Level Threshold Guide

FORGE · WEC14 Level/Threshold Guide — compiled 2026-09-29. Every claim below is traced to the specification, a named mark scheme, or a named examiner report (series + question), via research/veridian/WEC14-verified-facts.md unless a different source is named inline. No verbatim mark-scheme bullet list is reproduced at length — each is paraphrased; quotes are kept under ~15 words with attribution, exactly where precision matters (an N.B. gate, an examiner-report diagnosis). The illustrative "just misses / just clears" answers throughout are new prose written for this document from the real credited content — never copied from a mark scheme or from any existing lesson file — the same discipline content/veridian/question-bank/wec14.ts's own level-exemplar states for itself. Not affiliated with or endorsed by Pearson Edexcel.

#WEC14 Level/Threshold Guide (Unit 4: Developments in the Global Economy)

#0. What this document is, and isn't

forge-exam-technique-guide.md already states WEC14's tariff shapes: which parts are flat-additive, which are levels-based, and the real mark ranges each level covers. This document does not restate that — it starts one layer down, at the boundary itself. For each real threshold on this paper's levels-based components, it asks one question: what, concretely, is the difference between an answer that lands on one side of the line and an answer that lands on the other, and what does the mark scheme or the examiner report actually say separates them? Where the evidence supports a worked pair of answers — one that just misses a level, one that just clears it — this document writes that pair, built from real, named mark-scheme content. Where the evidence runs out before a threshold can be illustrated this way, it says so rather than inventing the missing half.

Two existing full worked exemplars already do part of this job for two specific real questions — question-bank/wec14.ts (June 2021 Q9, debt relief and development) and curriculum/wec14/balance-of-payments-exchange-rates-and-competitiveness.ts (June 2022 Q9, international competitiveness). Both are referenced below rather than reproduced. This document's own worked walk-throughs use different real anchors — October 2021 Q10 (current account deficit), June 2025 Q10 (public expenditure), October 2021 Q9 (specialisation and trade) and two Q7(e) fragments — so that the guide adds new ground rather than re-presenting what those two files already cover in full.

---

#1. Which parts of this paper actually have levels at all

Not every mark on WEC14 is level-marked, and a threshold guide is only meaningful for the parts that are. Confirmed directly from the spec-derived breakdown (paper-anatomy.ts) and cross-checked against every sampled mark scheme:

So the whole of this guide concerns 54 of the paper's 80 marks (Q7(e) plus the two Section C essays) — the rest of the paper has no level boundary to describe at all. This is worth stating plainly because it is easy to assume the 8-mark Section B part (which also carries an Evaluation component) works the same way; it doesn't — it is scored as four independent named marks, not a level.

---

#2. Section C essay — the KAA ladder (12 marks, Level 1–4)

Anchor question, real and verbatim: "Evaluate the disadvantages of a current account deficit to a developing country of your choice" (WEC14 mark scheme, October 2021, Q10 — confirmed as Q10, not Q9, by independent re-verification against the primary Pearson PDF). This essay's real KAA indicative content credits nine distinct mechanisms, paraphrased here rather than quoted at length: needing to finance the deficit (risking reserve depletion or IMF borrowing); a link between weak competitiveness and lower growth/rising unemployment; a goods-deficit forcing higher interest rates or asset sales; a foreign-currency-denominated debt burden that worsens if the currency depreciates; over-reliance on consumption alongside an uncompetitive manufacturing base; a net leakage from the circular flow that shrinks output, employment and income through the multiplier; a temptation toward protectionism that itself distorts comparative advantage; and a loss of investor confidence during a downturn cutting off foreign financing. A developing-country N.B. gate also applies (see §4).

The Level 1/Level 2 boundary (marks 3 → 4). Level 1 credits isolated, generic statement(s) with no developed mechanism and no named country; Level 2 requires the answer to be anchored in a real country and to carry at least one point through a genuine two-stage chain.

• Just misses (2–3/12): "A current account deficit means a country imports more than it exports. This can be bad for a developing country's economy because money is leaving the country and it might have to borrow more." — no country is named, and "money leaving the country" never resolves into a named mechanism the mark scheme actually credits (financing risk, competitiveness, the multiplier).
• Just clears (4/12): "Argentina's current account deficit means it has been relying on foreign borrowing and reserves to finance the gap between what it earns from exports and what it spends on imports; if lenders lose confidence, Argentina may need to raise interest rates or sell assets to keep financing it." — a real country is named, and one bullet (financing pressure → higher rates/asset sales) is carried through a first causal step, even though nothing beyond it is developed yet.

