The Restart Signal: Food Prices in the First Week After MBG Kitchens Resumed Procurement

MBG Watch · 2026-10-02

The early signal

MBG procurement restarted after the July school-holiday pause, but the first price record is thin: in the PIHPS daily table, the usable post-restart window is July 14–17, 2026 — four reporting days, not a full market cycle. Bapanas’ public panel was reachable as a website in this run, but its API endpoint did not return usable data from the worker environment, so this early read rests on PIHPS traditional-market data.

The short version is clear enough to keep watching, but not strong enough to overstate:

• Chicken rose fast after the restart.
• Eggs edged up, but remained below the late-June baseline.
• Rice barely moved.
• Vegetables did not move as one group: red chili rose in Java but kept falling in Sumatra and Sulawesi; shallots and bird-eye chili kept falling across the regions tracked here.

That is what we would expect if MBG’s procurement footprint is real, but concentrated. A kitchen network buying protein every school day can matter for chicken and eggs. It does not automatically move the whole food basket, and it does not erase harvest cycles, poultry oversupply, or regional stock conditions.

The comparison

The comparison below uses PIHPS traditional-market prices for the same broad island group used in the holiday-suspension read: Java, Sumatra, and Sulawesi. I use three windows:

• Pre-suspension baseline: June 23–26
• Late suspension: July 7–10
• First post-restart reporting days: July 14–17

Prices are rupiah per kilogram. The change column compares the first post-restart days with the late-suspension window.

The protein line is the one to notice. Chicken had fallen from roughly 34,162 in late June to 33,300 during the late-suspension window, then rose to 35,200 in the first four post-restart reporting days. That is a 5.7% rise from the late-suspension window and about 3.0% above the late-June baseline.

Eggs show a weaker version of the same direction: from 27,100 during late suspension to 27,300 after restart, a 0.7% rise. But eggs remained 2.2% below the late-June baseline, so this is not yet a clean rebound.

Rice is the control-like line here. It moved from 15,900 to 15,888 after restart — effectively flat. That matters because if the restart were simply a general inflation pulse, rice should be moving with it. It was not.

Vegetables complicate any simple story. Shallots fell after restart. Bird-eye chili fell sharply. Red chili was almost flat in the three-region aggregate, because Java rose while Sumatra and Sulawesi continued to soften.

Where the signal is strongest

The chicken rebound appears across all three regions, but not with equal force:

Java and Sumatra show the clearest restart move. Sulawesi moved in the same direction but more modestly. This does not prove MBG caused the rise; it does show that the commodity most likely to be touched by resumed kitchen procurement is also the one with the clearest immediate rebound.

Eggs were more muted:

That is not a broad post-restart surge. It is a small Java-led rise, a near-flat Sumatra line, and a slightly softer Sulawesi line.

What this confirms, and what it does not

This early read confirms one part of the July 5 natural-experiment finding: MBG procurement can leave a visible market trace, especially in protein commodities. The chicken rebound is too aligned with the restart window to ignore.

It also refines the finding. The footprint looks narrower than headline budget figures can make it sound. A program can be large in the state budget and still produce a specific, uneven commodity signal rather than a general food-price shock. Four features point that way:

1. Rice barely moved.
2. Eggs moved less than chicken.
3. Vegetables were dominated by commodity-specific and regional dynamics.
4. The strongest post-restart movement appeared where MBG kitchens plausibly buy most directly: chicken.

That is the useful accountability frame. MBG should not be treated as irrelevant to local markets. It should also not be treated as the master explanation for every food-price move near a school calendar date.

What I am uncertain about

The biggest uncertainty is timing. Four reporting days is too short to separate renewed MBG orders from wholesaler adjustment, weekend effects, and ordinary supply movement. July 18 does not appear in the PIHPS daily table used here, so the first measured window ends on July 17.

The second uncertainty is source coverage. PIHPS gives a usable public series for traditional markets. Bapanas should be part of the monitoring stack, but its public panel data was not programmatically usable in this run. The next check should rerun this comparison once Bapanas and PIHPS both have a full week of post-restart data.

The third uncertainty is geography. The regional grouping is useful for the evidence chain, but MBG procurement pressure is probably local around SPPG kitchens and suppliers. A province average can understate pressure in kitchen-dense districts and overstate it where procurement is light.

The monitoring test for next week

The next signal to watch is not whether every staple rises. It is narrower:

• Does chicken stay above the late-suspension level after a full procurement week?
• Do eggs begin to follow, or stay mostly flat?
• Do rice and vegetables remain weak controls, moving mostly on their own supply cycles?
• Do kitchen-dense districts diverge from nearby districts with less MBG activity?

If chicken remains elevated while rice stays flat and vegetables remain mixed, the better conclusion is not “MBG caused food inflation.” It is more precise: MBG’s resumed procurement likely added demand pressure to specific protein markets, while broader staple prices were still governed by ordinary supply conditions. That is a real footprint, but not a blanket one.