A.W. Phillips's original empirical finding was a stable relationship between:

Money supply and the interest rate

Stagflation — simultaneous high inflation and high unemployment — is best explained by:

Cost-push inflation, which raises prices while contracting output and employment

The original simple Phillips curve

A perfectly flexible-wage classical economy

Automatic stabilisers are valued over discretionary fiscal policy chiefly because they:

Respond to the economy with essentially zero decision lag

Always raise output permanently

Are set once and never change again

Require legislative approval before every use

The Lucas Critique argues that:

Behavioural relationships estimated on past data embed the past policy regime, so they are unreliable for predicting the effect of a genuinely new policy

Fiscal policy is always more effective than monetary policy

The money supply has no effect on prices

Adaptive expectations are always correct

Under the policy-ineffectiveness proposition, a fully anticipated, systematic monetary expansion affects:

Only the price level, not real output or employment

Both output and prices permanently

The expectations-augmented Phillips curve becomes vertical at the natural rate of unemployment once:

Expected inflation catches up to actual inflation

The government runs a permanent budget surplus

The money supply is held constant forever

Wages are fixed by the government

Which of the following are components of the policy lag that delays discretionary stabilisation policy?

Implementation/impact lag — the response working through the economy

A permanent, unavoidable rise in the price level

Inflation and the Phillips Curve, Business Cycles, Monetary-Fiscal Policy and the Rational Expectations Critique

Unit 2 closes by connecting the goods-and-money-market model to inflation, cyclical fluctuations, and policy — and to the rational expectations critique that unsettled the standard policy story in the 1970s. This chapter covers inflation and the Phillips curve, business cycles, monetary and fiscal policy, and rational expectations.

1. Inflation and Phillips Curve Analysis

A.W. Phillips's original empirical finding — a stable, negative relationship between wage inflation and unemployment in the UK — was reinterpreted as a trade-off between price inflation and unemployment, apparently offering policymakers a menu: accept a bit more inflation to buy a bit less unemployment, or vice versa. Demand-pull inflation arises when aggregate demand outruns aggregate supply at (near) full employment; cost-push inflation arises from a rise in the cost of production (wages, raw materials, energy) independent of demand conditions, shifting aggregate supply inward even as output and employment fall — a combination often called STAGFLATION, which the simple original Phillips curve cannot accommodate at all.

⚠️ The 1970s stagflation was the empirical crack, not a theoretical afterthought

The original Phillips curve implies inflation and unemployment cannot rise together. The 1970s oil-price shocks produced exactly that combination — high inflation AND high unemployment simultaneously — which the simple curve could not explain at all and which motivated Friedman and Phelps's expectations-augmented Phillips curve (see below): the trade-off exists only in the SHORT run and only for a given, fixed level of expected inflation.

2. Business Cycles

A business cycle is the recurring, though not perfectly regular, pattern of expansion (recovery and boom) and contraction (recession/depression) in aggregate economic activity, conventionally marked by four phases: expansion, peak, contraction/recession, and trough, before the next expansion begins. Real business cycle theory attributes fluctuations to real shocks (technology, productivity) propagating through an otherwise efficiently-clearing economy, while Keynesian-inspired theories attribute them to demand-side shocks and nominal rigidities that prevent quick market clearing — the multiplier-accelerator interaction (from the previous chapter) is one classic Keynesian mechanism for generating cyclical, even oscillatory, paths of income.

Leading, coincident and lagging indicators — variables such as new orders and stock prices tend to turn before the cycle (leading), industrial production and employment turn roughly with it (coincident), and average duration of unemployment and interest rates on business loans turn after it (lagging) — a standard applied-macro classification.
Automatic stabilisers — progressive taxation and unemployment benefits dampen cycles automatically, without new legislation, by pulling more tax revenue in booms and paying out more transfers in slumps.

3. Monetary and Fiscal Policy

Fiscal policy uses government spending and taxation to shift the IS curve; monetary policy uses the money supply (or a policy interest rate) to shift the LM curve. Effectiveness of each depends, as the IS-LM chapter showed, on the interest-elasticity of investment and money demand — a very interest-inelastic investment demand (steep IS) weakens monetary policy just as a very interest-inelastic money demand (steep LM) weakens fiscal policy. Policy lags — recognition lag (realising a problem exists), decision lag (agreeing on and legislating a response) and implementation/impact lag (the response actually working through the economy) — mean policy, especially fiscal policy which typically needs legislative approval, often arrives after the shock it was meant to counter has already changed character.

ℹ️ Automatic stabilisers act without any lag at all

Because progressive taxes and unemployment benefits respond mechanically to income and employment as they change, they act with essentially zero decision lag — this is precisely why they are valued as a first line of defence against a downturn, ahead of discretionary policy that must still be recognised, decided and implemented.

4. Rational Expectations Hypothesis and Its Critique

The rational expectations hypothesis (Muth, applied to macroeconomics by Lucas and Sargent) holds that economic agents form expectations using all available information EFFICIENTLY, so their forecast errors are unbiased and uncorrelated with any information known at the time the forecast was made — agents do not repeat systematic mistakes. Its sharpest policy implication is the LUCAS CRITIQUE: any econometric model estimated on past data embeds the PAST policy regime inside its estimated behavioural relationships (like the Phillips curve), so using that model to predict the effect of a NEW policy is invalid, because rational agents will change their own behaviour in response to the new, anticipated policy rule itself. A stronger companion result, the policy-ineffectiveness proposition, says that a systematic (fully anticipated) monetary policy has no real effect on output or employment, only on the price level — only genuinely UNANTICIPATED policy can move real variables, even temporarily.

