Adaptive mechanisms to refund emissions payments

Abstract

Carbon pricing is often discussed as a means by which to reduce power sector emissions, but typically faces opposition on the grounds that such schemes will increase wholesale electricity prices and have adverse impacts on the wider economy. Consequently, there is renewed focus on emissions pricing initiatives that have muted price impacts, with a Refunded Emissions Payments Scheme one such example. Under this mechanism, collected emissions payments are refunded to generators in proportion to energy output, with the net effect of the scheme resulting in each generator being subject to a net liability per MWh that is proportional to the difference between it’s own emissions intensity and the average emissions intensity of energy produced by regulated plant. A limitation of the scheme is the average emissions intensity of energy produced by regulated plant is only known after energy has been delivered, therefore there is some uncertainty with respect to each generator’s short-run marginal cost at the time of production. The presented framework addresses this issue by developing adaptive recalibration strategies that seek to make these net liabilities explicit. A rule-based updating strategy is developed, along with a highly generalisable model predictive control framework for updating scheme parameters. A case study using an agent-based framework is used to evaluate these mechanisms in the context of Australia’s National Electricity Market, and demonstrates how a policymaker could use such an approach to transparently update scheme parameters, and also accomplish auxiliary economic objectives.