The Level 2/Level 3 boundary (marks 6 → 7). This is the boundary the archive's most-repeated finding is about: a two-stage chain — "X causes Y, so Z happens," and stop — caps an answer at Level 2 regardless of how accurate the economics is. This is not a one-off phrasing; it is confirmed in near-identical wording across four series and three different essay topics (terms of trade and aid/development, January 2021; income inequality, January 2024; restated in January 2025's general summary). Level 3 needs a third stage — the chain has to end in a named variable moving in a stated direction, not a vague "the economy suffers."

• Just misses (5–6/12): "Argentina's weak international competitiveness is part of why the deficit persists, and if a country isn't competitive its growth suffers." — two stages (weak competitiveness → "growth suffers"), and "growth suffers" is exactly the kind of unnamed, non-directional outcome the two-stage-chain finding describes; it stops one step short of a scored variable.
• Just clears (7/12): "Argentina's weak international competitiveness constrains export revenue, which — because government spending and net exports are components of aggregate demand — feeds through to slower real GDP growth and, via lower derived demand for labour, a rising unemployment rate." — the same starting mechanism now runs three stages to two distinct, named, directional outcomes (GDP growth; unemployment), which is what the mark scheme's own bullet on this point is actually built around.

The Level 3/Level 4 boundary (marks 9 → 10). Level 4 is not "the same argument written more confidently" — it requires genuinely covering enough of the essay's distinct bullets, each still integrated with the named country, not one mechanism restated three ways. A Level 3 answer of 7–9/12 has developed one or two of the essay's nine real mechanisms into full chains; a Level 4 answer adds at least one further, genuinely distinct mechanism from elsewhere in the same indicative content, still tied to Argentina throughout.

• Just misses (9/12): the Level 3 answer above, extended only by restating the competitiveness point from a slightly different angle ("this also makes Argentina's exports less attractive relative to competitors") — real elaboration, but the same single mechanism, not a second one.
• Just clears (10/12): the same answer, now adding a genuinely separate real bullet — the multiplier-leakage mechanism: "the deficit also represents a net leakage from Argentina's circular flow of income, which — independently of the competitiveness channel above — further depresses output, employment and income through the multiplier." Two distinct, developed, country-anchored mechanisms is the real difference the boundary is drawing, not extra length on one.

---

#3. Section C essay — the Evaluation ladder (8 marks, Level 1–3, no Level 4)

Anchor question, real and verbatim (with a genuinely different marking nuance from §2's anchor): June 2025's Q10, on Italy — "Between 2002 and 2022 Italy's public expenditure as a proportion of GDP increased from 47.1% to 56.7%. Evaluate the likely economic effects of an increase in public expenditure as a proportion of GDP. Refer to a developed country of your choice in your answer." This mark scheme carries a rule not found elsewhere in the sampled archive: "Award positive effects as KAA and negative as evaluation (or vice versa)" — the KAA/Evaluation split here is not fixed to sentiment, so what actually earns the Evaluation marks is not "a negative point" but a genuinely conditional, tested claim, whichever direction it argues. Its real Evaluation content (paraphrased) includes: the effect's significance depends on the scale of the increase relative to GDP; the multiplier's actual size, which could make the same policy negligible or large; whether GDP itself is moving faster than spending; the economy's spare capacity/LRAS elasticity; whether tax revenue is also rising, offsetting the budget effect; the risk that crowding out reduces long-run private investment; and which specific area of spending increased (transfer payments risk work-incentive effects a different area would not).

The Level 1/Level 2 boundary (marks 3 → 4). Level 1 credits a generic evaluative gesture — "it depends," with no stated condition. Level 2 requires one real, named condition that would actually change the verdict.

• Just misses (2–3/8): "The effects of Italy's higher public spending depend on the situation and could be more or less significant in practice." — a real evaluative instinct with no content behind it; nothing here is a specific, checkable condition.
• Just clears (4/8): "The size of this effect on real GDP depends heavily on the multiplier: if Italy's multiplier is low, the same 9.6-percentage-point rise in spending could produce only a small rise in output." — one real, named mechanism (the multiplier) is turned into an actual condition under which the claimed effect would be small rather than large.

The Level 2/Level 3 boundary (marks 6 → 7). This is where "an informed judgement" — the exact examiner-report phrase, confirmed in near-identical wording four years apart (January 2021; January 2025) — becomes the deciding factor, and it is a qualitative requirement, not an additional quantity of evaluative writing. A Level 2 answer raises one or two real conditions in parallel; a Level 3 answer tests a second condition directly against the first and closes with a stated verdict plus the condition it depends on.

