‘Slow Burn’

The following is an excerpt from “Slow Burn: The Hidden Costs of a Warming World” by R. Jisung Park, assistant professor in the School of Social Policy & Practice (©2024, Princeton University Press).

In October 2012, Superstorm Sandy hit the northeastern United States, flooding streets and homes, cutting power to millions, and causing extensive property damage from Florida to Maine. Some of the most enduring images from the storm’s aftermath come from New York City, where widespread power outages combined with flooding of subway tunnels to create a scene out of some dystopian horror movie. As a former New York City resident, it is difficult to fully scrub from my mind footage of entire blocks of Lower Manhattan underwater, or of sewage-infused storm water overflowing subway platforms.

Of course, such damages were only the more iconic tip of the iceberg: downed power lines meant over 6 million residents went without electricity across Washington, DC, New Jersey, Pennsylvania, and Connecticut; major transit systems and airports were shut down for days; thousands of homes and businesses suffered property damage. All told, official estimates put the economic damages from Hurricane Sandy at over $85 billion, making it one of the costliest hurricanes in history.

What is less often appreciated is how disruptive Hurricane Sandy was for young people. The storm shut down all 1,750 New York City public schools, halting learning for over 1 million students for an entire week, likely causing pedagogical ripple effects that lasted far longer.^[1] Schools in Washington, DC, Boston, Philadelphia, and Providence, Rhode Island, were temporarily shuttered as well.

When environmental disasters hit, the damage to physical capital—so-called brick and mortar assets—is usually quite apparent. Downed power lines; bridges washed away by floods; the charred remains of homes and cars in the wake of wildfires. We see such footage in the news all the time.

Less obvious is the harm done to human capital, which, broadly speaking, can be thought of as the capacity of human agents to engage in productive economic activity. In an increasingly knowledge-based economy, the skills and capacities learned in school comprise an asset base every bit as essential as the bricks in a building or the steel rods in a factory.

Could it be possible for largely invisible disruptions to human capital formation to have economic implications that are deeper and more impactful than first meets the eye? This is important because educational attainment and the skills that schooling and training confer have become an increasingly valuable input to success in the modern economy. This is true both in terms of aggregate economic performance—the relative wealth and poverty of nations—as well as for individual economic mobility, that is, for the chances of moving up the economic ladder within a given society. As we will discuss in greater detail in later chapters, the implications of climate change for economic inequality may hinge in part on how readily individuals can access the education and skills necessary to compete in an increasingly technologically complex and automated workplace.

Physical capital assets tend to be much easier to value monetarily, in part because they usually comprise tangible quantities traded on the market. You could probably look up the estimated trade-in value of your car in a matter of minutes on your smart phone. Even in the case of bridges or airports, which are not bought and sold on markets like homes and cars, the cost of repair—in terms of materials and construction labor—can be readily calculated.

Human capital, on the other hand, is often less tangible, and certainly far less fungible. Indeed, there are a great many such intangible assets that are vital to human flourishing. These include not only educational achievement and job-specific skills, but also mental health and physical vitality (what economists call “health capital”). One can also think of things like the air and water filtration benefits of nature (known as “ecosystem services or natural capital”), and even the historical significance of a particular building or monument or the degree of social cohesion or sense of safety one feels in a neighborhood, as forms of intangible capital that, while not actively traded on markets, are foundational to our quality of life.

What do we know about the effects of climate on such intangible, nonmarket assets? Increasingly, the evidence suggests that headline damage estimates—which typically focus on physical capital—may miss a wide swath of economic harm.

If a storm knocks all the apples out of a fully grown apple tree before they are ready to be picked, the losses are significant, but with some care the harvest can bounce back the next season. If a storm damages the roots of a growing seedling, however, it may eventually mature to produce fewer or more often diseased apples every year than otherwise might have been the case. Over the lifetime of an apple tree, the difference in the cumulative number of harvested apples may turn out to be very different between the damaged seedling and the average tree. Could there be something similar about the way children learn and grow that makes them especially susceptible to adverse shocks—natural disasters included?

Economists Isaac Opper, Lucas Husted, and I were interested in what happens in the case of less obviously catastrophic natural disasters. We were particularly interested in how the human capital consequences of such run-of-the-mill natural disasters might stack up in dollar terms.^[2]

Fortunately for us, the US government takes great pains to document the property damage associated with all manner of natural disasters, ranging from hurricanes and floods to tornadoes and wildfires. Because federal aid dollars are administered on the basis of assessed physical property damages, government agencies spend considerable time and resources trying to tally the dollar value of damaged and lost property for any given natural disaster.

Unfortunately for us, there is far less information on the human capital damages. So we pulled together data on standardized student achievement for nearly all third through eighth graders in the United States (available thanks to many years of data collection and harmonization on part of researchers at Stanford), as well as information on high school graduation and post-secondary enrollment rates across the country between 2008 and 2018. We combined this with information from the universe of presidential disaster declarations over a roughly thirty-year period, as well as detailed local property damage assessments.

We measured the impact of natural disasters of varying magnitudes on these two measures of human capital: student achievement, in the form of gains in test scores, and educational attainment, in the form of changes in high school graduation and college enrollment. We exploited the fact that many places in the United States are hit by multiple types of natural hazards of varying intensity, repeatedly over time. For example, we could compare the gains in learning that occur in a particular school district in the Chicago area in an average school year, to the gains that occurred in that same Chicago district in school years where students were exposed to a natural disaster, also controlling for factors that may have affected learning across the board (like the Great Recession), as well as potential trends in disaster frequency and local economic and demographic trends.

Our hypothesis was that, by being forced to cope with a potentially wide range of additional disruptions to learning, students who experienced natural disasters during the school year may eventually perform worse on exams aimed at testing their knowledge of the subjects studied during that time.

Consistent with other work, we find that, across a range of disaster types, natural hazards can indeed be highly disruptive to learning. Test scores drop significantly in school years where students experienced a natural disaster. College enrollment drops as well. In both cases, disasters in future years have no impact on current test scores or college enrollment, which is what we would expect if year-to-year variation in disasters was in fact more or less randomly assigned.

Through this analysis, we wanted to provide at least a ballpark sense of how bad the educational impacts are for a given amount of property damage. To do this, we borrow from previous studies that carefully assess the economic benefits of education—both in terms of educational attainment (years of education) and the content of that education in the form of standardized achievement.

Natural disasters that cause on average $10 to $100 per capita in physical property damage appear to result in human capital damages of approximately $41 per capita ($245 per student). The effects appear to scale more or less proportionately with the magnitude of disasters, at least when measured in terms of physical destructiveness. Disasters that cause more than $500 per person in assessed physical capital damages lead to human capital impacts on the order of $268 per person, or $1,520 per student.

This suggests that the learning losses may be in the same order of magnitude as the physical capital destruction. Zooming out, it seems possible that, when it comes to climate change, the less-readily monetized impacts on intangible capital—whether on student learning or health or something else like ecosystem services—may be of a magnitude warranting legitimate concern.

^[1] Rajashri Chakrabarti and Max Livingston, “The Impact of Superstorm Sandy on New York City School Closures and Attendance,” Liberty Street Economics, December 19, 2012, https://libertystreeteconomics.newyorkfed.org/2012/12/the-impact-of-superstorm-sandy-on-new-york-city-school-closures-and-attendance/.

^[2] Opper, Isaac M., R. Jisung Park, and Lucas Husted. “The effect of natural disasters on human capital in the United States.” Nature human behaviour (2023): 1-12.