This report explores three stabilisation levels at the lower end of the range explored in the literature: 450, 510 and 550 ppm CO2−e. The level of global mitigation effort is an important determinant of overall economic impact.
Stabilisation at these levels requires a fundamental shift in emission trends. Once that occurs, the differences — in terms of aggregate economic impacts — are relatively small.
This report uses a suite of models spanning global, national, sectoral and household scales to provide an integrated set of projections across these four dimensions.
In economic terms, climate change results from the externality associated with greenhouse gas emissions: it creates costs not paid for by those who generate the emissions.
The policy challenge is to reduce greenhouse gas emissions. Policy options include imposing a price on emissions, establishing technology and efficiency standards, supporting research and development of low-emission technologies, and instituting direct command and control regulation of emission sources.
An efficient policy reduces the overall economic cost of achieving any given emission reduction objective. Pricing emissions — by introducing a trading scheme — is a cost-effective option. A broad-based price measure ensures all emitters face the same cost for any extra unit of greenhouse gas emitted. This encourages uptake of the lowest-cost opportunities to reduce emissions.
The Government’s planned Carbon Pollution Reduction Scheme will price emissions. Large emitters will need a permit for every tonne of greenhouse gas they emit. Permits will be traded in a market, with emitters competing to buy the number of permits they require. At any given time, some emitters will find it cheaper to reduce emissions than to buy permits (DCC, 2008a).
An important determinant of the permit price will be the scheme cap. This sets the upper limit on the number of permits available in the market. The Government will set the cap, taking into account the costs and benefits of different emission trajectories, and the evolving international framework for mitigation action. The international context is crucial, as the level of global action determines the climate change risks Australia will face. This report, and the economic modelling it presents, positions Australia’s action within a global framework to reduce emissions and stabilise atmospheric greenhouse gas concentrations.
This chapter outlines the analytical framework of this report. It introduces the reference and policy scenarios modelled, and the assumptions regarding global and Australian mitigation action. It also describes the economic models used. Finally, it provides guidance on interpreting the modelling results.
Without new climate change mitigation policies, global greenhouse gas emissions are projected to rise to double current levels by the late 2030s, then nearly double again by 2100. As a result, greenhouse gas concentrations increase from around 430 ppm CO2−e today to 1,560 ppm by 2100.
Wide uncertainties surround these projections and their economic impacts. Nevertheless, climate science provides a strong case for achieving global emission pathways that reduce the probability of extreme, irreversible damages from climate change (Box 2.1). Such pathways aim for low and stable long-term greenhouse gas concentration targets, with minimal or no overshooting.4
Aggregate global emissions ultimately have to fall to less than one quarter of current levels to allow and maintain stabilisation of greenhouse gas concentrations in the atmosphere (Pearman, 2008). Faster emission reductions would achieve lower concentrations and limit overshooting; delayed or gradual reductions lead to higher concentrations. This report explores three stabilisation levels: 450, 510 and 550 ppm CO2−e. These levels are at the lower end of the range in the literature (IPCC, 2007b). This is important because the level of global mitigation effort is an important determinant of overall economic impact.
Box 2.1: Climate change risks at different stabilisation levels
The global average surface temperature has risen 0.8°C since 1850, and will rise further in the coming decades as a result of emissions that already have occurred.
Climate change risks increase with additional warming. Climate science provides strong evidence of system thresholds, beyond which adverse impacts could increase in a non-linear way (Pearman, 2008).
Without new policies to reduce emissions, the reference scenario sees continued strong growth in global emissions. As a result, the concentration of greenhouse gases in the atmosphere rises strongly to 1,560 ppm CO2−e by 2100.
This corresponds to an increase in global average temperature of more than 5°C above pre-industrial levels by 2100, and substantially more in subsequent centuries (to 8°C or more above pre-industrial levels).5 A temperature increase of this magnitude brings very high risks of extreme and irreversible climate change impacts, including the loss of complete ecosystems such as the Great Barrier Reef; severe water availability problems; significant and widespread food shortages; large areas of Australia’s coastline permanently or periodically inundated; and greater international instability (Pearman, 2008).
Stabilisation of atmospheric concentrations requires significant cuts in global greenhouse gas emissions. The stabilisation level depends on how soon emissions peak and how quickly they decline. Lower stabilisation levels require global emissions to peak within the coming decade and fall well below current levels by 2050 (IPCC, 2007b).
