Australian economic models of greenhouse abatement
Introduction
What are the costs and benefits of reducing greenhouse gas emissions in the energy sector? What models are currently being used, or could be used, to answer this question, and what are their respective assumptions and limitations? The answers to these questions are important for national greenhouse response strategies and in the regular deliberations of the Conference of the Parties (COP) to the Framework Convention on Climate Change. In July 1996 at the second COP the Australian Government aligned itself with the Organisation of Petroleum Exporting Countries (OPEC) by refusing to accept legally binding targets and timetables for reducing greenhouse gas emissions. The basis for this position was apparently some runs on an economic model of greenhouse response.
Therefore this paper aims to evaluate the roles and limitations of the most prevalent type of computer model used in Australia to evaluate the economic implications of greenhouse gas abatement in the energy sector. This type of model examines the economic impact of these responses on the whole nation or on groups of nations and is known as a ‘top–down’ model (defined in the next section). Similar models are used in other countries.
The paper does not attempt to analyse the equations or the details of the economics in the models1. Instead it offers a nonmathematical explanation of the basic assumptions of ‘top–down’ models for policy makers, political scientists, environmental managers, businesses and community groups which are concerned about the costs and benefits of greenhouse abatement in the energy sector2. To assist in bridging the gaps between the various disciplines interested in these models, a glossary is provided as an Appendix A.
In focusing on greenhouse abatement, this paper shows that many important assumptions underlying the models are matters for debate by these wider interest groups, instead of ‘technical’ matters for economic modellers whose skills involve specific mathematical and computational techniques. Hence the paper avoids jargon where possible, defining the economic and mathematical terms which are introduced in a manner which reveals their underlying assumptions.
A model is just a simplified representation of reality. Well-known examples are the medical model of achieving health, Newton's laws of motion, or any map or equation. A computer model could be defined as a model which, from a set of initial assumptions, provides a logical description of how a system performs, and hence can be represented on a computer. Computer models are potentially very useful for analysing complex systems with many variables and many interrelationships between these variables (Meadows and Robinson, 1985). Examples of complex systems are an ecosystem, the economy of a nation-state, or the interactions between economics and ecosystems in a particular resource industry.
The assumptions and quantities in computer models must be specified precisely and this can be valuable to modellers in clarifying their thinking, even before computer analysis takes place. A computer can manipulate much more information than the human mind and can keep track of many more interrelationships. If constructed correctly, computer models can process a very complicated set of assumptions to draw logical, error-free conclusions. Sometimes they can elucidate a counter-intuitive result. Computer models can easily test a wide variety of different conditions and policies, thus providing a form of social experimentation that is much less costly and time-consuming than tests within the real social system (Meadows and Robinson, 1985).
However, in practice many computer models fall short of this potential. This is often the case in economics, where questionable assumptions are buried in the currently dominant paradigm and where comparison with empirical observation is generally inadequate. In particular, it is pointed out in Section 3that some assumptions are so important to the neoclassical economics paradigm that they are held even in cases where they conflict with clear empirical evidence.
Unlike scientific models of the greenhouse effect3, economic models of greenhouse abatement tend to lack mechanisms for describing the relationships between different variables and so, for simplicity, they tend often to assume that these relationships are very simple (e.g. linear) and that the unknown parameters can be determined empirically. In applied mathematics, this approach is known as ‘parametrising our ignorance'. It may lead to oversimplification and the false belief that the mechanisms are understood.
Section snippets
‘Bottom–up’ models
‘Bottom–up’ or ‘engineering’ models of greenhouse abatement start with data on the cost and performance of specific technologies in the energy sector. Given scenarios for future growth in the demand for energy services (see Appendix A), these models focus on the least-cost provision of technologies to meet these scenarios. But, they do not attempt to examine the impact of changes in the energy sector on the rest of the economy.
As a result of this approach and because most of them do not assume
One-sided cost–benefit analysis
Economic models attempt to calculate the cost of greenhouse abatement, but rarely weigh this against the benefits. The latter include both the direct economic benefits of implementing an ecological sustainable energy system — which may include net employment creation (Moeller, 1985; ACF and ACTU, 1994), growth in exports of new technologies (Blakers and Diesendorf, 1996), and reduced imports of fossil fuels — and the environmental and health benefits of reducing human-induced global climate
Conclusion: what is a good model?
All the seriously flawed assumptions discussed in Section 3are made by MEGABARE; most are made by the other general equilibrium models, Orani/Monash and G-Cubed; and a few are made by IMP. Yet, even within the limited framework of general equilibrium models, several of these shortcomings could be readily rectified, namely the assumption of constant returns to scale (Section 3.3, the assumption that GDP=welfare (Section 3.7, the sensitivity to different ways of spending a carbon levy (Section 3.9
Acknowledgements
I thank Clive Hamilton, Joel N. Swisher and an anonymous referee for valuable comments. However, responsibility for views expressed and any errors is the author's.
Mark Diesendorf is Professor of Environmental Science and Director of the Institute for Sustainable Futures at the University of Technology, Sydney. He has a BSc (Hons) in Physics from the University of Sydney and a Ph.D. in Applied Mathematics from the University of New South Wales. His current research interests are energy, transport and greenhouse response, and processes for achieving ecological, economic and social sustainability. He is a Board Member of the Sustainable Energy industries
References (56)
- et al.
The cost-effectiveness of CO2 emission reduction achieved by energy conservation
Energy Policy
(1993) The greenhouse effect: the fallacies in the energy efficiency solution
Energy Policy
(1990)How can a ‘competitive’ market for electricity be made compatible with the reduction of greenhouse gas emissions?
Ecological Economics
(1996)The impact of carbon taxation on the UK economy
Energy Policy
(1994)Energy efficiency and economic fallacies
Energy Policy
(1990)- et al.
Closing the efficiency gap: barriers to the efficient use of energy
Resources, Conservation and Recycling
(1990) - et al.
Reducing Australian energy sector greenhouse gas emissions
Energy Policy
(1994) - et al.
Technical evidence for assessing the performance of markets affecting energy efficiency
Energy Policy
(1994) Barriers to improvements in energy efficiency
Energy Policy
(1991)- et al.
‘Normal’ markets, market imperfections and energy efficiency
Energy Policy
(1994)
Exploring the gap: top–down versus bottom–up analyses of the cost of mitigating global warming
Energy Policy
A scenario for the expansion of solar and wind generated electricity in Australia
Australian Journal of Environmental Management
Market transformation strategies to promote end-use efficiency
Annual Review of Energy and the Environment
The costs of limiting fossil-fuel CO2 emissions
Annual Review of Energy and Environment
Cited by (0)
Mark Diesendorf is Professor of Environmental Science and Director of the Institute for Sustainable Futures at the University of Technology, Sydney. He has a BSc (Hons) in Physics from the University of Sydney and a Ph.D. in Applied Mathematics from the University of New South Wales. His current research interests are energy, transport and greenhouse response, and processes for achieving ecological, economic and social sustainability. He is a Board Member of the Sustainable Energy industries Council of Australia Inc., a member of the Australian Cooperative Research Centre on Renewable Energy, a former President of the Australasian Wind Energy Association and has served on the Committee of the Australian and New Zealand Solar Energy Society. He is co-editor and principal author of the transdisciplinary book, ‘Human Ecology, Human Economy: Ideas for an Ecologically Sustainable Future’, Sydney: Allen & Unwin, 1997. In all he has authored or co-authored three books and monographs, has edited five books, and has published over 70 research papers and book chapters, 23 conference papers and over 50 popular articles.