Deadly Nepal Flood and the Price of Staying Standing
Key Takeaways
- •The Aug. 26 avalanche along the Nepal-China border caused flooding that killed at least 543 people, with more than 1,500 missing as of Aug. 28, while destroying at least 35 bridges, 40 kilometers of road, and over 430 megawatts of hydropower capacity — about one-eighth of Nepal's installed total.
- •A 2025 World Resources Institute analysis of $133 billion in adaptation investments across 12 countries estimated over $1.4 trillion in benefits over 10 years, exceeding $10.50 per dollar invested.
- •UNEP estimates developing economies will need $310-365 billion annually for adaptation by 2035, but international public adaptation finance totaled only $26 billion in 2023.
- •Nepal's annual GDP is roughly $45 billion, while its National Adaptation Plan identifies $47.4 billion in priority adaptation costs through 2050, of which only $1.5 billion is expected from domestic sources.
- •Glaciers across the Hindu Kush Himalaya lost ice 65% faster in 2011-2020 than in the previous decade, affecting river systems that nearly two billion people depend on.

By Gonzalo Delacámara
Climate change has acquired a problem of persuasion. Those who work on natural resources and climate risk are routinely cautioned against sounding apocalyptic. Doom, the argument goes, is a poor currency for engagement: repeat the word "catastrophe" often enough and urgency curdles into fatalism.
That caution is not without merit. But occasionally events refuse to abide by the rules of good climate communication. Sometimes the catastrophe actually happens.
What unfolded along the Nepal-China border, including in Tibet, on Aug. 26 is one such occasion. A large rock-and-ice avalanche detached in the Himalayas and entered the river system below, generating a debris-laden flood wave that moved rapidly downstream. The resulting surge reportedly raised the Trishuli River by as much as nine meters within 30 minutes. By the afternoon of Aug. 28, at least 543 people had been confirmed dead, while more than 1,500 remained missing or unaccounted for on both sides of the border. The event fits a wider pattern: a comparable rock-and-ice collapse in the Chamoli district of Uttarakhand, India, in February 2021 destroyed hydropower projects and killed more than 200 people, and glacial-lake outburst floods have recurrently struck Himalayan river valleys in recent decades.
The economic toll is no less sobering. At least 35 bridges and some 40 kilometers of road were reportedly destroyed, and more than 430 megawatts of operational generating capacity were affected — roughly one-eighth of Nepal's installed hydropower capacity. The exposure is structural rather than incidental: hydropower supplies the overwhelming majority of Nepal's electricity, so damage to river-basin infrastructure translates directly into energy-system risk.
These are not merely entries on a disaster-loss ledger. They are parts of an economic system that have abruptly stopped performing their functions. That raises a deceptively simple question: how do we value what allows life itself to carry on?
Economics is relatively well equipped to value realized losses. We can estimate the capital destroyed, output forgone, and the cost of restoring disrupted services. The harder problem is valuing the counterfactual: the losses that resilience prevents.
The reinforced bridge generates no reconstruction bill. The early-warning system that keeps a community out of harm's way leaves no rubble against which to measure its return. Resilience is systematically undervalued because much of its return is realized in absence: the loss that never occurs, the disruption that never materializes.
Infrastructure, moreover, derives its economic value not merely from the capital it embodies but from the flows it sustains. Before earlier flooding disrupted the Rasuwagadhi-Gyirong corridor, the crossing handled some $609 million in imports during Nepal's 2024/25 fiscal year — roughly a quarter of the country's imports from China. When a critical node fails, the shock propagates through supply chains, prices, and employment. Climate disasters destroy more than stocks of physical capital; they impair the economic and social functions from which that capital derives its value.
We continue to describe such events as "natural disasters," as though the adjective absolved us of responsibility. The underlying hazard may be natural; the scale of the resulting risk increasingly is not. Risk is shaped by exposure and vulnerability — by what we protect and what we leave exposed.
Attributing this particular collapse to anthropogenic climate change would be premature without formal analysis. The broader physical trend, however, is well established. Rising temperatures are altering high-mountain environments as glaciers retreat, permafrost degrades, and slopes become less stable. Across the Hindu Kush Himalaya, glaciers lost ice 65% faster in 2011 to 2020 than in the preceding decade. The region's cryosphere regulates river systems that nearly two billion people depend on.
