Every year, somewhere between July and October, a strange ritual unfolds across Southeast Asia. Parents check air quality apps before checking the weather. Schools hand out masks like party favours. Joggers reconsider their life choices. And in the middle of it all, a sullen grey sky settles over millions of people who did not light a single match. This is transboundary haze, the region's most reliable unwelcome guest, and it has been showing up on schedule for more than a quarter of a century.
Put simply, transboundary haze is smoke pollution that crosses national borders and done so for the past 65 years in this part of Southeast Asia. It is born mostly on the Indonesian islands of Sumatra and Kalimantan, where forests and peatlands are cleared and burned to make way for plantations, and it is carried by seasonal winds across the Strait of Malacca and the South China Sea into Singapore, Malaysia, Brunei, southern Thailand and sometimes beyond. The villain of the story is peat. Indonesia holds the largest expanse of tropical peatland on Earth, roughly 22 million hectares of waterlogged, carbon-rich soil that took thousands of years to build. Drain it for farming and it turns from a wet sponge into dry fuel. Once it ignites, it smoulders underground, sometimes for months, laughing at fire crews and refusing to go out until the rains finally arrive. A peat fire does not just burn a field; it burns centuries of stored carbon and exhales it as a toxic cocktail of fine particulate matter, carbon monoxide and greenhouse gases.
The history is longer and gloomier than most people remember. The haze made international headlines in 1997 and 1998, when a ferocious El Niño dried out Indonesia and fires raged across an estimated ten million hectares. Smoke blanketed the region for weeks, a ship collision in the Strait of Malacca was blamed on poor visibility, and the economic damage was put at around nine billion US dollars. Scientists later estimated that the fires released between 0.8 and 2.6 billion tonnes of carbon, the equivalent of somewhere between a seventh and two-fifths of the planet's entire fossil fuel emissions that year. One burning season, in one corner of the world, briefly rivalled the industrial output of continents. The region responded with the ASEAN Agreement on Transboundary Haze Pollution in 2002, a solemn promise to cooperate on prevention and monitoring. Then came the awkward part: Indonesia, the country at the heart of the problem, did not ratify it until 2014, twelve years later and well after the haze had returned.
It returned in style. In June 2013, Singapore's Pollutant Standards Index shot past 400, a record that landed far into the hazardous range and turned a gleaming, efficient city into a place where people stayed indoors and bought out every N95 mask in sight. Then 2015 arrived and made 2013 look like a rehearsal. About 2.6 million hectares of Indonesia burned. The World Bank put the damage at roughly 16 billion US dollars, more than the cost of rebuilding after the 2004 Aceh tsunami. Researchers at Harvard and Columbia estimated that smoke exposure contributed to around 100,000 premature deaths across Indonesia, Malaysia and Singapore. On some days that autumn, Indonesia's fires were emitting more daily greenhouse gases than the entire United States economy. Another severe season followed in 2019, with 1.65 million hectares burned and costs of roughly five billion dollars, and again in 2023, when El Niño reminded everyone that the problem had never really gone away. It had merely been waiting for dry weather.
So why does it keep happening? The honest answer is that fire is cheap and the people who profit from it rarely breathe the smoke. Clearing land with a match costs a fraction of clearing it with machinery, and the ash even fertilises the freshly exposed soil. Palm oil is the great economic engine behind much of this clearing. Indonesia and Malaysia together produce roughly 85 percent of the world's palm oil, and Indonesia alone has well over sixteen million hectares planted. It is the most productive vegetable oil crop on the planet, yielding several times more oil per hectare than soy or rapeseed, and it hides in about half of all packaged supermarket products, from instant noodles and chocolate to shampoo and, awkwardly for the eco-minded, some cleaning products. Palm oil is not the devil. Poorly governed palm oil expansion onto drained peat is. The crop can be grown responsibly on degraded land; it simply is not always grown that way, because the cheapest route to a new plantation often runs straight through someone else's forest.
This is where corporate governance walks into the room, preferably with a better set of excuses. Plantation concessions are frequently held through layers of subsidiary companies, shell entities and nominee owners, so when satellites spot fires inside a concession, finding out who actually answers for them can feel like unwrapping a Russian doll. Companies tend to blame smallholders, smallholders blame the companies, and everyone blames the weather. Many of the big names have adopted No Deforestation, No Peat, No Exploitation pledges, and several have genuinely cleaned up their own estates. Yet pledges are only as strong as the audit trail behind them, and a company can have a spotless sustainability report while the suppliers feeding its mills quietly keep the matches handy. Consumer-country regulation has been slow to close the gap too. The European Union's deforestation regulation, designed to keep forest-destroying commodities out of its market, has been delayed more than once, which tells you something about how readily good intentions give way to lobbying.
