Two weeks ago, two climate stories crossed my desk that seemed, at first glance, to describe different planets.
Writing in the Financial Times, Simon Mundy warned that a strengthening El Niño could intensify extreme heat, disrupt harvests, raise commodity prices and deepen food insecurity. A day later, Bjørn Lomborg argued in a Wall Street Journal opinion piece that, despite dramatic wildfire headlines and smoke drifting south from Canada, the world is not experiencing a generalized increase in wildfire activity. Global burned area, he noted, has been declining.
Since then, reality has complicated both narratives. Germany has deployed remote-controlled armored vehicles to fight fires in forests littered with unexploded World War II ammunition. Lake Powell and Lake Mead have reached record lows. Yet global wildfire data continue to challenge the notion that ever more of the planet is simply burning.
If the objective is to defend a preferred climate narrative, there are facts available for either side.
If the objective is to govern, we need all of them.
Both climate stories can be true at once, but the governance challenge begins where the headlines end.
The wildfire metric problem
Lomborg’s central factual point deserves to be taken seriously.
A landmark 2017 study found that global burned area declined by roughly 24 percent between 1998 and 2015, driven largely by changes in agriculture, landscape fragmentation and human use of fire in savannas and grasslands. More recent data have likewise shown unusually low global biomass-burning emissions during portions of 2026.
But those facts do not establish that wildfire risk is declining everywhere.
Global burned area is heavily influenced by enormous grassland and savanna ecosystems. Reduced burning there can outweigh increases in smaller forest regions containing more carbon, property and human settlement. Other research has found that the frequency of the most energetically extreme wildfire events more than doubled between 2003 and 2023, particularly in boreal and temperate conifer forests.
Total area burned is not the same as fire intensity, forest loss, smoke exposure, property damage or risk to human communities.
Germany illustrates the distinction. Annual wildfire acreage remains small by global standards and highly volatile, yet several of its largest fire years of the past two decades have occurred since 2018.
That does not prove a single cause. It demonstrates something more useful: a favorable global aggregate can coexist with a deteriorating regional risk. The statistic that matters depends on the decision being made.
A recent Wall Street Journal story about a fire in Germany’s Hürtgen Forest makes the governance dimension even clearer. The fire consumed only a few hundred acres, but nearly 2,000 people were evacuated. One reason was extraordinary: the forest remains littered with unexploded ammunition from World War II. Shells were detonating as the fire moved through the area, preventing firefighters from safely reaching portions of the blaze.
Germany has responded with remote-controlled armored firefighting vehicles. Experts also point to less dramatic solutions: replacing highly flammable pine forests with mixed woodland, clearing old munitions, improving access and using controlled burns.
Hotter and drier conditions can make a forest more combustible. They did not put artillery shells beneath the Hürtgen Forest, determine its tree composition or decide how quickly Germany would clear former military lands.
Climate influences the hazard. Governance shapes the outcome.
Rivers are infrastructure
Europe’s rivers tell much the same story.
Exceptionally low water on the Rhine and other major waterways is forcing shipping adjustments and putting pressure on industrial supply chains. Germany has responded with additional rail and road capacity, low-water shipping measures and closer coordination among government and industry.
The significance is not simply environmental. The Rhine is infrastructure. It moves chemicals, fuel, minerals and industrial inputs through the center of Europe’s largest economy. A low river can therefore become a transportation problem, an energy problem, a manufacturing problem and ultimately a growth problem.
But observing the event is only the beginning.
An event is not a trend. A trend is not a cause. A cause is not a policy.
A low river is an observation. Whether such events are becoming more frequent is a trend question. How much warming contributes is a causal question. What governments and businesses should do about it—transportation redundancy, infrastructure investment, water management—is a governance question.
Those questions are connected. They are not interchangeable.
The Colorado River is the governance stress test
No example illustrates this more clearly than the Colorado River.
On August 21, the Department of the Interior reported that Lake Powell and Lake Mead had both recently reached record lows, with their combined contents lower than at any point since before Lake Powell began filling in 1963. The river supports more than 40 million people, 30 tribal nations and millions of acres of farmland. New federal operating rules call for Lower Basin deliveries to be reduced by 1.25 million acre-feet in both 2027 and 2028.
Climate clearly matters. Warmer conditions can reduce snow retention, increase evaporation and leave less runoff reaching reservoirs.
But climate change did not create the Colorado River’s governance problem.
The 1922 Colorado River Compact was negotiated using observations from an unusually wet period. Over the century that followed, cities expanded, farmland was irrigated, reservoirs were built and legal rights accumulated around expectations of a river more abundant than the longer hydrologic record would support.
Now that inherited system is colliding with a warmer and drier hydrologic regime. Since 2000, system water use has exceeded inflow in most years; average inflow from 2000 through 2024 was only 12.9 million acre-feet, well below the roughly 18 million acre-feet observed during the period that informed the original compact.
