The $5.7 Trillion Shadow: El Niño's Long-Term Economic Damage
Published: May 22, 2026 · 10 min read
For decades, the standard estimate of the 1997-98 El Niño's economic cost was $45 billion — a figure from the World Bank that covered direct damages: destroyed infrastructure, crop losses, disaster relief. It was a big number, but manageable.
In May 2023, a paper published in Science by Christopher Callahan and Justin Mankin of Dartmouth College dropped a number that made the old estimate look like a rounding error: $5.7 trillion. Not $45 billion. $5,700 billion. A factor of over 100x.
The difference wasn't in what they counted. It was in when they counted until.
The Old Way: Counting Broken Things
Traditional disaster economics works like an insurance adjuster: you tally the physical damage. How many houses flooded? What's the replacement cost of destroyed crops? How much did the government spend on emergency aid?
This approach makes intuitive sense. It also misses the entire story.
The World Bank's $45 billion estimate for 1997-98 captured the immediate wreckage: roads washed out in Peru, drought-killed cattle in Indonesia, flooded homes in California. These are real costs. But they're the costs of the event itself — not of what happens next.
The New Way: Following the Money for Years
Callahan and Mankin did something simple in concept but computationally enormous: they analyzed GDP data for dozens of countries over multiple decades, separating El Niño years from non-El Niño years, and tracked the economic trajectory for five years after each event.
What they found rewrites how we think about climate economics:
- The 1997-98 El Niño: $5.7 trillion in global income losses by 2003 (five years post-event)
- The 1982-83 El Niño: $4.1 trillion in losses
- Average El Niño event: $3.4 trillion in persistent losses
- Projected 21st-century cost of amplified ENSO: $84 trillion — even under current emissions pledges
These losses persisted for at least five years, and in some cases up to 14 years or longer. The economy doesn't bounce back — it ratchets to a lower growth trajectory and stays there.
Why Growth Stays Depressed
The mechanism isn't mysterious once you think about it:
Diverted investment. Money spent on disaster recovery is money not spent on technology, infrastructure upgrades, education, or R&D. A government rebuilding flood-damaged roads is not building a new port. A farmer replanting lost crops is not investing in irrigation.
Compounding effects. If a country loses 2% of GDP one year, it doesn't just "catch up" the next year. The lost 2% would have generated its own growth. That compounding loss accumulates over time.
Unequal burden. The world's poorest tropical nations bore the brunt. Peru, Indonesia, Brazil, and Ecuador lost 5% to 19% of GDP per capita in the aftermath of strong El Niño events. The United States — a rich, diversified economy — still saw GDP roughly 3% lower than it would have been five years after the event.
In other words: even the world's largest economy isn't immune. It just absorbs the hit better.
The Comparison That Should Worry Everyone
The Great Recession of 2008-09 caused global GDP to contract by roughly $2 trillion. Callahan and Mankin's estimate suggests the average El Niño causes $3.4 trillion in persistent losses — significantly more than the worst financial crisis in a generation.
This doesn't mean an El Niño event feels like a financial crisis in real time. The Great Recession was sharp and fast — banks collapsing, markets panicking. El Niño's economic damage is slow and distributed: a farmer in Zambia loses a harvest, a fishery in Peru closes for six months, a California highway washes out. None of these individually trigger a market panic. But they add up, year after year.
| Event | Estimated Global Economic Loss | Timeframe |
|---|---|---|
| 2008 Great Recession | ~$2 trillion | 1-2 years |
| 1982-83 El Niño | $4.1 trillion | 5+ years |
| 1997-98 El Niño | $5.7 trillion | 5+ years |
| 2023-24 El Niño | $75 billion (direct only) | Long-term effects unknown |
The $84 Trillion Question
The Dartmouth paper's most alarming projection isn't about the past. It's about the 21st century.
Using climate models that simulate how ENSO behavior changes as the planet warms, they projected that even under current emissions reduction pledges, the amplified ENSO cycle could cost the global economy $84 trillion by 2100. That's roughly the current size of the entire global economy.
This number is contested. Cambridge macro-economist Kaimar Mohaddes questioned whether the magnitude was realistic, noting it approaches the cost of the Great Recession multiple times over. Stanford's Marshall Burke, however, said the authors "make a compelling case" — emphasizing that the study captures opportunity costs that previous research ignored entirely.
What This Means for 2026
If the developing 2026 El Niño reaches the intensity of 1997-98 — and the subsurface data suggests that's a real possibility — the economic math changes. We're not talking about whether a bridge washes out or a harvest fails. We're talking about whether the global economy's growth trajectory gets ratcheted down for the next half-decade.
Three things make the 2026 scenario potentially worse than 1997:
1. Higher baseline debt. Many developing countries entered 2026 with higher sovereign debt than in 1997. Less fiscal space means less ability to absorb the shock without cutting investment.
2. Food system fragility. The 2023-24 event demonstrated how quickly an El Niño-triggered rice shortage can cascade through global markets. India's export ban sent prices up 38% in months. A stronger event in 2026 would test reserves that are still recovering.
3. Insurance withdrawal. Major insurers have pulled out of California, Florida, and parts of Australia in the last five years. When the next El Niño-fueled disaster hits, more of the damage falls on households and governments rather than being absorbed by insurance pools.
The Dartmouth study's core insight is that treating El Niño as a series of disconnected weather disasters — each with its own cleanup bill — fundamentally misunderstands how the damage works. It's a persistent drag on economic growth, not a one-time shock. And with the 2026 event building, that distinction matters more than ever.
Data sources: Callahan, C.W. & Mankin, J.S. "Persistent effect of El Niño on global economic growth." Science, May 18, 2023. DOI: 10.1126/science.adf2983; World Bank disaster damage estimates; WMO 2023-2024 El Niño event summary; Bloomberg economics coverage of Dartmouth study (May 2023).