Pakistan declared a national emergency this week as extraordinary monsoon rains and flooding continued across large parts of the country. More than a thousand people have reportedly died since the monsoon began, including hundreds of children, while roads, bridges, homes and agricultural land have been destroyed. The government estimates that around 33 million people, roughly one in seven Pakistanis, have been affected in some way. The scale of the emergency is still becoming clear, and continued rainfall means the final human and economic costs will almost certainly be larger than what can be measured tonight.
It is tempting to describe this simply as a natural disaster. The phrase is increasingly inadequate. Rain is natural. The distribution of vulnerability is political.
Two countries can experience similar physical hazards and suffer radically different consequences because their infrastructure, wealth, geography, institutions and preparedness differ. A storm becomes a humanitarian catastrophe partly through the interaction between weather and the society the weather encounters. That means asking what caused Pakistan’s floods requires several levels of explanation rather than one convenient culprit.
Climate change belongs in that explanation. A warmer atmosphere can hold more moisture, increasing the potential intensity of extreme rainfall, while warming also affects glaciers and hydrological systems across South Asia. Pakistan is particularly exposed because of its geography, including the Indus river system, enormous populations living in flood-prone areas and mountain regions containing thousands of glaciers. But attributing one specific disaster entirely to climate change before detailed attribution research is completed would be too confident. Climate change changes the probability and intensity of many extreme events; it does not mean every raindrop can be assigned a carbon-emissions certificate.
That scientific qualification should not become an excuse to avoid the political question. Pakistan is experiencing a disaster in a world that has already warmed because of greenhouse gases accumulated over generations. The countries that contributed most to those historical emissions are generally much wealthier than Pakistan and possess considerably greater resources for adaptation and recovery.
This creates a distributional problem that climate negotiations have discussed for years without resolving adequately. The people who contributed least to creating a global externality can be among those least capable of protecting themselves from its consequences.
Climate change therefore violates one of the intuitions underlying ordinary ideas of responsibility: the people paying the cost are not necessarily the people who produced the harm.
If a factory pollutes a river and destroys crops downstream, identifying responsibility is comparatively straightforward. The factory produced the pollution, the farmer experienced the damage, and institutions can theoretically require compensation.
Climate change is different because billions of decisions across two centuries contribute to one atmospheric stock. Coal burned in nineteenth-century Britain, twentieth-century American industrialisation, European consumption, contemporary Chinese manufacturing, oil production in the Gulf and emissions from consumers everywhere become physically mixed in the atmosphere.
There is no single defendant.
That does not mean there is no responsibility.
It means responsibility is distributed.
International climate politics has struggled with this through the principle of common but differentiated responsibilities. Every country has some responsibility to address climate change, but those responsibilities differ according to historical contribution and capacity.
That principle makes intuitive sense. Asking Pakistan and Britain to bear identical costs for climate mitigation would ignore enormous differences in historical emissions and income. At the same time, emerging economies now produce a substantial share of current emissions, meaning historical responsibility cannot become a permanent argument that only wealthy Western countries need to act.
Climate justice therefore operates across at least three dimensions: who caused the existing stock of emissions, who is producing emissions now, and who possesses the financial capacity to respond.
Those answers are not identical.
That is what makes international agreement difficult.
Pakistan’s floods bring a fourth dimension into the argument: who suffers the damage.
The geographical distribution of climate vulnerability is largely unrelated to the historical distribution of emissions. Small island states can face existential sea-level risks despite negligible emissions. Poor agricultural economies can experience severe droughts despite tiny carbon footprints. Pakistan can face destructive floods while its historical contribution to global warming remains relatively small.
This creates the case for climate finance extending beyond mitigation. Wealthier countries have repeatedly promised financing to help developing countries reduce emissions and adapt to climate change. Adaptation includes flood defences, drought-resistant agriculture, resilient infrastructure, early-warning systems and urban planning capable of reducing the damage produced by extreme weather.
These investments can appear expensive before disaster.
After disaster, the alternative becomes visibly more expensive.
That is the economics of resilience.
A bridge built to withstand a one-in-one-hundred-year flood may cost more than a weaker bridge. For decades, the additional expenditure can look unnecessary. Then the extreme event occurs and the more expensive bridge is still standing.
Resilience therefore suffers from the same political problem as deterrence and pandemic preparedness: successful prevention is difficult to observe.
Politicians receive credit for building something visible.
They receive much less credit for infrastructure failing to collapse.
This can create systematic underinvestment in resilience because the benefit is probabilistic and often arrives beyond the electoral cycle.
Developing countries face an additional constraint. Even governments that understand the value of adaptation may lack the fiscal capacity to finance it.