The federal government’s new framework appropriately acknowledges that uncertainty. It establishes shorter operating periods, objective reservoir triggers and the ability to adjust rules as hydrology and negotiated agreements evolve.
To say simply that climate change is drying up the Colorado River therefore misses half the story. To say the problem is merely overuse or a century-old allocation system misses the other half.
Climate change did not write the Colorado River Compact or build an economy around its assumptions. But a warmer climate is exposing the fragility of those decisions.
That is where climate risk becomes a governance problem.
Markets do not need a single climate verdict
The same distinction matters to investors because markets rarely experience climate risk as a single global event.
A wildfire can leave global burned acreage largely unchanged while destroying billions of dollars of property in one region. A falling river can barely register in a global equity index while sharply increasing costs for a manufacturer dependent on Rhine shipping. A water shortage can be manageable for one city and devastating for a farmer holding a junior claim.
Aggregate conditions can remain benign while concentrated risks deteriorate beneath the surface.
Investors therefore do not need to decide whether “the world is on fire.” They need to determine whether a particular asset, company, supply chain or portfolio is exposed to a changing distribution of risk.
A climate risk can be financially material without being globally catastrophic.
This is why both easy climate narratives are inadequate. Treating every fire, drought or heat wave as definitive proof of catastrophe eventually undermines credibility. But treating favorable global statistics as permission to ignore changing regional risks is no more useful.
The appropriate posture is urgency without alarmism: acknowledge what is improving, confront what is deteriorating, distinguish events from trends and global averages from regional consequences, and make decisions proportionate to the risk.
What better governance looks like: the Rhine
The encouraging news is that this kind of governance is not theoretical.
After extreme low water disrupted the Rhine in 2018, Germany’s federal government, major industrial users, shipping interests and ports developed a coordinated low-water action plan. It has produced better forecasting, low-water-optimized ships, additional logistics capacity and more structured contingency planning. When water levels again became exceptionally low this summer, federal and state authorities coordinated road, rail and waterway capacity to help keep supply chains functioning.
That architecture would have been familiar to Elinor Ostrom.
Ostrom, who received the 2009 economics prize for her analysis of the governance of the commons, spent much of her career studying how communities manage shared resources such as forests, fisheries and water systems. Her central insight was that effective governance need not mean choosing between an all-powerful central government and an unfettered market. Durable solutions often emerge through polycentric governance: multiple centers of decision-making operating at different levels, each contributing information, incentives, monitoring and the ability to experiment and adapt. She later applied that logic explicitly to climate change, arguing that complex collective-action problems are better addressed through efforts operating simultaneously at local, regional, national and international scales.
The Rhine is moving in that direction. River authorities understand hydrology. Shippers know where logistical bottlenecks emerge. Manufacturers know which inputs become critical. Governments control infrastructure and regulation. Ports, railroads and trucking networks provide alternatives when river capacity falls.
A logical next step is to bring insurers, reinsurers and capital providers more explicitly into that architecture. They can help identify where losses concentrate, price vulnerability and direct capital toward resilience investments with the greatest economic return.
That is Ostrom’s insight translated into climate resilience: do not wait for one institution to solve the whole problem. Build a system in which institutions with different knowledge and capabilities can respond together.
The goal is not to predict exactly how low the Rhine will be in 2035. It is to know what happens when particular thresholds are crossed, where redundancy is required, who acts and who pays.
The Colorado River requires the same basic approach under far more difficult political and legal circumstances: common information, realistic allocations, predetermined triggers and rules capable of adjusting as conditions change.
This is also why mitigation and adaptation should not be treated as competing ideologies. Mitigation seeks to limit the future growth of the hazard. Adaptation reduces current vulnerability. Germany does not need to wait for global emissions to fall before improving forests or shipping resilience. The Colorado River states cannot wait for certainty about future warming before reconciling water consumption with available supply.
Nor does good governance require perfect knowledge.
We do not need to know precisely how much a river will fall in 2040 to make infrastructure less vulnerable today. We do not need agreement on every climate model to allocate scarce water more realistically. And we do not need to settle the political debate over climate change before recognizing financially material risks.
In my previous post, I offered a simple rule:
Investors should neither impose an unrelated political agenda through a portfolio nor ignore financially relevant risks because those risks have become politically controversial.
The same principle applies here. Climate change does not become immaterial because some advocates overstate it. Nor does every climate policy become prudent simply because the underlying risk is real.
The objective of governance is not to eliminate uncertainty. It is to make institutions less fragile in the face of it.
That is manageable. Along the Rhine, elements of that model are already in place.
The question is whether we can do it consistently, before crisis forces the issue.
Two climate stories can be true at once.
The responsibility of governance is not merely to recognize both. It is to act intelligently.
