OVERVIEW

Every Industry needs to purchase goods and services (Commodities) from other Industries (and even from itself!) to operate their business. Formerly known in IMPLAN as Intermediate Expenditures, Intermediate Inputs are purchases of non-durable goods and services such as energy, materials, and purchased services that are used for the production of other goods and services, rather than for final consumption. These are known as Industry Spending Patterns.

IMPLAN has developed regionalized Industry Spending Patterns by taking national-level spending patterns and combining it with regional industry-level data and Trade flow data. These regionalized Industry Spending Patterns reflect the Industry demand that is sourced locally. This article defines a few key terms and describes the data sources and processes IMPLAN employs to obtain regionalized Industry Production Functions, as well as techniques for estimating more specialized Industry Spending Patterns for Electrical Power Industries.

KEY TERMS

GROSS INPUTS

The total dollars spent on a given Commodity by an Industry are known as Gross Inputs. Total Gross Inputs, therefore, represent the total dollar value of all Intermediate Inputs purchased by the Industry. 

This data can also be described using Gross Absorption Coefficients which represent the percentage of each Industry’s Output that is spent on each given Commodity (good or service) as part of its annual operations. 

REGIONAL INPUTS

Intermediate Inputs cannot all be sourced from within the region; some Intermediate Input purchases are imported from other regions (as captured by the Regional Purchase Coefficient (RPC). The total dollar amount spent locally on a given Commodity is known as Total Regional Inputs and is equal to the Gross Inputs multiplied by the RPC.  

Regional Absorption Coefficients reflect the proportion of Intermediate Inputs expenditures that are purchased locally. Regional Absorption Coefficients are calculated as Gross Absorption multiplied by the RPC. 

DATA SOURCES

The Bureau of Economic Analysis’ (BEA) Benchmark I-O Tables (which are released every five years), are the primary source for IMPLAN’s Industry scheme and Intermediate Input data. However, IMPLAN has developed techniques to maintain a larger list of Industries than currently available from BEA data (as described in Industry Scheme) as well as more specialized Industry Spending Patterns for Electrical Power Industries, described below. 

As mentioned previously, these national-level production functions are combined with sub-national data on Value Added and total Output by industry. Read more information about the data sources and estimation processes for this data in Value Added by Industry Data and Industry Output Data. IMPLAN also incorporates Trade data. Read more about how this is done in Estimating Trade Flows.  

UPDATING AND REGIONALIZING NATIONAL INDUSTRY SPENDING PATTERNS

While IMPLAN's Industry Spending Patterns are based on the BEA Benchmark I-O Tables (augmented for greater industry detail as described in Industry Scheme), the Gross Absorption Coefficient values for all regions change year to year as they are forced to conform to new Industry Output and Value Added totals and new commodity supply, demand, and trade values each year, as follows:

  1. We start with the latest BEA Benchmark-based Gross Absorption values.
  2. We derive current Industry Output and Value Added, which dictate total Intermediate Inputs (II) for each industry. We adjust the BEA Benchmark-based Gross Absorption values upward or downward proportionately to conform to the new total II values for each industry.
  3. To this we add Final Demand for each Commodity to arrive at Total Commodity Demand.
  4. Using the Benchmark Byproducts Matrix and current Industry Output, we derive current Industry Commodity Supply.  
  5. To this we add Institutional Sales and Imports to obtain Total Commodity Supply by Commodity.
  6. The columns of the USE matrix (industry purchases of commodities) were forced to sum to Industry Output less Value Added (column controls) in Step 2. Now, the rows of the current USE matrix are forced to sum to Total Commodity Supply less Final Demand (row control totals). This will be the first time the gross absorption coefficients will change non-proportionately between the commodities.
  7. This column and row balancing is repeated until no further adjustment is needed. This method is known as a RAS.

Gross Absorption Coefficients for a particular Industry will also vary across regions because the ratio of VA to Output varies from region to region, which forces the national Gross Absorption Coefficients to adjust (so that the sum of an Industry's Gross Absorption Coefficients plus that Industry's total VA = 1.0). The assumption is that the local Output and VA data are correct, and the national coefficients need to adjust to fit the local situation. 

Finally, applying the Regional Purchase Coefficients (RPCs) from the Trade Flow data to the Gross Absorption Coefficients yields Regional Absorption Coefficients by pulling out the imports (which also vary by region). Regional Absorption Coefficients, therefore, represent the percentage of each Industry’s Output that is spent on local production of each given Commodity as part of its annual operations, and are affected by new data on Foreign Trade obtained annually.  

ELECTRICAL POWER INDUSTRY SPENDING PATTERNS

While the Bureau of Economic Analysis’ Benchmark I-O Tables are IMPLAN’s primary source of Industry Spending Pattern data, there is growing need to provide more detail behind the generation and sale of electricity in ways that help businesses, analysts, and policy makers understand the change in both the technologies and economics that underpin the consumption of electricity. For that reason, IMPLAN Group, LLC contracted with Skip Laitner of Economic and Human Dimensions Research Associates, both in 2014 and 2019, coinciding with releases of the 2007 and 2012 Bureau of Economic Analysis (BEA) Benchmark Input-Output tables, to break out the BEA Benchmark’s single aggregate “Electric power generation, transmission, and distribution” Industry into eight electrical power generation Industries and one electric power transmission and distribution Industry. In 2024 – coinciding with the release of the 2017 BEA Benchmark I-O tables, IMPLAN did this work once more to develop nine separate Industry Spending Patterns (vectors of coefficients representing each Industry’s operation expenditures per dollar of Industry Output): 

