Contents - Previous - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

13 Greenhouse gas emission abatement in Australia

Abatement of energy sector emissions
Economic impact of abatement strategies
Non-energy emission abatement
Australia's international role
Carbon taxes, externalities and other policy instruments
References

Hugh Saddler

Greenhouse gas emissions have been at the forefront of public policy debate in Australia for over three years. In October 1990 the federal government adopted an 'interim planning target' - to reduce emissions of greenhouse gases to levels 20 per cent below 1988 emissions by 2005. The reduction of 20 per cent relative to 1988 levels was first proposed at an international conference held in Toronto, Canada in 1988; it is referred to as the Toronto target in this chapter.

Gases controlled by the Montreal Protocol on Substances that Deplete the Ozone Layer, that is, chlorofluorocarbons and related compounds, were not included, but the government had previously announced that the use of CFCs would be eliminated by 1995. The undertaking with respect to other greenhouse gases was subject to the important qualification that it would not proceed with measures which have net adverse economic impacts nationally or on Australia's trade competitiveness.

The eight Australian state and territory governments have important policy powers relating in particular to the electricity and gas industries and to the control of pollution. Their concurrence and participation is virtually a prerequisite for the realization of the federal target. The eight governments have agreed to participate with the federal government in the development of a national greenhouse response strategy. A draft strategy document was released for public comment in June 1992.

Although it is normally viewed as a developed country, Australia's economy is heavily dependent on exports of raw and partly processed commodities. Among the most important Australian exports are coal, liquefied natural gas (LNG), and alumina and aluminium metal smelted with coal-fired electricity. Consideration of greenhouse response strategies has therefore been strongly influenced by concerns about the possible effects of any abatement measures on Australia's international competitiveness as a supplier of fossil fuel intensive commodities.

A steady stream of reports and studies from government, business and nongovernmental organizations has provided the material for an enthusiastic public policy debate. Many of the reports have sought to estimate the costs to the Australian economy of reducing carbon dioxide emissions by changes in energy supply and use, focusing in particular on the costs of meeting the Toronto target. This narrow emphasis on a single goal reflects partly the concerns of special interest groups who fear the impact of achieving the target on their activities. Most studies have sought to estimate the macro-economic effects of a carbon tax on fossil fuel use that would suppress demand for fossil fuels in 2005 to the Toronto target level. The narrow focus of so many of these studies has had two unfortunate effects. First, it has encouraged an 'all or none' view of the desirability of implementing emission reduction policies. And second, it has led to neglect of policy instruments other than a carbon tax.

Nevertheless, the studies have greatly improved our understanding of the workings of the Australian energy system and its interaction with the wider economy. Very much less is known about non-energy related sources of greenhouse gas emissions, which are the principal sources of the other important anthropogenic greenhouse gases such as methane and nitrous oxide.

In this chapter, I examine the cost and scope of emission abatement measures available in Australia. I also review estimates of the effect on the Australian economy of achieving various levels of abatement.

Abatement of energy sector emissions

Estimated carbon dioxide emissions from the use of energy in Australia in1989, and 'business as usual' projections for 2005 are shown in Table 13.1. The energy use figures shown are the quantities of energy consumed at the point of end use, whereas the carbon dioxide emission figures embody the allocation of emissions arising from energy transformation processes (such as electricity generation and oil refining) required to deliver energy to the respective end uses. However, emissions associated with the transformation of energy for export (such as those associated with liquefaction of natural gas) are included in the agriculture and mining sector.

Table 13.1 Estimated energy-related CO₂ emissions in Australia

Data are for financial years running respectively from July 1988 to June 1989 and from July 2004 to June 2005. Carbon is elemental carbon as carbon dioxide, that is, 0.273 of the mass of carbon dioxide.
Source: Australian Commission for the Future (1991).

These projections imply high rates of growth in energy consumption. The main assumptions which underlie this projection include: continuing high rates of net immigration and hence population growth; continuing structural shifts in the economy towards energy-intensive materials processing activities, particularly primary aluminium and paper; and rather modest increases in energy use efficiency.

