Everything, on one sheet

The card

GNI = GDP + NPIA (net primary income from abroad). NPIA > 0 → GNI > GDP; NPIA < 0 → GNI < GDP.
PPP: the exchange rate where an identical basket costs the same in both currencies — "parity" is the whole point.
Recession = negative real GDP growth for two CONSECUTIVE quarters. One bad quarter alone is not a recession.
CPI = basket price change weighted by spending share, not a simple average. Demand-pull: AD right, output up. Cost-push: SRAS left, output down.
Q12(d) "examine" (8 marks) is KAA-only — no evaluation credited. Save conditional judgement for Q12(e) and the essay.
The Q13/14 essay itself is marked on two independent bands, not one: 12 marks KAA (4 levels: 1-3, 4-6, 7-9, 10-12) + 8 marks Evaluation (3 levels: 1-3, 4-6, 7-8). An unconditional "X is/isn't reliable" verdict caps Evaluation at the middle band regardless of KAA strength — the fix is always a stated "only if..." condition, not more content piled onto the same flat claim.
GDP/GNI's limitations (distribution, informal economy, composition, subjectivity of 'living standards' itself) don't make it useless: it's standardised, simple to construct, internationally comparable, and dynamic enough for real-time policy use — the Q13 living-standards essay credits Evaluation marks on BOTH sides, not just the limitations list.

Why it works — Why GNI can exceed — or fall short of — GDP: deriving net primary income from first principles

GDP and GNI start from the same total economic activity and split it two different ways, and the difference between them is entirely mechanical once you see the split, not a fact to memorise. GDP asks a TERRITORY question: how much was produced within this country's borders this year, regardless of who owns the factory, the land, or the labour that produced it? A foreign-owned car plant operating inside the country counts fully in that country's GDP, even though the profit it generates ultimately belongs to a foreign shareholder. GNI asks an OWNERSHIP question instead: how much income did this country's OWN citizens and firms actually receive, wherever in the world they earned it? A citizen working abroad and remitting wages home counts in their home country's GNI, even though the work itself — and the GDP it created — happened inside a different country's borders. Because GDP and GNI are answering two genuinely different questions, there is no reason to expect the same number from both, and the entire gap between them has an exact name: net primary income from abroad. "Primary income" here means the return to a factor of production — wages to labour, profit/interest/dividends to capital — as distinct from a transfer like foreign aid, which isn't a return to a factor at all. Net primary income from abroad is the primary income this country's own citizens and firms earned abroad, MINUS the primary income earned inside this country by foreign-owned citizens and firms and sent back to THEIR home country. GNI = GDP + net primary income from abroad — and that's not a rule to memorise, it falls straight out of the two definitions once territory and ownership are kept separate. This is also exactly why the direction of the GDP/GNI gap is predictable, not random, and case-specific rather than a fixed rule: a country whose citizens work abroad in large numbers and remit wages home, or whose firms and citizens hold large amounts of overseas investment earning profit, interest and dividends, has a large primary-income inflow — net primary income from abroad is positive, and GNI runs above GDP. A country that instead hosts large amounts of INWARD foreign investment — foreign-owned factories and subsidiaries whose profits are repatriated abroad rather than reinvested or spent domestically — has a large outflow instead, net primary income from abroad is negative, and GNI runs below GDP. Both are real, common, opposite-sign versions of the exact same mechanism, which is exactly why "GNI is usually higher than GDP" isn't a safe generalisation to memorise: it's true only for the first kind of economy, and the direction has to be reasoned from the underlying primary-income flows each time — precisely the reasoning step examiner reports confirm most candidates skip even when they can correctly read the direction straight off a data table.

Traps — 7

ppp-defined-without-parity

Examiner reports confirm students consistently describe PPP using only "the ability to buy goods with a currency" and drop the comparative, parity element entirely — confirmed directly: "many weaker students simply ignored the word 'Parity'" (January 2021 examiner report, Q12a). The fix is structural, not just a missing word: state the comparison explicitly — an identical basket priced in two different currencies — not just purchasing power in the abstract.

gdp-gni-per-capita-vs-per-capita-itself

Asked to explain the DIFFERENCE between GDP per capita and GNI per capita, many candidates explain what "per capita" means — dividing by population, a step both terms already share — instead of explaining the GDP/GNI distinction itself, where the actual difference lives (confirmed: January 2021 examiner report, Q12b). The worked chain above deliberately keeps the two steps — GDP→GNI via net primary income, then ÷ population — visibly separate for exactly this reason: 'per capita' never changes which of GDP or GNI sits in the numerator.

growth-rate-vs-real-gdp-itself

Asked to "explain real GDP growth rate," most candidates define real GDP correctly but drop "growth rate" entirely, losing the second knowledge mark (confirmed: January 2020 examiner report, Q12b). Real GDP is a LEVEL, a total in £bn; the growth rate is the percentage CHANGE in that level between two periods — two different numbers, and a question naming the rate specifically wants the second, not the level restated.

q12d-examine-does-not-want-evaluation

Section C's Q12(d) carries 8 marks and the command word "examine" with reference to a source — a tariff that looks essay-sized, but is confirmed directly from mark schemes across multiple series as KAA-only (Knowledge, Application, Analysis): up to three levels available purely on those three assessment objectives, with zero evaluation credit available at this sub-question at all. Several series separately confirm candidates losing marks and time by writing an evaluative conclusion here regardless. Save the conditional-judgement move for Q12(e) (14 marks, KAA + Evaluation) and the Q13/14 essay (20 marks, KAA + Evaluation) — the only two places in this paper's Section C/D where evaluation actually earns credit.

