« Previous | Contents | Next »
Listen
EXECUTIVE SUMMARY
Introduction
This project uses an experimental energy-economy-environment computable general equilibrium ( CGE) model of the Scottish economy ( AMOSENVI) to conduct illustrative simulations of the economic and environmental impacts of various options to reduce the generation of CO2 emissions (as the main greenhouse gas) in Scotland. These simulations are illustrative in nature because the process of appropriate database development and model specification for a comprehensive and accurate analysis of climate change policy issues for Scotland, while advanced relative to many regional (and even national) economies, is still in its very early stages. One of the key objectives of this project is to illustrate the potential value-added to Scotland's analytical capacity if further investment is made (by both the policy community, particularly in terms of data provision, and the research community, with public support, for example by seeking support from the research councils) in developing an appropriate CGE modelling framework
Objectives of the research
The objectives of this project were:
1. To review the literature on the macroeconomic impacts of mitigation against climate change.
2. To model the impact of climate change mitigation policies on the Scottish economy.
The literature review is reported in Technical Appendix 1; it has informed development of the existing (experimental) energy-economy-environment computable general equilibrium ( CGE) model of the Scottish economy ( AMOSENVI) used under Objective 2, and will continue to inform future developments of the modelling framework. The CGE simulations carried out under Objective 2 were agreed with the Scottish Government project management team in the context of the current Government Economic Strategy and National Performance Framework in order to inform the forthcoming Scottish Climate Change Bill. However, given the early stage of development, and consequent experimental nature of the current AMOSENVI model, the simulations carried out are illustrative in nature. In this context, a more fundamental objective of this project was to explore how far this model can be used to answer questions regarding the economic impact of reducing greenhouse gas emissions in Scotland, and to identify further steps in the development of the AMOSENI framework.
Simulation results
Five sets of illustrative simulations using the AMOSENVICGE modelling framework have been carried out and report the impacts of: increased energy efficiency; population and demographic change; the imposition of costly requirements on households to reduce energy use; and the impacts of increasing renewable energy supply. In the case of the latter, we also offer a detailed input-output analysis of the impacts of changing the structure of the electricity supply sector in favour of renewables. The motivation for carrying out and reporting this additional analysis is that we are able to examine a more detailed disaggregation of the electricity supply sector using input-output than CGE, as the current AMOSENVI framework only distinguishes between renewable and non-renewable generation at a very aggregate level.
Findings of energy efficiency simulations
We simulate a very simple increase in energy efficiency in the production sectors of the Scottish economy. This allows us to identify the key drivers of what are known as 'rebound' effects. We do not attempt to consider how the efficiency improvements may be achieved (this will be the focus of future research). The results of the energy efficiency simulations highlight the issue that, because of the system-wide response to falling actual and effective energy prices, particularly in an economy like Scotland (a producer and exporter of energy), reductions in energy consumption due to increased efficiency are likely to be partially or even wholly offset by increased demand for energy ( i.e. rebound effects will occur). The more extreme variant of rebound effects, backfire, where energy consumption actually increases in response to increased energy efficiency, with consequent increases in the level of CO2 emissions and the CO2 intensity of GDP/production, is more likely when the Scottish energy supply sectors, particularly the electricity sectors, are targeted directly. This is due to the fact that these are heavily traded and the responsiveness of demand to falling prices is therefore relatively high. Our long run analysis suggests that a 5% increase in energy efficiency directed at the Non-Renewable Electricity sector will increase GDP by 0.5%, but total CO2 emissions in Scotland by more, 1.18%, with the implication that the CO2 intensity of Scottish GDP will rise by 1.29%. If a 5% increase in energy efficiency is directed at the Renewable sector instead, our analysis suggests a smaller increase in CO2 emission (0.07% in the long-run, but only a 0.05% increase in GDP, so, again the long-run impact on the CO2 intensity of GDP would be an increase of 0.02%. However, if the same proportionate increase is directed at any other of the 25 sectors modelled, (smaller) increases in GDP would be accompanied by reductions in CO2, even though rebound effects are present in all cases, leading to a reduction in the CO2 intensity of Scottish production.
Our analysis actually suggests that, in terms of reducing the CO2 intensity of Scottish production (if not the level of CO2 emissions), improving labour productivity may actually be a more effective form of technological progress to focus attention on. However, we qualify our results not just with respect to the quality of currently available data, but by the fact that we do not attempt to consider the precise form of efficiency improvements, the costs involved in introducing them or the use of any resulting revenues. These factors may have a significant impact on results. However, at present the specification of the AMOSENVI framework is not sufficiently sophisticated to effectively account for them (though some broad brush analysis has previously been attempted in carrying out comparable analysis for the UK in a project commissioned by DEFRA). Here, in our initial work for Scotland, we instead focus on isolating and examining the basic system-wide response to improved energy efficiency ( i.e. the basic drivers of rebound and backfire effects) on the basis that it is necessary to understand these before introducing more complex, albeit very policy relevant issues.
Findings of population/demographic simulations
The results of the population/demographic change simulations suggest that population decline and ageing has a significant impact on the Scottish labour market, on economic activity, as well as on energy use and CO2 generation. Our central case of a 1.7% decline in total population and 15% decline in working age population between 2000 and 2050 produces a decline in GDP of 9.30% and a fall in CO 2 generation or 8.76%. The CO 2 intensity of production thus increases. Energy (both electrical and non-electrical) demands fall in this scenario. The functioning of the labour market and possibilities for in-migration of labour are the key factors influencing environmental impacts. This is because, as working population falls, the labour market will tighten, pushing wages up. The greatest impact will be felt in labour intensive sectors, where output prices will increase and competitiveness reduce to the greatest extent. Our analysis suggests that, because of the precise structure of the Scottish economy, and particularly the export intensity of more directly and indirectly labour intensive sectors, the export price index is particularly badly hit leading to falling competitiveness. With higher values for net migration to Scotland, the economic impact of ageing can become positive, and, although this will tend to increase CO 2 emissions, the faster rate of GDP growth means that the CO 2 intensity of Scottish production falls.
