Sensitivity of collective action to uncertainty about climate tipping points

Abstract

Despite more than two decades of diplomatic effort, concentrations of greenhouse gases continue to trend upwards, creating the risk that we may someday cross a threshold for ‘dangerous’ climate change^1,2,3. Although climate thresholds are very uncertain, new research is trying to devise ‘early warning signals’ of an approaching tipping point^{4,5,6,7,8,9,10,11}. This research offers a tantalizing promise: whereas collective action fails when threshold uncertainty is large, reductions in this uncertainty may bring about the behavioural change needed to avert a climate ‘catastrophe’⁵. Here we present the results of an experiment, rooted in a game-theoretic model, showing that behaviour differs markedly either side of a dividing line for threshold uncertainty. On one side of the dividing line, where threshold uncertainty is relatively large, free riding proves irresistible and trust illusive, making it virtually inevitable that the tipping point will be crossed. On the other side, where threshold uncertainty is small, the incentive to coordinate is strong and trust more robust, often leading the players to avoid crossing the tipping point. Our results show that uncertainty must be reduced to this ‘good’ side of the dividing line to stimulate the behavioural shift needed to avoid ‘dangerous’ climate change.