Tax Multiplier Calculator
Compute the Keynesian tax multiplier from the marginal propensity to consume (MPC), measuring how a tax change affects total economic output. The result is negative because tax increases reduce output.
Last updated: May 2026
Compare with similar
About this calculator
The formula is Tax Multiplier = −MPC / (1 − MPC), where MPC (marginal propensity to consume) is the fraction of an additional dollar of disposable income that households spend rather than save, with MPC ∈ (0, 1). The negative sign reflects that a tax increase reduces aggregate demand: every $1 of additional taxes removes $MPC from initial consumption (households save 1−MPC of their disposable income), and that $MPC reduction in spending then triggers further reductions through the spending chain, multiplied geometrically by 1/(1−MPC). The total multiplier on output is −MPC/(1−MPC). Compare to the government-spending multiplier 1/(1−MPC), which is positive and larger in magnitude — because government spending injects the full dollar into the economy directly, while a tax change only affects spending indirectly through the MPC. Edge cases: MPC = 0 gives multiplier 0 (no effect, since households save 100% of disposable income); MPC = 1 gives a division by zero (multiplier diverges to −∞), but in practice MPC is bounded below 1 by some savings rate. MPS = 1 − MPC is the marginal propensity to save. Empirical estimates of US MPC: roughly 0.6–0.7 for permanent income changes, 0.2–0.4 for temporary tax rebates (households save more of one-off windfalls). Real-world multipliers are smaller than simple Keynesian theory predicts because of monetary-policy offset, leakage to imports (open-economy MPC), and supply-side responses. Modern macroeconomics views multipliers as state-dependent: larger in deep recessions, smaller near full employment.
How to use
Example 1 — typical MPC. MPC = 0.8 (households spend 80% of additional disposable income). Step 1: 1 − MPC = 1 − 0.8 = 0.2. Step 2: tax multiplier = −0.8 / 0.2 = −4. Verify: a $1 tax increase reduces total output by $4 in the simple Keynesian model — the initial $0.80 spending reduction triggers a cascade ($0.80 × 0.8 = $0.64 next round, then $0.51, etc., summing to $4 over infinite rounds). Compare to government-spending multiplier: 1/0.2 = +5, larger in magnitude than the tax multiplier ✓. The difference (5 vs 4) is the additional $1 that government spends directly versus the $0.80 that taxpayers would have spent. Example 2 — lower MPC. MPC = 0.5 (households spend half, save half). Step 1: 1 − MPC = 0.5. Step 2: tax multiplier = −0.5 / 0.5 = −1. Verify: a $1 tax increase reduces output by $1 in this case — modest multiplier reflecting low spending propensity ✓. This case might represent a high-income household receiving a windfall who saves much of it, or an economy with strong wealth effects where additional income is more likely to be saved. The lower MPC also implies a smaller government-spending multiplier: 1/0.5 = +2. Higher MPC means larger multipliers in both directions — fiscal policy is more powerful in MPC-high economies but also produces larger contractions from tax hikes.
Frequently asked questions
Why is the tax multiplier smaller in magnitude than the government-spending multiplier?
Government spending of $1 increases aggregate demand by the full $1 directly, then triggers the same MPC-based spending cascade as tax cuts do — giving a multiplier of 1/(1−MPC). A tax cut of $1 puts $1 of disposable income in households' hands, but only MPC of that is spent (the rest is saved), so the initial demand impulse is $MPC, not $1. The subsequent cascade is the same in both cases, but starting from a smaller initial impulse. The ratio: tax multiplier / spending multiplier = −MPC / 1 = −MPC. So for MPC = 0.8, the tax multiplier (−4) is 80% the magnitude of the spending multiplier (+5). This is the basis for the Keynesian argument that government spending is more 'efficient' fiscal stimulus per dollar than tax cuts. In practice, the difference is small for high MPC and large for low MPC, and is qualified by many real-world complications (timing of spending, distributional effects, behavioural responses) that the simple model ignores.
How do real-world multipliers compare to the simple Keynesian formula?
