Introduction: Two Diagrams, One Price Level
Inflation is a sustained rise in the general price level. That definition is simple. Identifying which curve moved is not — and getting it wrong leads directly to the wrong policy.
If prices rise because buyers are chasing too few goods, the cure is to cool demand: raise interest rates, tighten fiscal policy. If prices rise because production has become more expensive, raising interest rates will suppress output and employment without addressing the cause. The same policy is either the remedy or the poison depending on a diagnosis that must be made in real time, with incomplete data.
This is not an academic distinction. It is the single most consequential judgement a central bank makes.
1. Demand-Pull Inflation
Demand-Pull Inflation — a rise in the general price level caused by an outward shift of aggregate demand, with aggregate supply unchanged. Sometimes described as “too much money chasing too few goods.”
Recall AD = C + I + G + (X − M). Anything that raises a component shifts AD right.
The mechanism
AD shifts right along an upward-sloping short-run aggregate supply curve. Two things happen simultaneously:
- Real output rises (firms produce more).
- The price level rises (firms bid for scarce inputs; workers bid up wages).
In demand-pull inflation, prices and output move in the same direction. Inflation rises and unemployment falls. This is the co-movement the original Phillips curve described. It is the only circumstance under which a policymaker faces a genuine inflation–unemployment trade-off, and it is why cooling demand works: you give up output, and you buy back price stability.
Causes
- Expansionary monetary policy — lower interest rates, quantitative easing, credit growth
- Expansionary fiscal policy — tax cuts, spending increases, direct transfers to households
- A consumption boom driven by wealth effects (rising house or equity prices)
- Export surges from strong foreign demand or currency depreciation
- Rising inflation expectations, which are self-fulfilling
The classical view: the Quantity Theory
Milton Friedman’s famous claim — that inflation is always and everywhere a monetary phenomenon — follows if you assume V is stable and Y is fixed at full employment. Then M and P move one-for-one.
Between 2008 and 2020, central banks expanded the monetary base enormously through quantitative easing. Inflation persistently undershot target. Japan expanded its balance sheet beyond the size of its economy and could not escape deflation. Velocity collapsed in almost exact proportion to the base expansion. MV = PY is an identity — it cannot be false. But as a theory of inflation it requires V to be stable, and V is not.
2. Cost-Push Inflation
Cost-Push Inflation — a rise in the general price level caused by an inward (leftward) shift of short-run aggregate supply, arising from an increase in the cost of production at every level of output.
The mechanism
SRAS shifts left. Now:
- The price level rises.
- Real output falls.
In cost-push inflation, prices and output move in opposite directions. Inflation rises and unemployment rises.
This is why cost-push inflation is a policy nightmare. There is no trade-off to exploit — both objectives deteriorate together. A central bank that raises rates deepens the recession. A central bank that does nothing risks unanchored expectations. This dilemma has a name.
Causes
- Wage-push — nominal wages rising faster than productivity
- Import-price push — currency depreciation raising the domestic cost of imported inputs
- Commodity shocks — oil, gas, food, industrial metals
- Supply chain disruption — the dominant channel in 2021–22
- Higher indirect taxes — VAT rises, carbon taxes, tariffs
- Profit-push — firms with market power widening margins under cover of a general price rise
3. Stagflation
Stagflation — the simultaneous occurrence of stagnant or falling output, rising unemployment, and high inflation. It is the direct macroeconomic consequence of a large, persistent negative supply shock.
The shock: Following the Yom Kippur War, OPEC imposed an embargo and cut production. Crude oil prices roughly quadrupled within months.
The transmission: Oil is not merely a consumer good. It is an intermediate input to virtually all production — transport, plastics, fertiliser, electricity. A quadrupling of its price shifts SRAS sharply left across the entire economy simultaneously.
The outcome: Advanced economies experienced double-digit inflation alongside deep recession and sharply rising unemployment — a combination the prevailing Keynesian models of the day treated as essentially impossible.
The intellectual consequence: Stagflation destroyed the naive Phillips curve as a policy menu, and vindicated the Friedman–Phelps critique in the eyes of most of the profession. It ushered in rational expectations, the natural rate hypothesis, and eventually central bank independence.
Friedman, M. (1968), “The Role of Monetary Policy,” American Economic Review 58(1); Phelps, E. (1967), Economica 34(135)
The argument: Independently, Friedman and Phelps argued that the Phillips curve trade-off exists only when inflation is unanticipated. Workers care about real wages. If they come to expect 5% inflation, they demand 5% higher nominal wages, and the trade-off vanishes.
The formalisation — the expectations-augmented Phillips curve:
π = πe − β(u − un) + shock
Unemployment can be held below the natural rate un only by continually surprising people with ever-accelerating inflation. In the long run, the Phillips curve is vertical at un.
