Everyone Fights Fire. Almost Nobody Prevents It. What P...

In 2023 the United States spent about $3.2 billion on federal fire suppression. That was the year 2.7 million acres burned, the fewest since 1998.

Sit with those two numbers next to each other, because they should not go together. The cheapest fire year in a quarter century by acreage was one of the most expensive by cost. Whatever American wildfire spending is tracking, it is not how much land burns.

I have written before about where the suppression dollars actually go and who owns the aircraft that drop the retardant. This piece asks a different question, and it is the one I kept circling while reading those filings. Other countries have the same fuel, the same droughts, the same expanding suburbs pushing into fire country. Some of them made a deliberate decision to spend differently. What did they actually do, and did it work?

The answer is more interesting than either the boosters or the skeptics will tell you. It is also less flattering to the United States than I expected, and more complicated than a headline allows.

The American shape: an emergency budget for a chronic condition

Start with what the US pattern actually is, because it is often described loosely.

Federal firefighting costs, tracked by the National Interagency Fire Center in a table anybody can download, first crossed a billion dollars in 2000. They hit a record of about $4.4 billion in 2021. In 2022 they ran about $3.6 billion, and in 2023 about $3.2 billion.

The institutional consequence lands on one agency in particular. Fire operations consumed roughly 16 percent of the Forest Service's appropriated budget in 1995. By 2015 that had passed 50 percent, and the agency projected it would reach something near two thirds by 2025. That last figure deserves a caveat I want to state plainly rather than bury: it was a 2015 projection, not an audited result, and it should be read as the agency's own forecast at the time.

Still, the direction is not in dispute. An agency that spends half or more of its money fighting fire is not primarily a land management agency anymore, whatever its charter says.

Here is the structural part that matters for the comparison. Suppression is closer to an emergency expenditure. When a fire is burning, aircraft fly and crews deploy and the money is found. Prevention is a discretionary line item that competes with every other discretionary line item, in a budget cycle, against constituencies who all have a case. One of those two things is politically easy to fund in the moment. The other is not.

That asymmetry is not unique to America. What is unusual is how few countries have tried to correct it on purpose.

Portugal: the clearest deliberate reversal I could find

Portugal is the case worth studying closely, because it is the one place where you can watch a country decide to change the ratio and then read the numbers afterward.

The reason it happened is grim. In 2017 the area burned in Portugal exceeded 560,000 hectares. To put that in proportion, the OECD notes that this was about 56 percent of the total burned area in the entire European Union that year, in a country that occupies a small fraction of the EU's land. The June and October fires together caused more than 100 casualties, injured more than 200 people, and left more than 2,700 homeless.

The financial damage was itemized afterward, and the table is worth reproducing because it shows where wildfire cost actually lands. Total estimated cost came to about EUR 1,456 million, made up of EUR 634 million in forestry, EUR 311 million in commercial and industrial activity, EUR 210 million in agriculture, EUR 102 million in private homes and assets, EUR 94 million in network infrastructure, EUR 64 million in cleanup, EUR 39 million in emergency operations, and EUR 2 million in public assets.

Notice what is small in that list. Emergency operations and rescue, the part everybody photographs, is EUR 39 million out of EUR 1,456 million. The cost of a fire is overwhelmingly the things it destroys, not the effort to stop it.

Portugal then did something governments rarely do, which is change the budget ratio rather than announce a plan. According to the OECD's case study on Portugal, in 2017 only 20 percent of wildfire management funding went to prevention. By 2021 that share had risen to 46 percent, amounting to EUR 145 million allocated to prevention activities.

Twenty percent to forty six percent in four years. Whatever else you think of it, that is a real reallocation, not a press release.

The institutional machinery came with it. Portugal created an agency for integrated rural fire management, established in 2019 as a result of the reform process, whose job is coordination across the many bodies that already existed.

The honest part of the Portuguese story

I am not going to present this as a solved problem, because it is not, and the sources are candid about that.