Global mitigation action to achieve stabilisation at 450 ppm CO2−e could limit global average warming to around 2°C above pre-industrial levels. This threshold is most frequently mentioned in the scientific literature as the limit beyond which ‘dangerous’ climate change may occur. For Australia, this level of warming is likely to involve substantial changes to natural and agricultural production systems due to the combined effects of higher temperatures and lower rainfall across much of the nation. Risks from bushfires and other extreme weather also increase, particularly in coastal and rural regions (Pearman, 2008).
Stabilisation at 550 ppm CO2−e could limit global average warming to around 3°C above pre-industrial levels. Changes projected under a 450 ppm scenario are likely to occur sooner and become more severe in a 550 ppm world. Between 20 and 30 per cent of all species are projected to face a 50 per cent likelihood of extinction under this scenario (IPCC, 2007c). Coastal communities, agriculture and infrastructure would all face significant risks of adverse impacts (Pearman, 2008).
This report uses scenarios to explore the potential economic effects of climate mitigation policy on Australia. Each scenario represents, in a stylised way, a different possible future. The scenarios are illustrative and do not represent the official policy or negotiating position of the Australian Government, are not an official Government or Treasury forecast, and are not an official projection of Australia’s future greenhouse gas emissions.
The reference scenario provides the starting point for the analysis. It presents a plausible future path for economic growth, population levels, energy consumption and greenhouse gas emissions in a world without climate change. It does not include new policies to reduce emissions or climate change impacts.6
Two scenarios — CPRS −5 and CPRS −15 — examine the potential costs of the Government’s proposed Carbon Pollution Reduction Scheme. Australia’s action takes place within a simple multi-stage global policy framework. Australia and other countries listed in Annex B of the Kyoto Protocol take comparable action from 2010; and developing countries gradually adopt emission reduction obligations from 2015 to 2025.7 National emission pathways gradually diverge from reference scenario emission levels towards substantial reductions in the long term.
The CPRS −5 scenario is consistent with stabilisation at around 550 ppm by 2100; the CPRS −15 scenario is consistent with stabilisation at around 510 ppm.
Two further scenarios — Garnaut −10 and Garnaut −25 — were developed jointly with the Garnaut Climate Change Review. These more stylised scenarios assume united global action, with all countries taking on emission reduction obligations from 2013. This represents an optimal post-2012 agreement. National contributions are based on a contraction and convergence approach, whereby the allocation of emission rights among countries converges from current levels to equal per capita rights by 2050 (Garnaut, 2008a).
The Garnaut −10 scenario is consistent with stabilisation at around 550 ppm by 2100; the Garnaut −25 scenario concentrations peak above 500 ppm, then decline to around 470 ppm by 2100 (consistent with stabilisation at 450 ppm soon thereafter).
Note: Temperature change is from pre-industrial levels, and based on the median estimate of climate sensitivity.
- Assumes comparable mitigation effort is sustained in the post-2050 period.
Source: Treasury estimates from MAGICC (concentrations and temperatures) and GTEM (global emissions).
Additional sensitivity scenarios show how mitigation costs might change with different technology availability and performance, and timing and coverage of mitigation policies.
2.2.1 Global emission pathways
Global emissions in the policy scenarios diverge greatly from the reference scenario trend. To allow stabilisation at 550 ppm, global emissions in the CPRS −5 and Garnaut −10 scenarios peak around 2020-2025 and fall to below 2000 levels by 2050. Lower stabilisation levels require global emissions to peak sooner and fall more rapidly: for example, global emissions in the Garnaut −25 scenario peak in 2012 and fall to around 50 per cent below 2000 levels by 2050 (Chart 2.1).
The global emission pathway determines the overall cap on emissions within the international emissions trading scheme, so it determines the global emission price. Because Australia is linked to the international market, it also determines the Australian emission price. Emission prices drive mitigation activity in the economy and are an important determinant of Australia’s costs.
The world could achieve any given stabilisation goal via many different global emission pathways. This report uses the so-called ‘Hotelling rule’, so the global emission pathway is consistent with an efficient distribution of mitigation effort over time, and an efficient market in which permits can be used or banked for future use (Hotelling, 1931; Garnaut, 2008a). The value of a permit (the emission price in real terms) rises at 4 per cent per year, which is assumed to be the real rate of return on a comparable asset.8
Note: Modelled emissions are shown for the reference scenario, while allocations (policy targets) are shown for the policy scenarios. Actual emissions differ from allocations because permits can be banked for future use.
