In addition to the totaled quantities available on the Cumulators sheet and the Cost Outputs sheet, the Energy Policy Simulator (EPS) calculates numerous other outputs that are useful for understanding the full range of impacts of the modeled policy package. A variety of these outputs are available on this sheet.

Changes in GDP and Emissions due to Energy Prices

In this section, the model estimates the impact that changes in fuel prices (as a result of the user’s policy package) will have on pollutant emissions and on GDP. Note that this is not a holistic measure of the impacts of the policy package on GDP because many policies would affect GDP positively or negatively in ways other than by affecting fuel prices. However, as the model only looks at first-order economic effects (as discussed above), and these GDP impacts would require the consideration of second-order and higher-order economic effects (including questions such as what the government does with increased tax revenue), the model is not equipped to make calculations on policy packages’ overall effects on GDP.

First, BAU GDP projections are taken in as input data. They are modified based on fuel prices and an elasticity of GDP with respect to fuel cost. (Fuel prices can vary by sector; we use industry sector fuel prices for this estimate, as industry has a larger impact on GDP than buildings, transportation, or electricity supply.) The difference between the modified and BAU GDP levels gives us the change in GDP, as shown below:

change in GDP

Next, we calculate the change in pollutant emissions due to the change in GDP. Ordinarily, the model calculates pollutant changes in a detailed, sector-by-sector manner. However, we do not know which sectors are responsible for the observed change in GDP, so we simply use economy-wide metrics here.

First, we take in a BAU economy-wide CO2 emissions intensity, and we use the BAU case’s ratio of other pollutants to CO2 to establish the economy-wide emissions intensities of those pollutants with respect to GDP. We also look at the difference in pollutants between the policy and BAU cases to find the percentage by which emissions of each pollutant were reduced by the policy package. This is used to reduce the economy-wide emissions intensities. Multiplying the change in GDP by the reduced emissions intensities of each pollutant gives the change in emissions of each pollutant due to GDP changes, in the policy scenario. See the structure below for details:

change in emissions due to change in GDP

We add the change in emissions to the total pollutant emissions to obtain the “Total Pollutant Emissions after GDP Adjustment.” We also report this output converted into CO2e terms (by multiplying by GWP values of the user-selected timeframe), and we divide emissions by GDP to obtain the “Total CO2e Emissions per Unit GDP after GDP Adjustment.” The structure is shown below:

total emissions after GDP adjustment

We do not use the GDP-adjusted variables in other model equations, nor do we include them on the Web Application Support Variables tab. This is because, as noted above, we are not accounting for all of the policy package’s effects on GDP when determining the change in GDP. We are only looking at the effect of changes in fuel prices on GDP. Therefore, the outputs such as change in pollutants due to changes in GDP are similarly incomplete. This structure primarily exists in the model to be ready to support future model extensions that might refine the “change in GDP” calculation to include more policy effects.

Social Benefits

The model estimates monetized social benefits of two types: avoided climate damages and avoided mortality (early deaths) due to particulate air pollution. On the climate side, we take the difference in CO2e emissions between the BAU and policy cases, and we multiply it by the Social Cost of Carbon, the U.S. government’s projections of the value to society in avoiding a given quantity of CO2e emissions. The structure is shown below:

monetized benefits from avoided climate damages

To estimate benefits from reduced mortality, we begin by finding the difference between emissions (disaggregated by pollutant) between the policy and BAU cases. We then multiply by a “Social Cost of Health Impacts by Pollutant,” which is from input data. The input data combine population-weighted exposure levels, an epidemiological concentration-response function, and a value of a statistical life (VSL) figure to determine the mortality-related cost of the different types of pollutants that contribute to particulate concentrations in the atmosphere. We also use a VSL figure to display the particulate mortality benefits in the form of lives saved. (This lives saved figure includes only avoided mortality from reduced parituclate pollution, not avoided mortality from reduced climate damages.) The relevant structure appears in the following screenshot:

monetized benefits from avoided mortality due to particulates

Finally, we sum the two types of social benefit, when represented in economic terms (rather than lives saved). We also calculate the social benefit per unit emissions abatement caused by the policy package. The structure is shown below:

total social benefits of emissions reductions

Some users might want to know the amount of emissions from the production of energy (in any form, such as heat, electricity, light, or force). All emissions in the Energy Policy Simulator (EPS) are “energy-related” except industrial sector process emissions and anthropogenic LULUCF emissions. We include the effects of carbon sequestration, so sequestered CO2 is removed from the total emissions reported here.

We calculate energy-related emissions for both the BAU and policy cases. From “Total Pollutant Emissions,” we subtract anthropogenic LULUCF emissions and Industry sector process emissions. We then convert this to CO2e using the user-specified GWP factor. A screenshot of the structure for the policy case is shown below:

energy-related emissions excluding leakage

Miscellaneous Additional Outputs

This section contains a variety of outputs that may be of use for scenario analysis. These outputs need little calculation structure (or in some cases, they are simply shadow variables from elsewhere), so they do not merit their own sections on the “Additional Outputs” sheet.

Electricity capacity divided up by type (coal, wind, etc.) but not by plant quality level is provided here. Similarly, we provide electricity generation (i.e. output) by type but grouped across quality levels:

electricity capacity and generation by type

We total up the amount of primary energy from non-fossil sources, and we also express this as a percentage of total primary energy:

non-fossil primary energy

We also report the fuel use for each fuel-using sector (transportation, electricity, buildings, industry, and district heat) by fuel:

fuel use by sector

There is also a “Zero Line for Graphs” variable in this section. Its value is always zero. It exists so that a line along the X-axis can be drawn on certain graphs in Vensim, which improves their appearance. It has nothing to do with graphs that appear in the web application interface.

Testing for Clean Power Plan Compliance

In the United States, the EPA released a “Clean Power Plan” that specifies particular emissions targets that must be met by the electricity sector. The plan was released after the release of our BAU input data, and so compliance with the plan is not part of the BAU case. This structure provides a simple, quick way to test whether a user-generated scenario (in the policy case) meets the Clean Power Plan emissions targets.

The plan has four periods during which targets are specified: cumulative emissions from 2022-2024, from 2025-2027, from 2028-2029, and 2030. We sum up the electricity sector emissions from each of these periods and compare them with the targets, which are specified in the “Clean Power Plan Compliance Schedule” input variable, to determine whether the targets were met. If all four targets are met, the “Clean Power Plan Compliance” variable has a value of 1 in 2030; otherwise, it has a value of zero in 2030. The relevant structure is shown below:

testing for clean power plan compliance