Deep Dive: Trade Theory and Comparative Advantage

Compiled 2026-09-29. Grounded in research/veridian/WEC14-verified-facts.md — specifically its 4.3.2.1 entry, its "Assessment structure" section (the Section A/B/C mark split, the AO weightings, and the Section C country-gate rule), and its "Per-lesson bullet-by-bullet coverage audits" entry for wec14-trade-theory-and-comparative-advantage (the October 2021 Q9 audit) — every specific fact and citation below traces to one of these, not to memory or to a different paper's material. Cross-read first, to keep this additive rather than a restatement: the live lesson content/veridian/curriculum/wec14/trade-theory-and-comparative-advantage.ts, read in full — its Kestria/Palmira worked chain, diagram, five-item trap-taxonomy, chain-drill, conditional-judgement drill, full five-level essay exemplar, and Ricardo/Heckscher-Ohlin/Krugman beyond-spec block are NOT reproduced here; the three existing WEC14 technique-hub documents in this folder (codex-definitions-and-application-bank.md, forge-exam-technique-guide.md, signal-frequency-and-boundaries.md), all read or grepped in full for the same non-duplication check; and content/veridian/question-bank/wec14.ts (checked directly — no entry for this spec point exists there to duplicate). No new primary source was fetched for this document. Its two genuinely new contributions are a short numeric derivation of the "no comparative advantage" boundary case (extending, not duplicating, the lesson's own second prequestion) and a synthesis that corrects one real gap in signal-frequency-and-boundaries.md's own frequency table for this exact spec point — both grounded in citations the facts bank already carries.

#WEC14 CODEX Deep-Dive — Trade Theory and Comparative Advantage (spec 4.3.2.1)

#What this document is, and isn't

The live lesson already derives the central result of this spec point from first principles: why comparative advantage, not absolute advantage, is the correct basis for specialisation, worked through a full two-country, two-good numeric chain with a diagram, a five-level essay exemplar, and a five-item trap-taxonomy tied to named examiner-report findings. It also already carries the complete, mark-scheme-audited benefits and costs of specialisation and trade (spec 4.3.2.1a), and a beyond-spec block tracing Ricardo through Heckscher-Ohlin to Krugman. None of that is repeated below.

What the lesson states but does not itself derive is why the "no comparative advantage" case it tests as a multiple-choice answer is actually true, and why the constant-opportunity-cost model it builds predicts complete specialisation when real production more often only supports partial specialisation. Both are genuine, spec-relevant sub-cases of 4.3.2.1(b) ("theory of comparative advantage… assumptions; limitations"), and both get a short numeric treatment here that the lesson's own prose only gestures at. This document also completes the beyond-spec theory with two brief, clearly-labelled illustrative sketches (Heckscher-Ohlin, Krugman) that connect back to the boundary case rather than restating the lesson's own paragraph on them, and it closes with a synthesis of every real exam appearance this spec point has produced across the archive — which turns out to correct one small, real gap in this course's own existing frequency table.

---

#Part 1 — Two worked sub-cases the lesson names but doesn't derive

#1. The boundary case: identical opportunity costs, worked through algebraically

The lesson's own second prequestion sets up Country M (4 labour-hours per tonne of steel, 8 per tonne of textiles) against Country N (6 hours per tonne of steel, 12 per tonne of textiles) and gives the correct answer — both countries' opportunity cost of steel works out to exactly 0.5 tonnes of textiles, so neither has a comparative advantage in either good. What the MCQ answer states but never derives is why that equality rules out mutual gains from trade, rather than merely making them harder to find.