oil-price-direction-reversed

On a question about FALLING oil prices, "quite a few confused answers" reversed the direction — suggesting more oil would be bought and sold and that oil-exporting countries' export VALUES would rise, when a falling price tends to cut export revenue for the exporter and cut import costs for the buyer, even allowing for some rise in the volume traded (confirmed: January 2021 examiner report, Q2/Q10 discussion). Whenever a price change and a quantity change both appear in the same reasoning chain, work out the effect on total VALUE (price × quantity) explicitly, rather than assuming the direction from the price movement alone.

generic-uncontextualised-effects-of-inflation

Effects of inflation written generically — without tying them to the specific country or data actually given — are capped at low evaluation levels; the mark scheme explicitly requires evaluation to be "critical of the impact of inflation" for the specific case in front of the candidate, not a textbook list recited from memory (confirmed: October 2019 examiner report, Q12d). The country-reference ceiling on the Q13/14 essay below applies exactly the same discipline at a larger scale.

public-sector-size-assumed-to-improve-living-standards

A country's public-sector spending share is itself a separate, examinable limitation on GDP/GNI as living-standards measures, confirmed directly from the mark scheme: "the size of the public sector spending, especially in developed countries such as Norway/Finland, which may or may not improve living standards" (Oct 2022 MS, Q13 KAA indicative content). All government spending counted in GDP is included at face value regardless of how effectively it is delivered — much of it (health, education, administration) has no market price and is conventionally valued at its cost of provision rather than its actual welfare impact — so a large public sector is not automatically evidence of correspondingly higher living standards, and two countries with identical headline GDP but very different public-sector shares can differ sharply in genuine living standards for the same figure. Keep the mark scheme's own "may or may not" framing — spending size alone doesn't determine the welfare outcome — rather than asserting the cost-valuation mechanism as the only cause.

The card

ILO: LFS survey — out of work for 4 weeks, willing to work in 2 weeks. Claimant count: benefit records only. Different criteria.
Unemployed = not working at all. Underemployed = working, but wanting more hours or a better-matched role.
5 causes, 5 mechanisms: frictional (search time), seasonal (calendar), structural (skills mismatch), demand-deficient (AD shortfall), real-wage inflexibility (wage stuck above equilibrium).
Unemployment rate = unemployed ÷ (employed + unemployed). Moving someone to 'inactive' changes the rate with zero jobs created.
Current account ≠ government budget balance — separate accounts, can move in opposite directions.
Effects-on-two-named-groups essay (workers AND public finances): capped at Level 3 unless BOTH sides are substantively developed — not just the country-reference rule, a second, separate ceiling.

Why it works — Why the unemployment rate can move without a single job being created or destroyed

The unemployment rate is not unemployed people divided by the whole population — it's unemployed people divided specifically by the labour force (employed + unemployed), and the economically inactive sit entirely outside that fraction. That structural fact has a direct, derivable consequence: moving a person between 'unemployed' and 'inactive' changes the unemployment rate even though total employment hasn't moved at all. Take an economy with 28.0 million employed and 2.0 million unemployed — a labour force of 30.0 million and an unemployment rate of 2.0/30.0 = 6.667%. Now suppose 300,000 of those unemployed people become discouraged workers — they stop actively seeking work, perhaps after months of rejected applications, and so no longer satisfy the ILO test, moving into the inactive category instead. Employment is still exactly 28.0 million — not one new job was created. But the labour force has shrunk to 29.7 million, and the unemployment rate is now 1.7/29.7 = 5.724% — a fall of nearly a full percentage point, purely from a definitional reclassification, in an economy that is arguably worse off than before the discouraged workers gave up. This is exactly what makes 'the significance of changes in the employment, unemployment and inactivity rates' a genuinely testable idea rather than spec padding: a falling unemployment rate driven by rising employment is unambiguously good news; a falling unemployment rate driven by rising inactivity can be the opposite; and the headline unemployment figure alone cannot tell you which one happened — reading the employment rate and the inactivity rate alongside it is what actually answers the question.

Traps — 8

ilo-vs-claimant-count

Confirmed directly in an examiner report: "Some candidates got confused with claimant count and were not able to access any marks" when the question specifically asked how ILO unemployment is measured (October 2023 ER, Q12b). The two measures test genuinely different definitions — see the teach section above — and a question that names one specifically expects an answer built from that one's own criteria, not the other's.

frictional-vs-underemployment

The most common wrong answer on a confirmed MCQ swapped these two: candidates picked 'temporarily unemployed while searching for a job' (frictional unemployment) when the correct answer described someone working, but wanting to work more (underemployment) — or the reverse (October 2020 ER, Q4). The fastest check: is the person in the stem actually working at all? If yes, it's underemployment or nothing; frictional unemployment is never the right answer for someone who currently has a job.

net-migration-defined-backwards

Weaker responses on net-migration questions either inverted the definition (describing emigration minus immigration — the reverse of the standard convention) or dropped 'net' altogether and described total immigration on its own (October 2020 ER, discussion of a net-migration/unemployment question). Net migration is immigration minus emigration — the direction sets the sign, and 'net' specifically signals that both flows, not just one, are meant to be netted off.

