Rational expectations

Based on Wikipedia: Rational expectations

The Theory That Changed How Economists Think About Thinking

In 1961, a young economist named John Muth asked a deceptively simple question: What if we stopped assuming that people are stupid?

For decades, economic models had treated expectations—what people believe will happen in the future—as somewhat arbitrary. Maybe consumers expect next year's prices based on last year's prices. Maybe businesses forecast demand by looking at recent trends. The specific mechanism didn't matter much; what mattered was that economists could plug in some formula and watch their models run.

Muth thought this was backwards. He proposed something radical: people form expectations rationally. They use all available information. They learn from their mistakes. They don't systematically get things wrong in predictable ways.

This idea—rational expectations—would go on to reshape macroeconomics entirely. It would challenge the effectiveness of government policy, upend conventional wisdom about inflation and unemployment, and earn its key developers a Nobel Prize. But at its core, it simply asked economists to respect the intelligence of the people they were modeling.

What Rational Expectations Actually Means

Let's be precise about what "rational" means here, because it's not quite what you might think.

Rational expectations doesn't claim that people have perfect foresight. It doesn't assume everyone has a PhD in economics or access to the same supercomputers that central banks use. It makes a much more modest claim: on average, people don't make systematic errors.

Think about it this way. If you're trying to predict tomorrow's weather, you might be wrong. You might guess sunny when it rains. But if you're rational in the economic sense, you won't be wrong in a predictable direction. You won't consistently underestimate rainfall by 20 percent every single time. If you did, you'd eventually notice the pattern and adjust.

The technical way economists express this is elegant. The actual outcome equals what people rationally expected, plus some random error. That random error has an expected value of zero—it's just as likely to be positive as negative. It represents genuine surprises, information that literally could not have been known in advance.

This matters enormously for policy. If people make random errors, you can't exploit those errors systematically. But if people made predictable errors—if they consistently underestimated inflation, say—then a clever government might be able to use that predictability to manipulate outcomes. Rational expectations closes off that possibility.

The Death of the Phillips Curve (Sort Of)

To understand why rational expectations was so revolutionary, you need to understand what it demolished: the Phillips curve.

In 1958, economist William Phillips documented an apparently stable relationship between unemployment and wage inflation in the United Kingdom. When unemployment was low, wages rose quickly. When unemployment was high, wages stagnated. Later economists extended this to prices generally: low unemployment seemed to come with high inflation, and vice versa.

This was catnip for policymakers. It suggested a menu of options. Want lower unemployment? Accept a bit more inflation. Worried about rising prices? Tolerate higher unemployment. You couldn't have everything, but you could choose your preferred tradeoff.

The 1970s shattered this comfortable picture. The United States and other developed economies experienced "stagflation"—high unemployment and high inflation simultaneously. The Phillips curve seemed to have vanished.

Rational expectations explained why. The apparent tradeoff had been an illusion, or at best a temporary phenomenon that depended on fooling people.

How Expectations Kill the Tradeoff

Here's the logic, stripped of its mathematical complexity.

Suppose the government wants to reduce unemployment. Traditional thinking suggested it could increase the money supply, which would raise prices somewhat, making workers cheaper in real terms and encouraging businesses to hire more.

But this only works if workers don't notice. If workers expect inflation, they'll demand higher wages to compensate. Businesses, anticipating higher wages, won't hire more workers after all. The inflation happens, but unemployment stays the same.

Under rational expectations, workers do notice. They pay attention to monetary policy. They read the news. They understand, at least roughly, how the economy works. When the government expands the money supply, workers adjust their expectations immediately. The attempt to exploit the Phillips curve fails before it even begins.

The mathematics confirms this intuition. When you work through the equations carefully—allowing for rational expectations—the terms involving monetary policy cancel out. Only random shocks can push unemployment away from its natural rate, and those shocks are, by definition, unpredictable and unexploitable.

Robert Lucas and the Neutrality of Money

Robert Lucas, working at the University of Chicago in the 1970s, took Muth's insight and applied it to monetary policy with devastating effect.

His key contribution was showing that when people hold rational expectations, money is "neutral" in an important sense. Changes in the money supply don't affect real variables like output and employment—at least not in any systematic, exploitable way. Money only affects nominal variables like the price level.

This wasn't entirely new. Classical economists had long believed something similar. But Lucas showed that this neutrality held even in the short run, provided expectations were rational. Previous economists had assumed that money might be neutral in the long run but could have real effects in the short run, while people were still adjusting. Lucas demonstrated that rational people adjust immediately.

The practical implication was stark. A central bank that tries to stimulate the economy by printing money will fail. People will anticipate the resulting inflation and adjust their behavior accordingly. All the central bank accomplishes is higher prices.

