Gene Blevins / Reuters

There were the hurricanes that rained down biblical floods on Texas and Florida and devastated Puerto Rico and the Virgin Islands. There were the fires that smoked wine country and coated Montana and Oregon in ash, and the fires that are burning down houses in Santa Barbara. Then, there were the king tides that flooded Miami, the heat waves that seared the southwest, the tornadoes that scarred the southeast, and the rains that never came in the Cascades. No wonder the National Oceanographic and Atmospheric Administration has deemed this to be the second most extreme year, weather-wise, in the past century.

That extreme weather has taken a devastating and unknowable human toll, on families from San Juan to San Francisco. And it has taken economic one as well. It now seems a near certainty that 2017 will be the most expensive year in American history in terms of natural disasters—and a preview of the trillions of dollars of costs related to climate change yet to come.  

The effect is perhaps clearest in terms of property damage, in the United States’ territories as well as in the states, with governments, insurers, and individuals counting up the losses from torn-apart homes, flooded cars, downed bridges, destroyed electrical grids, and shuttered hospitals. Early in the fall, Hurricane Maria devastated the island of Puerto Rico, which had already suffered a decade-long recession. The government there has asked for $95 billion to rebuild the electric grid, infrastructure, and homes, and the storm caused an estimated $85 billion in insured losses. The credit-rating agency Moody’s puts the estimate of the storm’s damage at $40 billion in lost economic output and $55 billion in property damage, in a region with a GDP of about $100 billion a year. The numbers are similarly devastating in the Virgin Islands, which were hit by Hurricane Irma.

Though Harvey and Irma did less catastrophic damage on the mainland, that damage was more costly because of the value of the homes, businesses, and public infrastructure there. “We’re going to have to rebuild everything,” Bonnie Stephenson, the mayor of the tiny town of Rose City, Texas, told me when I visited, with mold growing up the sides of homes and heaps of mucked-out garbage lining the streets, six weeks out. The whole town would have to be gutted, reframed, and rebuilt, like much of southwest Texas and western Florida. The government estimates the damages from the storms and floods at $131 billion.

Then, there were the wildfires that lit up a number of dry western states this year, and some of them continue to rage today. Fires in California alone have caused at least $9.4 billion in damages. “These numbers not only represent staggering losses to tens of thousands of Californians,” Dave Jones, the state’s insurance commissioner, said in a statement. “The October wildfires that devastated whole communities and tragically cost 44 people their lives have now proven to be the most destructive and deadliest in our state’s history.” That number is likely to prove a fractional  estimate, not taking into account the damages from the uncontained fires still alight in southern California and not accounting for uninsured losses.

All told, there were at least 15 weather events costing at least a billion dollars each in 2017, the second-most since 1980, the government estimates. To be fair, weather gets costlier over time because the value of America’s homes and businesses and the economy itself gets bigger over time, as noted by Roger Pielke Jr., a political scientist at the University of Colorado Boulder. Still, tallying up the damages, the winds, floods, and fires are likely to end up wiping away 0.2 or 0.3 percent of the nation’s wealth—causing as much of a hit in percentage terms, in other words, as the Great Chicago Fire of 1871 or the Great Mississippi Flood of 1927. This year really was different.

Another, related way of looking at the toll is in terms of the costs to the government to fight fires, move families, save stranded individuals, and rebuild. There are the significant direct costs for disaster relief, for one. Congress has provided more than $50 billion in emergency spending related to storms and floods, including emergency nutrition assistance for Puerto Rico, debt relief for the federal flood-insurance program, and new funds for the Federal Emergency Management Agency. The forest service has also spent $3 billion on firefighting, and state and local governments have committed similar sums.

But that does not fully capture the way that storms, fires, and droughts act as a drain on public resources. There is also the cost of cuts to other agencies to free up money for disasters, and the burden of increased spending on social insurance and safety-net programs, such as food stamps. Consider this analysis of the California fires—just one catastrophic event among many this year: “The cost to contain and fight the fire and deal with the aftermath will be in the billions,” wrote Joel N. Myers, the chairman of the forecasting firm AccuWeather. “The loss in tax revenue from businesses no longer around, including the vineyards; the workers who have lost their jobs and can no longer pay taxes as well as other impacts will be quite costly. This will create a hole in the California budget, which may necessitate an increase in taxes. If California has to borrow more this might negatively impact its bond ratings and it will have to pay higher interest rates on all borrowings.” He estimates the impact at $70 billion to $100 billion.

Despite the prevalence and devastation of storms like Katrina and Sandy and Maria, as well as fires and tornadoes and earthquakes, the government does not have a full accounting of the budget effect of extreme weather—nor a sense of how more-frequent and more-intense storms might tax public coffers in the years to come. “Little is known about the fiscal costs of natural disasters, especially regarding social safety nets,” writes Tatyana Deryugina, an economist at the University of Illinois. She notes “substantial” increases in unemployment insurance and medical costs for a full decade after a hurricane, with the government spending $780 to $1,150 more per capita on social insurance for every $155 to $160 it spent per person on disaster aid. “This implies, among other things, that the fiscal costs of natural disasters have been significantly underestimated,” she concludes.

Then, there is the cost to growth and jobs—those headline economic figures—where this year and in others there tends to be something of a yo-yo effect: A storm hits, or a fire strikes. Businesses close and stay closed. Families flee, and damage from water, wind, ash, and debris keeps them away for some time. Jobs and economic activity disappears. But soon after, the rebuilding starts. In Houston, for instance, the surge of activity caused by Hurricane Harvey was obvious when I visited. Volunteer crews were pulling down and replacing rotten siding. Families were restocking their cabinets and closets. Car dealerships were advertising deals. People were scavenging for metals and repairable goods in the debris. Lowe’s and Walmart were crowded with people who had lost something, or everything, and needed it replaced. “It has been crazy in here, crazy all the time,” a shift manager at a Home Depot in Beaumont, a coastal town west of Houston, told me. “Shop-Vacs, drywall, electrical, appliances.”

