What the fiscal multiplier actually measures
Definition. The fiscal multiplier is the ratio of the change in GDP to the change in government spending (or tax cut) that caused it. A multiplier of 1.5 means $1 of spending produces $1.50 of GDP; a multiplier of 0.5 means it produces only 50 cents -- the rest is offset by crowding out private activity.
When a dollar of spending does more or less
| Setting | Approximate multiplier | Why |
|---|---|---|
| ZLB + slack (post-2008 era) | 1.5 to 2.0 | Central bank cannot offset; idle workers and capital absorb spending |
| Normal times, near capacity | 0.5 to 1.0 | Central bank partially offsets; some crowding out of private investment |
| Hot economy + tightening cycle | 0.3 to 0.7 | Central bank actively raises rates to offset; heavy crowding out |
| Very high debt + credibility doubts | Below 0.3, sometimes negative | Households save the stimulus expecting future tax hikes; bond markets revolt |
Why the type of spending changes the payoff
Spending type matters too. Direct government purchases of goods (infrastructure, defense procurement) typically have multipliers above tax cuts of equal size, because some of any tax cut gets saved rather than spent. Within tax cuts, those targeted at lower-income households spend more (higher marginal propensity to consume); corporate or high-income cuts spend less. Within direct spending, projects that absorb idle resources beat those that compete with existing private demand.
Why the same stimulus helps more in a slump
Judge which multiplier regime applies today
Checking a stimulus claim against the regime
Going deeper: the multiplier's geometric-series ceiling
Going deeper (optional). Up next: the textbook geometric-series ceiling that multiplier numbers come from, and why the real-world table above sits so far beneath it -- an optional aside you can skip on first pass and come back to anytime. Continue when you're curious.
Where do multiplier numbers come from in the first place? Start with the simplest chain. The government spends $1. Whoever receives it spends a fraction of it on other goods -- their marginal propensity to consume (MPC); the next recipient spends that same fraction again, and so on down the line. Summing the whole chain gives 1 + MPC + MPC^2 + MPC^3 + ..., a geometric series that converges to 1 / (1 - MPC). At an MPC of 0.8 that ceiling is 1 / (1 - 0.8) = 1 / 0.2 = 5; at an MPC of 0.6 it is 1 / (1 - 0.6) = 1 / 0.4 = 2.5. This is the frictionless upper bound -- what the multiplier would be if every dollar recirculated forever with nothing leaking out of the chain.
Now compare that ceiling to the regime table above: even its high end, 2.0, sits far below the frictionless 5. The gap is every leakage the simple chain ignores. Households save part of each dollar rather than spending the full MPC; taxes claw back another part; imports send some of the demand abroad; the central bank can raise rates to offset the boost; and heavier government borrowing can crowd out private investment. Which of these dominates is exactly what the regime table captures -- at the zero lower bound with idle resources, few leakages fire and the multiplier climbs toward the top of its real-world range; in a hot economy with the Fed tightening, monetary offset and crowding out dominate and it falls toward 0.3-0.7. The geometric series tells you the theoretical ceiling; the regime tells you how far beneath it the real multiplier lands.
Sit with the ideas.
Two governments each spend $100 billion on infrastructure. Country A's central bank is at the zero lower bound with idle resources and unemployment elevated. Country B's central bank is actively tightening to combat inflation with the economy near capacity. Which country gets a larger fiscal multiplier from the same $100B, and why?