Ecologically sustainable working groups

During 1991 the federal government coordinated a national process under the rubric of 'Ecologically Sustainable Development' (ESD). The ESO process involved collaboration between all levels of government, business and national environmental groups. It aimed to provide policy advice on how to implement ecologically sustainable development in Australia. As part of this process, the government commissioned a series of studies of the technological potential to improve energy use efficiency in selected energy services in the residential, commercial and manufacturing sectors (Ecologically Sustainable Development Working Groups 1991a). These are the only set of detailed 'bottom up' or 'engineering' studies of the potential for emission abatement by demand side measures so far undertaken for Australia as a whole.

The studies examined the present stocks of energy-using equipment relevant to the respective processes, assessed the energy savings available from the use of selected technologies for each service, and estimated the associated costs and benefits. The energy services assessed were as follows:

The ESD studies estimated that in 1988 these energy services accounted for emissions of about 118 million tonnes (MT) of carbon dioxide, that is, about 42 per cent of the total from the Australian energy system. They also projected future energy services. They assumed 'frozen efficiency,' that is, no change in 1988 technology. This assumption, plus various others as to economic and demographic trends, determined that emissions from these energy services in 2005 would total about 192 MT.

ESD Methodology
These studies assessed a wide variety of technical options for each of these services or groups of services, including process changes and fuel substitution as relevant efficiency improvements. They evaluated the economic benefits and costs of the various options exclusive of any economic benefit of emission abatement itself. Thus benefits included reduced energy consumption, and in some cases associated savings in maintenance costs, increased productivity of capital equipment, etc. They measured the benefits of energy savings in terms of a schedule of projected resource costs for electricity, gas, petroleum products and coal. Costs were taken to be the additional capital costs of the new equipment. All studies performed the analyses in terms of investment programmes over the period 1992 to 2005 affecting the present stock of equipment in the Australian economy. They assumed that benefits would continue to accrue beyond 2005, until the end of the economic life of the equipment concerned. They further assumed that any cost savings achieved by the use of more efficient equipment would be realized in the form of reduced energy consumption, and hence carbon dioxide emissions, rather than increased output. Finally, they discounted the streams of costs (mainly capital costs) and benefits (mainly energy savings) back to 1990 at a real discount rate of 8 per cent.

The studies were confined to measures which either show a net benefit or a relatively modest net cost. Technical options involving, for example, prematurely scrapping and rebuilding major industrial plant were not considered. Indeed, for many of the technical options, the ESD found that costs are relatively small (and hence net benefits are positive) because the technical improvements are realized only as new plant is built and new equipment purchased, either as replacement for existing obsolete plant equipment or in the course of expanding total output. Thus, there is an inverse relationship between average emissions per unit of output and total emissions in such activities as aluminium smelting. If the industry grows rapidly (several new smelters), then total emissions will grow strongly, but emissions per tonne of aluminium will fall. Conversely, with no growth in the industry, there would be no change in total emissions, but also no improvement in emissions per tonne of aluminium.

The studies were concerned with the technical potential for emission reductions, not with the practicalities of achieving that potential. Thus, in most cases very high rates of penetration of optimally efficient technologies in-purchases of new plant and equipment were assumed, even in cases (such as electric motors) where most new purchases are of low efficiency motors which are not the economically optimal choice at a discount rate of 8 per cent. No administrative or incentive costs for programmes which might be needed to stimulate the economically optimal choice of equipment were included in the cost-benefit assessment. For this reason, the estimates of net benefits! cost of emission abatement could be interpreted as understating the costs. On the other hand, in a number of the studies - notably those dealing with residential sector energy services - the assumed resource cost for electricity was substantially less than the true cost at the customer's meter, thereby understating the benefits of energy savings.

ESD results
A supply curve of the abatement measures associated with each of the eleven energy services studies is shown in Figure 13.1. It should be noted that for each service, the potential abatement shown as a single block is in turn the sum of a number of individual measures, each having its own net abatement cost which may be higher or lower, sometimes by a substantial margin, than the average of all measures applicable to the particular service. Table 13.2 summarizes the abatement potential and costs of each measure reflected in Figure 13.1.

The estimated total emission reduction potential from all of the energy services studied was 61 MT of carbon dioxide, assuming no change in the present mix of supply technologies. On this basis, emissions associated with the services studied in 2005 would be 13 MT higher than 1988 emissions of 118 MT, that is, an increase of 11 per cent. This level of emission abatement is equal to about 32 per cent of the projected frozen efficiency emissions from the services studied.

