Loading field map
Inflation—the sustained increase in the general price level—is not merely a mechanical consequence of current economic conditions. What people, firms, and financial markets expect inflation to be in the future profoundly influences the inflation that actually occurs. This subfield of macroeconomics studies the relationship between those expectations and realized inflation: how expectations are formed, how they affect wage and price setting, and how central banks can manage them to stabilize the economy. The central puzzle is that expectations are both a cause and an effect of inflation, creating a feedback loop that can amplify or dampen price movements.
The stakes are high. If a central bank cannot anchor expectations—convince the public that inflation will remain low and stable—then even a temporary shock can set off a self-fulfilling spiral of rising prices and wages. Conversely, if expectations are well anchored, the central bank has room to respond to recessions without triggering runaway inflation. The subfield thus sits at the intersection of monetary policy, labor markets, and the psychology of economic decision-making.
Before the mid-twentieth century, inflation was largely understood through the lens of the quantity theory of money: changes in the money supply drove changes in the price level. Expectations played little explicit role. The Great Depression and the subsequent dominance of Keynesian economics shifted attention to aggregate demand and the Phillips curve—the empirical relationship between unemployment and wage inflation. But the Phillips curve was treated as a stable trade-off, not as something that could shift with expectations.
The modern subfield emerged in the late 1960s and 1970s, when the stable Phillips curve broke down. Economists observed that inflation and unemployment could rise together (stagflation), a pattern the old framework could not explain. This crisis prompted a fundamental rethinking of how expectations enter the inflation process.
The first systematic attempt to incorporate expectations into macroeconomics was the adaptive expectations hypothesis. It assumed that people form expectations of future inflation based on past inflation, gradually updating their forecasts as new data arrive. For example, if inflation has been 3% for several years, people expect 3% next year; if it rises to 5%, they revise their expectation upward, but only partially and with a lag.
This approach was used to modify the Phillips curve. The key insight was that the old trade-off between inflation and unemployment held only in the short run. In the long run, if people adapt to higher inflation, the economy would return to a "natural rate" of unemployment—the rate consistent with stable inflation—but at a higher inflation level. This explained why stimulative policies could temporarily reduce unemployment but eventually produced only higher inflation.
Limitations: Adaptive expectations are backward-looking and mechanical. They assume people ignore all information except past inflation, and they systematically make predictable errors during periods of changing inflation. The model cannot explain why rational agents would persist in such errors, especially when large sums of money are at stake.
The rational expectations revolution, beginning in the early 1970s, transformed the subfield. It posited that economic agents form expectations using all available information, including knowledge of the structure of the economy and the likely actions of policymakers. People do not make systematic, predictable mistakes; their forecasts are correct on average, though subject to random errors.
Applied to inflation, rational expectations had radical implications. If the central bank announces a policy to reduce inflation, and the public believes the announcement, then expectations of future inflation will fall immediately. Wage and price setters will adjust their behavior accordingly, and inflation can decline without a prolonged period of high unemployment. Conversely, if the central bank tries to exploit a short-run trade-off by creating surprise inflation, the public will anticipate this and adjust wages and prices preemptively, neutralizing the effect on output.
This approach gave rise to the "Lucas critique" of policy evaluation: using historical data to predict the effects of a new policy is invalid if the policy changes how people form expectations. It also led to the "policy ineffectiveness proposition," which argued that only unanticipated monetary policy could affect real output—a controversial claim that sparked decades of debate.
Limitations: Rational expectations assumes an unrealistic degree of knowledge and computational ability. In practice, people have limited information, face uncertainty about the true model of the economy, and may rely on simple heuristics. The approach also struggles to explain why inflation expectations sometimes appear to be "sticky" or slow to adjust, even when policy changes are clearly announced.