Reporting on the agency's own targets indicates only about 20 percent of its fire prevention goals have been achieved, and an ambitious national plan to substantially increase the area treated by prescribed burning underachieved by a large margin. The country's national plan runs to 2030 with a stated intent of 60 percent of the annual budget on prevention and 40 percent on suppression, which tells you the reallocation is a direction of travel rather than a finished state.

There is also a driver Portugal cannot budget its way out of quickly. Between 1990 and 2017, the extent of eucalyptus forest in Portugal grew by 62 percent, and research cited by the OECD associates the expansion of large non native eucalyptus plantations with the spread and intensity of some of the country's most extreme fires. You can reallocate money in four years. You cannot reallocate a landscape in four years.

So the Portuguese lesson is narrower than the headline, and more useful for it. A country can move the prevention share of its wildfire budget by twenty six percentage points in four years if a disaster creates the political room. It will still take decades for the land to respond, and it will miss most of its interim targets on the way.

Australia: the country that treats fuel as an industrial process

Australia is the counterexample on prescribed burning, and specifically Western Australia is.

Western Australia's Department of Biodiversity, Conservation and Attractions aims to burn 7 percent of its forest estate every year. Compare that with New South Wales, where the National Parks and Wildlife Service target is about 135,000 hectares a year, which the same source puts at roughly 1.5 percent of the protected area estate, or about 2.5 percent if you count only the forested part.

Victoria is the instructive middle case. It set a target of 5 percent of all public land from 2010 to 2016, then shifted to a risk based rather than an area based approach.

Then the sentence that reframes the whole subject. In all Australian states except Western Australia, the annual treatment rate is much less than the area of wildfires, and it has been relatively stable in recent years.

Read that carefully, because it applies to almost everywhere including the United States. If you treat less land deliberately than burns accidentally, fire is setting your land management schedule and you are responding to it. Western Australia is one of the few jurisdictions anywhere that has flipped that relationship, and it did so by treating fuel reduction as a routine annual industrial operation with a quota rather than as a project.

Australia is also the place to be careful about overclaiming. The Black Summer fires of 2019 and 2020 happened in a country with the most developed prescribed burning culture in the developed world. Prescribed burning is genuinely contested in the Australian research literature, on effectiveness, on ecology, and on smoke. Nobody serious argues it makes a region fireproof. The argument is about whether it changes fire behavior enough, often enough, to be worth its cost and its risks.

The European move: stop buying twenty seven small air forces

The third approach is not about fuel at all. It is about capital equipment, and it is the one I think is most directly transferable to an American reader.

Aerial firefighting aircraft are expensive, they are needed intensely for part of the year, and no single medium sized country can justify a fleet sized for its worst week. The predictable result is that everyone buys too few and everyone is short at the same time.

The European response, rescEU, pools it. The European Commission has committed roughly EUR 600 million to acquire a dedicated fleet: 12 new firefighting aircraft distributed across six member states, namely Croatia, France, Greece, Italy, Portugal and Spain, plus helicopters based in Slovakia, Czechia and Romania. Member states are separately buying additional aircraft for their own national fleets on top of that.

It is worth being accurate about the timeline, since this is often reported as though the fleet exists. The first helicopter was delivered to Romania in January 2026. The first batch of the new aircraft is expected around the end of 2027. This is a program mid delivery, not a finished capability.

The idea, though, is the durable part. A shared asset pool sized for the continent's worst week is cheaper than 27 national fleets each sized for a bad Tuesday, and it works precisely because peak fire danger does not arrive everywhere simultaneously.

The uncomfortable fact about American prescribed fire

Here is where the comparison turned on me while I was researching it.

The United States is not bad at prescribed fire. It does an enormous amount of it. According to the survey run jointly by the National Association of State Foresters and the Coalition of Prescribed Fire Councils, the only nationwide accounting of prescribed burning across state, federal and private land, about 9.4 million acres were treated in 2020 and close to 10 million acres in 2021.