Figure 13.1 Cost of CO₂ emission abatement in Australia, selected energy efficiency measures

Table 13.2 Emission abatement potential of selected energy efficiency measures in Australia

All dollar values here, and throughout this chapter are Australian dollars unless otherwise stated.

Unfortunately, the ESD studies did not provide any reliable, consistent estimate of per centage abatement relative to a 'lousiness as usual' projection. Moreover, the potential of the measures studied may not be representative of the energy services not included in the studies, the most important of which are:

• all uses of energy in agriculture, fishing etc.;
• all uses of energy in mining;
• all uses of energy in transport;
• low temperature process heat in the manufacturing sector;
• residential space heating and cooling;
• residential cooking;
• electronic and other miscellaneous (mainly electrical) residential appliances.

The activity levels associated with some of these services, notably agriculture and transport, will grow rather more slowly than the activities included in the studies.

ESD transport sector study
The ESD Working Groups also studied the potential for carbon dioxide emission abatement in the transport sector (Ecologically Sustainable Development Working Groups 1991b). They considered a wide variety of technical and behavioural changes to transport in Australia. The existing transport system is dominated by road and air which account respectively for 85 per cent and 7 per cent of total carbon dioxide emissions. Any technical improvements in the fuel efficiency of road and air transport in Australia will be almost entirely dependent on imported technology. Consequently, the potential for efficiency improvements in these modes will be largely determined by international developments in motor vehicle and aircraft technology. The study made two different assumptions about the rate at which fuel saving technologies would be incorporated into mass produced vehicles and aircraft on a global basis, and incorporated these into 'low' end 'high' emission abatement scenarios. For rail transport, the main options for fuel consumption improvements derive from upgrading the permanent rail system on heavily trafficked routes and improved train signalling and control systems, based largely on indigenous technology. The study also incorporated assumptions about changes in urban passenger transport systems, which currently account for about 45 per cent of total Australian transport energy use. The measures considered included changes in urban form (consolidation of cities to increase residential densities) and greater use of mass transit, bicycles and walking. 'Low' and 'high' levels of abatement were again considered.

Estimated carbon dioxide emission abatement under these two scenarios were respectively 12 per cent for the 'low' ease and 27 per cent for the 'high' case below 'business as usual' projections for 2005. However, since the 'business as usual' projection is for an increase in emissions of 40 per cent, these figures still represent higher emissions than in 1988 by respectively 25 per cent and 2 per cent.

The transport study did not analyse the economics of the various measures considered and of the two scenarios as a whole. As noted earlier, the economics of technical measures for efficiency improvements will depend heavily on international technologies and will likely mirror the economics of such measures in Japan, North America and Europe, appropriately adjusted for the local price of fuel and spatial densities. The projected changes in urban transport systems would require very substantial redirections of investments in all types of urban infrastructure, that is, not just transport infrastructure. Most Australian cities have spare capacity in at least some components of their urban infrastructure. Measures to consolidate urban form would result in substantial capital savings from the reduced requirement for new infrastructure at the urban periphery. There would be a need for significant additional investments in public transport infrastructure, but these would be offset, at least in part, by reduced investment in urban arterial roads. Thus the overall outcome could be a net saving in capital investment (McGlynn, Newman and Kenworthy 1991).

Other sectors

The remaining categories of energy service have not been scrutinized systematically on a national basis as the services already described. Information about residential space heating and cooling and cooking is particularly deficient. Some studies have drawn general conclusions about emission reduction potential in the very large low-temperature industrial process heat category. A detailed engineering study of a representative sample of industrial plants in the food processing, paper and other industries concluded that cost-effective savings of the order of 10 to 20 per cent might be available on average (Warren Centre 1991). An abatement of 15 per cent would be equivalent to about 8 MT of carbon dioxide in this end use in 2005. Opportunities for greater use of cogeneration are also associated with the use of low temperature process heat. The ESD Working Groups estimated also that additional cost-effective opportunities for gas-fired cogeneration could yield an emission abatement of about 4 MT carbon dioxide (Ecologically Sustainable Development Working Groups 1992).

