The Data Door Closes: What the Kejagung's Halt on MBG Evidence Collection Means for Accountability

MBG Watch · 2026-07-17

The premise

On July 10, 2026, Kejaksaan Agung (Kejagung) issued Surat B-3256/F.2/Fd.2/07/2026, signed by Direktur Penyidikan Jampidsus Syarief Sulaeman Nahdi. The letter ordered every Kepala Kejaksaan Tinggi across Indonesia to cease data collection and documentation activities at Satuan Pelayanan Pemenuhan Gizi (SPPG) kitchens under the Makan Bergizi Gratis (MBG) program.

The instruction reversed an earlier directive — Surat B-2668/F.2/Fd.2/06/2026, dated June 15, 2026 — that had initiated the nationwide evidence-gathering effort. The reversal arrived twenty-five days later, and just ten days after the seventh suspect in the MBG corruption case was named: Brigjen Pol Lalu Muhammad Iwan Mahardan, an active-duty police general who served as Sekretaris Deputi BGN.

The official justification, stated by Kapuspenkum Anang Supriatna on July 13: "batas waktu pengumpulan data-data sudah selesai" (the data-collection timeframe has been completed) and the halt order was issued "supaya tidak disalahgunakan" (so it would not be misused). The collected data, he said, would be used to complete case files against the seven existing suspects.

Indonesian Legal Aid Institute (ILAJ) called the decision "pukulan bagi publik" — a blow to the public — arguing the investigation was being shut down before it reached the institutional architecture beyond the seven individuals.

This piece examines the timeline, the stated rationale against standard criminal procedure, the scope of what was and was not collected, the ILAJ critique, and what procedural closure means for the possibility of naming additional suspects — including beneficiaries of the "uang titik" (kickback) scheme documented in Lampung and the "ompreng" forced-purchase scheme run through SPPG kitchens.

The timeline: June 15 → July 10

The interval between the start order and the stop order was 25 days. The interval between the seventh suspect's naming and the halt was roughly 8 days.

The stated rationale: "timeframe completed" and "prevent misuse"

Kapuspenkum Anang Supriatna offered two pillars:

1. The data-collection period had ended. The June 15 letter did not specify a deadline in its public excerpts, but the July 13 explanation treats the activity as having a defined window that has now closed.

2. Prevention of misuse. Anang stated the halt order was issued "supaya tidak disalahgunakan dalam pelaksanaannya" — to prevent abuse of authority in the field.

A third, implicit premise: the evidence gathered is sufficient to complete case files (berkas perkara) against the seven named suspects.

Does a "time limit" on evidence collection in a corruption investigation of this scale hold up?

Under Indonesia's Criminal Procedure Code (KUHAP), there is no statutory maximum duration for the investigation (penyidikan) phase of a general crime. For Tipikor (corruption) cases, the Anti-Corruption Law (UU 31/1999 jo. UU 20/2001) grants KPK and prosecutors broad investigative powers but does not prescribe a fixed calendar window for evidence gathering. Investigation proceeds until the investigator considers the evidence sufficient to forward the case for prosecution (penyidikan selesai → penuntutan).

What is time-bounded are detention periods (Article 24–29 KUHAP) and the overall case timeline once prosecution begins. But the collection of documentary evidence and witness statements from hundreds of SPPG sites across 38 provinces is not, by statute, a calendar-capped exercise.

A 25-day nationwide data-collection window across thousands of kitchen sites — many in remote areas — is, by any operational standard, extremely short. The June 15 instruction required "menginventarisasi dan melaporkan berbagai permasalahan" (inventory and report various problems) at SPPG level. SPPG kitchens number in the tens of thousands. Even a rapid-assessment sweep would require weeks of field deployment, coordination with local prosecutors, and data aggregation.

The explanation that the "timeframe is complete" therefore reads as an administrative determination, not a procedural necessity. The question is what that determination was based on: a genuine assessment of evidentiary sufficiency, or a decision to cap the investigation's reach.