Adaptive versus rational expectations, and their Phillips curve implications
Expectations formation	Implication
Adaptive (backward-looking, based on past inflation)	A short-run Phillips-curve trade-off is exploitable for a while, since expected inflation lags behind actual, but only until expectations catch up — the expectations-augmented Phillips curve becomes vertical at the natural rate of unemployment once expected inflation equals actual.
Rational (forward-looking, using the full model and known policy rule)	Even a SHORT-run trade-off vanishes for any systematic, anticipated policy — the Phillips curve is vertical even in the short run against anticipated policy changes, leaving room only for surprises to move output/unemployment.

Key takeaways

Demand-pull inflation comes from excess aggregate demand; cost-push inflation comes from rising production costs independent of demand, and their combination (stagflation) is what the original Phillips curve cannot explain.
Business cycles pass through expansion, peak, contraction and trough; automatic stabilisers (progressive tax, unemployment benefit) act with zero decision lag, unlike discretionary policy.
Fiscal policy shifts IS, monetary policy shifts LM; each is weakened by an unfavourable slope of the other curve, and policy lags (recognition, decision, implementation) delay both.
The rational expectations hypothesis says forecast errors are unbiased and uncorrelated with known information; the Lucas Critique says past-data-estimated behavioural relationships are unreliable guides to a NEW policy's effect, since agents change behaviour in response to it.
The policy-ineffectiveness proposition: only unanticipated policy can move real output/employment, even temporarily — systematic, fully-anticipated policy affects only the price level.

Practice questions (9)

Attempt each one before opening the answer. Every explanation names the tempting wrong option as well as the right one, because that is where marks are lost.

A.W. Phillips's original empirical finding was a stable relationship between:
1. Wage inflation and unemployment
2. Money supply and the interest rate
3. GDP and government spending
4. Exchange rates and exports
Show answer
Answer: A — Wage inflation and unemployment
Phillips's original UK data showed a negative relationship between wage inflation and unemployment, later reinterpreted (and extended) to price inflation and unemployment.
Stagflation — simultaneous high inflation and high unemployment — is best explained by:
1. Cost-push inflation, which raises prices while contracting output and employment
2. Demand-pull inflation alone
3. The original simple Phillips curve
4. A perfectly flexible-wage classical economy
Show answer
Answer: A — Cost-push inflation, which raises prices while contracting output and employment
A cost-push shock (e.g. an oil-price rise) shifts aggregate supply inward, raising prices while output and employment fall — exactly the combination the demand-side original Phillips curve cannot generate.
Automatic stabilisers are valued over discretionary fiscal policy chiefly because they:
1. Respond to the economy with essentially zero decision lag
2. Always raise output permanently
3. Are set once and never change again
4. Require legislative approval before every use
Show answer
Answer: A — Respond to the economy with essentially zero decision lag
Progressive taxes and unemployment benefits change automatically as income and employment change, with no need for new legislation or a decision lag — unlike discretionary policy, which must be recognised, decided and implemented.
The Lucas Critique argues that:
1. Behavioural relationships estimated on past data embed the past policy regime, so they are unreliable for predicting the effect of a genuinely new policy
2. Fiscal policy is always more effective than monetary policy
3. The money supply has no effect on prices
4. Adaptive expectations are always correct
Show answer
Answer: A — Behavioural relationships estimated on past data embed the past policy regime, so they are unreliable for predicting the effect of a genuinely new policy
Because rational agents adjust their behaviour to a new, anticipated policy rule, an econometric relationship estimated under the OLD policy regime cannot be used unchanged to forecast the effect of the new one.
Under the policy-ineffectiveness proposition, a fully anticipated, systematic monetary expansion affects:
1. Only the price level, not real output or employment
2. Only real output, not prices
3. Both output and prices permanently
4. Neither output nor prices
Show answer
Answer: A — Only the price level, not real output or employment
Because rational agents anticipate a fully systematic policy and adjust prices/wages accordingly in advance, only the price level moves; only a genuinely UNANTICIPATED policy surprise can move real output or employment, even then only temporarily.
The expectations-augmented Phillips curve becomes vertical at the natural rate of unemployment once:
1. Expected inflation catches up to actual inflation
2. The government runs a permanent budget surplus
3. The money supply is held constant forever
4. Wages are fixed by the government
Show answer
Answer: A — Expected inflation catches up to actual inflation
With adaptive expectations, the short-run trade-off exists only while expected inflation lags actual; once expectations fully adjust (expected = actual), unemployment returns to its natural rate at whatever inflation rate prevails, tracing a vertical long-run Phillips curve.
In the standard four-phase description of a business cycle, the phase immediately after the peak is:
1. Contraction/recession
2. Expansion
3. Trough
4. Recovery
Show answer
Answer: A — Contraction/recession
The standard sequence is expansion, peak, contraction/recession, trough, then a new expansion begins — the peak marks the turning point from expansion into contraction.
Which of the following are components of the policy lag that delays discretionary stabilisation policy?
1. Recognition lag — realising a problem exists
2. Decision lag — agreeing on and legislating a response
3. Implementation/impact lag — the response working through the economy
4. A permanent, unavoidable rise in the price level
Show answer
Answer: A — Recognition lag — realising a problem exists; B — Decision lag — agreeing on and legislating a response; C — Implementation/impact lag — the response working through the economy
The first three are the standard three stages of policy lag; a permanent rise in the price level is not a lag at all but a possible (and not universal) consequence of policy action.
Name the hypothesis, associated with Muth, Lucas and Sargent, holding that agents use all available information efficiently so their forecast errors are unbiased and uncorrelated with known information.
Numerical answer — type the value.
Show answer
Answer: rational expectations hypothesis
The rational expectations hypothesis is the standard name; it underlies both the Lucas Critique and the policy-ineffectiveness proposition covered in this chapter.