The Flawed Methodology Behind Studies Measuring the Cost of Climate Change

New report shows climate studies blend line between correlation and causation, fail to account for human adaption

climate change
Getty Images
March 20, 2018

Many prominent studies used by the federal government to measure the costs of climate change rely on a flawed methodology that incorporates faulty economic models and fails to account for human adaption, according to a new report.

On Monday, a report released by the Manhattan Institute shed light on how the doom and gloom climate change scenarios painted by the federal government and those in academia are actually the result of  "laughably bad economics."

For instance, a study conducted by researchers from Stanford University and the University of California, Berkeley estimated that countries such as Mongolia would experience unbridled economic growth because of climate change, resulting in the average Mongolian earning four times more than the average American by 2100.

The Manhattan Institute's report, written by Oren Cass, outlines how studies attempting to gauge the economic and social costs of climate change are inherently speculative. Cass, a former domestic policy director for the Romney campaign, explains these studies require the difficult translation of forecasted temperature increases into measurable effects on weather, such as rising sea tides. Those effects must then be translated into societal impacts, such as depleted freshwater sources, which must be further translated to determine fiscal impact.

This process results in the over-reliance on statistics analyzing variations in temperature, and it creates a correlation-based approach rather than one based on causation.

The linking of rising temperatures to increased mortality rates and lower economic output, on the basis of historical correlations, is flawed and provides results that vastly exceed those "from more traditional analyses of climate change's expected effects on the physical world."

The aforementioned Berkley-Stanford study found warm countries experienced lower economic growth in abnormally warm years, while cold countries experienced higher levels of growth. The economic model designed by the California academics concluded unmitigated climate change would decrease gross domestic product (GDP) by more than 20 percent by 2100. The study did not take into account a country's existing resources, but rather only drew estimations by looking at the "historical relationship between temperature and a country's economic output."

The authors of the Berkley-Stanford study did not return requests from the Washington Free Beacon for comment.

Cass notes studies that rely upon historical correlations to portray dystopian consequences have become increasingly ascendant in the ever-polarizing conversation on climate change.

"These studies have gained rapidly in prominence," Cass wrote. "They now account for the overwhelming share of costs in climate assessments."

In his report, Cass also offers a searing indictment of studies conducted by the Environmental Protection Agency (EPA) and the Government Accountability Office (GAO), citing the often made mistake of not accounting for the variable of human adaption. While Cass himself notes that human adaption is difficult to predict, the willful avoidance of it by climatologists casts doubt on the accuracy of their doomsday-like prognostications.

A study conducted by the EPA in 2015 attempted to estimate the number of heat-related mortalities that will result from climate change-related temperature increases. The study, incorporating data from 33 cities across the United States, operated under the assumption that a day considered unusually hot in 2000 will cause a similar mortality rate in 2100. The study projected that abnormally warm weather in northern cities by 2100 will result in nearly 12,000 heat-related deaths annually.

Cass notes the EPA study does not, however, account for the fact that "southern cities such as Phoenix, Houston, and New Orleans, were already hotter in 2000 than northern cities are predicted to be in 2100." The EPA's predicted heat-related mortality rates in 2100 for northern cities is nearly 50 times higher than the actual heat-related mortality rates that southern cities experienced in 2000.

The EPA study estimates that by 2100, climate change will cost the U.S. economy between $1.3 trillion and $1.5 trillion annually. As Cass notes, at least 89 percent of this sum was derived from economic models utilizing temperature studies that do not account for human adaption and blend the line between correlation and causation.

The EPA did not respond to a request for comment on this story.

The Manhattan Institute report finds attempts to extrapolate future costs from present realities, often without taking into account a region's ability to adapt to changing temperatures, lead to "absurd estimates."

The irony, Cass points out, is studies that attempt to understand the qualitative costs of climate change often–however inadvertently–fail to account for future change and adaption.

Published under: Climate Change , EPA