Macro Revision Plan · Macro-Economics

A four-pass revision plan for the Intermediate Macro final — foundations, problem sets by topic cluster, timed past papers, then targeted repair — prioritised by the past-paper frequency analysis. Map the passes onto your calendar once the exam date is set.

How to use this plan

No exam date is baked in, so this is structured as four passes, not four weeks — map them onto your calendar once the date is known and shrink Pass 1 before you shrink Pass 3. The ordering inside each pass follows the frequency analysis: two-period consumption, production/TFP, investment/user cost and labor supply carry both 32-point long questions on every Moed paper, so they get first claim on your time. This is a hard exam and the syllabus is wide — the plan works by being ruthless about what earns marks, not by covering everything equally.

Use the site properly while you revise

Status pills: every problem-set and past-paper question has a status pill — mark each attempt (nailed it / shaky / wrong) as you go, so Pass 4 can target exactly the questions that hurt.

Dictionary hover previews: hover any linked term (like euler-equation or public-debt-dynamics) for the definition without leaving the page — faster than re-opening lecture notes mid-question.

Show-solution toggles: past-paper questions hide their solutions behind a per-question toggle. In Pass 3, answer on paper before toggling — the toggle is the honesty test.

🗣️ In plain English recaps: every macro solution ends with a plain-English recap. If the algebra clicked but the point didn't, read that box before moving on.

Pass 1 — Foundations (1–2 days)

Goal: know the toolkit and the map before touching any questions. This pass is deliberately short — reading is not revising.

The final is open-notes with the formula sheet attached, so nothing needs memorising — but you must know which block solves which question shape without hunting. That recognition is what Pass 1 builds.

Read the Macro Equation Sheet top to bottom. For each block, say out loud what kind of exam question it solves (the high-yield page tells you exactly which past-paper questions each block has powered).
Dictionary sweep of the high-yield terms — hover or open each and check you could explain it to a classmate: euler-equation, consumption-smoothing, marginal-product-of-capital, capital-accumulation-equation, labor-supply, income-effect, solow-residual, development-accounting, public-debt-dynamics, primary-deficit, job-finding-rate, separation-rate, gdp-deflator, cpi, purchasing-power-parity, gnp, polarization.
Skim the lectures where you feel weakest — likely candidates given what the exams test: Lec_02-Consumption and Saving, Lec_05-Investment, Lec_07-Labor Market, Lec_10-Fiscal Policy. Skim Lec_01-Data Review, Lec_04-Production, Lec_06-Equilibrium in the Goods Market, Lec_08-Labor Market Data, Participation & Unemployment and Lec_09-Inequality & Polarization only where a dictionary term didn't click.
Write the four "workhorse" results on one page, by hand, from the sheet: the Euler equation, the static labor-supply FOC, the user-cost condition, and debt dynamics. This page is your Pass 4 benchmark.

Time estimate: 4–8 hours. Do not let it become a week.

Pass 2 — Problem sets by topic cluster (4–6 days)

Goal: fluency on every question shape the exams recycle. Work in topic clusters, not problem-set order, so each model gets one concentrated block.

The TA and extra questions are your fresh practice — use them last in each cluster

You've only ever done the submission questions of each set, and the TA/extra questions are newly added to the site. That makes them the only substantial pool of questions you've never seen — the closest thing you have to fresh exam practice outside the past papers. In each cluster: redo the submission questions first (fast, to confirm the method), then do the TA/extra questions blind, on paper, before opening the solution. Mark every attempt with its status pill.