Findings of simulating income effects of costly requirements on households to reduce their energy use
The next scenario we attempted to simulate focussed on the labour market effects of costly requirements on households to reduce their energy use. At present the AMOSENVI model cannot be used to simulate policies aimed at changing household energy consumption behaviour. However, it can be used to examine the likely knock-on effects of reductions in household income that are likely to occur as a result. Therefore we simulate the economy-wide impacts of a reduction in household income (that may accompany/result from policy actions requiring households to reduce their energy use). We find that this will lead to a reduction in the level of CO2 emissions in the Scottish economy (up to 1.81% in the long-run for a 1% decrease in real household income), and also to the CO2 intensity of Scottish Production (-0.19% where real income falls by 1%), but this is at the cost of a contraction in GDP (-1.63% in the 1% scenario). The key driver of these results is out-migration from Scotland, due Scottish real (take-home) wages declining relative to those in the rest of the UK.
Findings of simulations increasing the share of electricity generated from renewable sources
In the final two sets of simulations, we attempted to model the impacts of increasing the share of electricity generated from renewable sources. Ideally, this should be done using a CGE framework, where more theory consistent supply and demand side behaviour can be modelled, and the economy's path of adjustment can be tracked. We do attempt a CGE analysis in our final set of simulations; however, the electricity sector is quite highly aggregated in the current AMOSENVI model. Therefore, we also carry out analyses of the impact of shifting the generation technology mix towards the target of 50% of generation from renewable sources by 2020, using an IO model that identifies eight different types of generation technology. We examine four illustrative scenarios where different types of generation technologies make up the 50% from renewable and non-renewable sources respectively, with complete removal of nuclear generation. While all the scenarios examined lead to between 3% and 9% reductions in Scottish CO2 emissions (taking income effects into account) in the long-run, our results suggest that the biggest gains in terms of CO2 reduction, 8.9%, are made when coal generation is phased out all together (although the introduction of carbon capture and storage may overturn this conclusion). However, again, while all scenarios generate long-run increases in GDP (between 0.27% and 0.63% above the baseline), the largest GDP gains and reductions in the CO2 intensity of production (-6.28%) are found where non-renewable electricity production is split between coal and gas generation, but with a relatively high share of production (10%) from marine technology. This is primarily because of the strong backward linkages that the marine generation sector has with sectors such as Construction and Communications, Finance and Business, all of which have relatively high GDP multipliers. However, it is important to bear in mind that these results will be sensitive to what is currently only an illustrative/experimental disaggregation of the electricity sector and input-output assumptions of fixed input relationships. As technologies mature ( e.g. marine generation), we may expect these to change over time, a factor that is not reflected in the results reported here.
In our final set of simulations, we move back to the AMOSENVI model to conduct a more sophisticated analysis of the impact of increasing reliance on renewable electricity generation technologies, but where we are only able to distinguish between aggregate renewables and non-renewables, and where the composition of these is fixed to that given by the 1999 input-output database. However, here we are able to consider how the growth in renewables may be induced (using subsidies). On the other hand, in the current CGE framework, we are not able to simulate the full 50% target for electricity from renewable sources by 2050 stated under the National Performance Framework. The maximum share of Scottish electricity generated from renewables that we are able to simulate is just over 20%. Our main finding in this scenario is that the proportion of electricity generated from renewable electricity technologies in Scotland from 10.40% to 20.06% reduces CO2 emissions by 4.15% in the long-run, but lowers Scottish GDP by 1.25% in the same time frame. When we allow total electricity generated in Scotland to be lower than in the base year, we can find scenarios in which electricity generated from renewable energy sources, and consequently, the falls in CO2 generation are greater. However, this comes at the cost of larger declines in GDP
Discussion of limitations of current CGE analysis
We present summary results of each of the scenarios modelled, along with a basic introduction to the CGE modelling approach, and offer our conclusions and recommendations in the main body of the report. This is followed by six in-depth technical appendices, providing more detail on the results of each of the simulation scenarios and a comprehensive review of wider developments in the energy-economy-environment CGE literature. It is important to bear in mind the illustrative/experimental nature of the current AMOSENVI model and the assumptions involved in each simulation reported (outlined in the main body of the report with more detail in the technical appendices).
Conclusions and recommendations
We emphasise that the analysis presented here is experimental and constrained by two broad factors:
- The need to further develop the Scottish input-output tables for the purposes of examining energy-economy-environment issues.
- The need to further develop the AMOSENVICGE modelling framework to look at a wider range of issues.
Both of these points are discussed in Section 7 of the report. We also offer some recommendations for strengthening Scotland's analytical capacity in this area. Development of the AMOSENVI framework is currently ongoing through various EPSRC and ESRC funded research projects being carried out by the regional and energy modelling teams at the Fraser of Allander Institute, University of Strathclyde. In terms of the development of the data infrastructure, this is an area where Scottish Government can play a direct role and it is important to note that the benefits of doing so would not be limited to better informing CGE models, as environmentally augmented input-output tables can be applied for a wide variety of analyses, including carbon accounting and footprint analyses.
« Previous | Contents | Next »