The Evaluation Toolkit: 6 Moves, 8 Sentence Stems

01The structure rule that governs everything below

This is pure exam craft. You are not opening a new topic. You are picking up the moves that earn the evaluation marks: the difference between a correct answer stuck at Level 2 and an A grade answer at Level 3.

The examiner rewards the same handful of moves "again and again" across every theme, so learn them once and reuse them everywhere.

02Mastery checklist: the six evaluation moves

Tick these from memory.

• "It depends on..." I can finish this sentence four ways: it depends on PED/PES (does the elasticity make the effect hold?), time period (short run vs long run), magnitude (how big is the multiplier, how modest the move?), and type of economy (small open vs large domestic; reserves rich vs high debt).
• Short run vs long run. I match the time horizon to the policy: supply-side is slow but durable; demand-side is fast but inflation-prone near capacity; interest-rate cuts hit a liquidity trap; a depreciation worsens the trade balance before it improves it (J-curve).
• Conflicting objectives (H2). I can show at least two genuine conflicts (growth/employment vs inflation; inflation vs unemployment; inclusive vs sustainable; competitiveness vs strong currency) and argue why one objective can't be pursued without cost.
• Government failure / policy constraint. I can name the machinery that limits a policy: time lags, debt and crowding-out, the liquidity trap / zero lower bound, the trilemma, and the policy-versus-cause mismatch.
• Weighing stakeholders. I name who gains and who loses (firms, workers, savers, borrowers, low-income households, exporters) and reach a net judgement, never a list.
• "To what extent." When the command word grades the answer, I grade it: strongest case → strongest counter → how far, anchored on a decisive criterion.
• The structure rule. 10m: knowledge + application + diagram, no evaluation. 15m: KP + evaluation + an evaluative conclusion that takes a stand. I never fence-sit and I never name a stand the body hasn't earned.

03The four-move upgrade: FLAT to EVALUATED

This is the single most important habit this week. Any knowledge point can be upgraded with the same four moves.

1. State the point (the flat KP: what the policy does, with the mechanism).
2. Raise a genuine limitation or condition (when does it not work, or work only weakly?).
3. Bring in a better-matched alternative or counterweight (what would work better, or what offsets this?).
4. Reach a supported mini-judgement (so, on balance, to what extent or under what condition does the point hold?).

Worked example: turning a flat point into an evaluated one

Context: a 15m essay asking whether expansionary fiscal policy is the best way to raise Singapore's economic growth, the kind of question behind 2018 P2 Q4.

04The eight sentence stems, keyed to real questions

Use these stems live in the exam. Each ends in a judgement, never a list.

05The evaluative conclusion: how to take a stand (15m only)

A 15m conclusion must do three things: (1) answer the exact question asked, (2) name the decisive criterion the body established, and (3) state the condition under which your verdict holds.

Templates

06Diagrams: what evaluation does and does not need

Evaluation marks are earned in prose. The evaluative payload is verbal (limits, alternatives, judgement), and there is no diagram unique to these sub-concepts (S17.9 and S17.12).

• In a 15m essay, the diagrams sit in the analysis (KP) section: correctly-shifted AD/LRAS/SRAS/MEI/forex for the policy being assessed. The evaluation section is prose: limits, alternatives, stand.
• Don't try to "draw" an evaluation. Do make sure the KP diagrams are correct, because a wrong diagram caps the whole answer before evaluation can rescue it.
• The one place a second diagram is the evaluation move: the demand-side versus supply-side contrast (an AD-right shift and an LRAS-right shift) shows supply-side policy hitting both channels while demand-side moves AD alone (S17.7, 2023 P2 Q6b).

07Singapore reflexes to deploy in evaluation

These are the recurring Singapore facts that power the evaluation moves.