generic-effects-not-linked-to-the-stem

A pattern confirmed independently across three different series: candidates give textbook effects of unemployment — falling incomes, rising benefits spending — without tying any of it to the specific data given in the question, capping the response at partial marks (October 2019 ER Q7; June 2022 ER Q12d; October 2022 ER Q12d). A correct general effect, unconnected to the country or figures in the stem, earns knowledge but not application.

trade-balance-vs-budget-balance

The single most repeated confusion in the whole archive, confirmed across three independent series in two different question formats: MCQ distractors mixing up a budget surplus with a trade surplus (January 2024 ER Q4; October 2022 ER Q4), and a Section C answer where candidates "confused this with fiscal balance" when the question asked about the net trade balance specifically (January 2022 ER, Q12d). They are answers to two different questions — see the teach section above — and no amount of exam-day pressure makes them the same account.

net-trade-balance-left-undefined

Confirmed independently across two series: "marks were awarded for a definition of the net trade balance, though few candidates offered one" (January 2020 ER, Q9), and "most struggled to define net trade balance" (January 2022 ER, Q12d) — a specific, named weak area, not a one-off. The safe definition to have ready: net trade balance is the value of exports of goods and services minus the value of imports of goods and services — stating 'value' and naming both goods and services, not just 'more exports than imports.'

unemployment-effects-assumed-permanent-and-flat

Every effect on workers and public finances taught in "What unemployment actually costs" above is presented as a flat, permanent consequence of unemployment rising — the real mark scheme expects that assumption to be tested, not repeated. October 2024 MS, Q13 credits eight separate Evaluation-band points against exactly this essay, none of which appear in the KAA teach content above. Workers-side: the rise may reverse in the short or medium term rather than persist; "Some of those unemployed may be employed in the informal sector meaning incomes may not reduce, although impact on public finances may continue" — so lost formal-sector employment doesn't necessarily mean lost income, even though it vanishes from the official figures; and workers' own savings can cushion the income loss the headline job loss implies. Public-finances-side: part of the rise in unemployment may reflect capital-intensive output replacing workers rather than a genuine fall in output, so aggregate demand — and the corporation tax revenue that depends on it — may not actually fall the way the KAA chain assumes; and some of the extra government spending (training support, help for affected firms) can be postponed rather than avoided, softening the immediate hit to the deficit even though the underlying pressure is real. Two further points apply to the whole essay, not one side: "Impact depends on the magnitude of the increase: 1.4 percentage points" — the verdict should scale with the SIZE of the change, not assume any rise is dramatic; and "Unemployment is a lagging indicator, so effects may not be immediately seen – SR vs LR arguments" — confirmed as the examined pattern directly in the October 2024 examiner report: "Evaluation included an attempt to discuss the short-run versus long-run impact and the magnitude of the change in unemployment." This is the single most common Evaluation gap: every KAA bullet taught, the whole Evaluation band left empty.

single-side-effects-essay

A second, separate ceiling sits right next to the country-reference one in this essay's own mark scheme: "NB Award a maximum of Level 3 for answers that consider only workers or public finances" (October 2024 MS, Q13). This is a ceiling on the WHOLE level, not a demotion to the bottom of it — the same logic as the single-conflict ceiling rule (see Macroeconomic Objectives and Policy). A flawlessly developed workers-only (or public-finances-only) answer, diagram and named country included, still cannot cross into Level 4.

The card

AD = C+I+G+(X−M). Movement along AD = price-level change only. Shift = any other determinant.
AD slopes down via: real balance effect, interest rate effect, international trade effect — NOT summed micro demand curves.
Wealth effect ≠ income effect: a STOCK change (assets), not a FLOW change (Yd) — existing owners only, not renters/first-time buyers.
Savings ratio = (S ÷ Yd) × 100. Net investment = gross investment − depreciation.
WEC12 essay: "maximum of level 3 if no reference to a specific country" (June 2024 MS) — name one and use it.
Rising C isn't unconditionally good: check the other AD components, import leakage/current account, spare capacity vs demand-pull inflation, and magnitude (June 2019 MS, Q14 Evaluation band).

Why it works — Why the AD curve slopes down — and why the microeconomic explanation doesn't transfer

A single market's demand curve slopes down because of diminishing marginal utility and the substitution effect: as a good gets relatively more expensive, buyers switch toward substitutes and get less extra satisfaction from each further unit anyway. Neither idea survives the jump to AD. There is no single good called 'the economy's output' that buyers substitute away from when its price rises, and there's no meaningful 'marginal utility of GDP' — AD's downward slope has to be derived from what actually happens to real spending when the ECONOMY-WIDE price level changes, and that turns out to run through three genuinely separate channels, none of them a scaled-up demand curve. First, the real balance effect (sometimes called the Pigou effect, after Arthur Pigou): households hold wealth in forms with a FIXED nominal value — cash, bank deposits, some bonds. When the price level falls, that fixed nominal wealth buys more than it did before — it is worth more in real terms — so households, feeling genuinely richer in what their existing money can buy, spend more (a rise in C). (Other material sometimes calls this channel 'a wealth effect' in passing, precisely because it works through households feeling richer — but don't reach for the house-price wealth effect this lesson teaches later when this is what's actually being tested. The two share a feeling, not a mechanism: the real balance effect is triggered by a change in the PRICE LEVEL revaluing a FIXED nominal stock of money and financial assets, and it explains a movement along AD; the wealth effect below is triggered by a change in the MARKET VALUE of a real asset like housing, with the price level unchanged, and it explains a shift of AD instead.) Second, the interest rate effect: a lower price level, with the nominal money supply held fixed by the central bank, means a given nominal money supply now represents a LARGER real money supply. A larger real supply of money relative to the demand for it pushes down the interest rate that clears the money market, and a lower interest rate encourages more borrowing for both interest-sensitive consumption (durables, cars) and investment (I) — this is the SAME 'interest rate' idea that shows up as an influence on C and I later in this lesson, but operating here through the price level specifically, not through a central bank policy decision (a distinction the trap below makes explicit, because they are commonly and wrongly treated as the same event). Third, the international trade effect: if the domestic price level falls while foreign prices and the exchange rate are unchanged, domestically produced goods become relatively cheaper than foreign goods — exports rise, imports fall, and net trade (X − M) rises. All three effects point the same direction — a lower price level raises real spending — which is exactly why AD slopes downward, and none of the three has anything to do with one good's marginal utility falling as its own relative price rises.