The lesson's own worked chain supplies the tool needed to show this, in its stage 5: a trade price only produces a genuine gain for both countries if it sits strictly between their two domestic opportunity costs. Applied to M and N, that condition asks for a price strictly between 0.5 and 0.5 — an empty interval, satisfied by no price at all. The one value that isn't excluded, 0.5 itself, changes nothing: trading a tonne of steel for exactly 0.5 tonnes of textiles is the same trade-off each country already faces producing it at home, so a country's post-trade consumption-possibility line has the same slope as its own production possibility frontier and passes through the same point — it coincides with the frontier rather than lying outside it, the same diagram-level test the lesson's own PPF comparison uses to confirm a genuine gain in the Kestria/Palmira case. Because the interval is empty, no trade price can ever move either country's consumption bundle beyond its own frontier. That is the full mechanism behind the January 2022 Q5 examiner-report finding (facts bank, 4.3.2.1) that candidates had to recognise specialisation and trade have nothing to offer once the two countries' opportunity-cost ratios are already equal — not because the theory breaks down, but because its own stated condition for a gain is, in this one case, impossible to satisfy.

#2. Constant vs increasing opportunity cost: why real specialisation is usually partial

The lesson's chain-drill states, correctly but only in prose, that a straight-line production possibility frontier — and the complete specialisation the Kestria/Palmira chain ends on — rests on an assumption of constant opportunity cost, and that real production more often shows opportunity cost rising as a country pushes further into one good, because the resources redeployed later are progressively less suited to it. The short illustrative schedule below (an original construction for this document, not Pearson data and not the course's own Kestria/Palmira figures) puts a number on why that difference matters for how far specialisation actually goes.

Opportunity cost rises with every step (0.4 → 0.8 → 1.8) because the schedule is bowed outward, not a straight line. Suppose the going trade price for cloth is 0.8 units of wine. Moving from A to C is worth doing: each of those units of cloth costs at most 0.8 wine to make at home, exactly what trade can fetch for it. Moving from C to D is not: the last ten units cost 1.8 wine each to produce, nearly triple what trade returns. A constant-opportunity-cost model — the lesson's own model — has no such stopping point and always predicts complete specialisation, because the cost of the good never rises no matter how much of it is already being produced. An increasing-cost model predicts partial specialisation instead, stopping once the country's own rising opportunity cost catches up with the world price — the real-world qualifier on the lesson's own result, not a contradiction of it: gains from trade still exist, they are simply smaller than the constant-cost model implies, and specialisation stops short of the corner the Kestria/Palmira chain runs all the way to.

#3. Beyond spec: where the opportunity-cost difference itself comes from

The lesson's beyond-spec block already introduces Heckscher-Ohlin (a country's comparative advantage tracks its relative abundance of factors of production) and Krugman's new trade theory (intra-industry trade under economies of scale) in prose. Two short points connect them more concretely to the sub-cases above without restating that paragraph.

Heckscher-Ohlin's mechanism can be sketched numerically in miniature: if a labour-abundant country faces a lower wage relative to the return on capital than a capital-abundant country does, and a good is produced mostly with labour while another is produced mostly with capital, the labour-abundant country's cost of the labour-intensive good is lower purely from its factor prices — before any difference in technology or skill is assumed at all. This is a conceptual sketch, not a claim about any real, verified pair of countries or goods; no primary source in the facts bank supplies numeric Heckscher-Ohlin data for WEC14, and none is invented here.

Krugman's resolution is worth stating precisely against the boundary case above, since it answers a question that case leaves open: if two countries' opportunity costs really were identical — the exact condition proven in §1 to rule out gains from Ricardian trade — classical theory predicts zero trade between them. Real trade between similarly-endowed, similarly-advanced economies is large and mostly in similar goods, which the opportunity-cost model in this lesson has no mechanism to explain. Krugman's answer doesn't need the two countries' opportunity costs to differ at all: if economies of scale mean no single firm can profitably produce every variety of a good just for its own home market, firms in each country specialise in a subset of varieties, and trade lets consumers in both countries reach more varieties than either country's producers could support alone — a real gain from trade that would survive even in the exact zero-comparative-advantage case this document derives in §1.

