that grew from this work rebuilt macroeconomics around optimizing agents and market clearing. It made expectations a first-class variable and forced the field to take seriously the idea that the public's beliefs about policy are part of the policy problem7. But it also produced uncomfortable conclusions—that central banks are nearly powerless—that sat badly with the observed fact that monetary policy clearly does affect output in recessions.

New Keynesian Sticky Prices and the Synthesis

The new Keynesian response, developed from the late 1970s through the 1990s, accepted rational expectations but restored a role for policy by explaining why prices and wages are slow to adjust. The central device was sticky prices: firms do not reset prices every instant because doing so is costly (menu costs, contracts, coordination problems). The canonical model, associated with Guillermo Calvo's 1983 formulation, assumed that in any period only a random fraction of firms can change their prices8. This made current inflation depend on expected future inflation and on the current gap between output and potential. The result was a Phillips curve with expectations at its center but with real effects from monetary policy, because not all prices adjust at once.

By the late 1990s, a new Keynesian synthesis had emerged: a dynamic general-equilibrium model with rational expectations, sticky prices, and a central bank setting interest rates. It became the standard framework for policy analysis and the basis for most central bank forecasting models. Its Phillips curve—often called the new Keynesian Phillips curve—made expected future inflation the dominant driver of current inflation, which meant that a central bank's credibility, its ability to convince the public it will keep inflation low, became a policy instrument in its own right1.

Inflation Targeting and the Credibility Framework

The stagflation experience and the rational-expectations critique converged on a practical conclusion: what matters most is whether the public believes the central bank will control inflation. If expectations are anchored—if people expect inflation to stay near target—then temporary shocks do not feed into wages and prices, and the central bank has more room to respond to recessions.

New Zealand adopted formal inflation targeting in 1990, and Canada, the United Kingdom, Sweden, and others followed through the 1990s. The framework combined a numerical target, public communication, and central bank independence. The theoretical case, developed by economists including Lars Svensson and Michael Woodford, was that a credible target anchors expectations and improves the short-run trade-off: a central bank that is believed can stabilize output with smaller inflation costs. Woodford's Interest and Prices (2003) provided the comprehensive theoretical treatment, showing how expectations management could substitute for actual interest-rate movements. The framework's test came in the 2008 financial crisis and the 2020s inflation surge, when central banks had to defend credibility while responding to large shocks—a tension the framework was designed to manage but not to eliminate.

Learning, Surveys, and the Return of Bounded Rationality

Rational expectations was always an awkward assumption: it required ordinary households and firms to know the true model of the economy5. From the 1980s onward, a learning literature—associated with economists such as Albert Marcet, Thomas Sargent, and George Evans—asked what happens when people form expectations by estimating models from data rather than knowing the truth. Learning models could generate persistent deviations from rational expectations and helped explain why disinflation might be costly even when announced. They also showed that some rational-expectations equilibria are unstable under learning, which raised questions about which equilibria are plausible.

A parallel empirical tradition used surveys of expectations—the University of Michigan household survey, the Livingston survey of professional forecasters, and later the Survey of Professional Forecasters—to measure what people actually believe. The findings were consistent: households' inflation expectations are systematically biased, disagree widely, and respond to salient prices like groceries and gasoline more than to official statistics. Professional forecasters are more accurate but still make persistent errors. This evidence fed into behavioral models in which expectations are formed with limited attention, overconfidence, or simple rules of thumb, and into policy work on how central banks should communicate with a public that is not paying close attention.

Heterogeneous Expectations and the Present

The most recent development takes the survey evidence seriously: if people disagree about inflation, a model with a single representative expectation is misspecified9. Heterogeneous-agent models, building on earlier work by economists such as William Branch and Bruce McGough, allow different groups to form expectations differently—some rational, some adaptive, some following simple rules—and study how the distribution of beliefs affects aggregate outcomes and policy transmission. This connects to the broader heterogeneous-agent macroeconomics that models inequality and household heterogeneity explicitly.

Today the field's main approaches coexist. The new Keynesian framework with rational expectations remains the benchmark for policy analysis, but it is increasingly supplemented by learning models, survey-based expectation measures, and heterogeneous-agent simulations. Central banks now routinely track household and firm expectations as policy indicators, and communication strategy is treated as a tool of policy. The open questions are substantial: how expectations are formed at the level of individual households and firms, why expectations sometimes become unanchored, how central banks can rebuild credibility after it is lost, and whether the standard models adequately capture the feedback between beliefs and inflation in a world of rapid information and shifting attention. The history of the subfield is, in large part, the history of economists learning that the public's beliefs are not a detail to be assumed away but a central part of the mechanism they are trying to understand.

References

  1. Inflation (introduction, Inflation expectations, View post-2000 to present, After 1700, Keynes and the early Keynesians, Monetarism, New Keynesians, Rational expectations theory), Wikipedia
  2. Classical dichotomy, Wikipedia
  3. Quantity theory of money, Wikipedia
  4. NAIRU (Origins), Wikipedia
  5. Rational expectations (History, introduction), Wikipedia
  6. Lucas aggregate supply function, Wikipedia
  7. New classical macroeconomics, Wikipedia
  8. Calvo (staggered) contracts, Wikipedia
  9. Heterogeneity in economics, Wikipedia

The frameworks of Expectations And Inflation

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1900-1936

    Quantity Theory of Money

    What determines the general price level?

    Fly to its tower in Nousfarer

  2. 1936-1967
  3. 1958-1968

    Phillips Curve

    Is there a stable trade-off between inflation and unemployment?

    Fly to its tower in Nousfarer

  4. 1967-1985
  5. 1972-1995

    New classical macroeconomics

    Can systematic policy affect real output?

    Fly to its tower in Nousfarer

  6. 1980-Present

    Survey-Based and Behavioral Expectations

    How do people actually form inflation forecasts?

    Fly to its tower in Nousfarer

  7. 1983-Present
  8. 1985-Present

    Learning in Macroeconomics

    How do people learn the true model of the economy?

    Fly to its tower in Nousfarer

  9. 1990-Present

    Inflation targeting

    How can a central bank anchor inflation expectations?

    Fly to its tower in Nousfarer

  10. 2000-Present

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Quantity Theory of MoneyKeynesian MacroeconomicsPhillips CurveExpectations-Augmented Phillips Curve…New classical macroeconomicsSurvey-Based and Behavioral Expectati…New Keynesian economicsLearning in MacroeconomicsInflation targetingHeterogeneous Expectations and Hetero…

Beyond the field

Where Expectations And Inflation's ideas came from in other fields, and what they led to there: the strongest 12 of 52 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Expectations And Inflation is a planet in Nousfarer

Its city has a tower for each of these 10 frameworks. Land, walk in, and climb one concept at a time.

Land on Expectations And Inflation

Related fields

Fields its ideas are linked to

More in Macroeconomics

See where Expectations And Inflation sits on the Map of Human Knowledge