Traps — 6

ad-curve-is-not-a-summed-demand-curve

The single most common wrong explanation for AD's downward slope is a scaled-up version of the microeconomic demand-curve story — diminishing marginal utility, or substitution toward a cheaper alternative. Neither applies: there's no single good called 'the economy's output,' and no substitute economy to switch toward. The real explanation runs through the real balance effect, the interest rate effect, and the international trade effect of a change in the domestic price level — three channels with nothing to do with one good's marginal utility.

wealth-effect-answered-as-microeconomics

Confirmed directly, and confirmed as recurring: "quite a few students answered this as a microeconomic question, and explored the impact of rising house prices on the supply and demand for houses" (Jan 2020 ER, Q12c, Canada context), when the question was testing the wealth effect on consumption/AD. The same house-price/wealth-effect stem recurs in at least three further independent series — Jan 2022 ER Q10, Jan 2023 ER Q10, and Oct 2024 ER Q7 — the single most-repeated confusion trap in this entire spec point.

existing-homeowners-not-the-whole-population

On the same wealth-effect question type, candidates who wrote about first-time buyers instead of existing homeowners scored zero, because the question specifically asked about existing owners (Jan 2022 ER, Q10) — and the direction matters, not just the group: a first-time buyer is made WORSE off by a house price rise (a higher cost to enter the market), not better off, so substituting one group for the other doesn't just miss the target, it can reverse the sign of the effect.

a-policy-rate-cut-shifts-ad-it-does-not-move-along-it

The interest rate effect that makes AD slope downward operates ONLY through a change in the price level (a lower price level → larger real money supply → lower market interest rate). A central bank CHOOSING to cut its policy rate, with the price level unchanged, is a completely different event — it's a determinant of investment and interest-sensitive consumption, exactly like a change in business or consumer confidence, and it shifts the whole AD curve rather than tracing a movement along it. The two only share the words 'interest rate;' the mechanism generating them is not the same.

no-country-reference-caps-you-at-level-3

Confirmed directly from a WEC12 mark scheme, and printed as a standing instruction on this paper's essay question specifically (not a soft steer): "NB Award a maximum of level 3 if no reference to a specific country" (June 2024 MS, Q13). An evaluation essay on AD's components — including this one's natural essay pairing on the wealth effect — needs a genuinely named country worked into the reasoning, not just theory that would apply to any economy interchangeably.

rising-consumption-treated-as-unconditionally-good

A real WEC12 essay question tests exactly this framing directly: "Evaluate the view that rising consumer expenditure will always benefit an economy" (June 2019 MS/QP, Q14, 20 marks: 12 KAA + 8 Evaluation). Its own Evaluation-band indicative content credits exactly the limits above and no others: other components of AD may be falling at the same time, some economies rely more on exports than on domestic consumption, a falling savings ratio funding the rise can raise personal debt and risk demand-pull inflation, and the net effect depends on the level of household incomes and on how significant a share of the economy consumption actually is (June 2019 MS, Q14 Evaluation indicative content — verified directly; that series' own examiner report is separately confirmed corrupted and unrecoverable in this archive, so no ER commentary on how candidates actually answered it is claimed here). An answer that lists the KAA-band benefits — jobs, growth, profits, tax revenue, living standards — and stops there can reach full marks on the 12-mark KAA band, but earns nothing on the separate 8-mark Evaluation band, which needs exactly one of these stated conditions, not a restatement of the benefits already given.

The card

AS = output supplied at a price level. Movement = price change only; shift = cost or capacity change.
SRAS: upward sloping (sticky wages). Shifts: raw-material/energy costs, exchange rate, tax RATE on production.
LRAS: classical = vertical at potential output. Keynesian = flat (spare capacity) → upward → vertical, same point.
LRAS shifts from: technology, productivity, education/skills, regulation/tax POLICY, net migration, competition policy.
Tax sits on both lists: a production tax RATE shifts SRAS; an investment-incentive tax POLICY shifts LRAS.
Diagram: y-axis 'Price level', x-axis 'Real output'. Draw only the curve asked for; no marks for prose.

