Reuse Before Replacement: The Existing-Infrastructure Test MBG Kitchens Need

MBG Watch · 2026-09-17

The premise

MBG’s next resilience question is not whether kitchens need stronger infrastructure. They do. The question is whether new resilience spending will be measured against the assets already nearby.

BGN has already put the scale on the public record. Its January 2026 note said it was targeting up to 40,000 Satuan Pelayanan Pemenuhan Gizi (SPPG) across Indonesia, with remote and difficult-access areas explicitly in scope, and that SPPGs are meant to serve students, pregnant women, breastfeeding mothers, and young children. The Public Works Ministry’s June handover record, reported by Tempo from Antara, described 222 completed SPPG facilities across 30 provinces, including 3T locations, with kitchens, washing areas, dry and wet warehouses, clean-water networks, wastewater treatment, panel rooms, and other support facilities.

That is not just a construction story. It is an operating-risk map.

If a kitchen loses power, if water pressure fails, if a ferry delay holds protein in the heat, if haze or flood blocks a supplier route, or if a public building nearby already has backup power and water that could support a safe temporary handoff, MBG’s resilience is not found in the label on a device. It is found in the record of what stayed working.

The useful crossing from Rupiah Stability Watch’s reuse-before-replacement argument is measurement discipline: outage hours, cold-chain excursions, generator hours, critical-load maps, fuel stock, battery health, water pumping, refrigeration, maintenance, and actual diesel displacement. MBG should borrow that discipline, not the branding.

What the evidence supports

The evidence supports a simple standard: before BGN treats new resilience procurement as food-safety protection, it should prove that existing assets have been mapped, tested, and found insufficient or useful.

BGN’s own 2026 governance page for the MBG technical guideline says the program is governed under Keputusan Kepala BGN RI Nomor 401.1 Tahun 2025 and that the guideline is in force. BGN’s January article says SPPG location decisions consider beneficiary density, geography, and service reach. WFP’s Indonesia country strategic plan for 2026–2030 frames MBG support together with supply-chain efficiency, monitoring, innovation, and anticipatory action for weather-related risks and shocks.

Those records point in the same direction: MBG is no longer only a meal-production program. It is becoming a distributed public-service network that depends on power, water, roads, ports, storage, sanitation, local labor, school feedback, and health supervision.

That makes an existing-infrastructure test practical. For each SPPG cluster, BGN can ask:

• Which nearby schools, puskesmas, posyandu, markets, public buildings, ports, ferry offices, warehouses, cold rooms, water systems, and PLN assets can support safe food operations during disruption?
• Which of those assets have actual backup power, fuel, water storage, drainage, refrigeration, communications, and staff access — not just a name on a map?
• Which routes have safe alternatives during flood, landslide, haze, heat, maritime delay, or security interruption?
• Which assets are legally and operationally available to MBG in an emergency, and who is responsible for maintenance, activation, cleaning, insurance, fuel, and payment?
• Which assets reduce food-safety risk in a measurable way: fewer temperature breaches, fewer missed deliveries, shorter outage hours, safer stop/go decisions, lower diesel exposure, or faster transfer to a certified kitchen?

The important word is “measurable.” A public building is not resilience because it is public. A generator is not resilience because it exists. A cold room is not food safety because it was purchased. A solar-battery unit is not protection unless the critical load, autonomy hours, maintenance record, and failure mode are visible.

What the evidence does not support

The record does not support a claim that Indonesia has a new MBG resilience-procurement scheme built around solar, batteries, cold rooms, generators, lockers, or sensors. I did not find a fresh Indonesia-specific procurement signal that would justify that allegation.

It also does not support the opposite comfort: that existing MBG infrastructure is already safe enough if formal facilities have been built.

BGN’s own September 2026 safety actions show why. On 16 September, BGN said it had again suspended 1,276 SPPGs that had not met Sertifikat Laik Higiene Sanitasi requirements; 821 were still in registration, 447 had registered but failed eligibility, and 8 had unconfirmed status. BGN also said 654 SPPGs in critical categories had been proposed for permanent closure if they did not meet requirements. The same release says BGN planned field checks on kitchen eligibility, cleanliness, and SOP implementation, and would redirect service temporarily to nearby SPPGs with SLHS and adequate capacity.

That last detail matters. Redirection is already a resilience action. It should have a public operating record: which kitchen absorbed which service area, what extra load it carried, whether delivery time changed, whether menu or temperature controls changed, and whether any beneficiary group missed service.

BGN’s 15 September note about a planned school rating application is also useful but incomplete. A star rating and complaint channel can surface late delivery, menu quality, and food-safety concerns. It cannot replace the infrastructure record. Schools can report that a meal arrived late or unsafe; they cannot be expected to infer whether a feeder line, water pump, generator battery, ferry delay, drainage failure, or cold-room maintenance gap caused the failure.

IDAI’s September warning sharpens the point. In Detik’s report, IDAI argued that food safety standards are strict and should not be handled by teacher tasting; it pointed to repeated poisoning incidents, wrong storage temperatures, cross-contamination between raw and cooked food, and higher risk for toddlers, pregnant women, elderly people, and immunocompromised groups. That is exactly why infrastructure claims must link to food-safety outcomes. The program’s risk does not stop at the kitchen wall.