#4. What's already fully covered, and where it lives

Spec 4.3.2.1(a) — the benefits and costs of specialisation and trade — is not re-covered here: the lesson's two teach chunks under "Two different questions about who's 'better' at making something" already carry the complete, mark-scheme-audited list (trade balance, growth and living standards; the economies-of-scale-to-competitiveness chain; allocative efficiency and the general price level; investment and employment in specialising industries; the foreign-currency-gap risk; occupational immobility and structural unemployment; supply-chain disruption; environmental and resource costs), each traced in the facts bank to the October 2021 Q9 mark scheme. Nothing in this document's Part 1 duplicates that list.

---

#Part 2 — Every real exam angle this spec point has actually been tested from

Section A (1-mark MCQ). Four separate series carry a confirmed, named MCQ on this spec point. October 2020 Q1 and January 2025 Q3 were both reported as well answered — the January 2025 examiner report specifically credits candidates with correctly working out that specialisation and trade raise total world output. January 2022 Q5 was the reported exception: a PPF/opportunity-cost table question candidates performed comparatively poorly on, because the correct answer required recognising the no-comparative-advantage boundary case worked through in Part 1 above. October 2024 Q4 was named the weakest-answered question of that entire series: candidates were reported unable to read an opportunity-cost table correctly in both directions at once — identifying which country held the lower cost in each of the two goods separately, rather than assuming one country must hold it in both.

Section C (20-mark essay). Exactly one confirmed appearance exists in the sampled archive: October 2021 Q9, "evaluate the possible benefits of specialisation and trade to a developed country of your choice" (12 KAA + 8 Evaluation marks), carrying the standard country-gate cap. The mark scheme's own note caps an answer with no named developed country at Level 3 of 4 (a maximum of 9 of the 12 KAA marks), excluding only the top level, Level 4 — not, as a looser reading might suggest, a cap at the lowest KAA level. The examiner report for that series recorded that most candidates could correctly name comparative advantage and economies of scale as benefits, but struggled specifically with the evaluation band, where generic points scored no higher than the lowest level. Read against the unit's own AO weightings (verified in the facts bank's "Assessment structure" section: AO4, evaluation, carries the largest single weighting in this unit at 30% of the unit's marks, ahead of AO1 at roughly 19%, AO2 at 22% and AO3 at 29%), this is a real, if narrow, finding worth naming plainly: the one time this spec point has been tested as a full essay rather than a quick MCQ, the recorded weakness sits exactly where the unit places its heaviest weighting, even though the same underlying mechanism is one of this paper's most reliably well-answered MCQ topics in every other appearance.

Section B (Q7, data-response). No confirmed appearance for this specific spec point exists anywhere in the facts bank's citations. That is stated here as a real, sampled gap rather than a claim that Section B never touches comparative advantage — the facts bank's own citations for this spec point are built from MCQ and Section C evidence only, and no search for a Section B appearance turned up a match.

A correction to this folder's own frequency table. signal-frequency-and-boundaries.md's topic-frequency table records spec 4.3.2.1 as seen in 4 of 14 series, tested only as a Section A MCQ. The facts bank's own later coverage audit (the source for Part 1's boundary-case derivation above) independently confirms a fifth series and a second format: October 2021 Q9's Section C essay, fully audited bullet-by-bullet and already taught in the live lesson. The accurate picture for this spec point is five of the fourteen sampled series, in two different question formats (Section A MCQ and Section C essay), not four series in one format. This is flagged here as a citation-grounded correction for whoever next revises that table — it is not edited in place in this pass, to avoid overwriting a shared file mid-batch.

---

#What isn't here

This spec point had, before this course's own research pass, one of only two pre-existing CODEX lesson builds in the whole WEC14 archive (the other being exchange rates/balance of payments), and its live lesson has since been audited bullet-by-bullet against its one confirmed Section C essay. That leaves comparative advantage the most thoroughly-covered topic in this paper by the time this document was written — which is exactly why this document is short: the two numeric sub-cases in Part 1 and the exam-angle synthesis in Part 2 are the real, citable material this pass found that the lesson states or implies but doesn't itself show in full. No further sub-case, statistic, or exam citation for this exact spec point was found in the facts bank beyond what is used above, so the document ends here rather than padding further.