Aggregate carbon conservation potential

Because the various studies referred to have not been performed on a single consistent set of baseline energy consumption data, and because definitions of 'lousiness as usual' are not consistent, it is difficult to sum the results to give a single figure for potential emissions abatement. However, subject to some assumptions, it can be estimated that the technical potential of these measures is for a carbon dioxide emission reduction in 2005 of between 71 and 86 MT, depending on which transport scenario is chosen, relative to the projected 'lousiness as usual emission' level of 383 MT shown in Table 13.1.

To this total, one could add a few additional million tonnes from savings in the unexamined residential sectors (space heating and cooling, cooking, appliances). However, even making allowance for these savings, the general form of the result is clear. Although very substantial reductions in carbon dioxide are technically available through the adoption of cost-effective or close-to-cost-effective measures to increase the efficiency of energy use, the reductions are somewhat less than the 102 MT required to stabilize emissions in 2005 at 1988 levels.

A recent critique of these studies has claimed that they overstate the potential and understate the cost of achieving such levels of emission reduction by demand side measures (ACIL Australia 1992). The critique points to imprecision and confusion in the definition of the 'lousiness as usual' case against which savings are measured and argues that the estimates of potential savings exaggerate what is achievable.

In essence, this criticism simply restates arguments about the existence, nature and causes of the 'efficiency yap' (Grubb 1990). As such, the criticism misunderstands the purpose of the studies, which are concerned with the technical potential for savings, not the savings achievable under the prevailing economic and policy environment. The savings identified are those which have been assessed as being cost-effective at a discount rate of 8 per cent, which is generally accepted as an appropriate rate for determining social costs and benefits, but is considerably lower than the discount rate commonly used for private and business decisions about purchases of energy-using equipment. To achieve the technical potential for energy efficiency improvements will require quite large and rapid changes in purchase decisions by energy users and will in turn require the implementation of a variety of new policies and programmes by governments, energy utilities and other parties. No estimate of the cost of incentives and other measures, likely to form part of such programmes, is included in the analysis.

Additional policy measures

It will be apparent from the figures cited, that - even assuming the full achievement of technical potential for efficiency improvement - stabilizing carbon dioxide emissions at 1988 levels by 2005 will require changes in other factors which affect the level of emissions from the energy system. Such changes could include:

1 a much lower level of immigration, and hence of population growth, affecting the absolute (as opposed to the per capita) rate of growth in economic output;

2 a great reduction in output from highly energy intensive export-oriented industries, notably aluminium production;

3 a reduction in carbon dioxide emissions associated with the energy supply system.

The first two of these options are not considered to be politically desirable or acceptable by the majority of Australians. In any case, unilateral action by Australia would have very little effect on global carbon dioxide emissions, since the growth in overall economic activity and aluminium production, and associated growth in carbon dioxide emissions, would simply occur elsewhere.

The third option, of reducing emissions associated with the energy supply system, is obviously the preferred method for achieving further reductions in carbon dioxide emissions. Australia has very large reserves of both black coal and brown coal (lignite), which are favourably located in relation to the major centres of population and economic activity. Consequently the electricity supply industry is heavily dependent on coal, which in 1988 accounted for 76 per cent of total electricity generated (Ecologically Sustainable Development Working Groups 1991c). Between 1978 and 1988 emissions of carbon dioxide per MJ of electricity generated decreased from about 310 tonnes to 270 tonnes, an improvement of about 13 per cent. This decrease was achieved largely by replacing older generating plant by large new 500 MW and 660 MW generating units. While there is probably some scope for modest further improvements in thermal efficiency of conventional coal fired generation, more substantial changes would require significant changes in technology.

A systematic review of available generation options, focusing on cost and carbon dioxide emissions, concluded that conventional black and brown coal fired steam turbine generation is the lowest cost, but most carbon dioxide intensive, option (Ecologically Sustainable Development Working Groups 1991c). Combined cycle gas turbine technology would be slightly more expensive and is a realistic alternative, making use of large uncommitted gas resources located off the south east and the north west shores of the country. Australia's natural resource endowments also make it well placed to utilize nuclear, wind, solar thermal and photovoltaic generation technologies, but these would all be significantly more expensive than coal and gas fuelled technologies in most parts of Australia.

The review also calculated abatement costs for these technologies, relative to conventional black coal fired steam turbine technology. Considerable uncertainty surrounds all the cost estimates because none of the alternative technologies are deployed on a commercial scale in Australia. The cost of gas is also uncertain.

