Adaptive investment behaviours enable earlier carbon peak in low-carbon electricity transition

Electricity planning often employs average production costs and efficiencies, assuming that energy companies are homogeneous. In reality, firms are heterogeneous in the scale, cost and efficiency of their power generation, leading to various adaptive behaviours such as differentiated investments in CO2 reductions and offsets. This has resulted in misjudgments about climate tipping points and unanticipated delays in phasing out coal-fired technology. Here we develop an agent-based model using company-level technology-specific installed capacity, cost and efficiency within China’s electricity sector. We estimate renewable energy expansion and carbon peaks by simulating companies’ adaptive behaviours. The results indicate that if such behaviours focused on technology stock and return-on-investment, China’s electricity sector could have reached carbon peak by 2019-2020, ahead of the national 2030 target. By analyzing adaptive investment behaviours, we pinpoint distinct profit strategies for companies with varying technology capacities. This enables tailored incentives, avoiding the fiscal inefficiencies of one-size-fits-all policies like uniform feed-in-tariffs. Our behavioural analysis further illustrates that the delayed phase-out of coal-fired technology, due to lock-in effects, could result in coal power overcapacity and hinder the operation of renewable energy generation. Policies to accelerate the phase-out of coal capacity are becoming increasingly imperative relative to incentivizing the scale-up of renewables.