Description of E3ME Global Edition


E3ME is a global macro-econometric E3 (Energy-Environment-Economy) model.  It was originally developed through the European Commission’s research framework programmes and is now widely used in Europe and outside Europe for policy assessment, for forecasting and for research purposes.

The E3ME model embodies three important strengths:

  • Integrated treatment of the world’s economies, energy systems, emissions and material demands, enabling the model to capture two-way linkages and feedbacks between these components. Key environmental factors, such as greenhouse gas emissions and resource use are represented explicitly in the model using physical units where appropriate.
  • A high level of disaggregation, enabling detailed analysis of sectoral and country-level effects from a wide range of scenarios. Social impacts (including unemployment levels and distributional effects) are important model outcomes.
  • Its econometric specification, addressing growing concerns over conventional macroeconomic models and providing a strong empirical basis for analysis. E3ME’s specification enables the model to fully assess both short and long-term impacts. It is not limited by many of the restrictive assumptions common to Computable General Equilibrium (CGE) models.

This model description provides a short summary of the E3ME model. For further details, the reader is referred to the model manual available online from www.e3me.com [1].

E3ME Structure

The structure of E3ME is based on the system of national accounts, as defined by ESA95 (European Commission, 1996), with further linkages to energy demand and environmental emissions.  The labour market is also covered in detail, with estimated sets of equations for labour demand, supply, wages and working hours.  In total there are 33 sets of econometrically estimated equations, also including the components of GDP (consumption, investment, international trade), prices, energy demand and materials demand. Each equation set is disaggregated by country and by sector.

E3ME’s historical database covers the period 1970-2010 and the model projects forward annually to 2050.  The main data sources for European countries are Eurostat, DG Ecfin’s AMECO database and the IEA, supplemented by the OECD’s STAN database and other sources where appropriate.  For regions outside Europe, additional sources for data include the UN, OECD, World Bank, IMF, ILO and national statistics.  Gaps in the data are estimated using customised software algorithms.

Key strengths of E3ME

In summary the key strengths of E3ME lie in three different areas:

  • the close integration of the economy, energy systems and the environment, with two-way linkages between each component
  • the detailed sectoral disaggregation in the model’s classifications, allowing for the analysis of similarly detailed scenarios
  • the econometric specification of the model, making it suitable for short and medium-term assessment, as well as longer-term trends

Application of E3ME

Although E3ME can be used for forecasting, the model is more commonly used for evaluating the impacts` of an input shock through a scenario-based analysis.  The shock may be either a change in policy, a change in economic assumptions or another change to a model variable.  The analysis can be either forward looking (ex-ante) or evaluating previous developments in an ex-post manner.  Scenarios can be used either to assess policy, or to assess sensitivities to key inputs (e.g. international energy prices).

For ex-ante analysis a baseline forecast up to 2050 is required; E3ME is usually calibrated to match a set of projections that are published by the European Commission and IEA.  The scenarios represent alternative versions of the future based on a different set of inputs.  By comparing the outcomes to the baseline (usually in percentage terms), the effects of the change in inputs can be determined.

It is possible to set up a scenario in which any of the model’s inputs or variables are changed.  In the case of exogenous inputs, such as population or energy prices, this is straight forward.  However, it is also possible to add shocks to other model variables.  For example, investment is endogenously determined by E3ME, but additional exogenous investment (e.g. through an increase in public investment expenditure) can also be modelled as part of a scenario input.

Model-based scenario analyses often focus on changes in price because this is easy to quantify and represent in the model structure.  Examples include:

  • changes in tax rates including direct, indirect, border, energy and environment taxes
  • changes in international energy prices
  • emission trading schemes

All of these can be represented in E3ME’s framework reasonably well, given the level of disaggregation available.  However, it is also possible to assess the effects of regulation, albeit with an assumption about effectiveness and cost.  For example, an increase in vehicle fuel-efficiency standards could be assessed in the model with an assumption about how efficient vehicles become, and the cost of these measures.  This would be entered into the model as a higher price for cars and a reduction in fuel consumption (all other things being equal).  E3ME could then be used to determine:

  • secondary effects, for example on fuel suppliers
  • rebound effects[2]

Limitations to the analysis

The main limitation of E3ME is the sectoral disaggregation of its sectors.  The industry classification is relatively detailed, covering up to 69 sectors at the NACE 2-digit Level for the European regions.  However, due to the availability of the data, it is not possible to go into this level of Detail for the world regions, as such they are disaggregated to 43 sectors. Furthermore, it is not possible for firm-based level, or very detailed product groups to be includet in the model.  For this type of analysis our recommendation is that the model (which provides an indication of indirect effects) is used in conjunction with a more detailed bottom-up or econometric analysis (which can capture detailed industry-specific effects).

The other main limitations to the model relate to its dimensions and boundaries.  Broadly speaking E3ME covers the economy, energy and material demands and atmospheric emissions.  While it is possible to provide an assessment of other policy areas, it is necessary to make assumptions about how this is translated into model inputs.  It should be noted that the E3ME_Global edition will eliminate one limitation of the current E3ME model which treats regions outside Europe as exogenous.  All world regions will be endogenously modelled in the new model version.


[1] In the example, the higher fuel efficiency effectively reduces the cost of motoring.  In the long-run this is likely to lead to an increase in demand, meaning some of the initial savings are lost.  Barker et al (2009) demonstrate that this can be as high as 50% of the original reduction.

