Defining the two inflations
Inflation is a sustained rise in the general price level of an economy, measured by the percentage change in a consumer price index. Economists separate it by source. Demand-pull inflation comes from the demand side: a rise in aggregate demand, where AD is C plus I plus G plus net exports, shifts the AD curve to the right along an upward-sloping short-run aggregate supply curve, pulling the price level up. Cost-push inflation comes from the supply side: a rise in the costs of production shifts the short-run aggregate supply curve to the left, pushing the price level up. In the IB model the long-run aggregate supply curve is vertical at potential output, so these are short-run movements around that capacity line. The claim to be evaluated is that cost-push inflation is the more damaging of the two. This essay argues that cost-push inflation usually is more damaging, because it raises prices while output and employment fall, but that the verdict is conditional on the size, persistence and cause of the inflation rather than fixed.
Why cost-push inflation tends to be more damaging
The core of the case is the effect on output. When costs rise, through higher energy and raw-material prices, faster wage growth, higher indirect taxes or a currency depreciation that raises import prices, the SRAS curve shifts left. With AD unchanged, the new equilibrium sits at a higher price level and a lower level of real output. The economy therefore suffers rising prices and falling output at the same time, the condition known as stagflation, which carries the costs of inflation and the costs of a recession together: eroded purchasing power, rising unemployment as firms cut back, and squeezed real incomes. The defining example is the oil shocks of the 1970s. When OPEC sharply raised crude prices in 1973 and again in 1979, production and transport costs rose across the United States, the United Kingdom and Western Europe, and those economies experienced high inflation and rising unemployment together for years. The policy trap is what makes this so damaging. To fight the inflation a central bank would raise interest rates, but that cuts demand on top of an already falling output, deepening the slump; to protect output it would loosen policy, but that risks entrenching the inflation. Demand-management tools, which work on AD, cannot address a problem whose source is the cost side.
Why demand-pull inflation is comparatively manageable
Demand-pull inflation arrives in a more benign form. Because it is driven by a rightward shift of AD, it appears when the economy is growing, with output and employment rising as prices rise. It is the inflation of a boom rather than a slump, so the human cost in lost jobs is far smaller in the short run. Just as important, the cause and the cure line up. Since the problem is excess demand, a central bank can raise interest rates, or a government can tighten fiscal policy, to shift AD back left and close the inflationary gap directly. The post-pandemic period of 2021 to 2022 shows both faces. Part of that inflation was demand-pull, as large fiscal stimulus and the release of pent-up spending pushed AD higher; that component responded to the tightening cycles of the Federal Reserve, the European Central Bank and the Bank of England. The part that was cost-push, the energy and food price spike after the invasion of Ukraine, was far harder to shift with interest rates and had to ease partly on its own as energy markets settled. The contrast within a single episode underlines the point: the demand-driven component was tractable, the cost-driven component was not.
Evaluation: when the ranking holds and when it does not
The claim should not be accepted as an absolute. Several conditions decide it. Magnitude matters most: a mild demand-pull inflation around a 2 percent target is benign or even useful, whereas a severe demand-pull episode that runs away into high or hyperinflation, as in Weimar Germany or more recently Argentina and Turkiye, is far more destructive than a mild cost shock. So at the extremes demand-pull can clearly be worse. Persistence matters too: a one-off cost shock that fades, with inflation expectations well anchored by a credible central bank, does limited lasting harm, while a cost shock that feeds into a wage-price spiral becomes embedded and much more damaging. The availability of supply-side responses also weighs on the verdict, because a government that can diversify energy supply or ease bottlenecks can shift SRAS back to the right and treat the actual source, something no demand tool can do. Finally, the two are not always cleanly separable, since a demand boom can raise costs and a cost shock can shift expectations, so real episodes are often a blend.
Judgement
On balance, cost-push inflation tends to be the more damaging of the two, because it raises the price level while output and employment fall, and because it disarms the demand-management tools that work cleanly against demand-pull inflation. The 1970s stagflation and the stubborn energy-driven component of the 2021 to 2022 surge both show why. But this is a tendency, not a law. The honest judgement is that cost-push inflation is usually more damaging at comparable rates, yet the ranking depends on the size and persistence of the shock, on whether inflation expectations stay anchored, and on the policy options available, so a severe demand-pull inflation can be worse than a mild cost-push one. The right policy response, and the right assessment of harm, follows the source of the inflation rather than a blanket ranking of the two.