Treasury climate modelling assumptions

STANDING COMMITTEE ON ECONOMICS
(Senate-Wednesday, 22 October 2008)

Department of the Treasury

Senator «MILNE» -I apologise for being a bit late and I hope I am not going to repeat what has already been asked. I want to ask about the assumptions that have been made. I take on board the statement that the assumptions outlined in the document have been judged by Treasury to be plausible central estimates. Which particular areas are you going to subject to sensitivity analysis? I heard one a little earlier, but could you run through for me which areas you will subject to sensitivity analysis? It seems to me that, for a number of these issues, it would make a lot more sense to put forward a number of different scenarios so that people could test against those rather than just provide one value for inclusion in the model. Could somebody tell me which areas you are going to subject or present in a sensitivity analysis way?

Ms Quinn-We have been tasked with looking at the long-term economic implications for different trajectories and targets. We are using a range of economic models. Different economic models provide different results and they have different input assumptions. So part of using the suite of models is to take on board how they deal with different levels of technology and different types of assumptions. So, even for a single scenario or a single set of assumptions, there is a range of results because we have a range of economic models that we are using. The different types of sensitivity analyses that we hope to explore-they relate to not only things to do with technology but also other components of modelling-are being provided to government for its report and will be in the public domain by the end of October. So exactly what will be presented et cetera is a matter for government.

Senator «MILNE»-Looking at your analysis, I would like to know whether Treasury assumes that an emissions trading system will reduce productivity.

Ms Quinn-When we think about the economics of introducing an emissions price or a carbon price into the models, the CGE models, as such, assume that the current world is in equilibrium and that introducing a relative price into those models moves resources from existing sectors to new sectors. Part of moving those resources from one place to another does typically result in losses of allocative efficiency, which some people commonly term as productivity losses. That is how the computation of general equilibrium models works. You have to keep in mind that they assume that the starting point is in equilibrium and that that may not be where we are now. So the analysis that we are looking at is relative to a world without climate change: what are the implications of putting a relative price into the economic models?

Senator «MILNE»-Why might we not assume an improvement in productivity, given that the whole world is moving to trying to achieve greater efficiencies in all sorts of ways? There are a number of analyses now about a whole wave of new green jobs with new green technologies, for example, which will lead to much more efficient resource use and greater economic activity in many areas. So why do you make that assumption necessarily that there will be a reduction in productivity?

Ms Quinn-Introducing a relative price into the economic models moves economic effort from one area of the economy to another area of the economy. As part of that shifting of effort, there may be an improvement in productivity in a sector that has an increase in demand, but there are other parts of the economy that have a reduction in demand. There is still a fixed number of people and a fixed number of resources in the economy, and moving things from one place to another does not necessarily improve the productive capacity of the world; you are just shifting things from one place to another. So, while there might be more people going to university studying issues to reduce greenhouse gases, there will be fewer people studying other things such as medical technology-as a hypothetical. So, typically, moving from one sector to another does not necessarily affect aggregate productivity assessments. There is a composition shift that happens with a relative price change.

Senator «MILNE»-Yes, that is true. But my view or my assumption would be that significant new investment in reducing energy use will result in an increase in productivity, as older machinery is replaced with newer machinery, and it is likely that there will be more productive machinery more rapidly. So, for a start, I would question, as an assumption, one of the central tenets about productivity. Secondly, in relation to population, I note that you assume that 150,000 people per year will immigrate to Australia between 2012-13 and 2049-50, with immigration rising each decade from 2050-51 to reach 200,000 per year in 2070-71, where it is then assumed to remain constant until the end of the century. Why do you make those assumptions? That does not seem a plausible immigration pattern to me.

Ms Quinn-The net immigration number is 150,000 people per year. These assumptions are made in relation to what has happened in history and partly reflect government policy as it exists today. The number of 150,000 people is less than the immigration numbers we have experienced in recent years and is well within the range of other estimates. The ABS, for instance, recently did an updated economic projection analysis for labour market developments and that included a range that went from 120, I think, certainly to 180, with 150 in the middle. The reason for raising the net immigration numbers through time was to keep the ratio of net migration roughly proportional to the number of people in Australia.

Senator «MILNE»-The assertion is that immigration levels will remain constant for the next 42 years, surge for the following 30 and then stabilise again. That might be a convenient thing for the modellers, but, in my view, it is not in any way likely to be representative of the real paths of immigration. Given everything that has been said about climate change and the fact that we already have 32 million climate refugees around the planet now, to assume that immigration growth will take that pattern does not seem plausible. So is this one of the ones that will be subject to a sensitivity analysis? It looks to me to be one of the crucial areas for which we should have a sensitivity analysis.

