Common Errors in Fiscal Multiplier Calculations

Why Fiscal Multiplier Calculations Go Wrong

Fiscal multiplier problems often look like a single substitution into a formula. The difficult part is usually deciding which change the formula describes, assigning the policy change its correct sign, and using the marginal propensities that match the model. A correct multiplier paired with a misread tax change can reverse the conclusion; a correct sign paired with the wrong marginal propensity can produce a plausible but incorrect number.

This tutorial focuses on diagnosing those calculation errors. It uses the basic short-run expenditure model in which prices are fixed, output adjusts to planned expenditure, and induced consumption and imports determine how much of each additional round of income is spent domestically. The results are model calculations, not guarantees about the size of a real-world policy response.

Model and notation: Let $Y$ denote equilibrium real output, $G$ government purchases, $T_0$ lump-sum taxes, $c$ the marginal propensity to consume out of disposable income, $t$ the proportional income-tax rate, and $m$ the marginal propensity to import. Assume $0\leq c<1$, $0\leq t\leq 1$, and $m\geq 0$. Treat prices, interest rates, and other autonomous expenditure as fixed, and assume that the induced-spending process converges. Define $D=1-c(1-t)+m>0$. A change in $T_0$ is a change in lump-sum taxes; it is not a change in the tax rate $t$.

With consumption $C=C_0+c(Y-T_0-tY)$ and imports $M=M_0+mY$, equilibrium requires planned expenditure to equal output. Holding autonomous terms other than the policy change fixed, the induced part of expenditure changes by $c(1-t)\Delta Y-m\Delta Y$. Rearranging the equilibrium change gives:

\Delta Y=\frac{1}{D}\Delta G-\frac{c}{D}\Delta T_0, \qquad D=1-c(1-t)+m.

The spending multiplier is $1/D$, and the lump-sum tax multiplier is $-c/D$. The minus sign is part of the tax multiplier: an increase in taxes reduces equilibrium output, all else equal. For a tax cut, $\Delta T_0<0$, so multiplying by the negative tax multiplier produces a positive output change.

A Short Derivation: Where the Tax Multiplier Comes From

The sign and the factor $c$ in the tax multiplier follow from how a lump-sum tax change affects consumption. With $G$ and the other autonomous terms fixed, the change in equilibrium output satisfies:

\Delta Y=c\big((1-t)\Delta Y-\Delta T_0\big)-m\Delta Y.

The term $-c\Delta T_0$ is the initial change in consumption caused by the tax change. The remaining terms describe how consumption and imports respond as output changes. Collecting the $\Delta Y$ terms gives:

\big[1-c(1-t)+m\big]\Delta Y=-c\Delta T_0.

Since $D=1-c(1-t)+m>0$, divide both sides by $D$:

\Delta Y=-\frac{c}{D}\Delta T_0.

This derivation explains two frequent traps. First, the tax multiplier is negative because a tax increase lowers disposable income and consumption. Second, the initial consumption response is $c$ times the tax change, not the full tax change: households do not spend every additional dollar of disposable income when $c<1$.

Worked Examples: Identify the Change Before Calculating

Worked Example: A Lump-Sum Tax Cut and the Sign of the Multiplier

In a fixed-price expenditure model with no proportional income tax and no import leakage, the marginal propensity to consume is $0.75$. The government cuts lump-sum taxes by $\$12$ billion. Find the change in equilibrium output. State the method and identify the sign error to avoid.

Method. With $t=0$ and $m=0$, $D=1-c=1-0.75=0.25$. The lump-sum tax multiplier is:

k_T=-\frac{c}{D}=-\frac{0.75}{0.25}=-3.

The policy change is a tax cut, so $\Delta T_0=-\$12$ billion. Therefore:

\Delta Y=k_T\Delta T_0=(-3)(-\$12\text{ billion})=+\$36\text{ billion}.

The two negative signs give a positive output change. As a check, the tax cut initially raises consumption by $0.75(\$12\text{ billion})=\$9$ billion. Each subsequent round produces $0.75$ of the previous round, so total output change is $\$9/(1-0.75)=\$36$ billion.

Common mistake avoided: Treating “cut by $\$12$ billion” as $\Delta T_0=+\$12$ billion. That would describe a tax increase and incorrectly give $-\$36$ billion.

Worked Example: Applying Tax and Import Leakages

Suppose $c=0.80$, the proportional tax rate is $t=0.25$, and the marginal propensity to import is $m=0.10$. A policy increases lump-sum taxes by $\$5$ billion. Find the change in equilibrium output. Assume all other autonomous expenditure is unchanged.

Method. First calculate the induced-spending denominator. The after-tax income share is $1-t=0.75$, so induced consumption per unit of output is $c(1-t)=0.80(0.75)=0.60$. Imports add another leakage of $0.10$:

$$ D=1-c(1-t)+m=1-0.60+0.10=0.50. $$

The tax multiplier is $-c/D=-0.80/0.50=-1.60$. Because this is a tax increase, $\Delta T_0=+\$5$ billion:

\Delta Y=(-1.60)(+\$5\text{ billion})=-\$8\text{ billion}.

Check using expenditure rounds. The initial consumption change is $-0.80(\$5\text{ billion})=-\$4$ billion. Each additional round of output induces domestic expenditure equal to $c(1-t)-m=0.60-0.10=0.50$ of the preceding change. The total is $-\$4/(1-0.50)=-\$8$ billion, matching the multiplier calculation.

