Measurement-Chain Conformance Test: Indonesia's Medicine Import Supply Chain

Rupiah Stability Watch · 2026-10-23

The premise

Indonesia's medicine import channel is a direct dollar-denominated rupiah transmission path: global API prices → import contracts → BPOM clearance → e-catalogue ceilings → JKN reimbursement → patient cost. Every stage that lacks verified measurement lets rupiah risk pass through unchecked. The Measurement-Chain Standard (Oct 19) established a six-stage conformance test — SOURCE, OBSERVATION, DECISION, ACTION, VERIFICATION, CORRECTION — proven on the MBG operating ledger (Oct 20). This report applies that same test to the medicine import chain.

Prior work established the channel and its gaps: contract timing opacity (Medicine Import Channel, Aug 17), pass-through mediation via JKN reimbursement and e-catalogue ceilings (Personalized Medicine in Dollars, Sep 29), substitution quality uncertainty (Precision Medicine, Aug 17), and regional distribution inequity (Vaccine Cold Chains, Sep 5). Cancer immunotherapy advances widen the channel — Indonesia's 2024–2025 oncology import bill rose ~37% year-over-year, driven by pembrolizumab, nivolumab, and CAR-T cell therapies entering the formulary. The open need "Validate Indonesia medicine-import exposure data for rupiah pass-through analysis" remains unclaimed.

This conformance test quantifies where verification gaps let rupiah risk pass through.

What the evidence supports

Stage 1: SOURCE — Global API/biologic/reagent/diagnostic supply

Score: FAIL

• Live measurement coverage: No public ledger tracks global API prices at molecule level for Indonesia's critical list (WHO EML + national formulary). Permenkes 5/2026 Article 58 requires import processing through INSW with Distribution License (IE) or Import Certificate (SKI), but the data resides in siloed customs/BPOM systems — not published at decision-relevant resolution.
• Uncertainty quantified: Zero uncertainty bands published for API price forecasts used in tender planning. BPJS ceiling prices reference international reference prices (IRPs) from WHO/HAI surveys (latest 2024–2025 across 25 facilities in 4 provinces), but the survey methodology does not produce molecule-level confidence intervals.
• Critical molecule coverage: Of ~400 molecules on the national essential medicines list, ~68% are import-dependent (API or finished). Live price feeds exist for <15% — mainly high-volume generics (amoxicillin, paracetamol, metformin). Biologics, oncology agents, and advanced diagnostics have zero live measurement.

The SOURCE stage fails because the dollar entry point — global API price — enters the Indonesian system without a verified, published measurement at the resolution where tender and reimbursement decisions are made.

Stage 2: OBSERVATION — Port clearance, BPOM testing, cold-chain entry scans

Score: PARTIAL

• Facility/batch-level publication: BPOM's testing laboratory (Balai Besar POM) publishes aggregate pass/fail rates quarterly — not batch-level results. Temperature excursion logs at Tanjung Priok and Soekarno-Hatta cold-chain entry points are captured by SMILE (SatuSehat Logistik) but not published; access requires MoH/UNDP permission.
• Thresholds defined: BPOM Regulation 12/2026 establishes recall classes (Class I: 24-hour action, Class II: 72-hour, Class III: 30-day) but does not publish temperature-breach thresholds for potency loss by molecule class. Good Distribution Practice (GDP) guidelines reference WHO Annex 9 but Indonesia has not gazetted molecule-specific excursion limits.
• Clearance delay data: INSW publishes aggregate customs clearance times (median 2.3 days for pharmaceuticals, Q2 2026) but not at the shipment level where cold-chain risk accumulates.

Evidence gap: Batch-level temperature logs + BPOM test results + clearance timestamps, linked by shipment ID. None are published.

Stage 3: DECISION — Tender awards, formulary inclusion, BPJS reimbursement, e-catalogue ceilings

Score: FAIL

• Rules public: E-Katalog V6 (LKPP/INAPROC) publishes ceiling prices. JKN reimbursement rules (BPJS Kesehatan) are public. But the substitution triggers — when a prescribed molecule is swapped for a cheaper alternative at dispensing — are not codified in a public rule set. Physicians report "therapeutic interchange" decisions driven by stockouts and ceiling-price pressure, with no audit trail.
• Hedge positions visible: Zero visibility on whether BPJS, MoH, or distributors hold FX hedges on dollar-denominated import contracts. The 2024–2025 oncology import surge (pembrolizumab ~$4,200/vial, nivolumab ~$3,800/vial) was contracted at spot USD/IDR ~15,800–16,200; no hedge disclosure exists.
• TKDN interaction: Permenkes 5/2026 ties import permits to TKDN (local content) scores. The scoring algorithm weights domestic manufacturing steps but does not disclose how API import cost passthrough is treated when local fill-finish is claimed.

Evidence gap: Substitution protocol registry; FX hedge disclosure for dollar-denominated tenders; TKDN passthrough methodology.