• Just misses (6/8): the multiplier point above, plus a second, independent condition stated in parallel: "It also depends on the spare capacity in the Italian economy — if the LRAS is inelastic, the extra spending is more inflationary than expansionary." Two real conditions are now on the page, but they sit side by side; nothing has tested one against the other, and there is no final verdict.
• Just clears (7/8): "…but Italy's budget position need not worsen even with a low multiplier and limited spare capacity, if — as the mark scheme itself notes for a rise of this real scale, 47.1% to 56.7% over twenty years — tax revenue has risen alongside spending. On balance, then, whether this 20-year rise in Italy's public expenditure has been a net benefit depends less on the multiplier in isolation than on whether tax revenue kept pace with it: if it did, the deficit-widening and crowding-out risks are contained; if it didn't, the debt-servicing and private-investment costs this analysis identifies are the ones that dominate." Two conditions are now tested against each other rather than listed in parallel, and the paragraph ends on an actual verdict with its own stated condition — the specific move that distinguishes Level 3 from "more evaluation points."

Why there is no Level 3/Level 4 boundary to describe. The Evaluation ladder on a Section C essay stops at Level 3 (7–8/8) by the mark scheme's own design — confirmed identically across every sampled series, and stated explicitly in paper-anatomy.ts's own spec-derived breakdown. There is no fourth stage of evaluative reasoning this paper credits with a mark that the Level 3 answer above hasn't already reached; a longer version of the same paragraph, however well argued, cannot move the Evaluation mark past 8/8. This matters at the level of technique, not just trivia: time spent writing a third or fourth tested condition after a genuine Level 3 judgement is already on the page is time better spent on the essay's KAA half, which — unlike Evaluation — really does have a Level 4 still to reach (§2).

---

#4. The country/development gate — a hard override, not a ladder rung

Everything in §2 and §3 describes a ladder an answer climbs by adding real content. The gate works differently: it is a ceiling imposed on top of the ladder, confirmed directly in the mark scheme's own printed instruction, regardless of how well the ladder has otherwise been climbed.

The real mechanic, verbatim and short. "Award maximum of Level 3 (9 marks) if a candidate does not refer to a developed country in their answer" (WEC14 mark scheme, January 2024, Q8) — confirmed in near-identical wording across at least 12 of the 13 mark schemes read (October 2020 through January 2025). Some essays stack two such gates on one question: January 2024's Q10 caps separately at Level 3 if no developing country is named, and separately again if the answer discusses economic growth rather than economic development specifically — two independent ceilings on one essay, each capable of holding an otherwise-Level-4 answer down on its own.

Applied to §2's own worked answer. The Level 4 paragraph built in §2 — two distinct, developed mechanisms, both explicitly run through Argentina — earns 10/12 KAA because Argentina is named and the mechanisms are tied to it throughout. The identical paragraph with every occurrence of "Argentina" deleted or replaced with "a country" is not a slightly weaker Level 4 answer; per the gate, it is capped at 9/12 regardless of how developed the two mechanisms are, because the essay's own mark scheme confirms this exact question carries a developing-country gate (verified against the mark-scheme bullet list itself, not inferred). The gate does not shave a mark or two off a strong answer — it overwrites whatever mark the ladder alone would have produced, down to 9, in one step.

Not universal — check the stem, not the pattern. October 2023's Q8 — "Evaluate factors that might cause a country's terms of trade to worsen" — carries no country-gate N.B. at all, because its stem asks generically about "a country," not "a country of your choice." The gate tracks the exact wording printed in front of the candidate that series, not a blanket rule that every Section C essay carries one. Treating the gate as automatic is itself a way to lose time rehearsing a country-naming habit a specific essay may not actually require — and, more dangerously, a way to assume an essay is safe from the gate when it explicitly is not.

---

#5. Q7(e) — both ladders cap at Level 3, and a real gap in what's verifiable

Q7(e)'s 14 marks split into 8 KAA marks and 6 Evaluation marks, each its own Level 1–3 band — confirmed structurally in paper-anatomy.ts and independently in the January 2021 examiner report's own description of this being "a significant change" for the paper. A genuine, stated gap: no source checked for this document — the verified-facts file, the curriculum lesson files, or the other technique-hub documents — states the exact mark sub-ranges (e.g., which marks fall in Level 1 versus Level 2) for Q7(e)'s 8-mark or 6-mark bands the way §2 and §3 state them for the Section C essay. This guide will not invent those ranges. What is verifiable, and load-bearing here instead, is what the examiner reports say actually separates a stronger Q7(e) answer from a weaker one.

The real, confirmed pattern: breadth without depth is the dominant failure, not wrong economics. "A common feature in responses was to try to cover as many measures as possible but without any real development in the analysis" (WEC14 Examiner Report, January 2021) — and, four years later, on a different topic entirely: "It is important that candidates select any two effects and develop their analysis by focusing on those points rather than trying to cover as many effects as possible" (WEC14 Examiner Report, January 2025, Ghana's proposed top income-tax rate). Because neither Q7(e) half reaches a Level 4 the way a Section C essay's KAA half does, the highest band available rewards two points, each carried into a genuine mechanism, over three or more points each stated and left thin.