Climate change represents a negative externality on an exceptional scale: the benefits associated with the activities that generated emissions have accrued unevenly, while the resulting damages are increasingly borne elsewhere. The geography of causation and the geography of consequence scarcely coincide. Nepal has contributed only a negligible share of cumulative global emissions, yet must allocate scarce resources to protecting infrastructure, livelihoods, and development gains against risks generated largely beyond its borders.
A second dimension of human agency is omission. Risk reflects not only what humanity has done — accumulating greenhouse gases — but what it has failed to do: decarbonize fast enough, reduce vulnerabilities, and mobilize adaptation finance at the scale required.
This makes the conventional treatment of adaptation as environmental expenditure increasingly untenable. Climate resilience is not an alternative use of capital; it is an attribute that increasingly determines whether capital continues to perform its economic function. Adaptation is not what economies spend on instead of investing. It is becoming part of what investment itself means.
The empirical case is compelling. A 2025 World Resources Institute analysis of 320 adaptation investments across 12 countries, representing $133 billion, estimated more than $1.4 trillion in benefits over 10 years — over $10.50 for every dollar invested. Those benefits extend beyond avoided damages to productivity, health, and broader development gains.
Yet capital remains strikingly scarce. UNEP estimates that developing economies will require $310 billion to $365 billion a year for adaptation by 2035. International public adaptation finance amounted to just $26 billion in 2023 — a figure parties to the U.N. climate convention agreed at COP29 in 2024 to scale up, alongside a broader goal of mobilizing at least $300 billion a year for developing countries by 2035. Calling this an "adaptation finance gap" almost understates the problem. It is a misallocation of capital.
Nepal makes the paradox painfully clear. Its annual GDP is around $45 billion. Its National Adaptation Plan puts priority adaptation costs through 2050 at $47.4 billion, of which only $1.5 billion is expected domestically.
Through the narrow lens of fiscal affordability, Nepal cannot afford adaptation. Through the economics of risk, the conclusion is almost the reverse: Nepal cannot afford the consequences of being unable to afford it. And through the economics of externalities comes a third conclusion: it makes little sense to expect countries bearing a disproportionate share of the damage to finance the adjustment alone.
Breaking this trap requires a broader conception of productive capital. Alongside physical, human, and natural capital, there is a case for resilience capital: the accumulated capacity of an economy to absorb shocks while preserving the essential functions on which economic and social life depend.
The word resilience itself deserves rescuing. Praising vulnerable communities for their "resilience" can amount to congratulating them for enduring conditions they should never have been expected to endure. The mistake is to confuse resilience with resignation.
Properly understood, resilience is a form of economic optionality: the capacity to preserve room to maneuver as conditions deteriorate. Alternative routes, reserve capacity, fiscal headroom, insurance, and early-warning systems take different forms but perform a common economic function: they preserve choices that a shock might otherwise eliminate.
This also challenges our understanding of efficiency. For decades, firms and governments have been encouraged to strip out slack, minimize inventories, consolidate suppliers, and maximize asset utilization. Under deep uncertainty, what appears efficient today may prove a source of fragility tomorrow. What looks like slack in a spreadsheet may be insurance embedded in the productive system.
This is not an argument for indiscriminate redundancy. As probability distributions change, so does the economically efficient level of redundancy. Efficiency asks how much output we can extract when things go right. Resilience asks how much we can preserve when they do not.
Walter Benjamin once wrote that "that things 'just go on' is the catastrophe." He targeted the comforting idea of progress as an uninterrupted continuum. Climate change poses a different but related problem for economics. Our models have long treated continuity as the baseline — infrastructure functions, electricity flows, supply chains hold. As those assumptions become less secure, continuity itself acquires economic value.
Yet much of economic analysis still proceeds as though continuity were the default. We discount future cash flows with extraordinary sophistication while often assuming the physical conditions required to generate them.
So the question is no longer merely what adaptation costs. It is what kind of future our prices are assuming — and who pays when those assumptions prove wrong.
Economics has become remarkably adept at valuing failure once it occurs. The harder task is to value the counterfactual: the losses resilience prevents, the productive capacity it preserves, and the continuity it makes possible.
In an unstable climate, continuity can no longer be treated as the default.
Gonzalo Delacámara is the executive director of Elevate Water Foundation and an adjunct professor at IE University in Spain.