• MAS managed float, with modest and gradual appreciation. The S$NEER basket-band-crawl; curbs imported and cost-push inflation but cannot fix a demand collapse or domestic demand-pull inflation (S17.4, the limitation that earns the mark).
• Small multiplier. High CPF MPS + high MPM; the load-bearing magnitude point in every SG fiscal essay (S17.1).
• Fiscal prudence, reserves and NIRC. The five-year balanced-budget rule, GIC/Temasek reserves, a COVID-19 package of about SGD$100bn funded without borrowing, so minimal crowding-out, unlike the US (over $37tn debt) or Japan (S17.2).
• Trilemma / impossible trinity. Free capital flows + managed exchange rate force SG to be an interest-rate taker; why interest-rate policy fails for SG (S17.4, the type-of-economy evaluation).
• Supply-side and SkillsFuture. Mid-Career Enhanced Subsidy, Workfare Skills Support, productivity grants; durable but slow, with SG land and scale limits (no wind or nuclear) (S17.6, the short-run-versus-long-run and "it depends" limits).
• Import dependence. Exports + imports >300% of GDP; food diversification (chicken from Brazil and Malaysia, rice from Thailand); why the exchange rate is the potent lever (S17.4/S17.5).

08Recommended practice: evaluation reps

Don't write full fresh essays. Drill evaluation reps only: for each question below, write just the evaluation paragraph(s) and the conclusion, applying a named template. This is the highest-value-per-minute work before the A Levels.

• Type-of-economy reps (15m evaluation + stand only, about 12 min each)

  • 2019 P2 Q4. Does SG's exchange-rate policy suit all economies? Write the "it depends on type of economy" evaluation plus a stand naming market size and import reliance as the decisive criterion (S17.4/S17.12, both).
  • 2018 P2 Q5. Why the exchange rate, rather than interest rates, for SG. Write the trilemma / high-MPM / FDI evaluation (S17.4/S17.3, both).

• Magnitude reps, the small SG multiplier (about 12 min each)

  • 2018 P2 Q4. Write the small-multiplier evaluation that ETG's knowledge map calls the central point; avoid the "SG as a large economy" error (S17.1/S17.9, both).
  • 2024 P2 Q4b. Write the US-versus-SG crowding-out comparison (reserves vs over $37tn debt) (S17.2, both).

• Short-run-versus-long-run reps (about 12 min each)

  • 2023 P2 Q6b. Write the "supply-side too slow for acute shocks, so demand-side better matched" evaluation (S17.6/S17.7, both).
  • 2021 P2 Q4b. Write the liquidity-trap and cyclical-versus-structural-matching evaluation for tax cuts (S17.3/S17.12, both).

• "To what extent" reps (about 12 min each)

  • 2024 P2 Q5b. Write the "investment is not a panacea" graded judgement (S17.6/S17.12, both).

• Stakeholder reps (about 12 min each)

  • 2019 P2 Q2. Write the foreign-worker-levy stakeholder weighting (firms vs workers vs long-run productivity) (S17.6, both).

• Conflicting objectives and Marshall-Lerner reps H2 only

  • 2017 P2 Q6. Write the growth-versus-inflation and deficit conflict evaluation for SG (S17.10). H1: skip, S17.10 is H2 only; substitute an extra type-of-economy rep.
  • 2016 P2 Q6a. Write the appreciation to current-account judgement qualified by PED / Marshall-Lerner and the J-curve. H1: skip, balance of trade and Marshall-Lerner are not in your syllabus.

• The 10m discipline check (do this once)

  • 2016 P2 Q5a. A 10m "explain the conflicts" part. Write it as knowledge + two diagrams with zero evaluation, to lock in the boundary (S17.10, H2; H1 substitutes a 10m both explain). The lesson: knowing when not to evaluate is itself an exam skill.

• Self-check

  • For any flat KP in your notes, run the four moves: state → limitation → alternative → judgement, until the upgrade is automatic. Then write three evaluative conclusions that each name a decisive criterion. If your conclusion would survive being moved to a different essay unchanged, it's a fence-sit, so rewrite it.