Real multipliers are typically smaller than the simple formula predicts due to several leakages and offsets not in the basic model. Leakage to imports: households spending their disposable income on foreign-produced goods reduces the multiplier proportionally — open-economy multipliers can be 20–50% smaller than closed-economy ones. Monetary-policy offset: central banks may raise rates to offset fiscal stimulus, undoing demand effects (the Fed under Volcker in the early 1980s explicitly offset fiscal stimulus). Crowding out: government borrowing can raise interest rates and reduce private investment, partially offsetting the stimulus. Supply constraints: at full employment, additional demand drives inflation rather than output, making the real multiplier near zero. Permanent vs temporary: households spend less of temporary tax cuts (like one-off stimulus checks) than permanent tax changes. Empirical estimates of US fiscal multipliers range from 0.2 (small, mostly inflation effects) to 2.5 (large, deep recession with monetary accommodation). The simple formula gives an upper bound assuming all conditions favour fiscal effectiveness.
Why does the marginal propensity to consume matter so much?
MPC governs how much of every additional dollar of income gets re-spent into the economy versus saved, and the geometric multiplier compounds this many times. At MPC = 0.5, the spending multiplier is 2; at MPC = 0.8, it's 5; at MPC = 0.9, it's 10. Small changes in MPC produce large changes in multiplier estimates, which is why estimating MPC accurately matters enormously for fiscal policy. MPC varies substantially across households: low-income and credit-constrained households have MPC close to 1 (they spend essentially everything they get); high-income households have MPC closer to 0.3 (they save most of additional income); MPC is also higher for temporary income (windfalls) than permanent, and for income shocks announced as permanent versus suspected to be temporary. This is why targeted stimulus to lower-income households (Earned Income Tax Credit, food stamps) typically has higher multipliers than across-the-board tax cuts that disproportionately benefit higher-income households who save more of the proceeds. Survey-based MPC estimates and randomised experiments with cash transfers help economists calibrate.
What are the common mistakes when interpreting the tax multiplier?
The biggest mistake is treating the simple Keynesian multiplier as reliable in all conditions — it's an upper bound for the most favourable case (closed economy, deep recession, monetary accommodation). Real multipliers vary 5–10× across conditions. The second is forgetting that MPC differs by income group; aggregate MPC values mask substantial heterogeneity. The third is applying the multiplier to the wrong tax base — payroll taxes have very different multipliers than capital-gains taxes, because the affected populations have very different MPCs. People also confuse the tax multiplier with the elasticity of tax revenue to GDP, or with the Laffer-curve-style argument that lower taxes can raise revenue (which is a supply-side argument, not a demand-side multiplier). The negative sign is also commonly forgotten: tax increases reduce output (negative multiplier), tax cuts boost output (negative multiplier × negative tax change = positive output effect). For policy debates, recognise that the simple multiplier is a starting point; estimated multipliers depend on empirical evidence and conditions, not just the formula.
When should I not use this calculator?
Do not use it for actual macroeconomic policy analysis — modern DSGE (dynamic stochastic general equilibrium) models and SVAR (structural vector autoregression) empirical models supersede simple Keynesian multipliers and produce state-dependent estimates ranging from 0.2 to 2.5 for US tax multipliers. The simple formula assumes a closed economy without monetary-policy offset; open economies with active central banks see much smaller real multipliers. It is not appropriate for permanent vs temporary tax changes — permanent tax changes have larger consumption responses than temporary stimulus checks, and the formula doesn't distinguish. Do not use it for distributional analysis — different income groups have very different MPCs, and a single aggregate MPC obscures distributional effects of tax policy. The formula gives undefined results for MPC = 1 (theoretical ceiling, never observed in practice). For fiscal-policy decisions, use CBO or institutional macroeconomic models that incorporate monetary policy reaction, supply constraints, distributional effects, and timing — not a single back-of-envelope multiplier. Finally, the simple Keynesian framework is contested; real-business-cycle and supply-side economists argue multipliers are much smaller than Keynesians estimate.