American and German researchers led by Jennifer Jacquet of New York University put together a collective-risk group experiment that is centered around climate change. Here’s how it worked. Each subject in groups with six participants was given a $55 operating fund. The experiment went 10 rounds, and during each round, they were allowed to choose one of three options: invest $0, $2.75 or $5.50 into a climate account. The participants were told that the total amount contributed would go to fund an advertisement on climate change in a German newspaper. If at the end of the 10 rounds, the group reached a target of $165 — or about $27 per person — they were considered to have successfully averted climate change, and each participant was given an additional $60 dollars. (If the numbers seem rough, it’s because I’m converting from euros — the currency used in the experiment — and rounding off.) If the group failed to reach the $165 target, there was a 90% probability that they wouldn’t get the additional payout. As a group, members would be better off if they collectively invested enough to reach that $165 target — otherwise they wouldn’t get the payout — but individually, members could benefit by keeping their money to themselves while hoping the rest of the group would pay enough to reach the target. (That’s the so-called free-rider phenomenon, and it’s a major challenge for climate policy.)

Here’s the twist, though: that $60 dollar endowment was paid out on three different time horizons. In one treatment, the cash was given to the groups the next day. In the second treatment, it was given seven weeks later. And in the third treatment, the cash was instead invested in planting oak trees that would sequester carbon — but since those trees wouldn’t be fully grown for years, all the benefit would accrue to future generations, not the current players in the experiment. The difference between that third treatment and the first and second is what’s known as “intergenerational discounting,” which happens when the benefits of an action in the present are highly diluted and mostly spread among many people in the future. Which, as it happens, is pretty much how climate policy would work.

Unsurprisingly, the more delayed the payout was, the less likely the experimental groups would put enough money away to meet the goal to stop climate change.