Baseline construction shapes estimates of mitigation cost, ambition and progress, while no-policy counterfactuals could vary with choices about policy, technology and development. Rather than relying on one constructed baseline, assessments and related conclusions should be compared across plausible baselines.
In 2020, the International Energy Agency (IEA) discontinued the Current Policies Scenario (CPS) from its World Energy Outlook, arguing that turmoil in energy markets and rapid changes in the policy landscape during the COVID-19 pandemic made it difficult to define a stable reference scenario. However, in 2025, the agency restored it as the world had passed through the pandemic and energy crisis, and there was merit in revisiting the scenario1. Yet meanwhile, the IEA warned that the CPS is neither a forecast nor business-as-usual (BAU) scenario and requires transparency about its assumptions1. That raises a more fundamental question about whether any single constructed future should be seen as the baseline against different climate actions.
Policy-frozen experiments are often used to evaluate counterfactual futures, but their assumptions still leave space for more than one possible future. Here, a no-policy scenario denotes a counterfactual that freezes specified climate-policy effects; BAU adds the claim that the counterfactual describes what would otherwise occur. Scenario design has started moving away from that framing: the priority set for the seventh phase of the Coupled Model Intercomparison Project (CMIP7) has no conventional no-policy baseline; its Medium scenario extends current policies, while its High scenario requires policy rollback2. However, many downstream assessments still rely on the same few conditional constructions as reference worlds.