National Economics' energy modelling systems
National Economics' energy modelling capability has been gradually developed over the last two decades. The energy demand and production model is an integral part of National Economics' Institute Multipurpose model (IMP model).
The energy modelling systems developed by National Economics have been applied in analysing numerous energy related issues including, over recent years:
- the impact of removal of cross-subsidies in electricity prices;
- forecasts of electricity demand and load growth;
- projections of greenhouse gas emissions;
- evaluating alternative power station options; and
- assessing the impact of increased penetration of energy efficient technologies and renewable energies on energy demand and supply and greenhouse gas emissions.
National Economics' energy demand and supply model offers flexibility in the mode of model use. It is directly linked to macroeconomic models of the national and state economies so that feedback effects between the energy sector (prices, investment) and the overall economy are effectively identified.
The alternative modes of model use include:
- energy demand and supply forecasting at the state/regional industry level;
- dynamic policy analysis; and
- end-use technology forecasts which disaggregate energy demand by service and efficiency.
The figure below shows the overall integrated structure of National Economics' energy modelling system. The system comprises of:
- the national macroeconomic and industry activity model;
- a state economic and industry activity model;
- an energy forecasting model (EFM);
- an energy technology model (ETM);
- an energy environmental impact model (ENVI); and
- energy production and pricing model.
The national and state economic models provide forecasts of industry output by sector, capital stocks, dwellings formation numbers and population by state and territory.
Table 1 shows the structure of the EFM. The main structural features of National Economics' energy forecasting modelling system are:
- primary and secondary energy is modelled at the regional (i.e. state) level;
- energy demand is modelled on an industry/sectoral basis by state (refer Table 1);
- major energy intensive projects are explicitly taken into account;
- environmental impacts in physical and dollar terms can be estimated;
- energy prices and technological trends determine substitution between fuels in the different sectors of the economy; and
- primary fuels used in electricity generation are modelled at the state level and include specific parameters for individual power stations within each state grid.
Table 1: Industry structure of the EFM |
|
|---|---|
Main sectors |
|
1. Primary |
|
Division A |
Agriculture |
Division B |
Mining |
2. Secondary |
Division C Manufacturing by 2-digit ASIC |
ASIC 21 |
Food, beverages, tobacco |
ASIC 23-24 |
Textiles, clothing, footwear |
ASIC 25 |
Wood, wood products, furniture |
ASIC 26 |
Paper, paper products, printing |
ASIC 27 |
Chemicals, petroleum and coal |
ASIC 28 |
Non-metallic minerals |
ASIC 29 |
Basic metals (iron and steel, aluminium smelting) |
ASIC 31 |
Fabricated metal products |
ASIC 32 |
Transport equipment |
ASIC 33 |
Other machinery and equipment |
ASIC 34 |
Miscellaneous manufacturing |
3. Tertiary sectors |
|
Division E |
Construction |
Division F |
Wholesale and retail trade |
Division G&H |
Transport and communication |
Division I |
Finance, property, business services |
Division J&K |
Public administration, defence, community services |
ASIC 37 |
Water, sewerage and drainage |
Division L |
Recreation, personal and other services |
4. Residential sector |
|
5. Energy supply sectors |
|
Electricity generation sectors |
|
By fuel type |
|
1. Gas |
Natural gas* |
2. Electricity*** |
Electricity |
|
|
The EFM produces detailed energy forecasts for each state and the Northern Territory. The energy consumption data contained in the EFM are based on Australian Bureau of Agricultural and Resource Economics (ABARE) historical energy statistics. These data are supplemented by information from individual generating and distribution utilities, as well as detailed consumption figures from various major energy users. Detailed energy consumption data by state, fuel type, and industry sector are updated on a financial year basis.
The econometric forecasts of energy usage are based upon a large-scale econometric model which relates state energy consumption to:
- state economic activity across 25 industry sectors;
- population, household size and customer numbers;
- appliance penetration and efficiency;
- real household disposable incomes;
- real energy prices;
- weather conditions; and
- major new industrial, mining and commercial developments.
The modelling approach taken explicitly identifies prospective shifts in the pattern of energy consumption induced by structural shifts in economic activity and real energy prices over the period.
