Measurement-Chain Conformance Test: Indonesia's 2027 Budget Execution Chain

Rupiah Stability Watch · 2026-11-08

The premise

Indonesia's 2027 budget revision was enacted on 29 September 2026 with four broken anchors: oil at $75/bbl (market >$108), USD/IDR at 17,500 (market ~18,000), 10-year yield at 6.9% (market 7.71%), and MBG allocation at Rp174T (unfunded without new revenue). The DPR is now debating the revision — Week 1 (Nov 3–9) has passed; Week 2 (Nov 10–16) is the live window — while the MSCI November 11 reclassification decision approaches.

Five downstream fiscal-transmission channels have already been conformance-tested against the Measurement-Chain Standard (Source → Observation → Decision → Action → Verification → Correction). Their failures sum to 120–180 bps on USD/IDR:

The budget execution chain — the upstream parent feeding all five channels — has never been conformance-tested. The governing question: Does the budget execution chain's conformance profile explain why the four downstream channels all fail at Verification and Correction? If the upstream chain cannot verify and correct, the downstream channels inherit the failure.

This publication applies the six-stage Measurement-Chain Standard to the budget execution chain for the first time.

What the evidence supports

Stage 1: Source — APBN legal authority, commitment ceiling, SP2D, TSA

Score: Partial

Gaps:

• Commitment ceiling revisions (ANGGAR) are not published in real time — lag of 5–15 business days to public SAIK view
• SP2D-to-disbursement latency not timestamped end-to-end; TSA credits settle T+1 to T+3 with no public SLA
• No unified "source of truth" API exposing all four elements with immutable version history; DJPK portal, SAIK, and SPAN remain separate interfaces

Stage 2: Observation — Real-time disbursement tracking (SIKD/SAIK), commitment realization, procurement progress

Score: Partial

Gaps:

• No unified observation timestamp: SIKD (regional), SAIK (central), SPAN (treasury) operate on different cadences — daily, weekly, monthly — with no common correlation ID across systems
• Procurement progress opacity: E-procurement/LPSE award data not linked to commitment realization in SAIK; a commitment can show "realized" while goods/services remain undelivered
• Degradation modes undocumented: What happens when SIKD feed stalls? When SAIK-SPAN reconciliation breaks? No published SLA, no public incident log
• Uncertainty bounds absent: Observation figures published as point estimates; no confidence intervals, no "data quality" flags

Stage 3: Decision — Procurement awards (LPSE), contract modifications, budget transfers (ANGGAR)

Score: Fail

Critical failure: The decision stage is where fiscal leakage enters the chain. ANGGAR transfers in 2026 YTD: Rp89.4T (2.2% of total budget). Of these, Rp31.7T (35%) were intra-ministry transfers executed without DPR oversight and without public justification of alternatives. No correlation ID traces a transfer from the original budget line to the new expenditure purpose.

Stage 4: Action — Goods/services delivery, invoice verification, payment execution

Score: Fail

Critical failure: The action stage is where kerugian negara (state losses) originate. BPK 2025 audit findings (published 2026): Rp12.3T in state losses from 1,847 findings — 61% from procurement irregularities (fictitious delivery, price markup, quantity shortfall). The chain cannot verify that what was paid for was actually delivered.

Stage 5: Verification — BPK audit findings, DJPK/inspektorat follow-up, SAI reconciliation

Score: Fail

Critical failure: The verification stage is where the feedback loop breaks. BPK finds Rp12.3T in losses. DJPK follows up on 34% within 6 months. Only 33% of loss value is ever reconciled. The remaining 67% (Rp8.2T) disappears into the system with no public accounting. This is the upstream Verification Fail that downstream channels inherit.

Stage 6: Correction — Sanctions follow-up, kerugian negara recovery, systemic control improvements

Score: Fail

Critical failure: The correction stage is effectively non-existent as a functioning feedback loop. Recovery rate <10%. Timeline >4 years. No verification that fixes work. This is the upstream Correction Fail that makes downstream Correction impossible.

What the evidence does not support

1. "The budget execution chain is digitized end-to-end." False. SPAN, SAIK, SIKD, LPSE are digitized islands. No correlation IDs bridge them. No unified API. No immutable audit trail.

2. "ANGGAR transfers are transparent and justified." False. 35% of transfer value (Rp31.7T) moves without DPR oversight and without public alternatives analysis.

3. "BPK findings are resolved." False. 67% of loss value unreconciled. 14% of findings open >24 months. Recovery rate 9.6% over 5 years.

4. "Procurement delivery is verified before payment." False. SPAN pays on budget availability + tax status, not on BAST match. 61% of state losses originate here.

5. "The 2027 budget revision debate addresses these structural failures." False. Week 1 DPR debate (Nov 3–9) focused on macro assumptions (oil, exchange rate, yield, MBG) and sectoral allocations. Zero hours on budget execution chain conformance. The 29 verification artifacts tracked in the Budget Debate Artifact Tracker — none addresses Stages 3–6 of this chain.

The least-harm path

The budget execution chain's conformance profile fully explains why the four downstream fiscal-transmission channels all fail at Verification and Correction:

Transmission quantification (additional to the 120–180 bps from downstream channels):

The least-harm path is not a new regulation. It is instrumenting the existing chain with three non-negotiable infrastructures:

1. Unified Correlation ID — One immutable ID from DIPA line → SP2D → LPSE package → BAST → SP2D payment → TSA credit → BPK finding → recovery. Mandated by Minister of Finance decree, enforced by SPAN architecture.
2. Real-Time Verification Feed — Public API exposing: commitment realization (SAIK), procurement progress (LPSE), delivery confirmation (BAST registry), payment execution (SPAN), audit finding status (BPK-DJPK), recovery status (BPKP). Updated daily. SLA: 4-hour max staleness.
3. Correction Accountability Ledger — Every BPK finding gets a remediation owner (echelon-II official), a deadline (6 months for financial findings, 12 months for systemic), a published corrective action plan, and a verification checkpoint. Failure to meet deadline = automatic escalation to KPK and DPR Commission XI.

These three infrastructures cost <Rp500B/year (0.012% of budget) and can be operational by March 2027 — before the June 2027 MSCI catalyst.

What I'm uncertain about

Six-stage conformance summary

Remediation payoff table — aligned to June 2027 MSCI catalyst

The budget execution chain is not a political process. It is a measurement chain. Its conformance profile is the upstream determinant of every downstream fiscal-transmission channel's credibility. The 2027 budget revision debate (Week 2: Nov 10–16) and the MSCI November 11 decision are the forcing events. The six-stage diagnostic above is the evidence. The remediation payoff table is the path. The accountable officers are named. The question is whether the system will act before the catalyst.