A near-miss/near-clear pair, built from two real Q7(e) fragments. October 2021's Q7(e) — "Discuss economic factors, apart from primary product dependency, that may constrain Zambia's economic growth" — credits, among other content, the point that the kwacha's response to a shock depends on the price elasticity of demand for Zambia's exports and imports, naming the Marshall-Lerner condition directly. October 2023's Q7(e), on raising Pakistan's top income-tax rate, separately credits a KAA point that tax revenue might rise via the Laffer-curve mechanism, and a distinct Evaluation point that revenue could instead fall if the rate is pushed past the optimum — each an example of a single mechanism carried to a named, conditional outcome rather than a listed consequence.

• Just misses: "Zambia's growth could also be constrained by its exchange rate, and by low government revenue, and by weak infrastructure." — three real topic areas named in one breath, none carried into a mechanism; this is the exact "cover as many measures as possible... without any real development" pattern both examiner reports independently describe.
• Just clears: "Zambia's exchange rate is a further constraint: how far a depreciation of the kwacha actually helps growth depends on the price elasticity of demand for Zambia's exports and imports — the Marshall-Lerner condition — so if that demand is inelastic, as it often is for a commodity exporter like Zambia, a weaker currency may fail to improve the trade balance at all." — one of the two points a Q7(e) answer needs, developed into a named, conditional mechanism rather than left as a label, matching the real credited content above rather than a generic list of "other constraints."

Honest limit on this section, stated rather than smoothed over. The "two-stage chain caps Level 2" finding in §2 is confirmed for Section C essays specifically, across four series. Whether the identical mechanism governs Q7(e)'s own Level 1/2 boundary was not independently checked for this document — it is plausible, since Q7(e) uses the same KAA vocabulary, but no examiner-report quote naming a two-stage-chain ceiling for Q7(e) itself was found in the sources checked. Treat that transfer as a reasonable inference, not a confirmed finding, until a Q7(e)-specific quote is located.

---

#6. All the real thresholds on this paper, in one place

---

#7. Sources and honest gaps

Primary sources used. research/veridian/WEC14-verified-facts.md — the assessment-structure section (spec-derived mark ranges for the Section C essay and Q7(e)); the 4.3.2 audit entry for October 2021 Q9 (specialisation and trade, referenced but not built into a full worked pair in this document, to keep scope honest); the 4.3.3 audit entry for October 2021/October 2024 Q10 (current account deficit, the anchor for §2 and §4); the 4.3.4 entry for January 2024 Q8 (the country-gate verbatim quote in §4); and the 4.3.5 audit entry for June 2025 Q10, Italy (the anchor for §3), including its flexible-marking N.B. content/veridian/curriculum/wec14/paper-anatomy.ts for the spec-derived level-range table and the structural KAA/Evaluation-ceiling asymmetry. content/veridian/question-bank/wec14.ts and content/veridian/curriculum/wec14/balance-of-payments-exchange-rates-and-competitiveness.ts were read in full to confirm this document's own worked anchors are genuinely different questions from their existing level-exemplars, not a restatement. content/veridian/technique-hub/wec14/forge-exam-technique-guide.md was read in full so this document does not restate its tariff-level summary or its own quoted examiner-report lines without adding a new layer to them.

Genuine gaps, stated rather than smoothed over:

• No source checked for this document states Q7(e)'s exact per-level mark sub-ranges (only that each band is Level 1–3 with a ceiling of 8 and 6 marks respectively). §5 works from examiner-report language about what separates the levels instead, and says so.
• The generic, boilerplate wording of Pearson's own Level descriptor table (the printed text describing what "Level 1," "Level 2," etc. mean in the abstract, independent of any one question's indicative content) was not found verbatim in any source checked for this document. Every threshold above is therefore described through mark ranges plus examiner-report and mark-scheme evidence about specific real questions, not through a quoted generic descriptor — a deliberate choice given the "no fabrication" rule this course holds itself to, not an oversight.
• The "two-stage chain caps Level 2 KAA" finding is confirmed for Section C essays across four series and three topics; its application to Q7(e)'s own KAA half in §5 is flagged as a plausible but independently unconfirmed inference, not a verified finding.
• The "informed judgement needed for Level 3 Evaluation" finding (§3) is confirmed twice, four years apart (January 2021, January 2025) — a real, repeated pattern, but not checked against the other eleven examiner reports in the archive, so stated as "confirmed twice," not "confirmed throughout."
• No likelihood percentage or grade boundary is stated anywhere in this document. Every mark range above is the spec/mark-scheme's own stated range, not a frequency claim.