Pakistan is already experiencing serious economic pressures. Foreign-exchange reserves are constrained, inflation is high, the government is dealing with external financing problems and negotiations with the International Monetary Fund have been politically important. Spending billions on climate resilience therefore competes with education, healthcare, energy subsidies, debt servicing and ordinary infrastructure.
A wealthy country can borrow more cheaply and absorb disaster costs more easily.
A poorer country can face the cruel situation where it cannot afford adequate adaptation before a disaster and cannot afford reconstruction afterwards.
This is why international climate finance should not be understood simply as charity.
There is a plausible argument that it is partial payment for a global externality whose costs are distributed independently of who produced it.
That argument will become increasingly important under the concept of loss and damage: assistance for countries experiencing climate harms that cannot realistically be prevented through adaptation alone.
Wealthy countries have historically been cautious about this language partly because accepting formal responsibility could create enormous financial expectations or even legal claims.
Developing countries understandably see the issue differently. If they face billions of dollars in damages from climatic conditions intensified by emissions produced disproportionately elsewhere, telling them simply to borrow more money for reconstruction can look like making victims finance the consequences of a problem they contributed relatively little to creating.
Pakistan may become one of the clearest political examples of that argument.
Yet there is a danger in moving too quickly from climate injustice to the conclusion that responsibility lies entirely outside Pakistan.
That would be equally incomplete.
Weather determines the hazard.
Governance helps determine vulnerability.
Where are homes permitted to be built? How are rivers managed? Are drainage systems maintained? Are floodplains protected from inappropriate construction? Are early warnings communicated effectively? Are emergency agencies properly funded? Are building standards enforced? Are forests and watersheds managed sustainably? Does public money intended for infrastructure actually produce resilient infrastructure?
Those are domestic political questions.
Climate change can increase rainfall intensity while poor land-use planning increases the number of people exposed to that rainfall.
Both mechanisms can operate simultaneously.
This is an important distinction because discussions of climate disasters can become politically convenient for national governments. Calling something an unprecedented climate catastrophe can correctly identify an external driver while also shifting attention away from failures that governments could have controlled.
The opposite political mistake occurs in wealthy countries, where commentators can focus entirely on local governance failures because acknowledging climate responsibility creates international obligations.
Each side has incentives to emphasise the causal explanation that reduces its own responsibility.
Good analysis should refuse both.
Pakistan did not create the global climatic conditions under which extreme rainfall risks are increasing.
Pakistani governments still possess responsibility for reducing the vulnerability of Pakistani citizens to those risks.
Those statements do not compete.
They describe different levels of causation.
This is one reason the language of natural disaster can be misleading. Earthquakes, floods and cyclones are physical phenomena. The number of people they kill depends enormously on political and economic systems.
Japan and Haiti can experience earthquakes with dramatically different consequences because building standards, infrastructure and emergency capacity differ.
The hazard may be natural.
The disaster is socially mediated.
That makes state capacity a climate policy.
A competent civil service is climate adaptation.
Reliable local government is climate adaptation.
Building-code enforcement is climate adaptation.
Accurate weather forecasting is climate adaptation.
Transparent infrastructure procurement is climate adaptation.
Climate policy therefore extends far beyond solar panels and carbon taxes.
It includes whether the state can perform basic administrative functions when environmental stress increases.
This is particularly important for countries like Pakistan because climate change acts as a threat multiplier. It does not necessarily create entirely new political problems. It intensifies existing ones.
Water scarcity becomes more severe.
Agricultural volatility increases.
Internal migration accelerates.
Electricity systems face greater stress.
Food prices become less stable.
Poor communities living in vulnerable areas absorb larger shocks.
Government budgets face repeated emergency spending.
Existing inequality determines who can recover.
A wealthy household can rebuild, relocate or purchase insurance.
A poor household may lose its home, livestock and income simultaneously.
Climate change therefore has distributional effects within countries as well as between them.
The poorest Pakistanis are not merely citizens of a relatively low-emitting country.
They are often also the people with the least ability to protect themselves inside that country.
Climate justice must therefore operate domestically too.
If international donors provide reconstruction money, who receives it?
Do politically connected regions obtain infrastructure first?
Are rural communities represented?
Are women able to access relief independently?
Do landless households receive support?
Does reconstruction reproduce the same vulnerable settlement patterns that existed before the flood?
These questions determine whether recovery builds resilience or simply rebuilds vulnerability.
The phrase “build back better” has become overused after disasters, but the underlying economic principle is sound. Reconstruction creates a rare opportunity to replace destroyed infrastructure with assets designed for future conditions rather than historical climate averages.
The danger is that emergency pressure encourages the fastest possible rebuilding.