For combined cycle gas turbines, the best estimate is a few tens of dollars per tonne of carbon abated. For wind and nuclear, the costs are between $100 and $300 per tonne. Advanced coal combustion technologies are among the least cost-effective options, with abatement costs of up to $500 per tonne. The cost is so high because of the relatively modest emission abatement available by use of these technologies. The economics of abatement are much more favourable relative to conventional brown coal fired steam generation, because this technology is intrinsically both more costly and more carbon dioxide intensive than black coal fired steam generation. It should be noted that these comparative assessments of generation technologies are based on costs at the power station bus-bar only. They do not take account of interconnected system characteristics which will affect the proportion of total demand which can be supplied by particular technologies, and the economics of doing so.

Two recent studies, using different modelling methodologies and somewhat different data estimates and assumptions, have concluded that the Toronto target for carbon dioxide abatement could be met. This goal would be achieved by: extensively substituting gas turbine combined cycle plants for coal fired steam generation technology; and by using some renewables such as wind and/or solar thermal energy technologies Cones 1992;

Australian Commission for the Future 1991). Both studies model the effects on carbon dioxide emissions of progressive introduction of more efficient energy using technologies and the replacement of coal by gas and renewables in electricity generation. Of course, as the electricity sector becomes progressively less carbon dioxide intensive, the emission abatement achieved by using electricity more efficiently is reduced. Because the two studies differ somewhat in their estimates of extent of demand side efficiency gains, they differ also in their estimates of the extent and cost of adjustments required on the supply side, particularly in electricity generation.

Given the relative costs of the respective technologies, there would be a net cost to the economy (excluding the benefit attributable to greenhouse gas emission abatement itself) of adopting this strategy. A large proportion of the current Australian coal fired generating capacity has been commissioned since 1975, and would normally be expected to have a life of 25 to 30 years. Meeting the target for emission abatement by 2005 would require prematurely scrapping a number of power stations, with an additional economic penalty. The penalty would not apply if the same, or even a more stringent abatement reduction, were to be met by a later date, say about 2015. This approach, however, incurs the additional ecological impact of releasing carbon dioxide emissions during the ten years from 2005 to 2015.

Economic impact of abatement strategies

One of the studies referred to above included estimates of the macroeconomic consequences of the proposed emission reduction measures (Australian Commission for the Future 1991). The ACF study is the one in Australia to date which integrates the results of 'bottom up', technology-specific energy modelling with a 'top down' macro-economic model. Two levels of abatement were analysed the demand side measures only, falling somewhat short of emission stabilization; and the demand and electricity supply measures combined, achieving the Toronto target, as described above. The major impact on the economy in both cases comes from the surge in investment expenditure needed to pay for the new, more efficient energy-using and energy-supply equipment. This strategy requires diverting economic resources from consumption to investment, with the result that consumption expenditures fall.

Since the ACF study found that most of the demand side measures were cost effective relative to expected energy prices, it is not surprising that aggressive deployment of these measures alone yields net macro-economic benefits over the long term, as measured by a slightly higher rate of GDP growth than in the base case. Achieving the Toronto target requires that Australia diverts a higher proportion of resources to energy-related investments and that the average unit cost of energy is higher than in the base case. Thus, GDP growth is lower than in the base case by between a half and one per cent, depending on assumptions about how the economy adjusts to change. This outcome is obviously a relatively small change, much less than the changes associated with the normal ups and downs of economic activity. One reason for the change being so small is the study's assumption that Australia takes a unilateral decision to limit greenhouse gas emissions. Domestically produced fossil fuel resources, particularly coal and petroleum, are therefore available to be exported if not required for domestic consumption.

Other economic studies have sought to model the effects of measures to reduce greenhouse gas emissions on the Australian economy or particularly vulnerable sectors of it, most of which have been summarized by the Industry Commission (1991). These studies share the following common features:

1 a 'top clown' approach to modelling the Australian energy economy;
2 the use of a carbon tax as the policy instrument by which emission abatement will be achieved;
3 a focus on the cost of achieving only the Toronto target, with no analysis of the costs of lower levels of abatement;
4 no allowance for improvements in technology which could increase the efficiency of energy supply and use without increasing costs.