Why it matters: This was published before the stagflation it predicted. It is one of the few genuinely successful out-of-sample predictions in the history of macroeconomics, and it is the reason both authors received the Nobel Prize. It also explains why modern central banks are obsessed with anchoring inflation expectations — the term πe is the one they can actually influence.
4. The Wage-Price Spiral
Cost-push inflation becomes self-sustaining when it triggers a feedback loop:
- A supply shock raises prices.
- Real wages fall. Workers bargain for higher nominal wages to restore them.
- Firms face higher labour costs and raise prices to protect margins.
- Real wages fall again. Return to step 2.
A spiral cannot continue indefinitely without monetary accommodation. If the money supply is held fixed, higher prices reduce real balances, demand contracts, and the spiral chokes off — in a recession. A wage-price spiral is therefore not simply a labour-market phenomenon. It is a statement about what the central bank is expected to do. This is the deepest reason central bank credibility matters: an anchored expectation of 2% inflation makes step 2 not worth taking.
5. Case Study: The 2021–2023 Inflation — Which Was It?
The demand-side story: Unprecedented fiscal transfers went directly to households — stimulus cheques, furlough schemes, expanded unemployment benefits. Households accumulated large excess savings during lockdown. When economies reopened, that saving met a supply side that had not yet recovered. Demand for goods, in particular, surged as consumption rotated away from services.
The supply-side story: Global supply chains fractured. Semiconductor shortages halted car production. Container shipping rates rose by an order of magnitude. Labour force participation fell. Then Russia’s invasion of Ukraine in February 2022 delivered a classic commodity shock to European energy and global grain markets.
The critical distinction from QE: The 2008–2020 quantitative easing created central bank reserves, which sat on commercial bank balance sheets. The 2020–21 fiscal response created household deposits, which people spent. Reserves are not purchasing power. Deposits are. This is why a decade of QE produced no inflation and eighteen months of fiscal transfers produced a great deal.
Bernanke, B., & Blanchard, O. (2023), “What Caused the U.S. Pandemic-Era Inflation?”, Brookings/NBER
The method: A structural model of wage and price determination, estimated on US data, decomposing inflation into contributions from labour market tightness, energy and food price shocks, supply chain disruption, and inflation expectations.
The finding: The initial surge was overwhelmingly driven by supply shocks and sectoral demand shifts — not by an overheating labour market. But as those shocks faded, labour market tightness became the dominant contributor to the persistence of inflation. The composition changed over time.
Complementary evidence: Shapiro (2022), working at the Federal Reserve Bank of San Francisco, decomposed PCE inflation into supply-driven and demand-driven components at the category level, by examining whether price and quantity moved in the same or opposite directions. Both components were substantial; neither story alone was sufficient.
Why it matters: The policy implication is subtle. A pure supply shock argues for looking through the inflation. A demand shock argues for tightening. When the shock is supply-driven but persistent enough to unanchor expectations, the central bank must tighten anyway — not to fix the supply problem, which it cannot, but to protect πe. This is precisely what happened.
The “greedflation” debate
A prominent argument held that corporate profit margins expanded during the episode, and that firms exploited a general inflationary environment as cover for price increases exceeding cost increases.
The economics is contested and worth stating carefully. Rising margins during a demand surge are exactly what a competitive model predicts — price rises above marginal cost when demand exceeds capacity, and that is the signal that induces expansion. Margins rising is not by itself evidence of market power. What would be evidence is margins remaining elevated after supply constraints ease, or margin expansion concentrated in the most concentrated industries. Evidence on both points exists but is not decisive.
6. Comparison Table
| Demand-Pull | Cost-Push | |
|---|---|---|
| Curve that shifts | AD shifts right | SRAS shifts left |
| Price level | Rises | Rises |
| Real output | Rises | Falls |
| Unemployment | Falls | Rises |
| Phillips curve | Movement along it | Curve shifts outward |
| Trade-off exists? | Yes | No — both objectives worsen |
| Policy response | Tighten monetary/fiscal policy | Supply-side reform; protect expectations |
| Classic example | Late-1960s US Vietnam-era spending | 1973 and 1979 oil shocks |
Q: Inflation is 8% and unemployment has just risen from 4% to 6%. Demand-pull or cost-push?
A: Cost-push. Prices up, output down. If it were demand-pull, unemployment would be falling. The co-movement of price and quantity is the diagnostic — and it is exactly the logic Shapiro’s Fed decomposition applies at the individual product category level.
7. Exam Technique
- Draw AD/AS correctly. Demand-pull: AD right, P↑ Y↑. Cost-push: SRAS left, P↑ Y↓.
- Label the new equilibrium and mark the change in the output gap.