Why it works — Why SRAS slopes upward: costs lag behind price

Start from a firm that has already signed wage contracts for the coming period — a standard short-run assumption, not a special case. If the general price level rises, the price the firm can charge for its output rises with it, but the wage it pays its existing workforce does not — it's fixed by the contract already in force. Revenue per unit rises while cost per unit stays the same, so profit per unit rises. A firm chasing profit responds to a wider margin by producing more: overtime, temporary staff at the going wage, or simply running existing capacity harder. That's the entire mechanism behind SRAS's upward slope — not 'more output somehow just happens at higher prices', but a specific, temporary profit incentive created by a real gap between how fast output prices move and how fast wage costs move. Because that gap is specifically a short-run phenomenon — wage contracts eventually expire and get renegotiated at whatever the new price level actually is — the incentive is self-cancelling over time, which is exactly the fact the next mechanism block builds on.

Traps — 6

draw-the-curve-asked-for-and-label-it-correctly

Confirmed across at least four series, and confirmed separately as its own distinct error: candidates draw the wrong-run curve, draw both SRAS and LRAS when only one was specifically asked for, or draw the correct curve and simply give it the wrong name — one report notes directly, "A few drew the SRAS and labelled it LRAS" (the same report confirms both classical and Keynesian LRAS shapes are accepted, so the risk isn't picking the 'wrong' shape, it's mislabelling a correctly-drawn one). The hardest version of this is the combined SRAS/AD/LRAS diagram used for output-gap questions: "Only a few candidates did this correctly" — most missed positioning LRAS correctly relative to the short-run equilibrium, and failed to label both the short-run equilibrium and the full-employment (potential) output level. The rule covering all three failure modes: draw exactly the curve named in the question, check which model (classical or Keynesian) it implies before choosing an LRAS shape, and when a combined diagram is asked for, practise placing LRAS relative to a GIVEN short-run equilibrium rather than drawing it on its own.

axis-labels-must-be-macro-not-micro

Confirmed in two separate examiner reports on this diagram type: axis labels reading 'price' or 'price of oil' instead of 'price level' lost marks, and so did "micro labelling for the axis" more generally. This is a macroeconomic diagram — the y-axis names the whole economy's price level, not one good's price, and the x-axis names real output (real GDP), not the quantity of one product.

no-marks-for-prose-on-a-draw-question

Confirmed independently across at least three series and two different diagram questions on this paper — not the same diagram tested twice: on Q10, "Many candidates also offered written explanations for this question, these are not required for 'draw' questions" (repeated in near-identical wording in both the October 2019 and January 2020 examiner reports), and separately, on a different SRAS/LRAS diagram question, Q8, "No further marks for additional text, which some candidates have included to support their diagram" (October 2022). Two different questions, the same rule every time: if a question's command word is 'draw', every available mark is already in the diagram itself — writing a paragraph next to it costs time and earns nothing extra.

tax-appears-on-both-lists

'Tax' is a named SRAS influence (2.3.3.2a) AND a named LRAS influence (2.3.3.3b) — genuinely, not a typo — through two completely different mechanisms. A tax on THIS PERIOD'S production (a specific duty on an input, say) raises unit cost immediately and shifts SRAS. A tax POLICY change aimed at investment incentives (a corporation-tax cut meant to encourage capital spending, for instance) works by changing the economy's future productive capacity, and shifts LRAS only once that investment materialises. Naming 'tax' as an influence without saying which channel is in play doesn't distinguish the two curves at all.

migration-effect-must-be-linked-to-a-stated-lras-shift

Confirmed directly on a real 4-mark question asking candidates to explain one possible effect of a change in net migration on LRAS: "A number of students did not fully address the question and offered impacts of net migration that did not link to a change in the LRAS curve" (January 2020 ER, Q7). A correct, real-world effect of migration — on unemployment, wages, tax revenue, whatever — earns nothing here on its own; the analysis marks are only available once the answer closes the loop and states explicitly what happens to the LRAS curve itself (which direction it shifts, and why), the same discipline this lesson's own conditional-judgement drills above are built to enforce.

no-named-country-caps-the-essay

A general WEC12 essay rule, not specific to AS but directly relevant whenever this topic is examined at length: "NB Award a maximum of level 3 if no reference to a specific country" is printed directly in the mark scheme. An LRAS essay with strong theory and a correct diagram but zero named real economy is capped below Level 4, regardless of how good the reasoning is otherwise.

The card

Equilibrium: I+G+X = S+T+M — the same condition as Y=AD, rearranged, not a separate rule.
Injections: investment, government spending, exports. Withdrawals: savings, taxation, imports.
Multiplier = 1/(1−MPC) = 1/MPW, MPW=MPS+MPT+MPM — one formula, not two (MPC+MPW=1 always).
ΔY = ΔI × multiplier. Multiply, never divide — the most common error on this topic.
Large multiplier needs small MPW. Works both ways — a fall in spending is multiplied too.
Show every step; state the unit (£m/£bn) in the final answer.

Why it works — Why injections equal withdrawals at equilibrium — an identity, not a coincidence

Real output Y can be counted two ways, and at equilibrium both counts must agree. On the expenditure side, everything produced is bought by domestic consumption, investment, government spending, or net overseas demand: Y = C + I + G + (X − M) — this is just AD's own components, the same AD from the aggregate demand lesson. On the income side, every pound of income received is disposed of in exactly one of three ways: spent on consumption (C — some of which is spending on imported goods, which is exactly why the expenditure side has to subtract M back out to isolate spending on *domestic* output), saved (S), or paid in tax (T): Y = C + S + T. At equilibrium these are the same Y, so C + S + T = C + I + G + (X − M). The C on each side cancels — it's literally the same consumption spending counted from two different angles — leaving S + T = I + G + X − M, and adding M to both sides gives S + T + M = I + G + X. Injections equalling withdrawals isn't a separate rule to memorise alongside "AD equals AS" — it's the identical equilibrium condition, rearranged. That's also why 2.3.4.3(b)'s "causes of change via AD/AS shifts" and 2.3.4.2(d)'s "impact of net injections/withdrawals" are describing one mechanism from two labels, not two topics: a rise in an injection (say investment) is, by definition, a rightward shift in AD, and both descriptions land on the identical new equilibrium.

