Cost-effectiveness and implications of GWPs and GTPs under alternative policy goals Andy Reisinger1 Keywan Riahi2 Oscar van Vliet2 1 New Zealand Agricultural Greenhouse Gas Research Centre 2 International Institute for Applied Systems Analysis (IIASA) Manuscript submitted to Climatic Change Work funded by NZ Ministry of Agriculture and various EU-FP7 programs Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

In a nutshell 100-year GWPs are not a cost-effective way of comparing GHGs if the main policy goal is to limit long-term climate change. Few studies have explored the cost and climate policy implications if other physically-based metrics were to replace GWPs. Determine the global cost-effectiveness of different metrics for the main policy goal of limiting radiative forcing in 2100 to 450 or 550ppm CO2-equivalent Evaluate influence of metrics on additional policy goals (realised warming, GDP, timing of CO2 emissions peak) Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

Evaluated alternative metrics: GWPs, fixed and time-dependent GTPs CH4 N2O Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

Sensitivity tests: technology/policy assumptions Assumed policy goal is to limit radiative forcing in 2100: 450 ppm CO2-eq (~2.7 Wm-2) 550 ppm CO2-eq (~3.8 Wm-2) Rate of improvement of agricultural mitigation potential No improvement / rapid improvement Additional technology from 2030 or 2070 to mitigate CH4 from enteric fermentation Policy treatment of agricultural GHG emissions Fully included in global mitigation efforts / fully excluded / excluded until 2050 Use integrated assessment model MESSAGE to determine cost-minimising abatement pathways over 21st century Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

Radiative forcing paths – all metrics/assumptions Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

CO2 and CH4 emissions – all metrics/assumptions Global agricultural marginal abatement costs from Beach et al. (2008) Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

CH4 emissions – detailed examples fixed 100-year GTPs: less CH4 mitigation time-dependent GTPs: less CH4 mitigation initially, more CH4 mitigation by 2100 Global agricultural marginal abatement costs from Beach et al. (2008) Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

Global agricultural CH4 emission pathways Global net costs MESSAGE results 450ppm CH4 mitigation determines ‘atmospheric space’ for CO2 emissions and hence total mitigation costs 550ppm Global agricultural CH4 emission pathways

Global agricultural CH4 emission pathways Global discounted net present value mitigation costs (2010-2100) Example for 550ppm CO2-eq stabilisation Global agricultural CH4 emission pathways

Global cost-effectiveness of metrics Fixed GTPs result in higher CO2 prices and higher total mitigation costs than GWPs, but lower prices/costs on CH4 Time-dependent GTPs (focusing on year 2100) result in lower CO2 prices and lower total mitigation costs than GWPs; prices and costs for CH4 are lower initially but (much) higher later Assumptions about agriculture mitigation potential have a larger effect on global costs than alternative metrics Different long-term stabilisation targets have a much larger effect than alternative metrics Excluding agriculture globally is by far the most costly ‘metric’ Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

What about other relevant policy goals? Realised amount of warming and overshoot GDP impacts Timing of cost-effective CO2 emissions peak Regional implications Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

Realised warming – rate and peak Peak: 2-2.25°C Rate (2020-2050): 0.2-0.26°C / decade Most of the differences are due to alternative assumptions, not alternative metrics Global agricultural marginal abatement costs from Beach et al. (2008) Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

GDP impacts Time-dependent GTPs result in lower aggregate costs, but greater GDP losses relative to BAU towards 2100 than GWPs Global agricultural marginal abatement costs from Beach et al. (2008) Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

Timing of cost-effective CO2 emissions peak Different metrics and assumptions allow the CO2 emissions peak to be delayed by almost 15 years But most of the delay comes from different assumptions; metrics have a smaller influence All pathways shown result in radiative forcing of 450ppm CO2-eq in 2100, but the timing of CO2 mitigation differs due to metrics and other assumptions Global agricultural marginal abatement costs from Beach et al. (2008) Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

Conclusions GWPs are not the most cost-effective metric to compare GHGs IF the main goal is to limit long-term radiative forcing in 2100, and to do so via cost-minimising global abatement pathways Fixed 100-year GTPs are even less cost-effective (+ 5 to 10%) time-dependent GTPs would be more cost-effective (- 4 to 5%) Cost implications of alternative metrics: smaller than alternative assumptions about future agricultural mitigation potential, and much smaller than choices of long-term target Other policy goals: different equivalent metrics do not result in equivalent other environmental outcomes – but differences are again smaller than those arising from other assumptions Regional implications: regional effect of metrics on production systems and land-use change – see second presentation please! Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |

Questions to the policy community How important is it for the policy process to have a metric that is optimal for a particular policy goal but, almost by definition, will do a poorer job for other policy goals? How sustainable is the implementation of a metric that implies escalating cost of CH4 emissions globally? What are the social/policy benefits and costs of continuously updating a metric to achieve optimality (however defined)? Are tests of metrics useful that assume full global and sectoral participation at full price levels? What are the implications of different metrics for regional and sectoral engagement to climate change? Copyright © 2010 New Zealand Agricultural Greenhouse Gas Research Centre 18 January 2019 |