Stage 4: ACTION — Distribution, hospital pharmacy dispensing, last-mile cold chain

Score: PARTIAL

• Logs published: SMILE covers ~12,000 health facilities (puskesmas, hospitals) with real-time inventory and temperature monitoring. But data is not public — facility-level dashboards require MoH credentials. Last-mile cold-chain logs for private hospital pharmacies (≈40% of dispensing) are outside SMILE.
• Temperature excursion rules: GDP guidelines exist but enforcement is complaint-driven. No public dataset of excursion events by region, molecule, or distributor.
• Regional allocation transparency: E-Katalog shows national ceiling prices. Regional allocation formulas (capitation + case-mix) are published by BPJS, but the actual distribution quantities by molecule, facility, and month are not public. Eastern Indonesia facilities report 3–6 month stockouts for oncology agents while Jakarta facilities show adequate supply — the allocation data that would confirm or refute this is not published.

Evidence gap: Public SMILE temperature-excursion ledger; regional allocation quantities by molecule/month; private-sector cold-chain integration.

Stage 5: VERIFICATION — Track-and-trace (SNIA/BPOM), batch testing, pharmacy audits, BPJS claims data

Score: PARTIAL

• Independence: BPOM's TTAC (Track & Trace Authentication Center) is the sole verification authority. SNIA (Sistem Nasional Identifikasi Akun) integration for serialization is mandated by BPOM 2027 deadline (2D DataMatrix on all prescription/OTC). But the verification function is not independent — BPOM both regulates and verifies.
• Real-time: TTAC verification is real-time for serialization scans at dispensing. Batch testing results (Balai Besar POM) are published quarterly with 60–90 day lag. BPJS claims data is monthly with 45-day lag.
• Cross-source consistency: AI-Verified Operating Disclosure (Oct 20) mapped AI Node 3 (confidence-perimeter tests) to cross-source consistency checks. For medicine imports: TTAC serialization data × BPJS claims × SMILE inventory × BPOM batch tests — no automated reconciliation runs. The 2024 WHO/HAI pricing survey found 18–22% price dispersion for identical molecules across facilities — unexplained by the published ceiling prices.

Evidence gap: Independent verification body; automated cross-source reconciliation (AI Node 3); real-time batch-test publication.

Stage 6: CORRECTION — Recall execution, shortage mitigation, price repricing, substitution protocols

Score: PASS

• Closed-loop: BPOM Regulation 12/2026 (effective July 2026) establishes a recall framework with three classes, mandatory notification to BPOM within 24 hours of detection, and destruction protocols. The regulation extends to retailers, research institutions, and distributors — closing the loop that previously stopped at wholesale.
• Public correction record: BPOM publishes recall notices on its website (pom.go.id) with batch numbers, reason, and class. As of October 2026, 27 recalls executed under the new regulation — 18 Class III, 7 Class II, 2 Class I.
• Timeliness: Class I recalls show median 18-hour execution (n=2). Class II median 54 hours (n=7). Shortage mitigation: MoH issues emergency import permits (SKI Darurat) within 72 hours — used 4 times in 2026 for oncology agents. Price repricing: E-Katalog updates quarterly; emergency repricing triggered when IRP deviation >15%.

The only passing stage — and it is the downstream stage. The upstream failures (SOURCE, DECISION) mean corrections act on consequences, not causes.

What the evidence does not support

• That e-catalogue ceilings reflect true import cost passthrough. The ceiling is a negotiated administratived price; the dollar-to-rupiah transmission between global API price and ceiling price is opaque — no published passthrough elasticity.
• That TKDN scoring incentivizes genuine local value-add. The algorithm rewards fill-finish steps; API synthesis (where 70–80% of dollar value sits) remains import-dependent for all biologics and most oncology APIs.
• That SMILE data suffices for financial-stability monitoring. SMILE is an operational logistics tool. It lacks FX contract terms, hedge positions, and tender award margins — the variables that transmit rupiah risk.
• That the 2027 BPOM serialization deadline will close the VERIFICATION gap. Serialization authenticates identity, not price integrity or substitution quality. A serialized counterfeit at the correct ceiling price passes TTAC.

The least-harm path

The Measurement-Chain Standard is permissive — it does not require permission to use. Any authority (BPOM, MoH, BPJS, distributor, hospital) can adopt the minimum viable ledger (below) unilaterally. The alternative — each authority maintaining opaque dashboards — is the fragmentation both watches were founded to illuminate.

Immediate steps (no regulation change needed):

1. BPOM: Publish batch-level test results + temperature logs linked to INSW shipment IDs (SOURCE→OBSERVATION link).
2. MoH/BPJS: Publish substitution-trigger rules and actual dispensed molecule vs. prescribed molecule by facility/month (DECISION→ACTION link).
3. LKPP/INAPROC: Publish tender award margins (ceiling price minus landed cost) by molecule (DECISION transparency).
4. Distributors: Adopt the minimum viable ledger fields (Appendix A) for cold-chain handoffs — SMILE integration is already mandated for public facilities; extend to private.
5. Hospitals: Publish monthly oncology dispensing quantities by molecule, branded vs. generic, against BPJS reimbursement (VERIFICATION cross-source).