Ms Quinn-As I said before, precise sensitivity analyses of things like that are part of our submission to the government for it to consider. The population numbers are part of the reference scenario for building our world without climate change, so these are projections about population in the absence of climate change. The way the economic modelling is framed is to imagine a world without climate change and then to imagine a world where climate change mitigation policy is being undertaken. In that world we would not be taking account of any implications, positive or negative, of potential climate change for population, in making those underlying assumptions.

Senator «MILNE»-Let me try another area in your models. Your modelling says that the price of coal, oil, gas and iron ore are expected to fall dramatically in the near future. You go on to say that the terms of trade are assumed to fall by around 25 per cent in the next few years, with oil prices assumed to fall by around 40 per cent over the next seven years. How plausible is that, given our recent experience? First of all, was ABARE in any way responsible for the basis of this assumption?

Ms Quinn-As is highlighted on page 15 of the Treasury assumptions book that was released by the government, the view about global energy prices was formed partly by examining Australian evidence but also by looking at the International Energy Agency projections; they do projections out to 2030 and, in some cases, beyond. Given that oil prices are driven by world economic developments, we assessed what the International Energy Agency had to say. ABARE have been involved in the steering committee for our part of the project and obviously we canvassed other opinions and views. We had to make assumptions about oil prices; they are assumptions rather than forecasts. Most people know that forecasting oil prices beyond a week or two-or even a day-can be quite difficult. The assumption that oil prices will come down over time was based on a view from the International Energy Agency about long-run marginal costs of extraction for oil resources, and that is what is reflected in the assumptions.

Senator «MILNE»-Do you accept, though, that there is also a growing body of evidence that oil has peaked globally and that, in the next seven years, we are going to see oil go through to $150 or $200 a barrel rather than fall by 40 per cent? Do you accept that there is a body of evidence out there to that effect?

Ms Quinn-I am aware of people who hold that view and the analysis around that. We have taken the assumption from the International Energy Agency.

Senator «MILNE»-Let me put it to you that, if prices of energy fall by the amounts forecast in your assumptions, the introduction of an ETS will have no noticeable impact on the price of energy; in fact, it appears under your scenarios that energy will be cheaper in the future with an ETS in place. Is that correct?

Ms Quinn-The analysis we are doing is developing a reference scenario where there is no climate change and then we are looking at the implications for introducing emissions pricing in Australia and the world. So we are looking at the difference from base, in a sense: what happens with an emissions price and what happens without an emissions price. We are not necessarily forecasting for the world in absolute terms; we are looking at what are the implications of an emission price in and of itself. So, in that world, putting a price on emissions typically would be expected to raise the price of emissions-intensive goods; energy is very emissions intensive.

Senator «MILNE»-I appreciate that, but you also make the assumption that you have falling underlying prices. Therefore, whilst you might put a carbon price on top of that, your overall energy price is going to be cheaper, according to your analysis.

Ms Quinn-It very much depends on the trajectories and targets in the global objective and the emission prices or carbon prices that are examined.

Senator «MILNE» -So will there be a sensitivity analysis as far as the oil price goes?

Ms Quinn-Once again, we have been looking at and examining a range of sensitivities. The advice that we have provided to government will be released as part of the government's report by the end of October.

Senator «MILNE» -So you are saying that those sensitivity analyses will be published?

Ms Quinn -I cannot be definitive about what will be published; that is in the hands of the government.

Senator «MILNE» -But at least they will be done; whether or not they will be published is a political decision.

Ms Quinn -Yes.

Senator «MILNE»-I have a lot of questions and will put some on notice. Senator Xenophon asked a couple of questions about carbon capture and storage that I would like to pursue. Table 21 shows that capital costs associated with, as yet, non-existent carbon capture plans are expected to decline at three times the rate of the decline in the cost of wind and biomass. Whilst MMA is cited as the source of this assumption, I would like to know how these cost de-escalators were estimated. Why would you assume that a non-existent technology, which is way behind schedule and costing many times more than people ever expected it to, would come down in price at three times the rate of the cost of wind and biomass, which already exist as proven technologies and are becoming cheaper all the time as they achieve critical mass?

Ms Quinn-Consistent across different consultants and different modelling in the public domain on this issue, typically it is assumed that, as new technologies go through the phase of development from scientific discovery or analysis to demonstration, to deployment, to commercialisation and to becoming an old technology, there are thoughts about a cost curve. Obviously, costs increase as people start thinking about how to build and implement these technologies. But, once they become commercially viable, costs start falling through time, as people and organisations learn by doing.