Order the clusters by exam weight:

Consumption & saving — PS2 (both questions, then the new extra material). Target: run the Lagrangian → Euler → substitute → solve pipeline without notes, for log and CRRA utility. This is the engine of a 32-point long question on every Moed paper.
Investment & goods market — PS4 (Q1–Q3 + TA questions). Target: given any five of { $p_k$ , $p_k^f$ , $r$ , $\delta$ , $\tau_K$ , $A^f$ }, solve the user-cost condition for the sixth; then shift $S$ / $I$ curves and sign $r$ confidently.
Labor market & taxes — PS5 Q1–Q5 + the new Q6–Q8, then PS6 Q1–Q5 (Q4–Q5 were TA-session — treat them as fresh). Target: derive $N(w)$ from the static FOC for any utility they throw at you, and narrate income vs substitution effects. Both Moed B long questions live here.
Fiscal & debt — PS6 Q3 (plus the tax questions already covered in the labor cluster). Target: the debt forecast recipe (Fisher for $r$ , compound growth for $g$ , then plug in) at speed.
Measurement — PS1 (Q1–Q4 redo, Q5–Q8 are TA questions you haven't done). Target: CPI vs deflator with a non-consumption good, PPP conversion, GDP/GNP/NFP signs. Cheap, reliable short-question points.
Production, TFP & allocation — PS3 (Q1–Q3 redo, Q4 growth accounting is extra), then PS7 (efficient allocation of capital across firms — brand new, never submitted, so it's all fresh material and it trains the allocation MCQ on the Sample Exam 2026). Target: back out $\alpha$ from a factor share, then TFP as the residual — and always differentiate the function actually given (the additive-form trap).

Time estimate: roughly a day per cluster; consumption and labor deserve the fullest days. 25–35 hours total.

Pass 3 — Past papers, hardest use (3–4 days)

Goal: exam performance, not understanding — understanding was Passes 1–2. The papers are scarce (five exist, and you've read solutions to some), so burn them carefully.

Moed A 2025 — timed and blind. 3 hours, on paper, formula sheet only, no site. Then mark yourself brutally against the worked solutions, one show-solution toggle at a time. Log every question's status pill.
Moed B 2025 — timed and blind, a day or two later. Same protocol. These two are the most recent and most predictive papers — treat them as the real thing.
Moed A 2023-24 and Moed B 2023-24 — work them question-by-question (untimed is fine; you've likely seen parts). Focus on the twists you haven't met: the additive production function, the consumption-tax Lagrangian, the Laffer-curve long question.
For anything that didn't click, read the 🗣️ In plain English recap under the solution, then the linked Macro Equation Sheet block, then the dictionary terms. If it still doesn't click, flag the topic for Pass 4 — don't stall the pass.
Sample Exam 2026 last, as the dress rehearsal. It's the only paper in your actual MCQ format — 20 questions, timed. Do it once everything else is done, in one sitting. MCQ distractors punish sign errors and dropped denominators, so slow down on the arithmetic you'd normally hand-wave.

Time estimate: ~4 hours per Moed paper (sit + mark), ~3 hours for the sample. 18–22 hours total.

Don't read solutions you haven't attempted

A past paper you've read is a past paper you've spent. The show-solution toggle exists so you can grade one question at a time without spoiling the rest — use it that way.

Pass 4 — Targeted repair (1–2 days, repeatable)

Goal: convert every red/amber status pill into green. This pass is a loop, not a stage — run it as many times as the calendar allows.

Sort your misses by the frequency table: a wobble on two-period consumption outranks a wobble on PPP, whatever order you happened to miss them in.
For each weak topic: re-derive the key result from memory on a blank page — the Euler equation from the Lagrangian, the labor-supply FOC to $N(w)$ , the user-cost condition from the firm's problem, and one full debt-dynamics forecast. Check against your Pass 1 page.
Redo the specific past-paper and problem-set questions you missed (the status pills are your queue). A question isn't repaired until you've re-done it cold, not just re-read it.
Rebuild speed on the surviving weak spots with any TA/extra questions you left unused in Pass 2.
Final evening: one last read of the Macro Equation Sheet and the plain-English recaps of the questions that hurt most. Then stop — sleep beats a fifth pass.

Time estimate: 6–10 hours per loop.

The honest summary

The exam is hard, but it is predictable: four topics power ~64 of 100 points every sitting, the same recipes recur with rotated unknowns, and you have a genuinely unseen question bank (the TA/extra questions) plus two blind-sittable 2025 papers. Foundations fast → clusters deep → papers blind → repair ruthlessly. That's the whole plan.