The scope question: What the data collection covered — and what it may have missed

The June 15 instruction targeted SPPG kitchens — the operational nodes where meals are prepared and distributed. This is also where two documented corruption mechanisms operated:

1. The ompreng scheme. Brigjen Iwan (suspect #7) allegedly established a company to sell food trays (ompreng) to prospective SPPG partners at prices he set, with Sony Sonjaya (former BGN Deputy Chief, suspect #2) providing the authorization. Every SPPG kitchen was compelled to purchase these trays. The scheme extracted an estimated trillions of rupiah across the network. (Documented in our pending publication "The Ompreng Toll.")

2. The "uang titik" kickback. In Lampung, business operators reported paying Rp 100–350 million per kitchen location to secure SPPG operating rights — a toll collected by intermediaries with access to the approval chain. Kejagung was reported to have deployed a special team to trace this in mid-June. (Inilampung, Monitor Indonesia, June 2026)

Both schemes operated through the SPPG layer. The data collection ordered on June 15 — inventories of kitchen operations, procurement records, approval documents — would, if fully executed, have produced the documentary substrate for both.

What remains unexamined in the public record

The seven named suspects (as of July 2) are:

1. Ir. Nanik S. Deyang — former BGN Deputy Chief (now BGN Chief)
2. Sony Sonjaya — former BGN Deputy Chief (applied for justice collaborator status; reportedly named 26–41 additional individuals across executive, legislative, and judicial circles)
3. Brigjen Pol Lalu Muhammad Iwan Mahardan — active Polri general, Sekdeput BGN, ompreng architect
4. Three other former BGN officials — linked to the electric motorbike procurement (25,000 units) and foundation-managed procurement
5. One private-sector actor — commissioner/controlling shareholder of PT IAT (electric motorbike contract)

What the public record does not yet show is whether the investigation reached:

• The foundation network. Sony Sonjaya's justice-collaborator application reportedly identified foundations (yayasan) as vehicles for inflated procurement. The June 15 data collection at SPPG level would capture kitchen-side procurement, but the foundation-side contracting — where budget inflation originates — sits upstream, in BGN central and provincial offices.

• The approving officials. The "uang titik" scheme implies officials with authority to grant or deny SPPG operating licenses extracted payment for that discretion. Those officials — at provincial and district levels, across health, education, and BGN channels — are not among the seven.

• The institutional chain. The corruption machine documented in our earlier publication ("The Seven Suspects") connects rigged kitchen approvals → inflated equipment contracts → foundation toll-taking. The seven suspects occupy positions within that machine. The machine's full geometry — who designed the approval workflow, who benefited from the "uang titik" flow beyond the kitchen-level collectors, whether the electric motorbike contract (Rp 1+ trillion) followed the same pattern — remains partially mapped.

If the data collection halted on July 10, the documentary trail at the SPPG layer was frozen at whatever stage it had reached in each jurisdiction. In some provinces, collection may have been thorough. In others, it may have barely begun. The halt order was blanket: seluruh wilayah hukum masing-masing (all legal jurisdictions).

The ILAJ criticism: "A blow to the public"

ILAJ (Lembaga Bantuan Hukum Indonesia) Chairperson Fawer framed the decision as damaging to public trust:

"Kami melihat ini sebagai pukulan bagi publik. Ketika Kejaksaan terus dihantam di media sosial, yang berpotensi melemah bukan hanya institusinya, tetapi juga semangat pemberantasan korupsi yang sedang berjalan."

The criticism is not that the seven suspects should not be prosecuted — it is that the institutional investigation appears to have been capped before the institutional architecture was exposed.

ILAJ's concern reflects a pattern observed in large-scale corruption cases: individual prosecutions proceed while the enabling system — the approval chains, the budgeting fictions, the cross-institutional referral networks — remains intact. The next iteration of the program simply reproduces the vulnerabilities.

Whether ILAJ's assessment is correct depends on what happens next. If the seven case files are completed and additional suspects are named from the collected evidence, the "blow" narrative weakens. If the seven become the final count, it strengthens.

What comes next: "Completing case files" — and then what?

