Marginal Propensity to Consume

The marginal propensity to consume (MPC) is the share of each additional dollar of income that a household or economy spends on goods and services rather than saving. If you earn an extra dollar and spend 80 cents, your MPC is 0.8. Aggregate MPC anchors the size of all fiscal multipliers—the higher the propensity to consume, the larger the stimulus from government spending or tax cuts.

Definition and measurement

MPC is straightforward in concept but subtle in measurement. When your income rises by $100, how much of it do you spend in the near term, and how much do you save?

The distinction between average and marginal matters. Your average propensity to consume (total spending divided by total income) might be 0.75—you spend 75 per cent of your income on average. But the marginal propensity is how you treat new income. You might spend only 60 cents of each additional dollar because you’re already meeting your basic needs and can afford to save the increment.

Measuring actual MPC requires observing households when their income changes unexpectedly—a bonus, a tax refund, a job change. Economists find that the MPC varies widely by circumstance. A low-income family receiving an extra $100 might spend $90 or $95 of it, because they have urgent needs and limited savings buffers. A high-income family might spend $30 and save $70. An economy’s aggregate MPC is a weighted average across all households and varies by wealth, age, confidence, and interest rates.

Why it matters for stimulus

The MPC is the throttle on fiscal stimulus. Suppose the government spends $100 billion on infrastructure. That money becomes income for workers and suppliers. With an MPC of 0.8, they spend $80 billion and save $20 billion. That $80 billion becomes income for retailers, manufacturers, and their workers, who spend $64 billion and save $16 billion. The chain continues, and total output increases by $100 billion ÷ (1 − 0.8) = $500 billion. This is the spending multiplier.

With an MPC of 0.5, the same $100 billion spending yields a multiplier of only $100 billion ÷ 0.5 = $200 billion. Lower consumption, lower stimulus.

A tax cut of $100 billion works through the same channel, but filtered through households’ consumption decisions. If the MPC is 0.8, they consume $80 billion of the tax cut and save $20 billion, so the initial demand impulse is only $80 billion—less powerful than direct spending. This is why the tax multiplier is always smaller than the spending multiplier: the MPC is both the numerator and the discount factor.

Variation by circumstance

MPC isn’t a constant. It depends on who receives the income:

Low-income households typically have high MPC—0.8 to 1.0—because their income barely covers necessities. An extra dollar goes to groceries, rent, or emergency repairs. They have few savings and limited ability to defer spending.

High-income households have low MPC—0.2 to 0.5—because they can easily save the increment. They’ve already met their consumption needs and can invest, build retirement savings, or purchase assets.

During recessions, MPC may fall sharply as households grow anxious about job security and pull back on spending even though their current income hasn’t fallen. This precautionary saving is one reason why tax cuts are less effective in deep downturns—the boost goes into savings rather than demand.

During booms, MPC may rise as confidence is high and credit is cheap, encouraging households to spend from future as well as current income.

Age and wealth matter too. Young households with few assets and high future earnings have higher MPC—they borrow and spend today. Retirees living off accumulated assets have lower MPC. Wealthy households with inherited or asset wealth consume a smaller fraction of their annual income increment.

The consumption function

Economists model this relationship through the consumption function, which expresses consumption as a function of income. The simplest version is:

C = a + c × Y

where C is consumption, a is autonomous consumption (spending that happens regardless of income, funded by savings or borrowing), Y is income, and c is the MPC.

If a = $50 billion and c = 0.8, then at income of $500 billion, consumption is $50 + 0.8 × $500 = $450 billion. At $600 billion income, consumption is $50 + 0.8 × $600 = $530 billion. The marginal propensity—the slope of the line—is 0.8.

This linear model is simplified; real behaviour is more complex, with non-linear responses to large income shocks and psychological thresholds. But it captures the essential regularity: households spend a baseline amount and a stable fraction of incremental income.

Policy implications

Understanding MPC helps explain why stimulus design matters:

1. Target low-income households. A tax cut for minimum-wage workers or direct payments to the unemployed have high MPC and generate large multipliers. The same dollar spent on tax breaks for the wealthy generates less demand stimulus.

2. Spending may be more efficient than tax cuts. Government spending delivers the full dollar directly to demand; tax cuts leak into savings. When the government is trying to boost output quickly, direct purchases are more reliable than hoping households will spend tax cuts.

3. Stimulus fades if confidence collapses. If a recession is accompanied by panic and rising precautionary saving, MPC falls sharply. Tax cuts become less effective; direct spending on visible projects (or transfers to those forced to spend immediately, like the unemployed) works better.

4. Interest rates matter. If the central bank raises interest rates, savings become more attractive, MPC falls, and multipliers shrink. This is how tight monetary policy can offset fiscal stimulus.

Empirical estimates

Academic estimates of aggregate MPC in developed economies typically range from 0.5 to 0.9, with most clustering in the 0.7–0.8 range. During normal times, households spend roughly three-quarters of a new dollar and save one-quarter. During crises, MPC can plummet to 0.3–0.5 as precautionary saving spikes.

The variation is important for policy. A government designing stimulus must estimate what the relevant MPC will be during the downturn, not in normal times. Pandemic-era stimulus in 2020 encountered high MPC among lower-income households (who had deferred spending) but lower MPC among the wealthy, who were already sitting on savings.

See also