Source: Treasury estimates from GTEM.
2.2.2 Australia’s emission pathways
The reference scenario and the four policy scenarios correspond to different medium and long-term emission targets for Australia (Chart 2.2). The gap between the reference and policy scenarios is large; the gaps between the policy scenarios are relatively small. This highlights the fundamental shift in emission trends required to achieve the stabilisation levels considered in this report. Once that occurs, the differences — in terms of aggregate economic impacts — are relatively small.
Note: Modelled emissions are shown for the reference scenario, while allocations (policy targets) are shown for the policy scenarios. Actual emissions differ from allocations due to banking of permits and international permit trade.
Source: Treasury estimates from MMRF; Garnaut (2008a).
Each trajectory has corresponding medium- and long-term emission reduction targets (Table 2.2). The targets are reductions relative to Australia’s emissions in 2000. Also shown are the per capita reductions. Because Australia’s population is projected to grow significantly, an absolute reduction of 5 per cent (in the CPRS −5 scenario) by 2020 corresponds to a 27 per cent reduction in per capita emissions. Each scenario’s name indicates the assumed national 2020 target.
Relative to 2000 levels
Source: Treasury estimates from MMRF; Garnaut (2008a).
In a trading environment, actual national emissions are unlikely to match national targets in any given year. If actual emissions are higher than the target, emitters could use previously banked permits or purchase emission rights in the global market; if actual emissions are lower, emitters could bank emission rights for future use or sell them in the global market.9 As a result, national emission trajectories and targets define Australia’s contribution to the global mitigation effort rather than Australia’s actual emissions.
2.2.3 Policy settings
International emissions trading and national emission targets drive emission reductions in the policy scenarios (Table 2.3). They act as a simple proxy for the possible policy mix. This stylised approach allows the comparison of alternative national emission trajectories. In reality, a mix of policies and programs is likely to be adopted at local, national, regional and global scales.
|CPRS −5 and CPRS −15||Garnaut −10 and Garnaut −25|
Australia’s emissions trading scheme
Starts in 2010.
Is based broadly on the Carbon Pollution Reduction Scheme Green Paper:
Starts in 2013.
Covers all emission sources.
Does not constrain international trade in permits.
Does not shield emission-intensive trade-exposed sectors (as emission price starts simultaneously in all countries).
Other Australian mitigation policies
Includes Renewable Energy Target of 45,000 GWh per year by 2020.
Does not have other mitigation policies. These cease with emissions trading.
Unified action from 2013 with national emission trajectories based on a contraction and convergence approach.
All countries converge from current levels to equal per capita emission rights by 2050.
This report uses a range of economic and other models. Models are a useful analytical tool because they account for many of the complex relationships between different activities in the economy, and between Australia and the world. This report uses economic models to assess four dimensions of mitigation policy in Australia:
- global — including the rate and pattern of economic growth, technology development and emissions. This determines the magnitude of climate change risks, scale of the global mitigation task, and trade and capital flows affecting Australia’s economy.
- national — including the overall performance of the macroeconomy and patterns of growth across industries, sectors, states and territories. This determines aggregate effects on regions, firms and households.
- sectoral — including likely technological developments and the timing and scale of opportunities to reduce energy use and emissions. This determines the rate and cost of emission reductions, and consequent effects on prices of goods and services.
- household — including detailed impacts on incomes, consumption and prices.
No single model adequately captures all four dimensions. Previous Australian studies of climate change mitigation policy typically focused on one in isolation from the others. This report uses a suite of models that together span global, national, sectoral and household scales to provide an integrated set of projections across all four dimensions (Box 2.2).
The results require careful interpretation.
- None of the models in this analysis incorporates the impacts of climate change itself. The models, therefore, reflect only the costs, not the benefits, of mitigation policy.
- Climate change operates over very long timeframes, so quantitative analysis must take a long-term view. As the timeframe lengthens, assumptions become more speculative.
- The models used in this report are not well suited to examining short-term adjustment paths, so may underestimate costs from changing capital and retraining workers. G-Cubed (one of the global models) incorporates some stylised adjustment costs, so provides some insights to adjustment processes, albeit at an aggregated level. The bottom-up models also provide insights to the adjustment process in the electricity generation and transport sectors.
