The Economics behind Catastrophe

Across Central Europe, record heat and drought have pushed the Danube river to historic lows. In Hungary, this has become more than an environmental problem: the country's Paks nuclear power plant, which normally provides around half of its electricity, has been forced to slash output as the river becomes too shallow to provide sufficient cooling water.¹ At the start of August it was producing around 240 megawatts against a normal output of 2,000. Romania has shut one of the two reactors at its Cernavodă plant, and in Serbia the Djerdap hydropower stations, which together supply around 18% of the country's electricity, have been cut to between a fifth and a third of their capacity.²

The response has been extraordinary. Hungarian companies have been asked to reduce electricity consumption, while Romania has resorted to blasting rocks in the Danube to redirect water towards the reactor still running.³

It is an almost perfect illustration of a problem economists are increasingly being forced to confront: disasters do not stay confined where they happen.

A drought can become an energy shock. An earthquake can become a supply-chain crisis. A nuclear accident can become a fiscal burden lasting decades. And a seemingly local catastrophe can alter investment, trade and government policy far beyond the geographical area in which it occurred. Forty years after Chernobyl, the most important lesson may therefore not be about nuclear safety. It is about how economies respond when something goes catastrophically wrong.

At its simplest, a disaster is a supply shock. Factories are destroyed. Workers are displaced. Transport networks break down. Energy production falls. Agricultural output declines. The economy suddenly has fewer productive resources available.

But this creates an unusual macroeconomic combination: output falls at exactly the moment the prices of essentials are likely to rise.

After a major disaster, supply contracts while demand does not necessarily disappear. Households still need food, electricity and housing. Firms still need inputs. Governments suddenly need enormous quantities of construction materials, medical equipment and emergency services.

The result can resemble the textbook supply-side nightmare, though in a narrower form than that phrase implies: real GDP falls, while the price pressure concentrates in food and energy rather than spreading across the whole basket. That is close to the pattern now unfolding along the Danube, where a drought is transmitting into the power system rather than into the general price level.

Recent research from the Bank for International Settlements finds that average-sized droughts can reduce GDP by around 2% over a four-year period, with wildfires and landslides also producing persistent losses. The same study finds that effects on headline inflation are usually small and short-lived, but that food prices rise by more and stay elevated for longer.⁴

This creates the counterintuitive possibility that a disaster can produce a short-term increase in measured economic activity during reconstruction without making society richer. The economy is spending resources replacing capital that already existed. GDP therefore does not capture the entire economic cost of catastrophe.

The harder question is whether a disaster permanently reduces the economy's productive capacity. Chernobyl is the case that answers it.

Chernobyl's huge consequences

At 1:23 a.m. on 26 April 1986, reactor four at the Chernobyl nuclear power plant exploded during a safety test.

The immediate physical consequences were devastating. But economically, the disaster was even more unusual because the damage did not end when the fire was extinguished.

Large areas of Ukraine, Belarus and Russia were contaminated. Hundreds of thousands of people were evacuated or resettled.⁵ Agricultural land became unusable. Businesses disappeared. Infrastructure had to be abandoned. The Soviet government was forced to divert enormous quantities of labour and public expenditure towards containment, evacuation and compensation.

The World Bank describes Belarus as the country most severely affected by the environmental, health and economic consequences of Chernobyl.⁶

By 1991, Chernobyl-related spending represented more than 22% of Belarus's national budget. Even by 2002, it remained above 6%.⁷

A disaster can turn a temporary shock into a permanent reallocation of resources.

Money that could have been spent on schools, infrastructure or productive investment instead had to be spent dealing with the consequences of an accident that had already happened.

The opportunity cost was enormous. And this effect is not captured particularly well by GDP.

Imagine two countries producing exactly the same amount of output. In one, the government spends billions building new railways and universities. In the other, it spends billions relocating populations, decontaminating land and maintaining exclusion zones.

Both expenditures can raise GDP, yet only one is an investment in future productive capacity.

When expectations reprice

But Chernobyl also demonstrated something even more important: how disasters change expectations.

Following a catastrophe, households and firms do not simply react to the damage that has already occurred. They begin pricing in the possibility that something similar could happen again.

This matters enormously for capital-intensive industries. Nuclear power is a particularly good example. Building a nuclear plant requires billions of dollars of upfront investment, while the revenues arrive over decades. Investors therefore care not only about expected returns but also about the probability of extremely large losses.

Chernobyl changed the perceived risk of nuclear energy across Europe. The disaster contributed to a political backlash against nuclear power, most dramatically in countries such as Italy, which abandoned nuclear generation following Chernobyl.⁸

After Fukushima in 2011, the effect was repeated on an even larger scale. This is an example of what economists might call a risk premium becoming economically consequential. The accident itself occurred in one reactor. But the change in perceptions affected investment decisions across entire energy systems. Once those decisions are made, they can be very difficult to reverse.

Fukushima: when an earthquake becomes an energy crisis

On 11 March 2011, Japan was hit by a magnitude 9.0 earthquake followed by a devastating tsunami.

The immediate damage was enormous. But Fukushima Daiichi created a second shock: Japan's nuclear power system was effectively thrown into reverse.

Before the disaster, nuclear reactors supplied around 30% of Japan's electricity.⁹ Following the accident, reactors across the country were gradually shut down as authorities reassessed their safety. Japan therefore faced a problem that went far beyond the damaged Fukushima plant. It suddenly had to replace a huge amount of electricity generation.

For an economy heavily dependent on imported energy, this meant increasing imports of liquefied natural gas, coal and oil. The country effectively replaced a domestic source of relatively low-carbon electricity with imported fossil fuels.