[2] See the E3ME manual


Liaise ownership:


Input variables:

The main dimensions of the model

The main dimensions of the E3ME model are:

  • 53 countries - EU28 and major world regions (see Table 1.1)
  • For European countries, 69 economic NACE Revision 2 sectors, including disaggregation of the energy sectors and detailed 38 service sectors
  • For regions outside Europe, 43 economic NACE Revision 1.1 sectors, including disaggregation of the energy sectors and 16 service sectors
  • For European countries, 43 categories of household expenditure
  • For regions outside Europe, 28 categories of household expenditure
  • 22 different users of 12 different fuel types
  • 14 types of air-borne emission (where data are available) including the six greenhouse gases monitored under the Kyoto protocol.



Output variables:

Standard outputs from the model

As a general model of the economy, based on the full structure of the national accounts, E3ME is capable of producing a broad range of economic indicators.  In addition there is range of energy and environment indicators.  The following list provides a summary of the most common outputs:

  • GDP and the aggregate components of GDP (household expenditure, investment, government expenditure and international trade)
  • sectoral output and GVA, prices, trade and competitiveness effects
  • sectoral international trade in bilateral format and can be presented by trade blocs
  • consumer prices and expenditures, and implied household distributional effects
  • sectoral employment, unemployment, sectoral wage rates and labour supply
  • energy demand, by sector and by fuel, energy prices
  • CO2 emissions by sector and by fuel
  • other air-borne emissions
  • material demands

This list is by no means exhaustive and the delivered outputs often depend on the requirements of the specific project.  In addition to the sectoral dimension mentioned in the list, all indicators are produced at the national and regional level and annually over the period up to 2050.

Documentation for the end user:

Scientific documentation:

E3ME as an E3model

The E3ME model comprises:

  • the accounting balances for commodities from input-output tables, for energy carriers from energy balances and for institutional incomes and expenditures from the national accounts
  • environmental emission flows
  • 33 sets of time-series econometric equations (aggregate energy demands, fuel substitution equations for coal, heavy oil, gas and electricity; intra-EU and extra-EU commodity exports and imports; total consumers’ expenditure; disaggregated consumers’ expenditure; industrial fixed investment; industrial employment; industrial hours worked; labour participation; industrial prices; export and import prices; industrial wage rates; residual incomes; investment in dwellings; normal output equations and physical demand for seven types of materials[1])

Energy supplies and population stocks and flows are treated as exogenous.

Figure 1.1 shows how the three components (modules) of the model - energy, environment and economy - fit together.  Each component is shown in its own box.  Each data set has been constructed by statistical offices to conform with accounting conventions.  Exogenous factors coming from outside the modelling framework are shown on the outside edge of the chart as inputs into each component.  For each region’s economy the exogenous factors are economic policies (including tax rates, growth in government expenditures, interest rates and exchange rates).  For the energy system, the outside factors are the world oil prices and energy policy (including regulation of energy industries).  For the environment component, exogenous factors include policies such as reduction in SO2 emissions by means of end-of-pipe filters from large combustion plants.  The linkages between the components of the model are shown explicitly by the arrows that indicate which values are transmitted between components.


Figure 1.1 E3 linkages in the E3ME model

The economy module provides measures of economic activity and general price levels to the energy module; the energy module provides measures of emissions of the main air pollutants to the environment module, which in turn gives measures of damage to health and buildings[1] (estimated using the most recent ExternE[2] coefficients).  The energy module provides detailed price levels for energy carriers distinguished in the economy module and the overall price of energy as well as energy use in the economy.

Technological progress plays an important role in the E3ME model, affecting all three Es: economy, energy and environment.  The model’s endogenous technical progress indicators (TPIs), a function of R&D and gross investment, appears in nine of E3ME’s stochastic (econometric) equation sets including trade, labour market activity, and prices. Investment and R&D in new technologies also appears in the E3ME’s stochastic energy and material demand equations to capture energy/resource savings technologies as well as pollutions abatement equipment. In additional, E3ME also captures low carbon technologies in the power sector through the Energy-Technology-subModel (ETM).

Treatment of international trade

An important part of the modelling concerns international trade.  The E3ME_Global model will move on from the current E3ME’s assumption of trade ‘pool’ to detailed bilateral trade between regions. Three key stages of international trade modelling in E3ME_Global are:

  • econometric estimation of regions’ sectoral import demand
  • econometric estimation of regions’ bilateral trade to get regions’ imports from different partner
  • forming regions’ exports from other regions’ import demand

First, the imports demand of a commodity for a region is modelled using three main factors:

  • activity, sales in the domestic market
  • three price effects: import price, price of sales to the domestic market and the relative price of the currency i.e. effects of exchange rate
  • technological progress measure which allow for the relative effects of innovations on trade performance and capture an important long-term dynamic effect on economic development

After region’s total import demand is formed, a bilateral trade modelling is introduced to captures price and other competitiveness effects between different regions. Region’s total imports demand is allocated to bilateral format on this basis. Finally, region’s total exports can be formed using bilateral trade information.

Econometric specification of E3ME

The econometric specification of E3ME gives the model a strong empirical grounding and means it is not reliant on the assumptions common to Computable General Equilibrium (CGE) models, such as perfect competition or rational expectations.  E3ME uses a system of error correction, allowing short-term dynamic (or transition) outcomes, moving towards a long-term trend.  The dynamic specification is important when considering short and medium-term analysis (eg up to 2020) and rebound effects[3], which are included as standard in the model’s results.


[1] Not available for regions outside Europe.

[2] http://www.externe.info/tools.html

[3] Where an initial increase in efficiency reduces demand, but this is negated in the long run as greater efficiency lowers the relative cost and increases consumption.  See Barker et al (2009).




Hector Pollitt (Director - International Modelling), hp@camecon.com Unnada Chewpreecha (Manager - International Modelling), uc@camecon.com



Publications and applications