Traps — 5

divide-not-multiply

The single most consistently mis-applied calculation on this whole paper, confirmed across four independent series. October 2020's examiner report states it directly: "The most common error was to divide the increase in government spending by the multiplier ratio, rather than multiplying the two." The same underlying error is separately confirmed in October 2021 ("many did not use the multiplier") and, in near-identical wording across two further series, January 2024 and June 2023: "Many candidates were unable to correctly calculate the MPC. This is because they were not always informed of the [multiplier] equation and hence did not arrive at the correct workings" (January 2024 ER, Q9, omitting the word "multiplier"; June 2023 ER, Q9 names it directly). ΔY = ΔI × multiplier, always: multiply the injection by the multiplier, never divide it.

stops-after-finding-the-multiplier

A two-stage calculation — find the multiplier, then apply it to the actual change in spending — where a documented pattern is stopping after stage one. One examiner report notes candidates who "calculated the multiplier for an additional mark but did not calculate the overall change in GDP after the investment" (January 2020 ER, Q11) — and, separately, that some "did not include the units as billions" even when the arithmetic itself was otherwise right. Finding the multiplier is never the final answer to a multiplier question — it's an intermediate result waiting to be applied.

workings-not-shown

Examiner reports across at least three series (October 2021, October 2022, October 2024) repeat close to the same advice: show every step of the calculation, because partial credit is available for a correct method even where the final figure is wrong — but only if the workings are actually visible on the page. A fully correct final answer earns full marks even with zero working shown — this exact paper's mark scheme says so explicitly ("NB: If correct answer (LE29.75bn) is given, award full marks regardless of working", June 2019 MS, Q11). What workings actually buy you is a fallback: if the final figure comes out wrong for any reason — the wrong operation, a slipped decimal, a units error — showing every step is the only way to still pick up the knowledge and application marks for whichever steps were correct. Never rely on reaching the right number in your head: show every step, so a single slip costs you one mark, not the whole question.

mpw-is-not-just-mps

MPW = MPS + MPT + MPM — all three, not just savings. Treating MPW as if it were MPS alone (the most intuitive of the three, since "saving" is the most familiar leakage) understates MPW and overstates the multiplier — see the worked MCQ below, where leaving out MPT alone turns a correct multiplier of 2.5 into a wrong answer of 4. Whenever a question gives all three propensities, sum all three before taking the reciprocal.

multiplier-explains-ad-not-as

The multiplier is a statement about how far AD shifts once an injection or withdrawal changes (spec 2.3.4.4) — it says nothing about a change in equilibrium output caused by AS shifting instead (spec 2.3.4.3b), for instance from a change in raw material costs or a productivity improvement. A rise in national income following better labour productivity isn't "multiplied" in this sense at all — that's an AS-side story, not an AD-side one, and the two shouldn't be blended into one explanation just because both eventually move the same equilibrium output figure.

The card

Actual growth: real GDP rises now (AD↑, SRAS↑, or a negative gap closing). Potential growth: LRAS/PPF shifts out (investment, FDI, innovation, labour force/migration, competition, productivity).
Output gap = (actual − potential) ÷ potential. Positive → inflation risk (near/beyond capacity). Negative → recession risk (spare capacity).
Benefits (living standards, jobs, profits, investment, tax revenue, reduced benefit spending, public services) follow any GDP rise — only POTENTIAL growth sustains them inflation-free.
Costs: opportunity cost, environment, trade deficit, inequality, inflation (actual growth outrunning potential).
Trend/potential output is estimated, not measured, and gets revised. 2.3.5 (growth theory) ≠ 2.3.1 (measuring GDP) — different points.

Why it works — Why the same 'benefit of growth' can be the trap in a different answer

An examiner marking a growth-benefits-or-costs answer is reading for one specific thing beneath the list: does the candidate treat 'growth' as a single undifferentiated blob, or do they say which kind of growth is doing the work? 'Growth raises tax revenue' is true whether the growth is actual or potential — but 'growth raises living standards sustainably' and 'growth causes inflation' are not both true of the same growth episode at the same time, because they depend on opposite conditions. A rise in real GDP that comes from AD alone, with potential output unchanged, delivers the benefits (more output, more jobs, more tax revenue) exactly as described — right up until it closes the output gap, after which the same mechanism that produced the benefit (more spending chasing more output) starts producing the cost instead (more spending chasing output that can no longer rise, so it shows up as price rises instead). Growth backed by potential growth doesn't hit that wall, because the ceiling has moved along with the demand pulling against it. What moves a mark from 'lists benefits, lists costs' (Level 2) to 'evaluates whether growth is beneficial' (Level 3/4) is exactly this: naming which type of growth is being discussed at the point a benefit or cost is claimed, not asserting both lists as if they always apply together.