This approach assumes that all markets for energy services are perfectly competitive, that is, energy is used throughout the economy with optimal technical efficiency and no cost-less opportunities for efficiency improvement are available. It follows that an increase in energy prices by means of a carbon tax is needed to induce any change in technical efficiency. Of course, this assumption varies sharply from the findings of the sectoral energy use studies described above.

Disagreement over this issue is by no means confined to Australia, but is a persistent theme around the world in debates over energy and greenhouse emissions (see for example Manne and Richels 1990; Williams 1990; Nordhaus 1991). Some have characterized this as a disagreement in perspective between economists and engineers. It could perhaps be more accurately described as a disagreement between those who sit behind desks and theorize about the economy; and those who go out to visit factories and building sites and talk to people who are making decisions about energy-using equipment.

The most detailed study of this kind was undertaken by the Industry Commission (IC) itself (1991). The IC estimated that a carbon tax of A$80 (1988 prices) per tonne of carbon (equivalent to about US$60 per tonne) would be required to reduce emissions to the Toronto target. A tax at this rate was found to reduce GDP by 2 per cent. The effect on the output of the energy industries was considerably greater - the output of the coal industry fell by 26 per cent, that of the electricity industry by 11 per cent, and of the gas industry by 19 per cent. This last result diverges strikingly from the results of the bottom up, technology oriented studies, which project an increase in gas industry output as gas is substituted for coal and electricity. It would appear that the substitution elasticities (between factors of production and between fuels) used in the macro-economic model used for this study do not accurately represent the realistic technical possibilities available in the Australian energy system. The IC recognizes the inadequacies of the modelling approach it has used, commenting that the work has been undertaken for 'illustrative purposes'.

Not surprisingly, given the different assumptions, this estimate of the GDP cost of abatement is somewhat higher than that from the previously cited ACF study. The Industry Commission study assumed that the revenue raised by the carbon tax would offset direct taxes (income tax and company tax). Somewhat surprisingly, it did not model the effect of offsetting the carbon tax by reducing another consumption tax or related tax, such as payroll tax, which is currently the focus of some policy debate in Australian politics. A recent study of the US economy concluded that a moderately severe gasoline tax (a form of one-sector carbon tax) would depress GDP and consumption if the revenue were used to reduce either the budget deficit or direct taxes, but would have virtually no effect on economic activity if used to reduce payroll taxes (Brinner et al, 1991).

Non-energy emission abatement

The other important anthropogenic greenhouse gases emitted in Australia are methane and nitrous oxide. Compared with carbon dioxide, little is known about either sources of, or possible abatement measures for these gases, particularly nitrous oxide. The most important source of methane, thought to account for about two-thirds of total emissions, is domestic livestock, principally cattle, sheep and pigs. Landfill (municipal garbage) is also an important source. Small quantities of methane are released as a result of coal mining and from the natural gas distribution system. Agricultural activities, in this case soil denitrification associated with both the application of nitrogenous fertilizers and the use of legumes in improved pastures, is thought to be the main source of nitrous oxide (Ecologically Sustainable Development Working Groups 1992). Measures which it is thought could contribute to reducing these emissions include: the use of rumen modifiers (anti-bloat capsules) with intensively reared cattle; a modest decrease in stocking rates on some pasture types used for extensively reared cattle; aerobic, rather than predominantly anaerobic treatment of piggery waste; and the optimization of application rates of nitrogen fertilizers. The ESD did not estimate the scope, let alone the cost of these measures if applied on a large scale.

Abatement of atmospheric carbon dioxide levels by increased tree growth in Australia has been analysed superficially. Over the last few years growing concern about deforestation and soil erosion stimulated a variety of government and privately supported programmes to reverse the trends of two centuries of European colonization. These programmes could perhaps stop and perhaps reverse the continuing emission of carbon dioxide previously sequestered in biomass in trees and in the soil. As such they can legitimately be seen as an important part of national activities to curb greenhouse gas emissions, although that is not the reason they were initiated. From a greenhouse perspective, therefore, they are costless measures. However, much more far-reaching tree plantation programmes would be required to make a significant contribution to offsetting carbon dioxide from energy related activities. No estimates are available yet of the possible cost and scale of such programmes in Australia.

Contents - Previous - Next