I want to be clear about something before I get into this. I actually think feeding children is a good idea. School meal programs, when done properly, work. Japan has been running one since 1947. It now reaches 99 percent of public elementary students, and research shows it effectively closes the nutrition gap between low and high income households. 56 percent of ingredients are locally sourced, supporting regional farmers. The program has a legal foundation, certified nutritionists in schools, and has been credited with giving Japan one of the lowest childhood obesity rates in the world. Brazil has had a mandatory school feeding program since 1955, and by law at least 30 percent of ingredients must come from local farms.
These programs work because they were built carefully, with legal frameworks, nutritional standards, and accountability structures. The idea is sound. The execution is what matters.
Indonesia’s MBG (Makan Bergizi Gratis) has the same stated goal. But somewhere between the idea and what actually got built, something went very wrong. And what makes it complicated is that the problems run so deep that stopping or seriously fixing the program has become almost politically impossible.
What’s Actually Wrong With It
Between January 2025 and April 2026, 33,626 students were poisoned from eating MBG meals. That’s across 31 provinces. There were 177 mass food poisoning incidents in 127 districts and cities. The monthly average for 2026 is higher than 2025, meaning this is getting worse, not better.
Why is this happening? Out of 11,592 kitchens operating under the program, only 198 have a valid hygiene and sanitation certificate. That’s 98 percent of kitchens operating without certification. The international food safety standard HACCP is met by exactly 26 kitchens. BGN (Badan Gizi Nasional, the agency running this) only made certification mandatory nine months after the program started, after thousands of kids had already gotten sick.
This is what happens when a program is designed to hit a coverage target as fast as possible rather than to actually work properly.
There’s also the distribution problem. 58.9 percent of MBG’s benefits go to Java. Urban areas get 80.4 percent. Rural areas get 19.6 percent. The regions with the highest stunting rates, Papua, Maluku, East Nusa Tenggara, are the ones least served. BGN admitted this was intentional: they prioritized densely populated areas because it was easier to implement. So the program that’s supposed to fix malnutrition is mostly going to the places that need it the least.
And then there’s the procurement side. BGN bought 21,801 electric motorcycles for around Rp42 million per unit, about Rp1.2 trillion total. All of them are reportedly still sitting in warehouses, not distributed to any kitchen. The Finance Minister had apparently rejected this purchase, but it went through anyway. The Director General of the Budget who approved it was later removed from his position. Members of parliament said they were never consulted. “If it had been brought to us, we would have rejected it,” said the Deputy Chair of Commission IX. On top of that there was a Rp113 billion event organizer contract and Rp100,000 socks!
And when the public started pointing this out, the government’s response was to look outward for credibility. Prabowo visited India and cited their PM POSHAN school feeding program as a model for MBG, then announced it would bring in Indian expertise. This did not land well. A TikTok video went viral showing kitchen conditions at the Karni Mata Temple, a rat temple where rodents are considered sacred and crawl freely through the food preparation area. The creator reportedly gagged from the smell. That became the public face of the Indian expert model. Around the same time, a Deputy House Speaker went viral for saying MBG doesn’t need nutritionists and that a high school graduate is enough to manage nutrition at scale. The professional nutrition association responded that Indonesia has plenty of nutritionists, just no system to deploy them. Which is a different problem, and not one looking to India was going to fix.
The legal framework for all of this? The presidential regulation governing how MBG works wasn’t issued until November 2025, nearly a year after the program launched and started spending tens of trillions of rupiah.
How Much This Actually Costs
The budget in 2025 was Rp71 trillion. In 2026 it was originally Rp335 trillion, recently cut to Rp268 trillion. Nearly a five-fold increase in one year. Rp268 trillion is roughly 7 percent of Indonesia’s entire 2026 national budget of Rp3,842 trillion. For one program!
To understand why that’s a problem, you need to know what condition the budget is already in. The 2026 deficit target is Rp689 trillion. Total government debt is now Rp9,920 trillion. Interest payments alone, just the interest and not the principal, are eating up 19 percent of all state revenue. The internationally accepted safe threshold for this is 7 to 10 percent. We’re at nearly double that!
The budget deficit in March 2026 was 140 percent higher than the same period last year, partly driven by aggressive MBG spending at the start of the year.
Researchers at The Conversation calculated that if MBG were actually targeted, only for those who genuinely can’t afford to eat and adjusted for actual school days, Rp110 trillion would be enough. The difference, more than Rp200 trillion, is the cost of running this as a universal program for political reasons instead of a targeted one for nutritional reasons.
Update (May 19, 2026): Finance Minister Purbaya Yudhi Sadewa confirmed at the May APBN press conference that the MBG budget ceiling has been cut from Rp335 trillion to Rp268 trillion, on direct instruction from Prabowo. He added the Rp268 trillion figure is still temporary and further savings are being calculated. “Stop blaming MBG. The President is improving the management of MBG and how they spend money,” Purbaya said. Whether further cuts actually follow remains to be seen.
So If It’s This Bad, Why Won’t Anyone Stop It?
The criticism is out there. The data is public. The KPK has weighed in. Economists have written about it. And yet the government only trimmed the budget. It didn’t stop the program. That tells you a lot.
There are four reasons, and they build on each other.
