IMPLAN has three core values: community, respect, and stewardship. The goal of being good stewards has taken us beyond examining only the economic impacts associated with changes in production to examining some of the environmental impacts associated with current and projected levels of production. Working with the U.S. Environmental Protection Agency (EPA), IMPLAN has bridged their emissions data to our Industries, thereby giving you the ability to see the environmental impacts tied to your economic impacts!
IMPLAN’s environmental data consist of Industry-specific coefficients of physical emissions or resource use per dollar of Industry Output. IMPLAN’s environmental data consist of ratios representing physical emissions or inputs per dollar of Industry Output, with the physical unit depending on the particular pollutant or input under consideration. While the ratios are Industry-specific, the EPA data have somewhat less Industry-specificity than the IMPLAN data, such that some EPA ratios are mapped to more than one IMPLAN Industry; in other words, some IMPLAN Industries have the same environmental ratios. These ratios can be applied to the IMPLAN modeling system in order to gain insight into the environmental impacts associated with economic impact scenarios.
The coefficients do not double count in that they do not include the emissions associated with the upstream industries/suppliers. For example, the emissions associated with the automobile manufacturing Industry do not include emissions associated with the production of car parts or other inputs produced by other Industries; the emissions associated with producing those inputs would be accounted for in the coefficients for those suppliers’ respective Industries. Basically, the emissions associated with the tire production or the manufacturing of headlights are not included.
Any emissions generated directly by households for example, emissions generated by households’ burning of wood in a wood-fired stove in the home - are not included in the EPA dataset. Therefore, they are not accounted for in the IMPLAN emissions data set. What this means is that all induced emissions in IMPLAN stem from household purchases of energy and of goods and services, all of which require energy to produce and thus create emissions. Any additional emissions generated directly by the household, those not associated with a purchase of energy, goods, or services, will not be accounted for.
For IMPLAN’s environmental ratios, we use the EPA’s Environmentally-Extended Input-Output model (EEIO) data (version 1.1) as suggested to us by Sandy Dall’Erba of the University of Illinois. Wesley Ingwersen served as our primary contact at the EPA. As per correspondence with Ingwersen, the current data set (as of the 2020 calendar year) are from various years but are all adjusted to represent physical units per 2013 dollars, and are available only at the national level from the US EPA Office of Research and Development (ORD): USEEIOv1.1 - Satellite Tables. An accompanying description of the methodology used can be found in Yang, Ingwersen, Hawkins, Srocka, and Meyer, (2017). As of 2020, the EPA is working on updated data, which we plan to incorporate into subsequent iterations of the IMPLAN environmental data.
Ingwersen confirmed that the emissions coefficients (emissions per dollar of Output) do not include the emissions associated with upstream industries (i.e., input suppliers); that is, the emissions coefficients for the automobile manufacturing Industry, for example, do not include the emissions associated with producing the engines, spark plugs, etc., but rather only include the emissions associated with assembling the vehicles and painting them and all the other aspects of the automobile manufacturing Industry.
Below is the list of the eight major groups of environmental data (“satellite accounts”). Each satellite account has a number of more-detailed “tags” that represent more specific types of the broader group. Also listed are the numbers of tags in each group and a few examples of each. Please note that the tag counts for water use and land use are misleading, as those two categories include the sector name in the tag, so the true count of tags in those categories is lower than stated here.
The satellite accounts are:
- Criteria Pollutants
- 267 tags
- Examples: Nitrogen Oxides, Chromium III, Cellosolve Solvent
- Greenhouse Gas Emissions
- 15 tags
- Examples: CO2, N2O, CH4, CF4
- Land Use
- 220 tags
- Examples: Urban Land, Urban Transportation, Rural Transportation, Coal Exploration Licenses
- Mineral Use
- 5 tags
- Examples: Boron, Gypsum, Salt
- Nitrogen and Phosphorus Release to Water
- 2 tags
- Includes: N and P
- Pesticide Emissions
- 747 tags
- Examples: Spirodiclofen/air, Norflurazon/water, Fluxapyroxad/groundwater
- Toxic Chemical Releases
- 408 tags
- Examples: 1,2-Dichloropropane, 1,2-Phenylenediamine, 1,3-Butadiene
- Water Use
- 247 tags
- Examples: Alpacas; On-farm surface source, Alpacas; surface water, Aquaculture; groundwater
Industries will be missing a Value per dollar of Industry Output where the EPA value is zero or indistinguishable from zero in a given satellite.
Just as the quality of an economic impact study relies heavily upon the assumptions and methodologies employed by the analyst, so goes with the quality of the associated environmental impacts. That is, the better the estimates of the changes in Output, the better the estimates of environmental impacts associated with such changes in Output. IMPLAN does its very best to provide accurate regional economic models, just as the EPA does its very best to provide accurate Industry-specific environmental coefficients. It is upon the analyst to understand the IMPLAN data and Input-Output modeling framework, including its limitation, and to employ best practices when using the IMPLAN application, in which case the resulting associated environmental impacts and/or footprints can be taken seriously.
When applying emissions coefficients to IMPLAN economic impact results, the emissions will only be those generated in the model study area (or in areas linked via MRIO); they will not include emissions generated in other regions not linked via MRIO. For example, inputs that are purchased from outside the study area will generate emissions in the region where the inputs are produced; if those regions are not included in the study area or linked via MRIO, those emissions will not be accounted for in the analysis. Also, weather patterns that distribute emissions from one region to another are not accounted for.
Any emissions generated directly by households (i.e., not accounted for by household expenditures on electricity and not associated with the production of other items purchased by households; for example, emissions generated by households’ burning of wood in a wood-fired stove in the home) are not included in the EPA data set and are thus not accounted for in the IMPLAN emissions data set. What this means is that all induced emissions in IMPLAN stem from household purchases – of energy and of goods and services, all of which require energy to produce and thus create emissions. Any additional emissions generated directly by the household – i.e., not associated with a purchase of energy, goods, or services – will not be accounted for.
The data is also slightly unclear regarding water. It appears as its own satellite account, Water Use, and a category, water. In the data for water, you will see two categories: resource and water. When the category is a "resource" this represents water usage. When it is "water" it represents releases of water back to a water source like a lake. To get the net water usage, you will need to subtract one from the other.
Yang, Y., W.W. Ingwersen, T.R. Hawins, M. Srocka, and D.E. Meyer, 2017. USEEIO: A new and transparent United States environmentally extended input-output model. Journal of Cleaner Production, 158: 308-318.
Written November 20, 2019
Updated May 25, 2021