Traps — 7

growth-theory-is-not-growth-measurement

2.3.1.1 ('economic growth') is about MEASURING growth — real GDP, GNI, PPPs, real vs nominal — and 2.3.5 ('economic growth') is about its CAUSES, benefits, costs and output gaps. They share a name and nothing else: a question naming 'economic growth' could be testing either, so check whether it's asking you to calculate or define a growth figure (2.3.1) or explain and evaluate why growth happens and what it does (2.3.5) before reaching for content from the wrong section.

how-not-why

Confirmed in an examiner report on a real growth-causes data-response question: candidates who correctly identified a genuine cause of growth from the source often "went on to explain how these would cause growth, rather than why" (January 2021 ER, Q12e discussion) — e.g. citing higher employment and real wages as a cause without tracing the actual mechanism (higher wages → consumer confidence → consumption → AD). Naming a cause is the knowledge mark; tracing its mechanism through to AD or LRAS is the analysis mark, and they are not the same step.

productivity-essay-misread-as-methods

Most candidates who attempted the productivity essay that series largely misread 'evaluate the importance of productivity for economic growth' as 'evaluate methods to increase productivity' — a close-sounding but genuinely different question (importance/consequence versus methods/cause) — and scored accordingly; it was also the less-popular of the two essay choices that series, with 67% of the whole cohort opting for the alternative title instead (October 2019 ER, Q13). Read the command word and the object of the sentence separately: 'importance ... for growth' asks what productivity DOES to growth, not how to GET more productivity.

costs-of-growth-needs-a-named-country

On a real costs-of-growth essay, most candidates did not fully apply their answers to a country of their choice, and a number "lifted the text from the stem in the question as application" instead of bringing in independent, country-specific knowledge (January 2020 ER, Q13) — capped below the top level even where the underlying theory was sound. That's not a soft steer: the real mark scheme sets a hard ceiling, printed directly on the paper — "Award maximum of Level 3 (9 marks) if candidate does not refer to a country/context in their answer" (January 2020 MS, Q13) — meaning Level 4 (10–12/12 KAA) is structurally unreachable without one, no matter how sophisticated the economics on the page. A named real country with a genuine detail about it, not copied from the question's own stimulus, is required for full application credit, not merely encouraged.

evaluation-named-not-developed

The same costs-of-growth essay's Evaluation band shows a distinct failure mode from the country-context one above: most candidates DID raise the right kind of point — weighing the cost's magnitude, or contrasting the short run against the long run — and still scored only Level 1 evaluation, because "the magnitude of the cost of rapid economic growth and the comparison between the short-run and long-run was regularly sited [cited] and rarely developed achieving level 1 evaluation" (January 2020 ER, Q13). Naming a legitimate rebuttal ('this cost is only short-run', 'this cost is offset by more tax revenue') is the Level 1 mention; stating how much it offsets, over what timeframe, and tied to the named country's own circumstances is what a Level 3 developed judgement actually requires — the same gap the worked chain above draws between K/An1/An2 and Eval, applied here to the Evaluation band's own internal levels instead of to KAA.

output-gap-stimulus-reading-traps

Two separate, both independently confirmed, traps show up whenever a question puts an output-gap figure (or two output-gap estimates) in the stimulus. First: candidates define the output gap correctly (knowledge) and correctly read the stem's own figure (application), but stop there — never developing the figure into its actual macroeconomic effect (inflation risk, or spare-capacity/unemployment risk) — a pattern confirmed independently across four series (January 2024 ER Q10, June 2023 ER, October 2021 ER Q10, October 2024 ER). Second: shown a real country's original and later-revised output-gap estimates for the same years, many candidates read the two series as 'original projected GDP versus final actual GDP' rather than engaging with the actual point being tested — that trend/potential output is an ESTIMATE that gets revised as more data and better methods become available, exactly the 'difficulties of measurement' sub-point, 2.3.5.4d (January 2021 ER, Q11). Naming the gap's sign isn't the analysis mark; and two output-gap estimates for the same period is almost always a measurement-difficulty point, not a before/after growth story.

lras-position-on-the-static-diagram

On the combined AD/SRAS/LRAS version of the output-gap diagram, this was reported as one of the single worst-performed diagram questions found across the whole archive: most candidates missed drawing LRAS to the left of the short-run AD/SRAS equilibrium and correctly labelling both the short-run equilibrium and full-employment output (October 2024 ER, Q8) — a related, separately-confirmed error is drawing SRAS correctly but labelling it LRAS (October 2022 ER, Q8). Whichever version of the diagram a question calls for, both the actual (or short-run) level and the potential (or full-employment) level need their own explicit position and label — the time-series diagram above shows the trend/actual version; the dedicated AD/SRAS/LRAS diagram later in this lesson works through the static version this trap is actually about.

The card

Six objectives: growth, low/stable inflation, low unemployment, current-account balance, balanced budget, income equality.
Conflicts: inflation-unemployment (SR Phillips), growth-environment, growth-current account, inflation-current account (export competitiveness), growth-inequality, growth-inflation (a deflationary policy that cuts inflation via the SRAS mechanism also cuts real output). One conflict caps KAA at Level 3 (up to 9/12) — not the bottom of it.
Supply-side = shifts AS/LRAS, not AD. Free-market: removes a government constraint. Interventionist: funds what the market under-provides.
Demand-side = shifts AD. Fiscal = government (G, tax). Monetary = central bank (interest rates, QE, lending criteria, reserve requirements).
No named country also caps Level 3 — on every Section D essay type, not just conflicts.

