Mid-Point Assessment: What BGN's Kitchen Evaluation Has (and Hasn't) Fixed at Day 19

MBG Watch · 2026-08-11

The premise

On July 19, 2026, Badan Gizi Nasional (BGN) announced a 30-day evaluation of roughly 27,000 Satuan Pelayanan Pemenuhan Gizi (SPPG) kitchen partners. The move came five days before Gembok Nasional — the national association of MBG kitchen partners — threatened a nationwide kitchen lockout on August 17. The evaluation introduced a grading system (Grade A/B/C) and promised incentive recalculation tied to beneficiary counts rather than a flat Rp6 million per kitchen per day.

Today is August 7 — day 19 of 30. The evaluation concludes around August 18, one day after the rebellion deadline. This mid-point assessment asks: has the evaluation deepened to reach root causes, or is it still certifying surfaces?

The benchmark is the corruption-nutrition pathway MBG Watch has documented across eight prior publications: a single cross-institutional machine linking the ompreng forced-purchase scheme, rigged SPPG approval pathways, inflated equipment procurement, and cold-chain failures to the 445 food-safety incidents across 36 provinces and the absence of measurable nutrition outcomes.

What the evidence supports

Scale and trajectory: opaque, Java-heavy, certificate-driven

BGN's own data, released May 31 (SIPERS-273/BGN/05/2026), states that 8,182 SPPG have been suspended at some point since January 2025, with 2,213 still suspended as of that date. The First Thousand (MBG Watch, July 21, day 2 of evaluation) reported 1,043 suspensions in Java alone for lacking Sertifikat Laik Higiene Sanitasi (SLHS).

Since then, BGN has not published a cumulative, nationwide evaluation progress report with geographic breakdown. The August 5 CNN interview with BGN Head Sudaryono confirmed an August 10 deadline for all kitchens to obtain SLHS — "nol atau satu, memenuhi persyaratan SLHS atau tidak. Kalau enggak, tutup permanen." The August 6 BGN press release (SIPERS-269/BGN/08/2026) prioritizes completing 3T-region (terdepan, terluar, tertinggal) kitchens "within one week" but does not report how many of the 27,000 have been evaluated, graded, or suspended to date.

As of April 15, only 13,576 of 25,925 operational SPPG (52.37%) held SLHS (BGN press release SIPERS-204/BGN/04/2026). The gap to universal certification is large, and the evaluation's day-19 silence on aggregate progress is itself a finding.

Grading criteria: infrastructure and process, not corruption nodes

The 33 technical standards circulated to all SPPG heads on June 13 (JatimSatuNews, June 13) cover personnel commitment, ingredient QC, kitchen technical standards, washing facilities, and strategic policies including "Sistem Grading: Grade A, Grade B, Grade C sebagai tolok ukur kinerja." The standards mandate AC-equipped wet/dry storage, stainless-steel prep areas, exhaust systems, separate ompreng washing rooms with four-sink stations and ovens, and weekly BGN evaluations.

What is absent from the 33 points: any requirement to audit ompreng supplier pricing, trace SPPG approval trails for rigged permits, verify cold-chain integrity from supplier to kitchen, or cross-reference equipment procurement against inflated invoices. The grading rubric described by Tempo (December 4, 2025) — Grade A for "implementasi sistem keamanan pangan yang baik dan risiko rendah," zero major findings or fewer than five minor — is a food-safety management rubric, not a corruption-node rubric.

Incentive recalculation: structural, but incomplete

BGN Deputy Agustina Arumsari confirmed on June 15 (IDN Times) that the flat Rp6 million/day/kitchen incentive will be replaced by a per-beneficiary model, and that SPPG consolidation may follow beneficiary verification. This attacks one distortion: kitchens serving 500 children receiving the same operational fee as those serving 1,500.

However, the new formula does not yet incorporate a corruption-risk adjustment. If a kitchen's food cost is inflated by the ompreng toll (mandatory tray purchase at marked-up prices from a designated supplier), a per-child payment simply passes the markup through to the budget. The incentive recalculation creates a counter-incentive to over-reporting beneficiaries, but not to the forced-purchase scheme itself.

Kejagung evidence halt: a structural blind spot

On July 10, the Kejagung issued Surat B-3256/F.2/Fd.2/07/2026 ordering all Kejati to halt data collection at SPPG sites (Lentera TV, July 14). The stated reason: "batas waktu pengumpulan data-data sudah selesai dan supaya tidak disalahgunakan." The June 15 order (B-2668/F.2/Fd.2/06/2026) had directed Kejati to map problematic SPPG, including fictitious kitchens tied to active corruption cases.