This surge in activity is financed, in part, with money pouring into a disaster area from insurers, volunteer groups, and the federal government. “FEMA has already filled hundreds of temporary positions to help rebuild communities impacted in Texas, Florida, Puerto Rico and the U.S. Virgin Islands,” wrote Rebecca Henderson, of the human-resources company Randstad Sourceright, in a research note. “Job openings for contingent talent have also spiked in the construction and hospitality sectors,” she continued, “and we expect to see similar demand from the engineering and environmental sectors.”

As such, the medium-term overall effect on GDP is often negligible, with a local economy slowing down to a halt and then speeding up again, and the national numbers rarely changing much. “Hurricanes Harvey, Irma, and Maria have devastated many communities, inflicting severe hardship,” the Federal Reserve said in a September statement. “Storm-related disruptions and rebuilding will affect economic activity in the near term, but past experience suggests that the storms are unlikely to materially alter the course of the national economy.”

But economists stress that GDP and jobs numbers are not necessarily a good way to capture the effect of extreme weather events—not this year, or any year, not for families and not for the government and not for the nation as a whole. “You’ve destroyed a bunch of housing stock and public buildings and disrupted for a short period of time people getting to work and dealing with tragedy, and then the rebuilding will start,” said Chris Varvares of the research firm Macroeconomic Advisers. “What you don’t measure is the welfare loss of what happened to people’s lives, their businesses, their pets, their photo albums. Part of the story is that GDP is perhaps not the best measure of the near-term impact of these storms, or the widening of the weather cycles.”

Natural disasters tend to have a disparate impact on the rich and the poor, for one, with not all families equally vulnerable to the terrible events and not all families equally capable of rebuilding. Studies have found that the rich and the poor are more likely than the middle class to live in flood zones, and are thus the most likely to be faced with property destruction and a mandate to try to rebuild. “Higher-income households often live in high-risk areas because of the aesthetic attributes of living next to water,” one study of flooding and income concludes. “Low-income households live in higher-risk areas than middle-income households in order to find affordable housing.” But the rich have greater capacity to flee a storm and to rebuild after it, given their likelihood of having insurance and their wealth. The poor, on the other hand, often get hit hard and struggle to recover. That seems true this year. Harvey has caused a spike in homelessness, and has hit the lowest-income communities the hardest. The Santa Rosa fires have been particularly devastating for people living in trailer parks and for migrant workers.

There is also the way in which natural disasters change the economic geography of a place, leading some families to flee and leaving others stuck. Higher-income residents of a given region tend to leave when devastation strikes, for instance. That might lead to higher earnings for them, but less economic vitality for the towns and cities that they left behind—something that seems a certainty in the case of Puerto Rico, with tens of thousands of people expected to relocate to Florida, New York, and elsewhere. “It is generally accepted among environmental geographers that there is no such thing as a natural disaster,” argues Neil Smith, an anthropologist and geographer at the City University of New York. “In every phase and aspect of a disaster—causes, vulnerability, preparedness, results and response, and reconstruction—the contours of disaster and the difference between who lives and who dies is to a greater or lesser extent a social calculus.” Who is exposed to storms, who is sheltered from tornadoes, who recovers from fires—all of these things are a matter of choice and public policy, in other words.

More broadly, catastrophic weather events change individuals’ and businesses’ economic decision-making—with a profound longer-term effect on the economy that is not always apparent in a jobs report or a GDP number. A family that lost its house might not be able to pay for a child to go to college. An entrepreneur might decide to retire early, rather than rebuilding. An insurer might decide to charge for more for building in a given region. Extreme weather, in some ways, acts as a tax on long-term economic vitality. “It affects people’s behavior and their investment decisions for many, many years, and that in turn affects economic growth,” says Solomon Hsiang, an economist at the University of California, Berkeley. “For the 10 years after a storm, there’s lower economic growth, in proportion to the intensity of the storm. It doesn’t show up as a huge spike, it’s much more like a gentle drift.” Part of the reason, he says, was that “people tend to shift their behavior away from things that look like investment, like education and health care, and tend to spend larger fraction of their earnings on things like consumption, things like food. Those are things they need in the short term, but don’t produce extra benefits in the long term.”

As hard as it might be to suss out the impact of extreme weather in 2017, yet harder is sussing out the impact of the changing climate, now and in the future—due to the difficulty of tying individual weather events to epochal changes like global warming, the inability of headline economic figures to capture the messy fullness of human life, and the inadequacy of the available data to measure changes in the natural and the economic world. “If people are giving you straight answers about this, they’re probably making it up,” Elizabeth Stanton, an economist at Global Development and Environment Institute at Tufts, told me when I asked her how much climate change had cost the U.S. in 2017. “I don’t think we can measure it, not at all. We’re missing vital information, and it’s impairing our ability to make decisions, decisions that are very important and very time-sensitive. It’s dangerous.”

But scientists have found a clear link between anthropogenic climate change and certain extreme weather events. Independent analyses have found that climate change intensified Harvey’s rainfall and has made similar storms wetter and more likely to occur, for instance. Hsiang and his co-authors have estimated that every degree Celsius the earth warms will cost the country more than a percentage point of GDP, worsening income inequality as well. If emissions are not contained, the climate effect would rival that of the Great Recession. Even those numbers are speculative and do not account for all the ways that a hotter planet might change human lives, researchers warn.

The country is pitching towards a more violent future, then, without a full sense of what storms, floods, and fires are costing it now and without a full sense of what is to come.