The New Keynesian approach emerged in the 1980s and 1990s as a response to both the empirical failures of pure rational expectations and the theoretical gaps in adaptive expectations. It retained rational expectations as a core assumption but introduced realistic frictions—most importantly, that firms cannot change prices continuously. Instead, prices are adjusted at staggered intervals (a "Calvo pricing" mechanism), so that only a fraction of firms reset their prices each period.
This yields the New Keynesian Phillips curve, which relates current inflation to expected future inflation and a measure of real economic activity (such as the output gap). The key difference from earlier models is that inflation is forward-looking: firms setting prices today care about the expected path of future costs and demand, because their prices will remain fixed for some time. This creates a role for monetary policy to manage expectations: by committing to a future path of interest rates, the central bank can influence today's inflation without changing current conditions.
The New Keynesian framework became the workhorse model for central banks. It explains why inflation can be persistent even with rational expectations (because firms are locked into past price decisions) and why credible central bank commitments can reduce inflation at lower cost than the adaptive expectations model would predict.
Limitations: The model's predictions depend heavily on the assumed frequency of price adjustment and the degree of forward-looking behavior. Empirical estimates vary widely. The framework also struggles to explain the large and persistent deviations of inflation from target observed in some episodes, such as the low inflation of the 2010s despite low unemployment.
A more recent strand of research relaxes the assumption of full rationality, drawing on psychology and experimental evidence. These approaches recognize that people have limited attention, rely on imperfect information, and use simple rules of thumb. For example, some models assume that only a fraction of agents update their expectations each period, or that people form expectations based on a combination of recent data and a long-run anchor (such as the central bank's target).
Behavioral approaches can explain phenomena that rational models find puzzling: why inflation expectations vary so much across households and firms, why they respond slowly to policy changes, and why they sometimes become unanchored during periods of high inflation. They also provide a rationale for why central banks communicate so carefully—because the public may not automatically infer the central bank's intentions from its actions.
Limitations: Behavioral models are often ad hoc, with many free parameters that can be tuned to fit specific episodes. There is no consensus on which psychological mechanisms are most important, and the models can be difficult to use for policy analysis because they lack the discipline of rational optimization.
These approaches are not simply a linear succession where each replaced the previous one. Adaptive expectations were largely abandoned as a theoretical foundation, but they remain useful for describing the behavior of some economic agents in empirical work. Rational expectations provide a benchmark for what would happen under ideal conditions, and most modern models incorporate them as a baseline. The New Keynesian synthesis is the dominant framework for policy analysis, but it is increasingly supplemented by behavioral elements to improve its fit with data.
The central tension in the subfield is between models that assume agents are fully rational and those that allow for systematic deviations from rationality. This tension is not resolved; it reflects a deeper disagreement about how to model human behavior in macroeconomics. Most practitioners adopt a pragmatic stance, using rational expectations as a starting point and adding frictions or behavioral features as needed to match empirical regularities.
The subfield today is organized around a few enduring questions. How do central banks anchor expectations? What determines the speed at which expectations adjust to new information? How should monetary policy be designed when expectations are imperfectly rational? And how do expectations interact with other forces, such as global supply chains, fiscal policy, and financial markets?
Central banks now routinely survey expectations of households, firms, and financial market participants, and they use these data to calibrate their policies. The concept of "forward guidance"—communicating the likely future path of interest rates—is a direct application of the subfield's insights: by shaping expectations, the central bank can influence current economic conditions.
The subfield also grapples with the experience of the 2010s and 2020s, when inflation remained stubbornly low in many advanced economies despite low unemployment and massive monetary stimulus, and then surged in the post-pandemic period. These episodes have challenged existing models and spurred new research on the role of global factors, the measurement of expectations, and the possibility that the relationship between expectations and inflation has changed.
No single approach has won universal acceptance. The field remains a vibrant area of theoretical and empirical work, with ongoing debates about the microfoundations of price setting, the formation of expectations, and the design of robust monetary policy rules. What unites the subfield is the recognition that inflation is not just a monetary phenomenon but a social and psychological one, shaped by what people believe about the future.