OVERVIEW
There are several Industries that are not fully covered by the Bureau of Labor Statistics’ Census of Employment and Wage (CEW) data and thus require different data sources and techniques.
These Special Industries include the following:
- Farm Industries
- Construction Industries
- Railroad Transportation
- Government
Using different data sources is one of two approaches that IMPLAN employs to account for exclusions and undercoverage by the CEW. Read more about this in Accounting for Non-Disclosures, Exclusions, and Undercoverage in Employment Data.
FARM INDUSTRIES
Farm Industries are unique from other Industries in IMPLAN as they often have different data sources and data processing techniques to capture the full economic activity in that Industry group. For example, while Total Industry Output (TIO) for most Industries is only available at the national level, state-level data from the USDA exist for Farm Industry TIO. On the other hand, the Bureau of Labor Statistics' (BLS) Census of Employment and Wages (CEW) data (our primary data source for most industries’ Wage and Salary Employment and Income) only has about 90% coverage for Farm Industries; thus, other data sources are used for these data elements. This section describes the data sources and estimation methods for Farm Industries.
BEA BENCHMARK I-O DATA
The BEA Benchmark Input-Output (I-O) tables, which are updated every 5 years, provide us with production functions (industry spending patterns) for each of IMPLAN’s 14 farm sectors. They also provide Output, Employee Compensation (EC), and Gross Operating Surplus (GOS) for each farm sector. However, the BEA Benchmark tables do not separate GOS into its component parts (Proprietor Income (PI) and Other Property Income (OPI)), nor do they provide Employment estimates. Thus, we combine the latest BEA Benchmark data with other data sources from the same year to calculate ratios that are used in the annual data production process, as described below.
Supplementary Data
BEA REA Data
From the Bureau of Economic Analysis’ (BEA) Regional Economic Accounts (REA) data series, we obtain national Benchmark-year values of Employee Compensation (EC) and Proprietor Income (PI). These data do not have the specific farm sectors broken out; rather, they are for the aggregate “Farm Total” sector.
BLS CEW Data
From the Bureau of Labor Statistics’ (BLS) Census of Employment and Wages (CEW) data series, we obtain national Benchmark-year values of Wage and Salary Employment and Income by IMPLAN farm sector.
USDA Census of Agriculture
From the U.S. Department of Agriculture’s (USDA) Benchmark-year Census of Agriculture, we obtain national farm counts (as a proxy for the number of proprietors) by IMPLAN farm sector.
Ratio Calculations
Every 5 years, with the release of a new set of BEA Benchmark tables, these four data sources (the BEA Benchmark I-O data, BEA REA data, BLS CEW data, and USDA Census of Agriculture data) are used in conjunction to calculate the following ratios for each of the 14 IMPLAN farm sectors:
- Wage and Salary Employment per Output
- Proprietors per Output
- EC per Wage and Salary Employment
- Proprietor Income per Proprietor
These ratios, inflated to the current IMPLAN data year1, are then used in the annual data production process, as described below.
OUTPUT
USDA ERS and NASS Data
Each year, we obtain estimates of state-level farm sector output from the USDA’s Economic Research Service (ERS) and National Agricultural Statistics Service (NASS). These data are current, empirical (i.e., based on observation and survey, rather than trend extrapolations), and at the necessary level of sector detail, so all other data elements for the farm sectors are based off of these output data.
Census of Agriculture Data
For some commodities, ERS and NASS do not maintain consistent definitions year over year. In such cases, we project values from the latest Census of Agriculture to estimate by-commodity output values.
Distribution to Counties
These state values are distributed to counties based on the ratio of county production to state production from the latest Census of Agriculture. For example, if County A has 10% of the state’s grain sales (or acreage if sales is not available, and farms if acreage is not available), then it gets 10% of the state’s output value for grain farming.
EMPLOYMENT AND LABOR INCOME
While the BLS CEW data are current and have the necessary sector detail, they omit many small farms and all proprietors, and thus only provide roughly 90% coverage of all farm employment. And while the BEA REA data provide full coverage of all employees and their labor income, the BEA REA data do not have the necessary sector detail nor proprietor employment estimates – they have a single “Farm Total” sector, and no proprietor data.
Thus, we use the following methodology to obtain annual estimates of farm employment and labor income:
- Calculate first estimates of Wage and Salary Employment for each place and farm sector by multiplying that place and farm sector’s annual Output by the Benchmark-based ratio of Wage and Salary Employment per dollar of Output.