The existing-infrastructure test

Before new resilience spending is counted as MBG protection, BGN should publish a minimum test for each kitchen cluster.

First, publish the critical-load map. This should identify the loads that must remain powered or backed up: refrigeration, water pumping, lighting for safe preparation, communications, wastewater treatment, food-temperature monitoring, and any digital reporting tools that are needed for stop/go decisions. Non-critical loads should be separated from life- and food-safety loads.

Second, publish the outage and excursion history. Each SPPG cluster should record power outages, water interruptions, cold-chain temperature excursions, route closures, delayed deliveries, and emergency transfers. The record should show duration, cause, corrective action, and whether food was served, discarded, rerouted, or held.

Third, publish the nearby-asset inventory. This is the reuse-before-replacement step. The inventory should include schools with usable kitchens, puskesmas or posyandu with reliable water and power, public buildings with backup systems, ports and ferry nodes, public warehouses, local cold storage, market facilities, water assets, and local grid constraints. For each asset, the record should say whether MBG can use it, under what authority, with which sanitation controls, and at whose cost.

Fourth, publish maintenance ownership. A generator without tested fuel, a cold room without preventive maintenance, a pump without spare parts, or a battery without state-of-health data is not resilience. The public record should name who maintains the asset, the last test date, the failure threshold, and the budget line that keeps it working.

Fifth, publish safe stop/go rules. The rule should be simple enough for a school, kitchen, health officer, and local government to understand: when food may still be served, when it must be discarded, when service shifts to another certified SPPG, when a route is suspended, and when beneficiaries must be notified that the meal will not arrive safely.

This is not a request for a beautiful dashboard. It is a request for a defensible ledger.

The fraud and safety risk

The risk is that resilience procurement repeats the ompreng pattern in a new language.

A program under pressure to scale can be tempted to count visible equipment: more trays, more rooms, more panels, more batteries, more generators, more sensors, more apps. The public sees objects. The budget sees units. Vendors see categories.

Children do not eat categories. They are protected only when the operating chain works: clean water, clean preparation, correct temperature, safe storage, reliable route, reachable school, credible complaint path, and fast stop/go authority when conditions fail.

That is why asset-first procurement is dangerous. It lets a system buy resilience before it has defined the failure it is trying to reduce. A cold room can reduce risk if it prevents temperature breaches for the foods actually used on that route. It can also become a budget object that fails quietly because maintenance, electricity, loading practice, and accountability were not built around it. A generator can reduce outage time. It can also create fuel exposure, noise, emissions, theft risk, and false confidence if the transfer switch, fuel stock, or maintenance contract is missing. A school kitchen can shorten hold time. It can also create new hygiene risk if staffing, certification, and cleaning responsibilities are unclear.

The least-harm path is not anti-technology. It is anti-pretending.

Use existing assets where they demonstrably reduce outage time, unsafe temperature exposure, route interruption, water failure, fuel dependence, or missed beneficiaries. Buy new assets only where the public record shows a specific gap that reuse cannot close. Count resilience only when the food-safety linkage is visible.

The minimum public record

Before any new resilience procurement is treated as a food-safety control, BGN should publish, at cluster level, at least:

1. A map of SPPGs, served institutions, and fallback SPPGs with SLHS status and capacity.
2. A critical-load list for each SPPG: refrigeration, water, sanitation, communications, monitoring, and lighting.
3. The last 12 months of outage, water-interruption, route-delay, and temperature-excursion records where available, with a clear note where records do not yet exist.
4. An inventory of reusable public and local assets within practical reach, including operating authority and maintenance responsibility.
5. Fuel, battery, and generator test records where backup power is claimed.
6. Water and wastewater continuity records where kitchens depend on pumps, tanks, or IPAL systems.
7. Safe stop/go rules for serving, discarding, rerouting, or suspending meals.
8. A post-incident record whenever service is redirected: from which SPPG, to which SPPG, for how many beneficiaries, for how long, and with what food-safety result.
9. A procurement justification for any new resilience device or retrofit that names the failure it is meant to reduce and the existing asset that was considered first.

This should be public enough for parents, schools, district health offices, auditors, and suppliers to know what is being claimed. It does not need to expose children’s identities or sensitive facility vulnerabilities. But it does need to be specific enough to prevent equipment from being counted as safety before it has survived an operating test.

What I am uncertain about

I am uncertain how much of this record already exists inside BGN, local governments, PLN, public works offices, ports, schools, and health offices but has not been made public. The absence of a public record is not proof of the absence of an internal one.

I am also uncertain how standardized the current SPPG asset handover and readiness checks are across ministries and regions. Tempo’s June report says BGN would likely inspect facilities before accepting them, while BGN’s later SLHS suspensions show that facility readiness and hygiene-sanitation compliance cannot be assumed from construction completion.

The final uncertainty is cost. Reusing existing infrastructure sounds cheaper, but only if maintenance, staffing, cleaning, fuel, water, and legal responsibility are counted honestly. “Free” public assets are often not free; the cost is just hidden in another office’s budget.

That is why the test matters. Reuse before replacement is not a slogan. For MBG kitchens, it is a way to keep procurement from outrunning proof.