Residential energy usage is determined by the number of households connected to the supply grid, and average consumption levels per connected household. National Economics' forecasts of domestic customer numbers are related to population growth and household formations, while average consumption per connected household is modelled against real household disposable incomes, real energy prices, weather conditions, and domestic appliance penetration and end-use efficiency.
The econometric models provide detailed information on the relationship between electricity demand and economic activity and prices. They do not, however, provide information on the market share of individual electrical technologies nor the efficiency of these technologies.
Hence, in addition to the `tops down' econometric forecasting system, the model also includes a detailed bottoms up end use or energy technology modelling system.
The energy technology modelling system (ETM) models state energy demand by energy service at a more aggregate level of industry structure. The main structural features of the ETM are:
- primary and secondary energy are modelled at the state level;
- energy demand is modelled on an industry basis;
- efficiencies of alternative energy end-use are renewable technologies are included by sector; and
- capital stocks, household numbers, appliance and equipment penetration ratios by sector.
Table 2 shows the structure of the state ETM model.
Table 2: Energy technology model structure |
||
|---|---|---|
Residential |
Space heating |
Dishwashers |
Commercial |
Heating |
Water heating |
Industrial |
Smelting |
Low temperature heating |
Transport |
Traction |
Road |
The end-use model disaggregates electricity consumption in each sector (e.g. lighting, process heat, water heating) by energy service and relates energy consumption to indicators of activity in each sector and the average and marginal efficiency of electrical technologies. The end-use model includes twenty-four residential end-uses, nine commercial and eight manufacturing end-uses. Fourteen of these end-use categories relate to electrical appliances and equipment.
End-use energy demand models have typically been used in policy studies and have not extensively been utilised for forecasting energy demand. This mainly reflects the availability of information on stocks of appliances and equipment and end-use consumption data in both Australia and overseas. End-use models require detailed information on:
- the stock of appliances and equipment;
- the efficiency of the stock of appliances and equipment;
- the penetration of energy services by type; and
- the utilisation of energy services by type.
The efficiency of energy services varies significantly with the appliance configuration, model type, size, and operating environment. For this reason, groups of energy using appliances and equipment are typically aggregated into major energy service groups.
The end-use modelling system is particularly important in modelling changes in the market shares of various energy services which will not necessarily be reflected in econometric cross price elasticity estimates.
For example, an increased penetration of gas hot water heaters in the residential sector, will provide a substantial boost to domestic natural gas consumption largely at the expense of electricity. This changing market share would not necessarily be accurately identified in the econometric equations, hence the necessity to reconcile the results of the econometric and end-use forecasting systems. This reconciliation process allows the coefficients in both modelling systems to be regularly updated to reflect changing technological and economic parameters through time. Each of the two modelling approaches has strengths and weaknesses and the combination of the two substantially improves the veracity of model outputs.
Accurate end-use forecasting requires extremely detailed information on appliance stocks, efficiency, penetration and utilisation. At present, reasonable data is available for the residential sector, however, both industrial and commercial energy service data remain relatively inaccurate. National Economics' energy modelling systems recognise end-use data imperfections, particularly in the industrial and commercial sectors, and combines the end-use approach with disaggregated econometric forecasting techniques in the model framework.
The energy modelling systems have been used extensively over recent years in the forecasting of both short and long term energy demands and also analysis of the impact of various institutional and policy changes currently occurring in the Australian energy sector.
Given the long lead times associated with the establishment of additional energy supply capacity, accurate long term energy demand forecasting is extremely important. Detailed supply capacity information is incorporated into the model structure for both the electricity and gas supply industries and in medium and long term energy demand forecasting, the IMP energy modelling system will identify the timing of additional supply capacity requirements for both electricity and natural gas.
The environmental model, ENVI, is integrated with EFM, ETM and National Economics' national and state models and can also be used in regional analysis. The model comprises a framework of linkages between economic activities and environmental impacts, for example the greenhouse gas and noxious emissions (units and where feasible values) associated with the production, transmission and use of energy services such as natural gas (primary) and electricity (secondary). The model covers economic activities outside the energy field, for example forestry, solid waste, water and waste water and is under continual development. The economic environmental impact of environmentally disruptive activities, environmental improvement activities (e.g. reforestation) can be estimated, for example the inputs of a reforestation program, by specification of the inputs (seedlings production and planting, cultivation, etc.) and outputs (timber, carbon dioxide absorption, water production effects, etc.) of reforestation.