ECON5002 · Macroeconomic Theory
Consumption and Investment
This topic opens the black box behind the two largest moving parts of aggregate demand. On the consumption side it moves beyond the Topic-1 Keynesian function (spending out of current income) to the forward-looking theories — Friedman's permanent-income hypothesis and Modigliani's life-cycle hypothesis, both of which say households smooth consumption, so the marginal propensity to consume out of a transitory windfall is close to zero. On the investment side it sets out the accelerator (investment is the derived demand for capital, driven by the change in expected output), the neo-classical user-cost / Jorgenson model and Tobin's q. A distinctive feature of ECON5002 is that the lecturer is openly critical of the interest-elastic investment-demand function: the capital-controversy critique is examinable, and an essay that simply asserts "lower interest rates raise investment" will under-score.
What this chapter covers
- 011. Why the simple Keynesian consumption function fails — a near-constant long-run APC and consumption smoother than income
- 022. Permanent-income hypothesis — C = k·Yᴾ, with income split into permanent Yᴾ and transitory Yᵀ
- 033. Short-run vs long-run MPC — the transitory MPC (k·j) is far below the long-run MPC (k)
- 044. Life-cycle hypothesis — flat consumption from lifetime wealth; the young borrow, the middle-aged save, the retired dissave
- 055. Relative-income (Duesenberry) — consumption is social and irreversible, so the MPC rises in downturns
- 066. The accelerator — net investment = v*·ΔYᵉ, simple vs flexible, and why investment is so volatile
- 077. Neo-classical investment — the user cost of capital v = r + δ, invest while MPK ≥ v, and Tobin's q > 1
- 088. The capital-controversy critique — why this course distrusts the interest-elastic MPK / investment-demand function
Permanent income: a temporary rebate vs a permanent pay rise
- +2A current-income change first moves permanent income by the fraction j, then consumption by k of that. So the short-run MPC = k·j = 0.8 × 0.25 = 0.20 — only a quarter of the long-run MPC of 0.8.
- +2Transitory rebate of $4,000: the surprise raises permanent income by ΔYᴾ = j × 4,000 = 0.25 × 4,000 = $1,000, so consumption rises by ΔC = k × ΔYᴾ = 0.8 × 1,000 = $800.
- +1The household therefore saves 4,000 − 800 = $3,200 — about 80% of the windfall is saved, not spent.
- +1A permanent pay rise eventually raises permanent income by the full $4,000, so consumption rises by ΔC = k × 4,000 = $3,200 — four times the response to the same-sized transitory amount.
- +1Because a temporary tax cut works through an MPC of only 0.20, its multiplier is small and the stimulus weak; a permanent cut works through 0.8 and is far more powerful — Friedman's case against fine-tuning demand with transitory measures.
Key terms
- Permanent-income hypothesis (PIH)
- Friedman's theory that consumption is a constant fraction k of permanent (expected long-run) income: C = k·Yᴾ. Because the MPC out of transitory income is near zero, temporary tax changes are mostly saved and have weak effects on consumption.
- Permanent vs transitory income
- Measured income splits into a permanent component Yᴾ (the steady flow your wealth and skills can sustain) and a transitory component Yᵀ (one-off surprises such as a bonus or a rebate). Consumption tracks only the permanent part; transitory income is largely saved or dissaved.
- Short-run vs long-run MPC
- Under the PIH with adaptive learning, the long-run MPC out of permanent income is k, but the short-run MPC out of a transitory income change is only k·j (where j is the fraction by which permanent income revises to a surprise) — far smaller, which flattens the short-run consumption function.
- Life-cycle hypothesis (LCH)
- Modigliani's theory that households smooth consumption over their whole lifetime out of lifetime resources (assets plus expected labour income). Income is hump-shaped over life while consumption is flat: the young borrow, the middle-aged save, and the retired dissave — so an ageing population lowers the aggregate saving ratio.
- Accelerator (simple vs flexible)
- Investment as the derived demand for capital: net investment = v*·ΔYᵉ, the desired capital-output ratio times the change in expected output. The simple accelerator assumes a constant v* and naive expectations; the flexible accelerator adds adaptive expectations, lags, variable capacity utilisation and a variable v. Because it depends on the change in output, investment is highly volatile.
- User cost of capital
- What it costs to use one unit of capital for a period: v = r + δ, the real interest rate plus the depreciation rate. In the neo-classical (Jorgenson) model firms invest until the marginal product of capital equals the user cost (MPK = v); a lower real rate or a depreciation allowance cuts v and raises desired capital.