Speed matters when people need homes.
Resilience matters when those homes may face another extreme event.
Again, policy involves a time-horizon trade-off.
Pakistan’s political crisis also complicates the response. The country remains intensely polarised following Imran Khan’s removal earlier this year. Khan continues mobilising large numbers of supporters against Prime Minister Shehbaz Sharif’s government, while economic conditions are already difficult.
A natural disaster does not suspend partisan politics.
It can intensify it.
Governments control relief resources.
Opposition parties scrutinise distribution.
Provincial governments may be controlled by different political parties from the federal government.
Every failure can become evidence of incompetence.
Every relief decision can become evidence of favouritism.
This is not necessarily cynical. Democratic opposition should scrutinise disaster response.
But extreme polarisation can reduce coordination precisely when coordination becomes most necessary.
Pakistan’s federal structure matters here because disaster management involves national, provincial and local institutions simultaneously. Sindh, Balochistan, Khyber Pakhtunkhwa and Punjab possess different political leadership, administrative capacity and exposure to flooding.
Federalism can improve disaster response because local governments possess information about local conditions.
It can also create coordination problems when responsibilities overlap.
Again, institutional design produces trade-offs rather than perfect solutions.
The floods will also have macroeconomic consequences. Agricultural losses can reduce food supply and increase prices. Destroyed roads and bridges disrupt trade. Reconstruction requires imports at precisely the moment Pakistan faces foreign-exchange constraints. Government spending needs increase while tax revenues may weaken in affected regions.
A humanitarian disaster therefore becomes a fiscal and balance-of-payments shock.
This is where climate vulnerability begins interacting with sovereign debt.
Countries facing repeated extreme-weather events may need to borrow repeatedly for reconstruction.
Higher debt increases interest costs.
Higher interest costs reduce fiscal space for adaptation.
Lower adaptation increases vulnerability to the next disaster.
A vicious cycle can emerge.
Climate change can therefore affect sovereign creditworthiness.
That has profound implications for international finance.
A country may pay higher borrowing costs partly because it is geographically vulnerable to a climate problem it did relatively little to create.
Then those higher borrowing costs make it harder to finance the infrastructure needed to reduce that vulnerability.
Markets price risk rationally from the lender’s perspective.
The resulting equilibrium can still be globally unjust and economically inefficient.
This is why development banks and concessional climate finance matter. They can provide long-term capital on terms that private markets may not offer, particularly for projects whose benefits are public and arrive over decades.
But financing alone is insufficient if domestic institutions cannot convert money into resilient infrastructure.
We return again to state capacity.
Climate justice without governance can waste resources.
Governance without international support can leave poor countries attempting to finance a global problem alone.
Both are necessary.
This is increasingly how I think climate politics needs to be discussed. The debate is often framed as a moral confrontation between countries that caused climate change and countries suffering from it.
The moral dimension is real.
But effective policy requires translating morality into institutions.
How much should wealthy countries contribute?
Through grants or loans?
Who qualifies?
How should historical emissions be measured?
How should current major emitters such as China be treated?
Who verifies that adaptation funding is actually used effectively?
What happens when damage cannot be prevented?
How should private insurance, development banks and national governments divide risk?
These questions sound bureaucratic compared with photographs of flooded villages.
They are where climate justice becomes real or remains a slogan.
Pakistan’s immediate priority is obviously rescue and relief. People need food, shelter, clean water, medicine and access to communities cut off by flooding. The international community should provide substantial assistance, and Pakistan’s government needs to coordinate that assistance transparently and effectively.
Later comes reconstruction.
Then comes the harder political question of whether the country and the international system learn anything before the next disaster.
Because this cannot be treated as an isolated emergency in a world where climatic extremes are becoming more consequential.
Governments need to stop treating resilience spending as optional expenditure that can be postponed until budgets improve.
Wealthy countries need to recognise that climate finance is not only about persuading developing countries to emit less.
Developing-country governments need to recognise that global climate injustice does not eliminate their responsibility for domestic planning and institutional competence.
Citizens need both arguments simultaneously.
Pakistan deserves international assistance because the burden it faces is grossly disproportionate to its historical contribution to global warming.
Pakistanis also deserve a state capable of using that assistance well.
One claim should not be used to silence the other.
Perhaps that is the broader political lesson from the floods.
Responsibility does not have to be exclusive.
A catastrophe can be intensified by global emissions, worsened by domestic vulnerability, constrained by poverty and shaped by political institutions at the same time.
Politics prefers one person to blame.
Reality usually gives us a causal chain.
If climate policy is going to become serious enough for the century ahead, governments will have to learn to govern the chain rather than argue endlessly over which single link deserves all the responsibility.