That is a very large number by any international standard.

Now the regional split, from the 2019 figures in the same survey series. The Southeast treated 5,933,022 acres. The West treated 3,705,213 acres. The Northeast treated 365,306 acres.

The Southeast, which is not where America's catastrophic fire losses concentrate, does the clear majority of America's prescribed burning. The West, which is, does substantially less on a far larger and more fire prone land base.

So the American problem is not ignorance and it is not a missing technique. The technique is practiced at scale, domestically, by Americans, in the same country. It is practiced least where it is needed most.

That reframing matters because it changes what the fix is. You do not need to import a foreign philosophy. You need to understand why a practice that is routine in Georgia and Florida is difficult in California and Oregon.

What actually differs, and it is mostly not knowledge

Reading across all of this, the real differences cluster in four places, and none of them is expertise.

Who owns the land. Prescribed burning is easiest where ownership is consolidated and the owner benefits directly, which describes southeastern timberland well. Portugal's structural difficulty is the opposite, a countryside of many small parcels with absentee or elderly owners, where getting anyone to act on a hectare is a coordination problem before it is a forestry problem.

Who carries the risk if it escapes. A prescribed burn is a deliberate fire. Someone has to be willing to be liable for the day it behaves badly. Regimes that make that liability manageable get more burning, and regimes that leave an individual holding unlimited downside get less. That is a legal design choice, not a scientific one.

Smoke. Deliberate smoke near populated areas is regulated, visible and unpopular in a way that wildfire smoke somehow is not, even though wildfire smoke is worse and lasts longer.

How the money arrives. Suppression tends to be funded as an emergency and prevention as a discretionary program. Portugal's reform is interesting mainly because it attacked this directly by moving the share, rather than by exhorting anyone to prefer prevention.

If you work in any organization with a maintenance budget and an outage budget, that last one should feel familiar. The outage always gets funded. The maintenance that would have prevented it competes with everything else, and it loses most years, and then the outage gets funded again. This is the same pathology at national scale, with pine trees.

What I would not claim

A few limits, stated up front rather than buried, because this subject attracts overclaiming from every direction.

I have not shown that prevention spending reduces burned area, and the sources here do not establish it. Portugal reallocated its budget and still burns. Australia has the most mature prescribed burning culture anywhere and still had Black Summer. Fire is driven by weather, fuel, ignition and where people build, and budget share is one input among several.

I have also not compared like with like on cost, and I want to be explicit about that rather than pretend. American suppression figures here are federal only, from the National Interagency Fire Center table, and exclude state agencies such as Cal Fire, which runs its own multibillion dollar operation. Portugal's figures are national wildfire management funding as tabulated by the OECD. These are not the same accounting boundary and I have not tried to force them into one.

What the comparison does support is narrower and, I think, more useful. Countries facing similar fire have chosen visibly different ratios of prevention to suppression. At least one moved that ratio substantially in four years. At least one region treats fuel reduction as a routine industrial quota rather than a project. A group of countries decided that expensive seasonal equipment is better pooled than duplicated. And the United States already does the most praised of these practices at large scale, in the wrong part of the country relative to its risk.

The question worth carrying away

If you own a home in fire country, the version of this that touches you directly is the insurance line, which is rising faster than inflation and faster than most raises, and which prices hazard whether or not any government reallocates anything.

But the general lesson is not really about fire. It is about the shape of a budget that funds consequences generously and causes reluctantly, and then treats the resulting bill as weather.

Portugal did not discover a new technique in 2018. It changed a percentage. That is available to almost any institution, including small ones, and it is usually the thing nobody proposes because the emergency is louder than the maintenance.

If you have ever tried to get a preventive budget approved in an organization that funds outages without argument, you already understand American wildfire policy better than most of its coverage does. The same instinct that makes a business defer maintenance until something breaks is the instinct I have written about in The W-2 Trap, which is largely about the difference between where money goes and where it would do the most good.