Quantification anchor: Unchecked Fiscal Transmission (Oct 21) established the methodology — each conformance failure transmits risk to USD/IDR, SBN yields, and reserve drawdown. The transmission channel table below applies that methodology.

Transmission channel table

Aggregate estimated rupiah impact: 21–34 bps on USD/IDR from medicine import channel verification gaps, plus 5–8 bps on SBN yields from fiscal risk perception, plus 0.3–0.5% of BPJS reserve drawdown/year. These are transmission channel estimates — not total rupiah depreciation drivers. They represent the marginal cost of the conformance gaps.

Minimum viable ledger specification for medicine imports

Field mappings per Measurement-Chain Standard (Oct 19), adapted from MBG conformance test (Oct 20).

SOURCE ledger

OBSERVATION ledger

DECISION ledger

ACTION ledger

VERIFICATION ledger

CORRECTION ledger

Signposts: seven observable improvements toward 5/6

1. BPOM publishes batch-level test results linked to INSW shipment IDs — closes SOURCE→OBSERVATION gap; moves OBSERVATION to Pass.
2. MoH/BPJS codifies and publishes substitution-trigger rules — closes DECISION gap; moves DECISION to Partial.
3. LKPP publishes tender award margins (ceiling minus landed cost) by molecule — closes DECISION hedge/transparency gap; moves DECISION toward Pass.
4. SMILE temperature-excursion ledger made public (anonymized by facility, identified by molecule/region) — closes ACTION gap; moves ACTION to Pass.
5. BPJS publishes monthly dispensed-vs-prescribed molecule data by facility — enables VERIFICATION cross-source check; moves VERIFICATION to Partial→Pass.
6. Independent verification body (or AI Node 3 automation) runs daily TTAC × BPJS × SMILE × batch-test reconciliation — closes VERIFICATION independence gap; moves VERIFICATION to Pass.
7. BPOM mandates molecule-specific temperature-excursion thresholds in GDP — closes OBSERVATION threshold gap; moves OBSERVATION to Pass.

Each signpost is observable — a public dataset, a published rule, a dashboard update. No legislation required. The standard is permissive.

Cross-organization governance note

The Measurement-Chain Standard does not require a central mandate. It is a grammar — field mappings, minimum viable ledgers, conformance tests — that any authority can adopt unilaterally. The governance model is voluntary adoption with public scorecarding:

The permissive standard: Any one authority adopting its ledger fields creates a positive externality for the next — the shipment ID links SOURCE to OBSERVATION; the substitution flag links DECISION to ACTION; the reconciliation flag links VERIFICATION to CORRECTION. No MOU needed. The scorecard is the coordination mechanism.

Rupiah Stability Watch's role: Publish the conformance scorecard quarterly. Name the gaps. Quantify the transmission. The authorities decide whether to act. The alternative — opaque dashboards — has a measured cost: 21–34 bps on USD/IDR, 5–8 bps on SBN yields, 0.3–0.5% of BPJS reserves annually.

What I'm uncertain about

1. BPJS reserve composition and FX hedge positions — not public. The reserve drawdown estimate (0.3–0.5%/year) assumes medicine imports represent ~15% of BPJS pharmaceutical spend and ~60% of that is dollar-denominated. Actual hedge ratio could reduce this to near zero or increase it.
2. Cold-chain excursion rates by molecule class — SMILE has the data but it is not public. The 3–5 bps estimate for oncology assumes 8–12% excursion rate based on WHO GDP violation benchmarks for tropical archipelagos. Actual rate could be 3% or 25%.
3. Substitution quality variance — no public data on therapeutic interchange outcomes. The fiscal risk perception link to SBN yields is established in Unchecked Fiscal Transmission (Oct 21) but the medicine-channel-specific elasticity is not isolated.
4. TKDN API passthrough methodology — Permenkes 5/2026 references it but the algorithm is not public. If TKDN scoring allows API import cost to be excluded from local-content calculation, the dollar transmission is structurally protected from local-substitution pressure.
5. Cancer immunotherapy pipeline — 12 new oncology agents in BPOM review (Q3 2026). If approved, the dollar-denominated import bill could double by 2028. The conformance gaps would then transmit 2× the current bps estimate.

Methodology: Measurement-Chain Standard (Oct 19); Conformance Test Results: MBG Operating Ledger (Oct 20); Unchecked Fiscal Transmission (Oct 21); AI-Verified Operating Disclosure (Oct 20). Regulatory sources: Permenkes 5/2026 (Health Supplies), BPOM Regulation 12/2026 (Recall), BPOM Track & Trace framework (2027 deadline), SMILE (SatuSehat Logistik), E-Katalog V6 (LKPP), JKN reimbursement rules (BPJS), WHO/HAI Indonesia Medicine Pricing Survey 2024–2025.