What is the fiscal multiplier?

The fiscal multiplier measures the change in GDP resulting from a one-dollar change in government spending or taxation. If the government spends $1 billion on infrastructure and GDP rises by $1.5 billion, the multiplier is 1.5. If GDP rises by only $0.5 billion, the multiplier is 0.5. The size of the multiplier determines whether fiscal stimulus is an efficient tool for boosting economic growth.

The multiplier operates through chain spending. When the government hires construction workers to build a highway, those workers receive income, which they spend on food, housing, and other goods. The businesses receiving that spending hire more workers and buy more supplies, generating additional rounds of spending. Each round is smaller than the last (because some income is saved, taxed, or spent on imports), but the cumulative effect exceeds the initial spending.

The multiplier is not constant; it depends critically on economic conditions. During recessions with idle resources and interest rates near zero, multipliers tend to be largest (estimates range from 1.5 to 2.5) because the spending employs otherwise unused labor and capital, and the central bank does not raise rates to offset the fiscal stimulus. During expansions with full employment and positive interest rates, multipliers tend to be smaller (0.5 to 1.0) because government spending crowds out private spending through higher interest rates and competition for scarce workers.

The type of fiscal action also matters. Government purchases of goods and services (direct spending) tend to have larger multipliers than tax cuts, because tax cut recipients may save a portion rather than spend it. Transfer payments to low-income households have relatively high multipliers because those recipients tend to spend a high share of additional income. The COVID stimulus debate illustrated these dynamics: direct payments to households with high propensity to spend produced rapid economic effects, while some programs aimed at higher-income recipients had smaller multiplier effects.