- Tobin's q
- The ratio of the market value of installed capital to its replacement cost. A firm should invest when q > 1 (the market values an extra unit of capital above the cost of building it) and run capital down when q < 1 — a financial-market restatement of the user-cost decision.
- Capital-controversy critique
- The lecturer's argument that the interest-elastic investment-demand / MPK function is theoretically unsound: with heterogeneous capital goods, capital must be measured in value terms, which already requires knowing the interest rate, so MPK is not independent of i and need not fall as K/L rises. Empirically the interest-investment link is weak, so essays should lean on the accelerator and expectations instead.
Consumption and Investment FAQ
What is the difference between the permanent-income and life-cycle hypotheses?
Both say households smooth consumption rather than spend out of current income, so they reach the same broad conclusion. The difference is the horizon and the measure of resources: the permanent-income hypothesis (Friedman) makes consumption a constant fraction of a statistical permanent income, while the life-cycle hypothesis (Modigliani) makes it depend on lifetime resources — accumulated wealth plus expected labour income — spread evenly over the years a household expects to live. The life-cycle version adds an explicit borrow-save-dissave pattern over the life span and a demographic corollary: an ageing population lowers the aggregate saving ratio.
Why does a temporary tax cut have little effect on consumption?
Because forward-looking households treat a one-off tax cut as transitory income. Under the permanent-income hypothesis the marginal propensity to consume out of transitory income is close to zero (k·j, far below the long-run k), and under the life-cycle hypothesis a windfall is spread thinly over all remaining years. Either way most of the cut is saved, so consumption barely moves and the multiplier is small. A permanent tax change, by contrast, raises permanent or lifetime income and has a much larger effect — which is Friedman's argument against fine-tuning demand with temporary measures.
Why is investment so volatile under the accelerator?
Because net investment depends on the change in expected output (v*·ΔYᵉ), not its level. If output simply grows more slowly, the required addition to capital falls, so net investment can drop sharply even while output is still rising. If output is merely expected to be flat, net investment is zero; if it is expected to fall, net investment turns negative (disinvestment, bounded by depreciation). This is why investment amplifies the business cycle and is hard to fine-tune, and why the accelerator combines with the multiplier to generate cycles.
What is the user cost of capital, and why is it not just the interest rate?
The user cost is the full cost of using a unit of capital for one period: v = r + δ, the real interest rate plus the depreciation rate. It is not the interest rate alone because capital wears out, so depreciation is a genuine cost of holding it. Firms invest until the marginal product of capital equals the user cost (MPK = v). Recognising the depreciation term also matters for policy: governments influence investment through v via depreciation allowances and profit taxes, not only through the interest rate.
Why does ECON5002 criticise the interest-investment relationship?
This is a lecturer's course, and the lecturer emphasises the capital controversies. The argument is that in a world of heterogeneous capital goods, capital must be aggregated in value terms — but that valuation already depends on the interest rate, so the marginal product of capital is not independent of i and need not fall smoothly as the capital-labour ratio rises. The interest-investment link is also empirically weak. So an essay that just asserts 'lower interest rates raise investment' under-scores; the marks come from stating the critique and relying on the accelerator and expectations instead.
Is this guide official or affiliated with the University of Sydney?
No. This is an independent AskSia study resource created to help students revise. It is not produced, endorsed by, or affiliated with the University of Sydney. Always check Canvas and the official unit outline for current dates, weights and assessment rules.
Exam move
Treat this topic as two halves that the final exam tests in mirror-image ways. For consumption, build a one-page comparison of the Keynesian, permanent-income and life-cycle theories and finish each line with its verdict on a temporary tax cut (large, small, small multiplier respectively), naming forward-looking expectations as the mechanism that shrinks the short-run MPC; be able to compute the short-run MPC as k·j and contrast a transitory windfall with a permanent rise. For investment, drill the accelerator numerically until you can show that net investment can fall sharply while output still grows, and learn the user-cost identity v = r + δ and Tobin's q > 1 rule. Above all, remember that this is a lecturer's course: whenever you write about the cost-of-capital view of investment, bring in the capital-controversy critique and the weak empirics, because that is where the marks sit and an uncritical 'lower i raises I' answer is explicitly penalised. As always, lead every essay with a correct, labelled diagram drawn from memory.