The macroeconomic consequences were substantial. Japan's GDP contracted at an annualised rate of 3.5% in the first quarter of 2011, while the economy contracted by 0.7% over the year as a whole.¹⁰ But the more lasting consequence was not the contraction itself. It was what Japan gave up over the decade that followed.

A 2019 study in Energy Policy estimates that had Japan and Germany cut coal and gas instead of nuclear after Fukushima, the two countries together could have avoided roughly 2.4 billion tonnes of CO₂ emissions and 28,000 air-pollution-related deaths between 2011 and 2017. The same study finds that Japanese emissions rose only until 2013 before falling back, as total energy demand declined and renewable generation grew.¹¹

This creates an uncomfortable economic paradox. A disaster can make one technology appear less attractive because its risks have become visible, while simultaneously making another technology more attractive despite its own external costs.

The decision to reduce nuclear power did not eliminate risk but redistributed it, from nuclear accidents towards fossil-fuel dependence, imported energy and climate-related risks.

The invisible cost: supply chains

Perhaps the most important lesson from Fukushima was that modern economies are networks, not collections of isolated factories. The Japanese automotive industry provides a particularly clear example.

Renesas Electronics, which then held more than 40% of the global market for automotive microcontrollers, had its Naka plant damaged by the earthquake. Those chips sit inside airbags, braking, steering and engine control systems, and no ready substitute existed. Carmakers well beyond Japan cut output while 2,500 engineers worked for close to a month to restore a single production line. Trial production resumed at the end of April; mass production only in June.¹²

A factory in one region can depend on a highly specialised component produced hundreds of miles away. If that supplier stops operating, an otherwise undamaged factory may also have to stop. This is the problem of interconnectedness. A disaster does not need to destroy an entire industry to disrupt it. It only needs to destroy a sufficiently important node within the network.

The same principle was visible during the COVID-19 pandemic, when shutdowns in particular regions propagated through international supply chains. Modern production has become remarkably efficient partly because firms have reduced inventories and specialised production geographically.

But efficiency comes with a trade-off. The more tightly connected a system becomes, the further a local shock can travel.

Recent research into disaster economics increasingly focuses on exactly this problem. A 2025 multi-regional model found that disasters can propagate through supply chains across regions, with the ability of unaffected areas to compensate constrained by their own production capacity and logistical limitations.¹³

The cheapest supply chain is not necessarily the economically optimal supply chain. A firm that sources a component from the cheapest possible supplier may minimise costs in normal times. But if that supplier is located in the world's only major production cluster for the component, the firm has effectively purchased efficiency by selling insurance.

The same logic applies to countries. A state that concentrates its generation in one river basin has bought cheap power by selling the same insurance.

The fiscal problem

Governments are often the ultimate insurer of catastrophic risk. When a disaster occurs, someone has to pay. Households may receive compensation. Businesses may need subsidies. Infrastructure must be rebuilt. Energy systems may need emergency investment. Workers may need support. But governments themselves have limited fiscal space.

This means a disaster can increase public debt at precisely the moment when tax revenues are falling because economic activity has been disrupted. The fiscal multiplier can therefore work in reverse. Lower output can lead to lower tax receipts. Higher emergency spending can lead to higher deficits. Higher deficits can lead to higher debt. And if investors perceive the disaster as a sign of deeper structural vulnerability, borrowing costs can rise too.

For smaller or poorer economies, this can become particularly damaging. A wealthy country can often borrow its way through a disaster. A highly indebted country may have to choose between reconstruction and everything else.

This is why disaster resilience is not simply an environmental policy but also a driver of fiscal policy.

The Chernobyl paradox

There is, however, one final lesson from Chernobyl that is easy to miss. The Soviet Union was already facing deep economic and political problems when the reactor exploded. It would be misleading to claim that Chernobyl caused the collapse of the Soviet economy. The Soviet system had structural weaknesses that had accumulated for decades.

But Chernobyl exposed them. A highly centralised government struggled with the information problem created by the disaster. The initial response was characterised by secrecy and delayed communication, while the eventual clean-up required enormous mobilisation of resources.

Institutions decide how long it lasts

The economic cost was therefore amplified by institutional weakness. This may be the most important variable in disaster economics. The size of the shock matters, but the quality of the response determines how long it lasts.

Two countries can experience equally severe natural disasters and experience completely different economic outcomes. One has strong infrastructure, deep financial markets, diversified energy supplies and effective institutions. The other has weak infrastructure, high public debt, limited insurance and little fiscal capacity.

Which brings us back to the Danube. It is ironic that Europe is currently confronting an energy crisis caused by the absence of the very resource needed to keep some of its nuclear reactors operating.

Nuclear power was designed to provide a stable, weather-independent source of electricity. But the plants still depend on the physical environment around them. The drought has exposed an uncomfortable reality: there is no such thing as a completely risk-free energy system.

Nuclear power carries low-probability, high-impact accident risks. Fossil fuels carry pollution, geopolitical and climate risks. Hydropower depends on rainfall and river flows. Renewables depend on weather and require grids, storage and backup capacity. Even the infrastructure built to make energy supply independent of the weather turns out to depend on it.

For decades, economic policy has often treated resilience as an additional cost: something governments and firms should pay for because catastrophe might happen. But increasingly, resilience looks less like an insurance premium and more like productive investment.

The alternative is to build an economy optimised for average conditions and discover its weaknesses only when those conditions disappear.

The real economic cost of a disaster is therefore rarely contained in the rubble it leaves behind. It lies in the factories that stop producing, the supply chains that break, the energy that becomes more expensive, the public money diverted from investment, the capital that becomes stranded, and the expectations that permanently change. More than that, disasters reveal how fragile the capital we thought we had actually was.

And as climate change makes extreme events more frequent, the question for economists is no longer simply how much the next disaster will cost. It is whether we can build economies in which the next disaster does not become an economic crisis.