Expectations and the great debates

On the expectations-augmented Phillips curve, π = π^e − β(u − u_n), policy can push unemployment below the natural rate only by making inflation higher than people expected. So the question “can policy move output?” becomes “can policy surprise people, and for how long?”. The answer depends entirely on how expectations are formed.

People who forecast inflation from this year’s and recent years’ inflation are using adaptive expectations (Hyderabad 2017): they correct their forecast by part of last period’s error.

This is the monetarist world of Friedman: policy can lower unemployment for a while, because expectations lag behind, but the gain disappears as they catch up, leaving only higher inflation. The weakness is that people make the same mistake over and over: if inflation keeps rising, adaptive forecasts are always too low.

John Muth (1961) proposed that people use all the information they have, including how the economy and policy work, so their forecasts are not systematically wrong (Hyderabad 2010). Errors happen, but they are random: the difference between the actual and expected price is a random error (UGC NET 2025). The theory is Muth’s, not Hansen’s (UGC NET 2025); Robert Lucas and Thomas Sargent built macroeconomics on it in the 1970s.

Put rational expectations together with flexible wages and prices that clear markets, and you get the new classical school: Lucas as its main figure, rational expectations, complete wage and price flexibility (UGC NET 2025). Sargent and Wallace (1975) drew the striking conclusion, the policy ineffectiveness proposition: a systematic monetary or fiscal policy, which people can anticipate, does not affect output or employment even in the short run (GATE 2025). Only surprises do, and a policy cannot surprise people systematically.

Its assumptions are rational expectations, full information and output at its natural level; “policy changes are not anticipated” and “nominal variables influence decisions” are not among them (UGC NET 2025). In a new classical world without surprises, all unemployment is voluntary.

The Lucas critique (1976). The coefficients of an econometric model, such as the slope of a Phillips curve, reflect how people expect policy to behave. Change the policy and they change their behaviour, so the old coefficients no longer hold. Predicting a new policy’s effects needs parameters that are stable under policy changes, meaning tastes and technology, not estimated reduced forms (UGC NET 2025).

Real business cycle theory (Kydland and Prescott, 1982) went further: fluctuations are not failures at all, but the efficient response of maximising households and firms to real shocks, chiefly to technology (GATE 2025). Money plays little part; without a supply shock there is only voluntary unemployment (UGC NET 2024).

New Keynesians (from the 1980s: Mankiw, Akerlof and Yellen, Blanchard and others) accepted rational expectations but gave microeconomic reasons for sticky wages and prices: menu costs, staggered contracts, efficiency wages, and imperfect competition (UGC NET 2023). With prices that don’t clear markets quickly, anticipated policy can matter again, and coordination failures can leave a competitive economy with too little demand even under rational expectations (UGC NET 2024).

So the chronological order is classical, Keynesian, monetarist, new classical, New Keynesian (UGC NET 2023). UGC NET 2023 also matched names to ideas: Keynes with liquidity preference, Kydland–Prescott with the time-inconsistency problem (their 1977 paper; the business-cycle paper came in 1982), Baumol–Tobin with the inventory approach, Friedman–Schwartz with the monetary history of the United States. And UGC NET 2024 ordered Smith’s growth theory (1776), the General Theory (1936), Solow’s growth model (1956), the Phillips curve (1958) and Muth’s rational expectations (1961).

A business cycle is the rise and fall of output around its trend: expansion, peak, recession (when output declines; CUET 2021), trough. Investment is the most volatile component of GDP over the cycle (Hyderabad 2015). Before Keynes and after him, economists proposed many explanations, which UGC NET likes to match to their authors:

• Pure monetary theory: Hawtrey. Monetary over-investment: Hayek. Innovation theory: Schumpeter, not Jorgenson. Psychological theory: Pigou. Cobweb theorem: associated with Schultz, among others (UGC NET 2025, 2026).
• Multiplier–accelerator interaction: Samuelson (1939), later Hicks (1950) and Goodwin (1951). In order: cobweb, Samuelson, Hicks, Goodwin (UGC NET 2026). The interaction is also called the super multiplier (CUET 2019).

Samuelson’s model, from the investment lesson of the foundations module, shows how the multiplier and the accelerator together can turn a single rise in spending into cycles. Try c = 0.5 and v = 1 for damped cycles, and c = 0.6, v = 2 for explosive ones.