This doesn't mean monetary policy is useless. Lucas argued that consistent, predictable monetary policy helps the economy by reducing uncertainty. What doesn't work is trying to be clever, surprising people with unexpected money supply changes to juice growth. Rational people can't be surprised systematically.

Thomas Sargent and the Natural Rate

Thomas Sargent, another pioneer of the rational expectations revolution, connected these ideas to the concept of a "natural rate" of unemployment.

The natural rate is the unemployment level that an economy gravitates toward when there are no surprises. It's determined by structural factors: how easy it is to match workers with jobs, how much skills vary across the workforce, how much friction exists in labor markets. It has nothing to do with monetary policy or government stimulus.

In the short run, unemployment might deviate from the natural rate. A genuine surprise—an unexpected oil shock, an unforeseen pandemic—can push the economy off course. But rational expectations implies that policymakers can't deliberately engineer such deviations. Any policy that's systematic enough to be effective is also systematic enough to be anticipated.

Sargent's work had an important implication for what governments should and shouldn't try to do. Trying to push unemployment below the natural rate through monetary or fiscal stimulus is futile at best. At worst, it just generates inflation. Better to focus on structural reforms that actually lower the natural rate itself—improving education, reducing labor market frictions, making it easier for businesses to hire and fire.

The Expectations That Fulfill Themselves

Rational expectations also highlights a troubling possibility: self-fulfilling prophecies.

If everyone expects a recession, they might cause one. Consumers cut spending because they expect hard times. Businesses postpone investment because they expect lower demand. Banks tighten lending because they expect more defaults. Each of these actions, taken in response to pessimistic expectations, makes the pessimism come true.

Conversely, if everyone expects prosperity, they might create it. Optimistic consumers spend freely. Confident businesses expand. Banks lend generously. The expectations become reality.

This creates a strange situation where expectations aren't just predictions about the future—they're forces that shape the future. The economy isn't just a machine that people observe from outside. It's a system that people are part of, and their beliefs change how it operates.

This has uncomfortable implications for communication by central banks and other policymakers. What they say might matter as much as what they do, because their statements shape expectations. A central bank that credibly commits to fighting inflation might actually face less inflation, because people expect less inflation and behave accordingly. A central bank with a reputation for being soft on inflation might find inflation harder to control, precisely because people don't believe it will be controlled.

Criticisms and Complications

Rational expectations has never lacked critics, and some of their objections carry real weight.

The most fundamental criticism is empirical: people don't actually seem to form expectations rationally. Survey evidence consistently shows that people's forecasts are biased in predictable ways. They tend to extrapolate recent trends too far into the future. They anchor on salient numbers. They underreact to some information and overreact to other information.

Behavioral economists have documented dozens of cognitive biases that violate rational expectations. People use mental shortcuts. They suffer from overconfidence. They weight losses more heavily than equivalent gains. They discount the future inconsistently. The list goes on.

Defenders of rational expectations have responses to these criticisms. One is that individual irrationality might not matter much in aggregate. Even if each person is somewhat irrational, their errors might cancel out. The market as a whole might behave as if participants were rational, even if no individual participant is.

Another defense is that even if expectations aren't perfectly rational, models that assume rationality might still be more useful than alternatives. If you don't assume rational expectations, you need some other theory of how people form expectations. And that theory will inevitably be somewhat arbitrary. At least rational expectations provides a clear, consistent benchmark.

The Legacy

Whether or not people actually think rationally, the rational expectations revolution permanently changed economics.

Before Muth, Lucas, and Sargent, economists often built models where the government could systematically fool people. After their work, such models seemed naive. Any policy rule that's predictable enough to be modeled is predictable enough for people to anticipate. If your model assumes that people can be fooled in the same way repeatedly, you probably have a bad model.

This insight survives even if you reject the strongest claims of rational expectations. Even if people aren't perfectly rational, they're probably not perfectly foolish either. They learn. They adapt. They pay attention to things that affect their lives. Any economic policy that depends on people not catching on is probably a bad policy.

The theory also shifted attention toward credibility and expectations management. Central banks today spend enormous effort communicating their intentions, not just executing their policies. They understand that what people believe about future policy matters as much as the policy itself. This emphasis on expectations is a direct legacy of rational expectations theory.

Perhaps most importantly, rational expectations forced economists to take seriously the idea that the people they model are thinking beings who respond to the models themselves. Economics isn't like physics, where the objects of study don't read physics journals. Economic actors are aware of economic theory, at least in broad strokes. They adjust their behavior based on what they think policymakers will do. Any economic model that ignores this feedback loop is missing something essential.

Muth's original paper made this point with characteristic modesty. He wasn't claiming that people were perfectly rational calculating machines. He was simply observing that economic theories that assume systematic irrationality are odd theories indeed. Why would people persistently get things wrong in ways that are obvious and costly? At some point, wouldn't they learn?

That question, once asked, was impossible to ignore. And economics has been different ever since.