C = 1,200 + 0.75·Yd means 1,200 of spending at zero income and 75 cents of each extra dollar
The linear consumption function has two parameters. The intercept a is autonomous consumption — spending that happens whatever income is, financed at low incomes by saving or borrowing. The slope b is the marginal propensity to consume, the share of each additional dollar of disposable income that is spent.
C = a + b·Yd
At a disposable income of 20,000 ($ billions), consumption is 1,200 + 0.75 × 20,000 = 16,200 and saving is the remaining 3,800. The average propensity to consume is 16,200 ÷ 20,000 = 0.81 — above the MPC of 0.75, as it always is while a is positive.
Worked example
a = 1,200, b = 0.75, Yd = 20,000
C = 1,200 + 0.75 × 20,000
C = 16,200
(saving = 20,000 − 16,200 = 3,800)
Below 4,800 of income this economy consumes more than it earns
The break-even income is where the consumption line crosses the 45-degree line — where consumption equals income and saving is exactly zero. Setting a + bY = Y gives Y = a ÷ (1 − b): 1,200 ÷ 0.25 = 4,800 in the example.
Y* = a ÷ (1 − b)
Below the break-even income, consumption exceeds income and saving is negative; above it, saving is positive and rises by 1 − b = 0.25 for every extra dollar. On the chart the gap between the two lines is saving.
Break-even
a + bY = Y
Y (1 − b) = a
Y = 1,200 ÷ 0.25
Y = 4,800
With investment of 3,800, planned spending equals income at 20,000
Add planned investment to consumption and you have planned spending in a closed economy with no government: AE = a + bY + I. Output settles where planned spending equals income, the point where the AE line crosses the 45-degree line. Solving a + bY + I = Y gives Y = (a + I) ÷ (1 − b) = (1,200 + 3,800) ÷ 0.25 = 20,000.
Y = (a + I) ÷ (1 − b)
At that income saving is 3,800, exactly equal to investment — the same equilibrium seen from the other side. The factor 1 ÷ (1 − b) = 4 is the spending multiplier: raise investment by 100 and equilibrium income rises by 400.
Equilibrium
1,200 + 0.75Y + 3,800 = Y
0.25Y = 5,000
Y = 20,000
(and S = −1,200 + 0.25 × 20,000 = 3,800 = I)
A change in income moves you along the line; a change in wealth moves the line
The consumption function isolates one cause of spending: current disposable income. A change in income is a movement along the line. Everything else that changes spending at a given income shifts the line — rising house or share prices (the wealth effect), lower interest rates, easier credit, and more optimistic expectations all raise a; higher taxes lower disposable income, which moves you along the line rather than shifting it.
Keeping the two apart is most of what the function is for. When spending rises faster than income, as it did in the United States when household wealth climbed, the right reading is an upward shift, not a higher MPC.
Keynes’s line predicted a falling APC; long-run data did not show one
Keynes proposed the function in 1936 with three claims: the MPC lies between 0 and 1, the APC falls as income rises, and current income is the main driver of spending. Cross-section data of the time bore him out. But when Simon Kuznets assembled long-run US data, the APC stayed roughly constant across decades of rising income — the consumption puzzle.
Milton Friedman’s permanent-income hypothesis and Franco Modigliani’s life-cycle hypothesis resolved it: households consume out of expected lifetime resources, so a short-run function fitted to one year looks Keynesian while the long-run relationship is close to proportional. The calculator’s straight line is the short-run tool; it should not be stretched across decades.
Two observations fix the line exactly; more observations need a regression
With two observations the slope is b = ΔC ÷ ΔY and the intercept follows from either point: a = C₁ − b·Y₁. The default observations — 10,200 of consumption at 12,000 of income and 16,200 at 20,000 — recover a = 1,200 and b = 0.75 exactly.
b = (C₂ − C₁) ÷ (Y₂ − Y₁); a = C₁ − b·Y₁
Real data never sit on one line. With a series of years the usual method is an ordinary least-squares regression of consumption on disposable income, which the linear regression calculator on this site will fit; the two-point fit here is exact for the two points and says nothing about how well a line describes the rest.
Sources & References
Figures on this page are checked against primary, authoritative sources. Links open in a new tab.
What the calculation assumes
- Consumption is linear in disposable income with a constant MPC.
- The equilibrium is the closed-economy Keynesian cross with no government, and investment does not depend on income.
- All figures share one unit and one period.
Where it stops
- A straight line fitted through two points is only as good as those points; regression on many observations is sturdier.
- Wealth, interest rates, credit and expectations shift the function; it is not a forecast of spending.
- The Keynesian cross assumes prices are fixed and capacity is idle.