Commentary / World

What global warming? Pass me a blanket

by Cass R. Sunstein


“Global warming strikes America! Brrrr!” So tweeted Missouri Representative Vicky Hartzler last week, as much of the U.S. experienced extreme cold. (In Buffalo, it was a full Snowpocalypse.) Do frigid temperatures give you doubts about global warming?

You wouldn’t be alone. When people think the day’s weather is exceptionally cold, research shows, they’re less likely to be concerned about global warming. And when the day seems unusually hot, concern jumps.

Notably, this effect can be found among Republicans and Democrats, men and women, young and old. Strange but true: As the question of climate change receives sustained attention in Washington in the coming months, public opinion is likely to be affected by short-term swings in the weather.

To study this phenomenon, Eric Johnson, Ye Li and Lisa Zaval of Columbia University’s Center for Decision Sciences, asked almost 600 Americans two questions. The first was whether they considered the local temperature, on the day of the survey, to be colder or warmer than usual (on a five-point scale from -2, meaning “much colder,” to +2, “much warmer”). The second question was whether they believed that global warming is happening and whether they were concerned about it (on a 4-point scale from 0, “not at all convinced/worried” to 3, “completely convinced/a great deal worried”).

The researchers found that when people felt the day was warmer than usual, they were significantly more likely to believe in and worry about global warming than when they considered the day to be unusually cold. The effect was substantial, with a “much colder” day producing a full one-point decrease in both belief and worry. (The researchers found the same basic results in Australia.)

You might wonder whether they got causation backward. Maybe the day’s temperature didn’t affect people’s thinking about global warming; maybe people who worry about global warming were more likely to perceive the temperature as warmer.

But after checking actual temperatures on survey days, Johnson and his colleagues found that, sure enough, the weather mattered. And even when the researchers went out of their way to inform respondents that minor fluctuations in weather are to be expected during climate change, the day’s temperature affected their answers.

A follow-up study found that, on exceptionally warm days, people were also far more likely to donate money to a charity concerned about global warming, and they were likely to donate more money as well — 500 percent more than on cold days.

What’s going on here? The best explanation probably involves “attribute substitution,” a pervasive phenomenon described by Daniel Kahneman, a behavioral scientist who won the Nobel Prize in economics. This occurs when we assess a difficult question having to do with science, politics or statistics (say, the risk of violent crime in our neighborhood) by asking an easier one (whether we can recall recent cases of violent crime).

Johnson and his coauthors think that people subconsciously use the current local temperature (which is easy to know) as a clue to whether global temperatures are increasing (a much harder question). That’s unfortunate. When it comes to deciding whether climate change is occurring, and worthy of concern, “Brrrr!” isn’t much of an argument.

Cass R. Sunstein, a former administrator of the White House Office of Information and Regulatory Affairs, is the Robert Walmsley university professor at Harvard Law School and a Bloomberg View columnist.

Coronavirus banner