Common mistake avoided: Using $1-c=0.20$ as the denominator. Here income taxes reduce the share of an output change that reaches disposable income, and imports are an additional leakage; the correct denominator is $0.50$.

Worked Example: Distinguishing the Multiplier from the Output Change

A model has $c=0.60$, $t=0.25$, and $m=0.05$. Government purchases rise by $\$6$ billion, while lump-sum taxes are cut by $\$3$ billion. Calculate each policy’s contribution to the change in output and the combined change. Do not report a multiplier as though it were an output change.

Method. Calculate the common denominator using the model’s marginal propensities:

$$ D=1-c(1-t)+m =1-0.60(0.75)+0.05 =1-0.45+0.05 =0.60. $$

The spending multiplier is $1/0.60=5/3$, while the lump-sum tax multiplier is $-0.60/0.60=-1$. The purchases change is $\Delta G=+\$6$ billion, so its contribution is:

\Delta Y_G=\frac{1}{0.60}(+\$6\text{ billion})=+\$10\text{ billion}.

The tax cut is $\Delta T_0=-\$3$ billion, so its contribution is:

\Delta Y_T=(-1)(-\$3\text{ billion})=+\$3\text{ billion}.

Adding the two contributions gives $\Delta Y=+\$10\text{ billion}+\$3\text{ billion}=+\$13\text{ billion}$. Check directly:

\Delta Y=\frac{\Delta G-c\Delta T_0}{D} =\frac{6-0.60(-3)}{0.60} =\frac{7.8}{0.60} =13\text{ billion dollars}.

Common mistake avoided: Calling $5/3$ the output increase. It is the spending multiplier, measured as output change per dollar of purchases change. The purchases increase is $\$6$ billion, so the associated output change is $\$10$ billion.

Marginal Propensities: Use the One the Model Calls For

A marginal propensity is a response to an additional unit of income or another specified change. It is not interchangeable with a multiplier. In this model, $c$ is the fraction of an additional dollar of disposable income consumed. The marginal propensity to save out of disposable income is $1-c$ when disposable income is allocated only between consumption and saving. If $c=0.70$, for example, that saving propensity is $0.30$; it is not the tax multiplier.

When proportional income taxes apply, an extra dollar of output yields only $1-t$ dollars of disposable income. Induced consumption per dollar of output is therefore $c(1-t)$. If imports rise with output, $m$ is also a leakage from domestic expenditure. That is why the denominator is $1-c(1-t)+m$, not simply $1-c$. For a lump-sum tax change, however, the initial consumption response is $c\Delta T_0$: the proportional rate $t$ does not replace $c$ in the numerator.

Key distinction: A multiplier is a ratio that describes the model’s response per unit of a policy change. The policy change, such as $\Delta G$ or $\Delta T_0$, is an input. The output change, $\Delta Y$, is the multiplier applied to that input, with its sign and units carried through.

A Reliable Calculation Check

Before accepting a result, make four checks. They do not replace the model calculation; they help catch interpretation and arithmetic errors.

Name the policy variable.
Decide whether the change is in purchases, lump-sum taxes, or a tax rate. Do not use a lump-sum tax multiplier for a tax-rate change.

Write the change with a sign.
A tax increase has $\Delta T_0>0$; a tax cut has $\Delta T_0<0$. A purchases increase has $\Delta G>0$.

Match the formula to the model.
Use the stated $c$, $t$, and $m$. Check whether proportional taxes and imports are included before choosing the denominator.

Calculate the output change and check its direction.
Multiply the relevant multiplier by the policy change. A tax increase should produce a negative output change in this model; a tax cut should produce a positive one.

The direction check is a diagnostic, not a substitute for calculating the magnitude. For example, a positive output change after a tax increase signals a likely sign error, but the correct negative value still depends on the specified marginal propensities and leakages.

Check Your Understanding

For each problem, state the relevant policy change and use the model specified. All dollar amounts are in billions. Give output changes to the nearest cent in billions.

In a model with $c=0.70$, $t=0$, and $m=0$, lump-sum taxes increase by $\$10$ billion. Find $\Delta Y$. (Answer: $-\$23.33$ billion)
A model has $c=0.80$, $t=0.20$, and $m=0.10$. Lump-sum taxes are cut by $\$5$ billion. Find the tax multiplier and $\Delta Y$. (Answer: tax multiplier $=-1.7391$; $\Delta Y=+\$8.70$ billion)
With $c=0.60$, $t=0.10$, and $m=0.20$, government purchases fall by $\$3$ billion. Find the spending multiplier and the output change. (Answer: spending multiplier $=1.5152$; $\Delta Y=-\$4.55$ billion)
In a model with $c=0.75$, $t=0.20$, and $m=0.05$, lump-sum taxes rise by $\$4$ billion. Find the denominator, tax multiplier, and output change. (Answer: $D=0.45$; tax multiplier $=-1.6667$; $\Delta Y=-\$6.67$ billion)
A model has $c=0.65$, $t=0.20$, and $m=0.05$. Purchases increase by $\$2$ billion and lump-sum taxes are cut by $\$2$ billion. Find the combined change in output. (Answer: $+\$6.23$ billion)

What You'll Learn