Why it works — Why a policy that helps unemployment mechanically raises inflation, in the short run

Start from an economy below full employment — a negative output gap, the state a reflationary policy is normally designed to fix. A reflationary demand-side policy shifts AD rightward. Because short-run aggregate supply slopes upward rather than sitting vertical, that single AD shift lands on a NEW intersection with SRAS that has both a higher price level and higher real output at once — the two aren't separate effects of the policy, they're two readings off the same intersection point. Producing that extra real output requires firms to hire more labour (derived demand: nobody demands labour for its own sake, only because of demand for what it makes), so unemployment falls at the exact same moment the price level rises. Nothing about this is a coincidence tied to one particular policy choice — it's what moving along ANY given upward-sloping SRAS curve via AD necessarily does, in both directions: a deflationary policy pushes back down the same curve, lowering the price level and lowering output (raising unemployment) together. That symmetry is the short-run Phillips curve's entire content, and it's why the trade-off is described as structural rather than a policy-specific side effect.

Traps — 8

one-conflict-caps-the-whole-level

Confirmed directly in the October 2023 mark scheme, exact wording: "NB Award a maximum of Level 3 for answers that consider only one conflict." Read this precisely: it is a ceiling on the ENTIRE level, not a demotion to the bottom of it. A single-conflict answer — even with a perfect diagram, precise mechanism and a named country — cannot cross into Level 4 (10–12/12 KAA), but it CAN still climb to the top of Level 3 (up to 9/12) on the strength of its depth. Reading this as 'one conflict = Level 3 entry' (the bottom of the band) — an error independently found in this course's own prior material for this exact topic — costs marks in the wrong direction: it undersells a strong single-conflict answer, and it can wrongly suggest that bolting on a second, thin, barely-mentioned conflict is worth more than it is. What actually unlocks Level 4 is genuine development of a SECOND conflict, not just naming one.

evaluation-must-address-the-claimed-conflict-not-restate-a-downside

Confirmed in an examiner report: candidates who argued 'growth causes inflation and environmental damage' as their evaluation — without addressing whether the SPECIFIC conflict the question named would actually hold in the given context — were marked as off-target evaluation. Evaluation on a conflicts question means stating the CONDITION under which the named conflict holds or breaks down (see the conditional-judgement drill below), not restating growth's downsides in general terms.

no-named-country-caps-the-essay-too

Confirmed directly in a mark scheme, on a supply-side essay: "NB Award a maximum of level 3 if no reference to a specific country." This is the exact same ceiling logic as the single-conflict rule above, applied to a different missing ingredient — and every Section D essay type checked in this course's research carries some version of this requirement, whether the essay is about conflicts, supply-side policy, or demand-side policy. The rule is general to Section D essays, but it applies to THIS conflicts essay just as much as to the supply-side and demand-side essays below, which is why it belongs here rather than waiting for those later sections.

picking-two-conflicts-isn't-enough-if-both-are-shallow

Confirmed in the October 2023 examiner report's own discussion of this essay: most candidates that series DID pick two valid conflicts — the ceiling rule above was well known — but still capped themselves below Level 3 KAA, because both conflicts were argued as shallow, two-stage chains of reasoning rather than developed to the depth a single well-argued conflict would need on its own. Avoiding the single-conflict trap is necessary, not sufficient: two conflicts named but not developed scores worse than the mark scheme's ceiling rule might suggest, because breadth without depth still fails the underlying KAA test each individual conflict is separately marked against.

dont-open-by-defining-and-listing-policies

Confirmed, close to verbatim, in an examiner report on a supply-side essay: candidates "typically started by defining supply side polic[ies], and often listed them" — examiners explicitly advise against this, because it "waste[s] a lot of time in the exam doing this for little reward." A definition earns at most the first knowledge mark; a list earns nothing extra beyond the first item named. Go straight to development: pick two policies, trace each one's mechanism through to the specific objective or variable the question actually asks about.

generic-whole-economy-answer-not-the-asked-variable

Confirmed in an examiner report on a supply-side/unemployment essay: "many candidates explored the impact of supply side policies on the economy as a hole [whole] rather than focusing on the impact on unemployment" — the question specifically asked about unemployment. If a question names a specific objective or variable, every paragraph needs to land back on THAT variable — a generic 'supply-side policy is good for the economy' essay answers a question that wasn't asked.

generic-time-lag-evaluation-earns-almost-nothing

Confirmed, close to verbatim, in an examiner report: "Merely saying that supply side polic[ies] have a 'time lag' will earn a level 1 evaluation mark." The phrase 'time lag' by itself is not evaluation. Explaining HOW the lag operates in the specific case — an infrastructure project's cost is spent immediately, but the extra capacity (and the cost reduction it eventually enables) only arrives years later, by which point the conditions that motivated the policy may already have changed — is what actually moves an answer up the evaluation levels.

monetary-policy-is-not-the-government's-to-claim

Confirmed in an examiner report on a case study that explicitly stated the central bank, not the government, sets interest rates: candidates who wrote about "the government" using monetary policy to raise consumption scored no credit for that part of the answer. Fiscal policy is a government decision (spending, taxation); monetary policy is a central bank decision (interest rates, QE, lending criteria, reserve requirements). In most modern economies the two are institutionally separate specifically so monetary policy isn't driven by short-term political incentive — mixing up which institution does which is directly penalised, not treated as a rounding error.