2022 NAICS Code   Description 

221111                    Electric power generation, hydroelectric
221112                    Electric power generation, fossil fuel (e.g., coal, oil, gas)
221113                    Electric power generation, nuclear
221114                    Electric power generation, solar
221115                    Electric power generation, wind
221116                    Power generation, geothermal
221117                    Power generation, biomass
221118                    Electric power generation, tidal (other)
2212                       Electric power transmission, distribution (and administration) 

The Industry Spending Patterns consist of coefficients based on the detail from the latest BEA Benchmark USE table for the commodities being purchased and include coefficients for 3 components of Value Added: Employee Compensation, taxes on production (TOPI), and gross operating surplus (GOS). These are then adjusted to meet the new IMPLAN Industry scheme. The aggregate expenditures are benchmarked to published accounts of the U.S. Energy Information Administration (EIA) for the year 2022. This paper covers the major steps necessary to generate the requisite production functions regardless of whether privately-owned or government-owned utilities, using several data sources. 

STEPS IN THE DEVELOPMENT OF THE NINE PRODUCTION COEFFICIENTS 

Due to large divergences across the various data that are available on energy production, it is first necessary to converge to a reasonable pattern of nine vectors of production coefficients in a series of steps that are described below. 

Step 1. Converging to Aggregate Categories of Expenditures

The first area of focus is converging Intermediate Inputs (II) and Value Added (VA) expenditures as they are broken out according to the aggregate set of costs within the eight categories of generation technologies, and also to the aggregate of Transmission, Distribution, and Administration (TDA) costs. Ultimately, the goal is to have all costs sum to the Revenue from Sales of Electricity to Ultimate Customers from the Energy Information Administration, our proxy for Output. 

We first scale the various expenditures from EIA table 8.3 – Revenue and Expenses for Investor Owned Utilities to the Total Gross Output (TGO) accounting from table 2.3 - Revenue from Sales of Electricity to Ultimate Customers. This gives us an indicative, but an initial, allocation of total electricity expenditures. Next, we focus on the initial break out of the generation versus TDA expenditures as suggested by the EIA’s Annual Energy Outlook. This provides an overall aggregation while the next step gives us a calculated “Production/TDA” split to help us allocate: (a) purchased power expenditures (essentially power generated by independent producers sold wholesale to electric utilities); (b) other categories of intermediate inputs; and (c) taxes paid by the various producers. This is true for both generation and TDA. 

At the same time, we rely on the KLEMS data to suggest a labor compensation proportion of total expenditures (part of the VA component). We allocate initial expenditures across two columns (one for aggregate generation costs and the other for aggregate transmission and generation costs) and six rows of II and VA. The Intermediate Inputs include (i) the cost of fuel, (ii) purchased power, and (iii) all other Intermediate Inputs. The VA categories include (iv) compensation of labor, (v) taxes, and (vi) GOS.  

Step 2. Allocating Generation Costs Among the Eight Technologies

With a working estimate of generation costs, we can now break out technology costs. EIA, National Renewable Energy Laboratory (NREL), and International Energy Agency (IEA) data provide the data by which we can calculate a cost of energy purchased by each industry, which is then used to split out the aggregate energy cost amongst applicable industries. Employee Compensation splits are based on Employee Compensation from the Bureau of Labor Statistics’ Census of Employment and Wages (CEW) data. EIA table 1.1 – Total Electric Power Industry Summary Statistics provides energy generation by type of electricity production, which is used as a distributor for all other categories of Value Added and Intermediate Inputs. 

Step 3. Allocation to the 546 IMPLAN Commodities

At this point, a base Industry Production Function for each electricity production industry is available with the following categories: Fuel Costs, Purchased Power, Other, Compensation, Taxes, and Surplus. We can now turn to the BEA Use Table for the most recent Benchmark and the IMPLAN bridge from the BEA Benchmark scheme to IMPLAN’s current industry scheme. 

Three fuel categories – coal, other fossil fuels, and nuclear – are allocated directly to key Industries within the IMPLAN structure. That is, coal is directly allocated to the IMPLAN coal mining Industry, the other fossil fuels are allocated to the IMPLAN oil and gas extraction Industry, and nuclear fuels are allocated to IMPLAN’s uranium-radium-vanadium ore mining Industry. At the same time, purchased power expenses are directly allocated to the IMPLAN’s electric power transmission and distribution Industry. All other Intermediate Inputs are allocated based on the matrix of expenditures from the BEA patterns as adjusted to the IMPLAN industry scheme. 

At the same time, the VA expenditures are allocated in a straightforward manner to ensure that labor and tax allocation are consistent with the KLEMS data described in Step 1. The GOS (consisting largely of profit and depreciation) provides the balance of allocation as a function of all remaining expenditures.  

Finally, IMPLAN adjusted the transmission and distribution Industry’s purchase of electricity to equal total Commodity Output of all electricity Commodities. We also removed the purchase of turbines by the transmission and distribution industry, and moved a portion of the transmission and distribution industry’s purchases of rail transportation and pipeline transportation to the fossil fuels generation industry.  

RELATED ARTICLES

Industry Output Data

Value Added by Industry Data

The RAS Method

Utility Purchases & Energy Rebates