Marginal propensity to consume measures how much consumption changes relative to an incremental change in disposable income over a stated period.
The marginal propensity to consume (MPC) measures the change in consumption associated with an incremental change in disposable income. If disposable income rises by 1,000 and consumption rises by 700 over the same measurement period, the measured MPC is 0.70.
MPC is a response ratio, not the share of total income that a household spends. It also is not a fixed personality trait: the result can differ by the type of income change, household liquidity, wealth, debt, expectations, and the period being measured.
Where:
The numerator and denominator must use the same household or population, comparable price basis, and compatible period. Using a monthly consumption change with an annual income change, or nominal spending with inflation-adjusted income, produces a ratio with no clear interpretation.
A household’s monthly disposable income rises from 5,000 to 6,000. During the same period, monthly consumption rises from 4,000 to 4,700.
If the rest of the additional disposable income is saved under the same definitions, saving rises by 300 and marginal propensity to save is 0.30.
This example does not say the household spends 70% of all income. Its average propensity to consume after the change is 4,700 / 6,000, or about 0.783. The average ratio and the marginal response answer different questions.
The standard identity begins with the simplified disposable-income allocation:
Taking changes gives:
Dividing by (\Delta Y_d) gives (MPC + MPS = 1). This is an accounting result when consumption and saving exhaust the defined change in disposable income.
The identity can appear to fail when data sources define consumption, income, or saving differently; when timing is mismatched; or when the estimate compares unrelated groups. Paying down principal, for example, may be treated as saving in an economic flow account even though a household informally calls it debt repayment.
Persistence of income: The permanent-income hypothesis predicts a larger consumption response to an income change expected to last than to a one-time windfall.
Liquidity and credit access: A cash-constrained household may spend more of an immediate payment because it could not borrow against expected future income. A liquid household has more ability to smooth consumption.
Wealth and debt: Cash holdings, asset values, debt service, and credit limits can affect whether additional income is spent, saved, or used to reduce liabilities.
Expectations: If a payment was fully anticipated, spending may change when the news arrived rather than when cash was received.
Household needs: Housing, food, health, dependents, and durable-goods purchases create different short-run responses across households.
Measurement window: A household might save a payment this month and spend it six months later. Short and long windows can produce different MPC estimates.
In a simple textbook consumption function, MPC is often assumed to lie between zero and one. That assumption is useful for model behavior but is not a universal empirical boundary for every household and interval.
MPC can exceed one when consumption rises by more than the measured income change, financed by borrowing, asset sales, prior saving, or the timing of a durable purchase. It can be negative when consumption falls despite an income increase, perhaps because a household repays debt, anticipates a future loss, or completes an earlier purchase cycle.
An unusual estimate does not automatically prove irrational behavior. It is a prompt to inspect definitions, timing, financing, and whether the income change caused the consumption change.
In the simplest Keynesian expenditure model, the spending multiplier is written as:
If MPC is 0.75, that formula gives a multiplier of 4. The result depends on strong assumptions: fixed prices, idle capacity, no taxes, no imports, no crowding out, a stable MPC, and repeated spending rounds. It should not be treated as a forecast of the effect of an actual policy.
Real fiscal analysis considers monetary policy, timing, household differences, imports, tax treatment, supply constraints, and how a measure is financed. Published multiplier estimates are therefore model- and context-dependent.
MPC helps economists interpret household responses to earnings changes, taxes, transfers, and economic shocks. Banks and analysts may use related cash-flow behavior when assessing resilience, but an aggregate MPC does not determine one borrower’s ability to pay. For investors, MPC can help explain consumption sensitivity, yet it is only one input among labor conditions, prices, wealth, credit, and policy.
This page is educational. MPC does not indicate whether a household should spend or save an additional dollar, and it does not provide individualized financial advice.