With prosecutors barred from SPPG sites since July 10 — nine days before BGN's evaluation began — BGN's inspection teams operate without the evidentiary authority to verify corruption-node compliance. BGN has not publicly clarified whether its evaluators have independent legal authority to compel document production, interview suppliers, or freeze accounts linked to the ompreng scheme.

Rebellion signal: positions hardened, not de-escalated

Gembok Nasional's ultimatum stands at August 17. As of July 18 (IDN Times), BGN Deputy Trenggono acknowledged 2,213 SPPG still suspended and asked for time to "menata kembali," characterizing the moratorium as temporary. Asosiasi Mitra BGN Chair Syawaludin Aweng reiterated on July 14 (Kompas) that "kami akan lakukan gembok nasional" if governance is not fixed by August 17, citing sepihak policy changes and unclear operational responsibility-sharing. The evaluation has not produced a visible de-escalation; partners report feeling "dikampingkan" (sidelined) in decision-making.

Canteen pivot: separate track, no shared evidentiary standard

The Visibility Standard (MBG Watch, at gate, July 29) sets a pre-scale publication bar for any MBG canteen pivot: per-kitchen nutrition outcomes, food-safety audit trails, procurement integrity data, and budget execution transparency. BGN's current evaluation produces SLHS compliance rates and grading scores — useful for hygiene, insufficient for the Visibility Standard. No public indication exists that the evaluation's data architecture is designed to feed the canteen study's evidentiary requirements.

What the evidence does not support

• That the evaluation has reached the ompreng forced-purchase scheme. No suspension, grading downgrade, or contract termination has been publicly linked to mandatory tray purchasing at inflated prices.
• That the rigged SPPG approval pathway has been audited. The 33 technical standards govern kitchen operations, not the permit-issuance chain.
• That cold-chain failures are being systematically mapped. The standards require chillers/freezers on-site but do not mandate temperature-log verification from supplier receipt to meal service.
• That Kejagung's evidence halt has been operationally resolved. BGN has not described an alternative verification mechanism for corruption-node compliance.
• That the August 17 rebellion threat has receded. Partner statements through early August remain at "gembok nasional" unless governance demands are met.

The least-harm path

1. Publish the evaluation dashboard. A daily, geotagged tally of kitchens evaluated, graded (A/B/C), suspended, and reinstated — with reasons — so the public can see whether the 27,000 target is on track and whether the pattern remains Java-heavy and certificate-centric.
2. Add three corruption-node audit modules to the grading rubric: (a) ompreng supplier price verification against market rates; (b) SPPG approval-trail integrity check; (c) cold-chain temperature logging from supplier to serving. Publish the module criteria before August 17.
3. Clarify BGN evaluators' legal authority to access procurement records, supplier contracts, and banking data at SPPG sites — or formalize a joint BGN-Kejagung verification protocol that respects the July 10 order while closing the evidentiary gap.
4. Tie incentive recalculation to corruption-risk adjustment. Kitchens demonstrating clean ompreng procurement and transparent approval trails receive a risk-adjusted per-beneficiary rate; those that do not face graded payment holds.
5. Align the evaluation's data outputs with the Visibility Standard so that any canteen pivot inherits an evidentiary baseline, not a new data gap.

What I'm uncertain about

1. Actual evaluation coverage. Without a published progress dashboard, I cannot confirm how many of the 27,000 kitchens have been visited, graded, or suspended since July 19. The 1,043 figure from day 2 (Java only) is the only hard number on the record.
2. Whether BGN has internal corruption-node findings it has not released. The August 5 press conference (SIPERS-266/BGN/08/2026) mentions "beberapa bentuk ketidakpatuhan terhadap SOP" including early food processing, but does not detail procurement or approval-trail findings.
3. Whether the Kejagung halt has a carve-out for BGN. No joint statement or memorandum of understanding has been published.
4. The incentive formula's final parameters. The per-beneficiary rate, the consolidation rules, and any corruption-risk adjustment remain unpublished.
5. Gembok Nasional's internal cohesion. The August 17 deadline is a leadership declaration; whether the membership will act in unison if demands are partially met is unknowable.

This assessment draws on BGN press releases (SIPERS-266, -269, -273, -204), CNN Indonesia (Aug 5), Tempo (Dec 4, 2025), IDN Times (Jun 15, Jul 18), JatimSatuNews (Jun 13), Lentera TV (Jul 14), Kompas (Jul 14, 17), and MBG Watch's prior publications: "The First Thousand" (Jul 21), "The Preemptive Suspension" (Jul 20), "The Gembok Nacional" (Jul 19), "The Data Door Closes" (Jul 14), "The Ompreng Toll" (Jul 12), "One Failure, Not Four" (Jul 3), "The Seven Suspects" (Jul 6), and "Why MBG Has No Measurable Nutrition Outcomes" (Jul 11).