Anang stated the collected data will be "didalami terkait tersangka korupsi BGN" — deepened in connection with the BGN corruption suspects. This is standard procedure: investigators consolidate evidence, draft investigation reports (BAP), and prepare the dossier for prosecution.

Two questions follow:

1. Does procedural closure on the seven foreclose additional suspects?

Legally, no. Article 120 KUHAP allows investigators to add suspects (tambah tersangka) if new evidence emerges during investigation or prosecution. The berkas perkara can be supplemented. The halt order stops new field data collection; it does not legally bar the prosecutor from naming additional suspects based on evidence already gathered, or from reopening collection if the case demands it.

Practically, however, a blanket halt order signals an organizational decision: the investigation phase is being treated as concluded for the current scope. Reopening it requires a new instruction from the Attorney General — a political act as much as a legal one.

2. What evidence would distinguish "procedural pivot" from "preemptive cap"?

The presence or absence of these signals in the coming weeks will clarify the meaning of July 10.

The verdict so far: Two interpretations, one evidence gap

Based on the public record as of July 14, 2026, two readings are consistent with the facts:

Interpretation A — Procedural pivot. The June 15 sweep achieved its operational purpose: it gathered sufficient kitchen-level documentation to substantiate charges against the seven. The halt prevents mission creep, protects field prosecutors from accusations of overreach (referencing the Central Java police letter warning personnel not to appear without proper procedure), and allows the prosecution team to focus on building airtight case files. Additional suspects may still emerge from the existing evidence; the door is not legally closed.

Interpretation B — Preemptive cap. The investigation was reaching the SPPG layer where the ompreng and uang titik schemes materially operated — and where the trail leads upward to provincial approvers, foundation operators, and the cross-institutional machine. Halting collection at ~25 days, days after the seventh suspect (the general who ran the ompreng scheme) was named, freezes the evidentiary frontier at the kitchen door. The "prevent misuse" rationale is a standard bureaucratic shield; the "timeframe complete" claim is administratively convenient but procedurally unusual for a corruption case of this scale.

What would distinguish them

We do not yet have:

• The actual number of SPPG sites where data was successfully collected vs. planned.
• The volume and type of documents seized (procurement contracts, approval letters, bank transfer records).
• Whether Sony Sonjaya's justice-collaborator application has been processed and whether his 26+ names are being pursued.
• Any public statement from Kejagung on whether upstream (foundation/provincial) investigation continues under a separate mandate.

Until those facts surface, both interpretations remain live. The burden of evidence lies with the institution that closed the door: to show what was collected, and to demonstrate that the seven case files do not merely conclude a chapter but also illuminate the rest of the book.

What I'm uncertain about

1. How much data was actually collected in 25 days. A nationwide SPPG sweep is a massive logistical operation. Without a province-by-province completion report, "timeframe complete" is an assertion, not a verified fact.

2. Whether the Central Java police letter (warning personnel not to attend prosecutors' summons without proper procedure) was a genuine procedural concern or a coordinated pushback. The letter circulated just before the halt order. Its influence on the AG's decision is not publicly documented.

3. The status of Sony Sonjaya's justice-collaborator application. If granted, his testimony could extend the suspect list far beyond seven. If stalled, the investigation loses its most direct pipeline to the institutional architecture.

4. Whether KPK will be invited into the upstream space. The electric motorbike procurement (Rp 1+ trillion) and foundation-managed contracts fall squarely in KPK's historical domain. A clean handoff would signal seriousness about the full machine; friction or silence would not.

5. What "misuse" (penyalahgunaan) the halt order was meant to prevent. The Kapuspenkum statement is generic. In the Central Java context, "misuse" could mean prosecutors overstepping — or it could mean the investigation itself was seen as politically inconvenient by actors with influence.

This analysis draws on public reporting from ANTARA, CNN Indonesia, Liputan6, Kompas, Tribunnews, Inilampung, and Monitor Indonesia (July 10–14, 2026), Kejagung press statements, and MBG Watch's prior publications on the seven suspects, the ompreng scheme, and the uang titik allegations. It will be updated as primary documents (the two surat edaran, investigation completion reports, justice-collaborator decisions) become available.