First, it got Prabowo elected. MBG wasn’t born from a technical study. It was a campaign promise, the most concrete and visible one from the 2024 presidential election. Prabowo ran in 2014 and lost. He ran in 2019 and lost. In 2024 he won, and MBG was one of the things that made his campaign tangible to ordinary voters. Cancelling it isn’t just a policy reversal. It’s admitting that the most concrete thing he promised people was something he couldn’t deliver.
Second, the program is now feeding the people around him. Indonesia Corruption Watch researched 102 foundation partners running MBG kitchens across 38 provinces. 27.45 percent had provable formal political affiliations. The party most represented was Gerindra, Prabowo’s own party, with seven foundations accounting for 25 percent of all politically affiliated foundations. The full ICW report also documents connections to the 2024 presidential campaign team, military veterans, and police-affiliated foundations. This program is distributing money to people in Prabowo’s political network. Stopping it means cutting off resources flowing to his own coalition.
The KPK has reported suspected trading of SPPG kitchen positions, people paying for the right to operate in certain areas. There are kitchens that were suspended for violations that are still receiving daily incentives of up to Rp6 million per day.
Third, the military and police now have a direct financial stake in it. As of late 2025, the TNI was managing 452 SPPG kitchens with a target of building 2,000. The National Police were managing 672 kitchens through foundations including the Kemala Bhayangkari foundation. Prabowo is a former general. The institutions that form the backbone of his political power now have a direct financial interest in keeping this program running.
Fourth, 62 million people are now eating from it. Per April 2026, 61.96 million people are receiving meals from this program every week. No politician wants to be the person who takes away lunch from 62 million Indonesians. The program has created a dependency that’s almost impossible to unwind, regardless of how poorly it’s managed.
A Question Nobody Seems to Be Asking
Here’s something that bothers me personally. Many Indonesian schools already have cafeterias. Some have been running functional food programs for years, pesantren included. So why did the government build a network of 28,000 centralized kitchens from scratch?
Think about what that model actually requires. Food gets cooked in a centralized location, then transported to schools. Every kilometer of that journey is a point where temperature control can fail, where contamination can happen, where something can go wrong. The more stops in the chain, the more risk. And that’s before you factor in traffic, late deliveries, or a kitchen that’s behind on its schedule.
A school cafeteria that cooks on-site doesn’t have those problems. The food goes from kitchen to student in minutes. Staff are accountable to the same institution. The principal can see what’s being served. There’s no logistics chain to corrupt.
Building centralized kitchens also means building new infrastructure, new procurement channels, new contracts, new vendors. It is, in other words, a far more expensive and complex system than improving what already exists. And complexity, in a program this large and this poorly supervised, is where money disappears.
If the goal were really just to feed children well, building on existing school infrastructure would have been the more logical, cheaper, and safer path. The fact that the government chose to build an entirely parallel system instead is worth thinking about.
Voices That Get Silenced
The reason a lot of people don’t know the full extent of these problems is that criticizing MBG has consequences.
A 2025 journalist safety survey of 655 journalists found 80 percent had self-censored and 72 percent had experienced editorial censorship, stories killed, taken down, or only permitted from certain angles. MBG was the most-censored topic.
A mother in Kampar who posted criticism about her child’s MBG portion on social media was threatened with prison and told to stop posting. Her child was subsequently removed from their PAUD kindergarten via a text message!
A vocational school student in Kudus who wrote an open letter to Prabowo suggesting his MBG allocation be redirected to teacher welfare instead received threatening DMs from someone claiming to be an SPPG kitchen employee, someone with a direct financial interest in keeping the program running.
In February 2025, thousands of students in Papua, across Jayapura, Jayawijaya, Nabire, and Dogiyai, demonstrated against MBG, asking for quality free education instead. Police intercepted them before they reached the protest location and transported them to police stations in trucks. Tear gas was used. In a video that went viral, a uniformed civil servant, the Secretary of the Nabide Education Office, was caught on camera kicking and stepping on a junior high school student sitting on the floor. Amnesty International Indonesia called it a clear human rights violation.
Where I Stand
I think the underlying idea of feeding children is worth pursuing. Stunting is a real crisis. Hungry kids learn worse. Japan, Brazil, and others have proven that school feeding programs can genuinely work, but only when they’re built with legal mandates, proper nutritional oversight, local sourcing, and accountability structures. Not when they’re built to be big, fast, and visible.
MBG is a corruption-ridden program being used to pay back political supporters, structured in a way that benefits Prabowo’s coalition more than the children it’s supposed to serve. It’s denting an already stressed national budget in ways that will take years to fully feel. And it should never have been rolled out nationwide at this speed and scale, without the frameworks that would have made it actually work.
The budget cut Purbaya just announced is the government acknowledging fiscal reality. But a trimmed version of a broken system is still a broken system.
Because it got rolled out fast, wide, and tied to everyone who matters politically, it’s now almost impossible to fix properly. You can’t remove the corruption without dismantling the political networks that depend on it. You can’t retarget it to the poor without taking it away from 62 million current recipients. You can’t slow it down without Prabowo admitting his flagship program failed.
So it will keep going. Probably in roughly its current form. With budget trims here and there to manage the fiscal pressure. With food safety scandals that get quietly forgotten. With critics being pressured into silence. And with a much simpler alternative, improving the schools and cafeterias that already exist, sitting unused because it wouldn’t have generated the contracts, the kitchens, the networks, and the political leverage that this one did.