Then comes the uncomfortable matter of corruption and politics, the two ingredients that turn a land-use problem into a recurring catastrophe. Permits to convert forest into plantation are valuable, and valuable things attract bribes. Several Indonesian regional officials have gone to prison over forest-conversion permits, including a governor of Riau province, the very heartland of the fires, who was convicted of accepting bribes to change forest zoning for palm plantations. Academic research on Indonesian districts has also found that deforestation and permit issuance tend to rise ahead of local elections, when campaigns need cash and favours are easiest to trade. In other words, the smoke is sometimes funded by the ballot box. Add decentralisation, which handed enormous land-licensing power to district heads after 1998, plus under-resourced enforcement and courts where big polluters have often enjoyed remarkable luck, and you get a system where the incentives to burn are strong and the odds of paying for it are weak.
Foreign policy, meanwhile, has produced some of the most elegant finger-pointing in diplomatic history. Indonesia treats the haze as a domestic land-management issue and bristles at being lectured by neighbours, and in 2015 a senior Indonesian official famously suggested that Singaporeans should be grateful for the many months of clean air they enjoyed the rest of the year. Fair enough, retorts the region, but those smouldering concessions are often owned or financed by companies with headquarters, bank accounts or shareholders in Singapore and Malaysia. Indonesia has a point when it says that its neighbours' corporate citizens are part of the supply chain. Singapore, for its part, passed the Transboundary Haze Pollution Act in 2014, which lets it pursue companies, even foreign ones, whose activities cause haze in the city-state, with penalties that can reach millions of dollars. It has since named and pressed firms linked to the fires, though enforcing Singaporean law on Indonesian soil remains a diplomatic and legal tightrope. Malaysia has its own historic burning problems in peat areas and a habit of closing schools first and asking questions later, while Thailand's northern provinces suffer annual smoke from crop burning that gets blamed on everyone except Thailand. Each government has a story in which the blame sits comfortably with someone else, which is exactly why a regional problem keeps being treated as a series of national ones.
And this is the real reason haze belongs in the global climate conversation. Peat fires are a climate disaster hiding inside a public health emergency. They release carbon that cannot be replanted away within a human lifetime, and they do so at moments when the world is promising to cut emissions. Indonesia has made real moves, including a moratorium on new permits in primary forest and peatland that was made permanent in 2019, a peat restoration agency created after the 2015 disaster, and rewetting efforts across millions of hectares. Those efforts are one reason burned area has been well below 2015 levels in several recent years. But progress is fragile, and a dry year with a strong El Niño still sends hotspots soaring. Climate action at the global level has the same disease in a larger body: pledges get announced in grand halls, while the people paid to turn them into enforcement are underfunded, underpowered or quietly working for the other side. Greed explains why the fire is lit, corruption explains why nobody is stopped, and politics explains why everyone agrees it is terrible and nobody agrees whose job it is to fix.
So how does it actually stop? Not with another summit communique, though the region has certainly tried that. It stops when the economics flip, so that leaving peat wet is more profitable than draining and burning it. That means rewetting and restoring peatlands at scale, backing it with continuous satellite monitoring that is public, so anyone can see which concession is burning today. It means holding the parent companies, and not just the local contractors, legally and financially responsible, and publishing who really owns every concession so the Russian doll can finally be unpacked. It means banks and investors refusing to lend to anyone with a fire record, and big buyers refusing to purchase from mills whose supply chains cannot be traced. It means paying smallholders a living wage for land-clearing alternatives rather than treating them as the convenient scapegoat, since the cheap fix of fire is often the only affordable option they have. It means ASEAN replacing polite cooperation with something that has teeth, including shared data and joint enforcement. And it means importing countries, including the ones with the cleanest skies, demanding forest-friendly oil and sticking to those rules when lobbyists come knocking.
Here is the cheerful part, because there is one. Haze is one of the rare environmental disasters with a plainly visible cause, a known fix and a ready-made list of people who could act on it. We do not need a miracle technology to put out a fire that humans choose to light. We need honest ownership records, working courts, consumers who read labels, and leaders who treat clean air as a shared asset instead of someone else's problem. The smoke has never respected a border, a passport or a diplomatic note, and it is not going to start now. Until the incentives change, the region will keep checking the air quality app each morning and hoping the wind blows somewhere else. The sooner we all admit that there is no somewhere else, the sooner the sky can go back to being blue.