The view that has been put to us by the people who know more about this than I do is that therefore new technologies have greater de-escalations in their capital costs once they become commercial and that more mature technologies have already had those capital cost reductions, so their future de-escalation will be less. In that context, wind, as you have said, has been implemented and currently is commercially viable in various parts of the world, as is biomass. So their de-escalation costs are expected to be less in the future, although they will still have reductions as new technologies and new types of wind turbines and biomass are introduced. However, as you have said, there is no commercial operation of carbon capture and storage; so the expectation is that, if it becomes viable or commercial, once it becomes deployed at a larger scale, it will experience the reductions in capital costs that have been found in other technologies.

These de-escalated costs are examined through looking at historical analysis and precedent for different types of technological advances, not just in the electricity sector but also in general engineering and other types of capital costs analysis. It is fairly well established that there is this cost de-escalator; the range and the extent is obviously a matter for judgement, and there is great uncertainty about what might happen on that front.

Senator «MILNE» -Who were the consultants on carbon capture and storage?

Ms Quinn -As I have said, MMA is the consulting company that is helping us more closely. We have also already received information from other parts of the electricity generation sector and have examined other analysis that is in the public domain from various modelling organisations; so we have tried to capture a view. In addition, the International Energy Agency has done quite a lot of analysis on the cost of different technologies across the board. We also got some advice from ACIL Tasman on the international costs of various technologies in different locations as well as the range of technology options. In addition, there is quite a lot of literature out there on the various options. We have used consultants to provide the very detailed information that is required for Australia, done literature surveys across the world and across different technologies, and taken submissions from people who wanted to provide us with estimates of particular things. We have had quite a few conversations with particular people involved in technologies, particularly some of the renewables; people have wanted to approach us and talk to us about those technologies. So we have been able to take that information on board also.

Senator XENOPHON-In terms of the assumptions at table 21, you have ‘ultra supercritical coal' at $2,255 per kilowatt with a 0.5 de-escalator. With ‘USC'-ultra supercritical coal-'with post-combustion capture', it is a bit more at $2,482 but with a 1.5 per cent de-escalator. I do not understand how an ultra supercritical coal plant with post-combustion capture will be cheaper to build than one without post-combustion capture. If you have the added extras, how can it eventually be cheaper?

Ms Quinn-According to the table, the ultra supercritical coal is $2,255 and the one at the bottom, the ‘USC with post-combustion capture', is $2,482.

Senator XENOPHON -But it will be cheaper eventually, won't it?

Ms Quinn-Eventually, over time, it will be cheaper, if these de-escalations continue. But that is partly because the de-escalation costs take account of the higher level; the 1.5 is on the $2,400. They do not necessarily cross over the time horizon that you are looking at.

Senator XENOPHON-I do not get it. How can something with post-combustion capture be cheaper to build than something without post-combustion capture, in a few years time?

Ms Quinn -It is not clear that it is.

Senator XENOPHON -It is. With a 1.5 per cent capital cost de-escalator, by about 2018-19 it will be cheaper.

Ms Quinn -I am happy to take that on notice, but I do not think that is the case.

Senator «MILNE»-The International Energy Agency released a report only yesterday on carbon capture and storage, which shows that it is even further away than they thought previously and that to make it commercial will require billions of dollars, yet you assume that you will have commercial plants available in 2020. However, I would like to know particularly why you assume that the costs of constructing the pipes necessary for carbon capture and storage would be paid for by anyone else but the generator. Why did you make the assumption in the costs that somebody else was going to pay for all the pipes?

Ms Quinn-We have treated equally in the model the cost of building transmission lines for some of the renewable technologies and the cost of building potential pipelines for carbon capture and storage. We have assumed that, instead of people having to pay upfront the full capital cost of the infrastructure, that full capital cost is turned into a transaction fee for transmission or for pumping the CO2. So, instead of being paid as a fixed cost in terms of the building of power plants, it is paid as a variable cost for every unit of carbon moved or electricity transmitted. This is an economic way of modelling and does not assume any particular ownership of the infrastructure. It is costing the infrastructure but not costing who owns the infrastructure.

In terms of the economic analysis that we have been doing, we are not looking at ownership structures-whether generator X or generator Y has to build pipelines or whether geothermal plant A or wind farm B has to pay. We are looking at the overall economic cost of the different technologies and ensuring that they are paying for the capital costs, the variable costs and the costs of infrastructure associated with that technology.

Senator «MILNE» -Why do you assume that the cost of CCS capital will be subsidised?

Ms Quinn -I do not think that is the case. The cost of transporting and storing carbon is being captured in our economic modelling.

Senator «MILNE» -But the reflection of the relative costs assumes that the generators are not paying for that.

Ms Quinn -No. The generators are paying for the variable cost of moving and transporting carbon. When they generate electricity and they incur capture costs they also have to pay for the storage of that technology. The storage and transportation costs include the cost of the capital of building the pipes. They do not pay for it upfront. There is not a billion dollars in one year that pays for the pipes. That billion dollars is spread out over the capital life of the pipes.