Fact-check notes and sources

Federal US suppression costs, including the record of about $4.4 billion in 2021, about $3.6 billion in 2022 and about $3.2 billion in 2023: the National Interagency Fire Center publishes the underlying table, Federal Firefighting Costs (Suppression Only). These are federal figures and exclude state agency spending.
2.7 million acres burned in 2023, the fewest since 1998: NIFC annual statistics, corroborated by secondary compilations of the same series.
Forest Service fire operations rising from about 16 percent of appropriated budget in 1995 to more than 50 percent by 2015, with a projection near two thirds by 2025: US Forest Service, The Rising Cost of Fire Operations. The two thirds figure is a 2015 projection and is presented here as such, not as an audited outcome.
Portugal's 2017 burned area exceeding 560,000 hectares, about 56 percent of the EU total for that year; more than 100 casualties, more than 200 injured, more than 2,700 homeless; and the EUR 1,456 million damage table: OECD, Taming Wildfires in the Context of Climate Change: The Case of Portugal, 2023, which adapts the damage table from San-Miguel-Ayanz et al., 2020.
Prevention share of Portuguese wildfire management funding rising from 20 percent in 2017 to 46 percent in 2021, totaling EUR 145 million: same OECD case study, based on agency figures from 2021.
Eucalyptus forest extent in Portugal growing 62 percent between 1990 and 2017, and its association with extreme fire behavior: same OECD case study.
The Portuguese agency for integrated rural fire management being established in 2019 as a result of the reform process: same OECD case study.
Portugal's national plan targeting 60 percent prevention and 40 percent suppression, and roughly 20 percent of prevention goals achieved so far: AGIF, Portugal Wildfire Management in a New Era, and reporting on the national plan. Treat the achievement figure as reported rather than audited.
Western Australia aiming to burn 7 percent of its forest estate each year, the New South Wales target of 135,000 hectares a year at roughly 1.5 percent of the protected area estate and about 2.5 percent of forests only, Victoria's 5 percent public land target from 2010 to 2016 and its later shift to a risk based approach, and the observation that in all states bar Western Australia the annual treatment rate is much less than the area of wildfires: Prescribed burning in Australian forests, CSIRO Publishing, which attributes the Western Australian figure to Bradshaw et al. 2018. A correction worth recording: an earlier version of this post cited an Ecology and Society article for these figures on the strength of a search result. I checked the page directly, found it is a different study about landowner risk perception, and replaced the citation with the source that actually contains the numbers.
The roughly 200,000 hectare annual scale of the Western Australian program: this figure comes from the South West Forests Defence Foundation, an organization critical of the program, and is presented here as their characterization rather than as an agency figure. The peer reviewed number above, 7 percent of the forest estate, is the one to rely on. The department's own website returns an automated block to non browser requests, so I could not verify its published target directly.
rescEU committing roughly EUR 600 million for 12 firefighting aircraft across Croatia, France, Greece, Italy, Portugal and Spain, plus helicopters in Slovakia, Czechia and Romania, with the first helicopter delivered to Romania in January 2026 and the first aircraft batch expected around the end of 2027: European Commission, rescEU and its announcement that production has begun.
US prescribed fire totals of about 9.4 million acres in 2020 and close to 10 million in 2021, and the 2019 regional split of 5,933,022 acres in the Southeast, 3,705,213 in the West and 365,306 in the Northeast: the National Prescribed Fire Use Report series produced jointly by the National Association of State Foresters and the Coalition of Prescribed Fire Councils. This survey is the only nationwide accounting spanning state, federal and private land, and it is a survey rather than a census.
What is deliberately not claimed: no causal link between prevention spending and reduced burned area is asserted here, because these sources do not establish one. US and Portuguese cost figures use different accounting boundaries and are not presented as directly comparable.

Everyone Fights Fire. Almost Nobody Prevents It. What Portugal, Australia and the EU Do Differently