Inflation traces back to three forces, usually acting together rather than alone: too much demand chasing a limited supply of goods, rising costs that businesses pass on to customers, and too much money circulating relative to what the economy actually produces. Economists label the first two demand-pull and cost-push inflation, and reach for the money supply, the domain of monetary policy, to explain why either kind can persist for years instead of fading in a quarter. Every major U.S. inflation episode is some mix of the three, and telling them apart matters because the Federal Reserve’s main tool, raising interest rates, works cleanly on only one of them.
Demand-pull: too much money chasing too few goods
Demand-pull inflation shows up when spending power grows faster than businesses can supply goods and services, so sellers raise prices instead of running out of stock. The clearest U.S. example has nothing to do with an oil shock or a war: it’s 1946 through 1948, when wartime price controls and rationing ended just as returning soldiers and years of forced household saving unleashed a wave of pent-up demand onto an economy still retooling from wartime production. The CPI jumped about 14% in 1947 and another 8% in 1948, among the sharpest peacetime inflation the country has ever recorded, driven almost entirely by demand outrunning supply rather than by any single cost shock. The 2021 reopening after COVID-19 lockdowns, with stimulus checks and enforced savings meeting a still-constrained supply of goods, followed a similar demand-side script before supply problems piled on top of it.
Cost-push: rising input costs passed on to buyers
Cost-push inflation starts on the supply side instead: the cost of producing or shipping goods rises across the whole economy at once, and businesses raise prices to protect their margins even as output slows. Oil is the textbook trigger, because energy costs touch nearly every product’s manufacturing and transportation. The October 1973 OPEC oil embargo roughly quadrupled crude prices within months, and the CPI rose about 11% in 1974 as the shock rippled through the economy; the 1979 Iranian Revolution repeated the pattern, pushing inflation to a peak of 13.5% in 1980. Cost-push inflation is what makes stagflation possible: rising prices and a slowing economy can occur together only when the shock originates in costs, not demand, since a genuinely overheated, demand-driven economy is a growing one.
Monetary inflation: too much money in the system
Behind both patterns sits a longer-run driver: how much money is circulating relative to the economy’s output of goods and services. Milton Friedman’s famous claim that “inflation is always and everywhere a monetary phenomenon” overstates it as a complete explanation, oil shocks and demand surges are real and distinct forces, but the underlying point holds up: sustained, multi-year inflation is hard to produce without money supply growth to fund it. The Federal Reserve controls the money supply indirectly, mainly by setting interest rates that make borrowing and spending cheaper or more expensive, which is exactly why the tool that ended the Great Inflation of the 1970s was Paul Volcker pushing the federal funds rate above 19% rather than any change to oil markets or consumer demand.
Built-in inflation: expectations and the wage-price spiral
Once demand-pull or cost-push pressure has been running for a while, a third, self-reinforcing mechanism can take over: workers and businesses start expecting inflation to continue and act accordingly. Workers negotiate raises to keep pace with expected price increases; businesses, facing higher labor costs, raise prices again to cover them, restarting the cycle even after the original demand surge or cost shock has faded. This wage-price spiral is part of why the 1970s inflation proved so hard to shake, and why the Fed now treats inflation expectations themselves as a policy target, not just the current price level.
What actually caused the 2021-2023 surge
The 2021-2023 inflation surge is a useful case study because it mixed all three forces rather than fitting cleanly into one box. Demand-pull came from stimulus spending and reopening demand after lockdowns; cost-push came from pandemic-era supply chain bottlenecks and the 2022 spike in oil and food prices following Russia’s invasion of Ukraine; built-in pressure showed up as businesses, having seen costs rise once, kept raising prices even as the original shocks eased. Annual CPI rose about 8% in 2022 (a monthly peak of 9.1% that June) before cooling toward 4% in 2023 as the Fed’s rate hikes worked through the economy. Untangling which force was doing the most damage, at any given moment, is exactly the job the Fed’s staff economists spend their careers on.
Why the distinction matters for fixing it
Raising interest rates cools demand-pull inflation directly, borrowing gets more expensive, spending slows, and prices stop climbing so fast. It does far less against a pure cost-push shock, since a higher interest rate doesn’t refill an OPEC oil field or unclog a shipping port; it can only fight a cost shock indirectly, by cooling demand enough elsewhere in the economy to offset it, at the price of higher unemployment. That’s the mechanism behind every recession the Fed has deliberately engineered to break an inflation cycle, and why economists watch not just how fast prices are rising but which of these three forces is driving it before predicting how painful the fix will be.