- The databases used in the whole-of-economy models aggregate firms into broad industry categories, so do not examine firm-level impacts.
- The models do not capture the effects of uncertainty, or non-market factors which can significantly affect economic behaviour.
Box 2.2: The suite of models
The Treasury’s climate change analysis centres on three top-down, computable general equilibrium (CGE) models developed in Australia: GTEM, G-Cubed and MMRF. CGE models are whole-of-economy, capturing interactions between different sectors. GTEM and G-Cubed model the global economy, whereas, MMRF is a model of the Australian economy, with rich industry sector detail. Economic modelling is inherently uncertain; using a suite of models provides a broader perspective on possible impacts.
Treasury has worked with economic modellers within and outside government to develop and improve the CGE models, including updating databases, enhancing model flexibility, and incorporating new research and analysis of sectoral growth prospects and mitigation potential.
Bottom-up models of electricity generation, transport and land-use change and forestry complement the CGE models. Detailed analysis of these sectors enriches the understanding of the economy’s likely response to climate change mitigation policy, particularly in the short to medium term. Analysis of the impacts on industry costs, consumer prices and household incomes uses the Treasury’s own models.
In combination, these models provide an integrated set of projections that are broadly consistent at the macroeconomic level and sufficiently detailed in key sectors.
Modelling results can be presented in different ways, which can influence how people intuitively interpret the results. For example, modelling results are often discussed relative to a reference scenario. This is a sensible approach when examining how a particular economic policy, or shock, will influence the economy in isolation from other events.
Results reported in this way do not mean the policy will have an absolute impact relative to the current world. For example, if GNP per capita in the policy scenario is 1 per cent lower than in the reference scenario in a particular year, this can nevertheless be consistent with continued strong growth in real GNP per capita in the policy scenario. The GNP per capita level is expected to be lower than it would otherwise have been, not lower than its current level. To illustrate this point, Chart 2.3 shows two measures (levels and deviation) of exactly the same result.
CPRS −5 scenario
Change from reference
Source: Treasury estimates from MMRF.
Changes expressed as a percentage of the reference scenario GNP level indicate the scale of the costs, and allow comparison across years and economies. On the other hand, differences between policy and reference scenario growth rates are appropriate for comparing long-term mitigation costs (Barker et al, 2006). This report uses both measures. In addition, it presents current levels where they provide an important or useful reference point for a particular effect.
Box 2.3: Measuring economic impacts
The modelling encompasses a wide range of variables that could be used to measure mitigation costs. This report focuses on gross national product (GNP) as the appropriate high level measure of economic welfare rather than gross domestic product (GDP). GDP measures output but GNP captures output and international income transfers. Reducing greenhouse emissions, in a cost-effective way, involves international trade in emission rights and influences Australia’s terms of trade. Given that context, GNP is a better measure of welfare, and better depicts the current and future consumption possibilities available to Australians.
An issue that has received a great deal of attention in the context of the economic costs and benefits of climate change is how to account for results over long timeframes, in particular how to discount the value of future events. This report focuses on mitigation costs alone, and reports all results in real terms, that is adjusted for inflation, so this presentation issue does not arise.
4 Overshooting occurs when atmospheric concentrations initially exceed then return to the target level. Overshooting is inherently more risky than approaching stabilisation levels from below (Pearman, 2008).
5 All temperature changes are based on the median estimate of climate sensitivity, calculated using the simple climate model MAGICC (Garnaut, 2008b). There is substantial uncertainty in such estimates (Pearman, 2008).
6 This study follows the modelling approach of previous studies in excluding the impacts of climate change, so the results can be compared (Box 3.2). Pre-existing mitigation policies such as the 9,500 GWh/year Mandatory Renewable Energy Target and the NSW Greenhouse Gas Reduction Scheme are included in the reference scenario.
7 Annex B of the Kyoto Protocol lists countries with quantified emission reduction commitments (‘Kyoto targets’), and includes the members of the OECD and economies in transition (Russia and other eastern European nations).
8 This rate of return embodies all the commercial risks of holding permits. The 4 per cent embodies a risk-free real rate of 2 per cent and a risk premium in permit markets of 2 per cent.
9 The Government proposes not to allow the export of Australian permits in the initial years of the Carbon Pollution Reduction Scheme, but favours open linking within the context of an effective global emission constraint (DCC, 2008a).