Heat Waves Kill Sick People. Cold Snaps Kill Healthy Ones

Detroit experienced record-breaking subzero temperatures on Jan. 6 Photograph by Joshua Lott/Getty Images

Heat waves usually get all the press, but cold snaps are deadly, too. Already the polar vortex is being blamed for at least 16 deaths—mostly from traffic accidents and snow-shoveling-induced heart attacks—and more are sure to come. In fact, if a National Bureau of Economic Research paper from a few years back is correct, cold snaps are far bigger killers than heat waves. The paper, written by Olivier Deschenes and Enrico Moretti, was unearthed this morning by economist and blogger Tyler Cowen.

Attributing deaths to meteorological events can be a tricky business (as one commenter on Cowen’s blog points out, if traffic accident deaths during a cold snap are blamed on the weather, why not count the lives saved because the weather keeps people at home who would otherwise be out on the roads?), but Deschenes and Moretti, looking at overall mortality figures among American whites, found that both heat waves and cold snaps clearly do cause spikes in death rates. The main cause wasn’t the car accidents that get reported on the news, though, but something that shows up only in epidemiological data.

In both heat and cold, deaths were caused mostly by heart attacks, strokes, and asthma attacks—results of the stress placed on the body’s cardiovascular and respiratory system by maintaining core temperature in extremis. And looking at mortality figures over time the economists found a key difference between extreme heat and cold: While heat waves were followed by periods of lower-than-usual mortality, cold snaps were not. In other words, more people than usual died during heat waves, but then fewer died after, wiping out any long-term effect. Not so with cold snaps.

What’s going on here? Deschenes and Moretti argue that the people who die during heat waves are usually those whose health is already seriously compromised—people who would have died soon anyway. The technical (and unsettling) term for this is the “harvesting effect.” Cold, by contrast, kills people who had years left to live, and it kills lots of them. The authors write:

“We estimate that the number of annual deaths attributable to extreme cold temperature in the white population is 14,380, or almost 360 deaths per cold day. This roughly corresponds to 0.8% of average annual deaths in the United States during the sample period. … [T]his total exceeds the annual deaths due to leukemia, homicide, and chronic liver disease/cirrhosis. The overall impact on longevity is substantial: the average person who died because of cold temperature exposure lost in excess of 10 years of potential life.”

Ten years is a lot of potential life to lose, enough to make one reconsider living in a cold climate. Is this, then, an upside to climate change, which promises hotter summers and warmer winters? Not necessarily: The two authors avoid the issue, and the exact relationship between global warming and extreme weather events remains unclear—there’s some evidence that cold snaps, counterintuitively, could be more common in a warmer future.

Looking backward, however, Deschenes and Moretti do argue that the steady movement of Americans from colder to warmer parts of the country in recent decades—especially among those older Americans more susceptible to extreme temperatures—has played a role in increasing life expectancy: 3 percent to 7 percent of the gains in longevity, they suggest.

So please, stay warm.

    Before it's here, it's on the Bloomberg Terminal.