Estimates of Economic Losses from El Niños Are Far-fetched

A recent study makes the provocative claim that some of the most intense past El Niño events cost the global economy from $4 trillion to $6 trillion over the following years. That’s two orders of magnitude higher than previous estimates, but almost certainly wrong.

One reason for the enormous difference is that earlier estimates only examined the immediate economic toll, whereas the new study estimated cumulative losses over the five-year period after a warming El Niño. The study authors say, correctly, that the economic downturn triggered by this naturally occurring climate cycle can last that long.

However, even when this drawn-out effect is taken into account, the new study’s cost estimates are still one order of magnitude greater than other estimates in the scientific literature, such as those of the University of Colorado’s Roger Pielke Jr., who studies natural disasters. His estimated time series of total weather disaster losses as a proportion of global GDP from 1990 to 2020 is shown in the figure below.

The accounting used in the new study includes the “spatiotemporal heterogeneity of El Niño teleconnections,” teleconnections being links between weather phenomena at widely separated locations. Country-level teleconnections are based on correlations between temperature or rainfall in that country, and indexes commonly used to define El Niño and its cooling counterpart, La Niña. Teleconnections are strongest in the tropics and weaker in midlatitudes.

The researchers’ accounting procedure estimates total losses from the 1997-98 El Niño at a staggering $5.7 trillion by 2003, compared with a previous estimate of only $36 billion in the immediate aftermath of the event. For the earlier 1982-83 El Niño, the study estimates the total costs at $4.1 trillion by 1988. The calculated global distribution of GDP losses following both events is illustrated in the next figure.

To see how implausible these trillion-dollar estimates are, it’s only necessary to refer to Pielke’s graph above, which relies on official data from the insurance industry (including leading reinsurance company Munich Re) and the World Bank. His graph indicates that the peak loss from all 1998 weather disasters was 0.38% of global GDP for that year.

As El Niño was not the only disaster in 1998 – others include floods and hurricanes – this number represents an upper limit for instant El Niño losses. Using a value for global GDP in 1998 of $31,533 billion in current U.S. dollars, 0.38% was a maximum instant loss of $120 billion. Over a subsequent 5-year period, the maximum loss would have been 5 times as much, or $600 billion assuming the same annual loss each year which is undoubtedly an overestimate.

This inflated estimate of $600 billion is still an order of magnitude smaller than the study’s $5.7 trillion by 2003. In reality, the discrepancy is larger yet because the actual 5-year loss was likely much less than $600 billion as just discussed.

Two other observations about Pielke’s graph cast further doubt on the methodology of the researchers’ accounting procedure. First, the strongest El Niños in that 21-year period were those in 1997-98, 2009-10 and 2014-16. The graph does indeed show peaks in 1998-99 and in 2017, one year after a substantial El Niño – but not in 2011 following the 2009-10 event. This alone suggests that financial losses from El Niño are not as large as the researchers think.

Furthermore, there’s a strong peak in 2005, the largest in the 21 years of the graph, which doesn’t correspond to any substantial El Niño. The implication is that losses from other types of weather disaster can dominate losses from El Niño.

It’s important to get an accurate handle on economic losses from El Niño and other weather disasters, in case global warming exacerbates such events in the future – although, as I’ve written extensively, there’s no evidence to date that this is happening yet. Effects of El Niño include catastrophic flooding in the western Americas, flooding or episodic droughts in Australia, and coral bleaching.

The study authors stand by their research, however, estimating that the 2023 El Niño could hold back the global economy by $3 trillion over the next five years, a figure not included in their paper. But others are more skeptical. Climate economist Gary Yohe commented that “the enormous estimates cannot be explained simply by forward-looking accounting.” And Mike McPhaden, a senior scientist at NOAA (the U.S. National Oceanic and Atmospheric Administration) who was not involved in the research, called the study “provocative.”

Next: Targeting Farmers for Livestock Greenhouse Gas Emissions Is Misguided