 --------------Senator «MILNE» -As to the Treasury modelling on the ETS that we were discussing previously, I note that Treasury says that limits are placed on the rate of take-up and total take-up of renewable energy capacity, reflecting resource availability and engineering and technical constraints. Those assumptions are represented in chart 7 and they are central to this whole modelling exercise. I would like to know the assumptions on which those constraints were based, the consistency or otherwise of those constraints with the increase in the MRET to 45,000 gigawatt hours and whether or not similar engineering and technical constraints to do with the installation of carbon capture were used, particularly given that carbon capture is going to require very high-grade steel. What are your assumptions in relation to these constraints? Was there consistency with the MRET? Were the same constraints applied to CCS?

Ms Quinn-I can answer some of those; others I will have to take on notice in terms of the technical detail. As you have correctly pointed out, in chart 7 of the Treasury assumptions book, we have provided some information about cumulative renewal capacity constraints. This information was provided to us by the electricity sector bottom-up modellers MMA, and they have based that analysis on a region-by-region examination of what is feasible both technically and practically in terms of the timing. You can see, for instance, that some of the constraints ease over time as infrastructure and other developments occur. We have also examined capacity constraints around carbon capture and storage in terms of when it might be possible for the technology to be deployed, around the cost structure of carbon capture and storage and around the implications for building infrastructure around carbon capture and storage. We have attempted as far as it has been possible within the constraints and time available for our project to treat all technologies equally. In terms of the precise details about the constraints, I would have to take that on notice. I do not have the details before me at the moment. I think there was third leg of that question, which I-

Senator «MILNE»-It was the CCS. I want to know why you would treat a proven renewable energy technology already in existence in exactly the same way as a technology that is unproven and not likely to be commercially viable before 2020, if ever, but certainly not before then.

Ms Quinn-As I said, making assumptions about the precise mix of technology or the precise path of the Australian or world economy is a very difficult task to undertake. We have tried to take the best advice-and have taken a very open-door approach-from people who have different views to examine the range of possibilities, and plausible people can disagree about various assumptions. We have taken the advice as provided to us by experts in the field that carbon capture and storage may well be viable. Whether or not it is commercial depends on design features of the Carbon Pollution Reduction Scheme and the possible carbon prices that are associated with that scheme.

The economic modelling takes assumptions about capital cost structures, variable cost structures and the demand for different types of electricity and then it chooses between those on the basis of that information. The economic models do not have a pejorative view about one technology over the other; it is purely based on the cost structures of the different technologies. We have tried to capture, as much as we can, the cost structures from the building of the equipment, the running of the equipment, the storage or the other infrastructure that goes with the equipment. But, as I said before, over long time horizons this can be quite a difficult exercise, and sensitivity analysis is important for looking at these questions.

Senator «MILNE» -What is the de-escalation rate for solar/thermal compared with carbon capture and storage?

Ms Quinn -I will take you to table 21, which looks at the de-escalation rates. A concentrated solar/thermal plant is assumed to have a capital cost de-escalate rate of 1.5 per cent per year between 2010 and 2020 for Australia. If this technology were to be chosen by the economic modelling, the assumption would be 1.5 per cent for the ultra super-critical post combustion capture black coal plant. These are capital cost de-escalators on the assumption that that technology is chosen. These are the possibilities that are provided to the economic model. Whether or not they are chosen depends on the location, the cost and the particular type of electricity that might be demanded of that region at that time. Not all of these technologies are chosen in different scenarios. It depends on the actual cost structure of the precise analysis that has or would be undertaken.

CHAIR -Last question, Senator «Milne».

Senator «MILNE»-If it the last question, I will go to why you are assuming that the impact of the 2006-07 drought will disappear by 2012 and hydro-dam levels will be replenished. Does that mean that future rainfall patterns have been factored into the forecast output from electricity, or are you just making an optimistic assumption that the drought is over and the dams are full by 2012? What is the basis of that assumption?

Ms Quinn-The reference case world is a world without climate change, as I mentioned before, and it is based on the typical assumptions that are made in economic analysis and in the budget forecasts. Looking out past the current forecasts and current projections, we typically assume the return to normal conditions. That is just a technical assumption that is applied-and we have applied the same approach to this. We have assumed a return to normal conditions; it is not a forecast of a return to normal conditions.

Senator «MILNE»-How can you assume a return to normal conditions if you accept climate change is real? That is the whole point of the ETS I would have thought. How can you make an assumption that it is going to return to past levels?

Ms Quinn-As I said, we have been asked to look at the mitigation costs of putting a price on emissions. We have not been examining a world of climate change impacts. We are looking at the economic implications of what happens if you put a price on emissions against a reference case without climate change in it.