- Know that a negative supply shock creates a recessionary gap with inflation — the policy dilemma.
- Be able to connect to the short-run Phillips curve: a supply shock shifts the SRPC; a demand shock moves along it.
“Evaluate the effectiveness of monetary policy in controlling inflation.”
- Distinguish demand-pull from cost-push with two accurate diagrams. Analysis.
- Argue that monetary tightening addresses demand-pull directly, by reducing C and I.
- Argue that against cost-push, monetary policy operates only on expectations, not on the cause — it cannot produce semiconductors or oil.
- Introduce the expectations-augmented Phillips curve and the natural rate. This is the single highest-value theoretical addition available.
- Evaluate with the 2021–2023 episode: Bernanke and Blanchard showing composition shifted from supply to labour tightness over time.
- Consider time lags (monetary policy acts with a lag of 12–24 months), the zero lower bound, and central bank credibility.
- Conclude conditionally: effectiveness depends on the source of the inflation and on whether expectations remain anchored.
- Describing inflation as “an increase in all prices.” It is a rise in the general price level. Relative prices change constantly.
- Confusing disinflation (falling rate of inflation) with deflation (falling price level).
- Treating a one-off price rise (a VAT increase) as inflation. Inflation is sustained; a level shift drops out of the annual rate after twelve months.
- Claiming QE caused the 2022 inflation without addressing why a decade of QE caused none. The reserves-versus-deposits distinction is essential.
- Drawing a supply shock as a shift in AD.
Summary
Demand-pull and cost-push inflation produce the same rise in the price level and opposite movements in output. That single asymmetry is the whole diagnostic, and it determines whether the correct policy response is to cool the economy or to protect expectations while the shock passes.
Real inflationary episodes are rarely pure. The 2021–2023 surge began as a supply shock, became a demand story as excess savings were spent, and ended as a question about labour market tightness — with the composition shifting under policymakers’ feet in real time. The Friedman–Phelps insight is what allowed central banks to navigate it: whatever the source, if πe becomes unanchored, the cost of restoring price stability rises steeply.
Exercise 1 — Should a Central Bank Ever Tighten Against a Supply Shock?
Standard theory says a central bank should “look through” a supply shock: raising rates cannot produce oil, and doing so merely converts a price problem into a price-and-unemployment problem. Yet in 2022, every major central bank tightened aggressively in the face of an energy shock.
Construct the argument for why tightening was correct. Your answer must explain what the tightening was targeting, given that it could not affect energy supply. Then construct the counter-argument, and identify what observable variable would have told policymakers, in real time, which course was right. Was that variable available to them?
📄 Read: Bernanke, B. S., & Blanchard, O. J. (2023). “What Caused the U.S. Pandemic-Era Inflation?” Brookings Institution / NBER Working Paper. Attend especially to their decomposition of the contribution of inflation expectations versus labour market tightness across sub-periods.
Exercise 2 — Is “Greedflation” an Economic Category?
Corporate profit margins expanded during the 2021–22 inflation. Critics called this profiteering. Defenders noted that rising margins during excess demand are exactly what any competitive model predicts, and are the signal that induces capacity expansion.
Design an empirical test that would distinguish “margins rising because demand exceeds capacity” from “margins rising because firms exercised market power under cover of general inflation.” What data would you need on industry concentration, cost pass-through, and the timing of margin normalisation? Is the distinction even well-defined, given that market power is itself what allows a firm to have a margin at all?
📄 Read: Shapiro, A. H. (2022). “Decomposing Supply and Demand Driven Inflation.” Federal Reserve Bank of San Francisco Working Paper 2022-18. Consider whether his price–quantity co-movement identification strategy could be adapted to test the margin question.
References
- Bernanke, B. S., & Blanchard, O. J. (2023). What Caused the U.S. Pandemic-Era Inflation? Brookings Institution / NBER Working Paper.
- Blanchard, O. J., & Summers, L. H. (1986). Hysteresis and the European Unemployment Problem. NBER Macroeconomics Annual, 1, 15–78.
- Friedman, M. (1968). The Role of Monetary Policy. American Economic Review, 58(1), 1–17.
- Phelps, E. S. (1967). Phillips Curves, Expectations of Inflation and Optimal Unemployment over Time. Economica, 34(135), 254–281.
- Phillips, A. W. (1958). The Relation Between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1861–1957. Economica, 25(100), 283–299.
- Shapiro, A. H. (2022). Decomposing Supply and Demand Driven Inflation. Federal Reserve Bank of San Francisco Working Paper 2022-18.
- Blinder, A. S., & Rudd, J. B. (2013). The Supply-Shock Explanation of the Great Stagflation Revisited. In Bordo & Orphanides (eds.), The Great Inflation. University of Chicago Press.