- Calculate national first estimates of Proprietor Employment by using a combination of total number of farms from NASS, various farm operation characteristics from the U.S. Census of Agriculture and ARMS, and NASS agricultural prices. Then, use IRS or CEW data (depending on the timeliness of each) to project the prior year's estimates of Proprietor Employment for the nation and states. State values are distributed to counties using various county-specific data elements.
- Calculate first estimates of EC and Proprietor Income for each place and farm sector by multiplying that place and farm sector’s annual Output by the Benchmark-based ratios of EC per Wage and Salary Employment and Proprietor Income per Proprietor.
- As with all IMPLAN data elements, all of the afore described data elements are geographically controlled so that the sum of all state-level values match their corresponding U.S. values, and all county-level values sum to their corresponding state-level values.
OTHER VALUE ADDED
The BEA releases state-level GDP data at the aggregate “Farm Total” level, that includes the break-out of GDP into EC, TOPI, and GOS. GOS consists of Proprietor Income (PI) and Other Property Income (OPI), so OPI is derived by subtracting our estimates of PI (described above) from GOS. These “Total Farm” control values are distributed to the IMPLAN sectors based on the latest BEA Benchmark’s characteristics for GOS and TOPI, and by using data from the USDA ERS Agriculture Resource Management Survey (ARMS), where available, as described next.
ARMS Data
The USDA ERS Agriculture Resource Management Survey (ARMS) reports some sub-components of the Value-Added components for certain commodities and certain states. We incorporate the ARMS data to help distribute the BEA’s Value-Added data (which are only at the “Total Farm” level) to the IMPLAN farm sectors.
The ARMS data are typically lagged one year relative to the IMPLAN Data Year, in which case we project them (based on farm sector Output). While the ARMS data are only used as distributors among the IMPLAN farm sectors, such projection is important to prevent a farm sector from seeing a significant drop in Output while receiving a large share of TOPI or OPI based on prior-year un-projected distributions. The ARMS data are not available for all states and commodities. For the states and commodities for which there are no ARMS data, we use Benchmark–based distributions. We use the following data elements from ARMS:
- Government payments, i.e., subsidies (a negative component of net TOPI)
- Real estate and other property taxes (a large component of gross TOPI)
- Interest (a component of OPI)
- Depreciation (a component of OPI)
Distribution to Counties
State-level OPI and TOPI estimates are distributed to counties based on EC and Employment, respectively.
NOTE
Because the farm sector data are largely derived, using lagged and/or more-aggregate data, and because farm sector Output can fluctuate in other ways not associated with labor (e.g., weather conditions), analysts with local agriculture data are encouraged to use their local data when setting up analyses in IMPLAN.
CONSTRUCTION INDUSTRIES
DEFINITIONAL NOTES
IMPLAN construction sectors are not NAICS-based, but rather are defined by the Census Bureau’s structure types. Because new construction (in contrast to maintenance and repair construction) is an investment expenditure, there is no intermediate demand for new construction – that is, new construction is not the part of any industry’s production function. Rather, all Output of the new construction industries goes to meet Final Demand.
OUTPUT
National Output values for the construction Industries come from the BEA’s Industry Output Series, which is the same source used for most IMPLAN service sectors, as described in Industry Output Data. These national values are distributed to states and counties based on Employment and Value-Added. For the new construction sectors, additional adjustments are made to balance Output with Final Demand.
EMPLOYMENT & INCOME
Because the BLS CEW data for the construction sectors are NAICS-based activities (e.g., Land subdivision, Drywall and Insulation Contractors, Flooring contractors) that are not unique to a single structure type and therefore do not map to unique IMPLAN construction sectors (which are based on structure type, as noted above), we only make use of the 2-digit-level CEW data for total construction Wage and Salary Employment and Income. These values are split among IMPLAN's construction sectors using various ratios from the latest BEA Benchmark I-O tables and construction-sector Output estimates.
The Wage and Salary Income values are converted to Employee Compensation (EC) using ratios from the BEA’s REA data, which provide Employee Compensation, Wage and Salary Income, and Proprietor Income values at the national, state, and county level for a single aggregate construction sector. The BEA REA Proprietor Income values are distributed among the IMPLAN sectors based on the Employee Compensation data.
Proprietor Employment for the construction sectors is estimated the same as for most other sectors, as described in Employment and Labor Income Data.
RAILROAD
Because the CEW data exclude most employees of railroads, IMPLAN utilizes wage and salary employment data from the U.S. Railroad Retirement Board. Labor Income, Proprietor Employment, and Output for the Rail Transportation sector are estimated the same as for most other sectors, as described in Employment and Labor Income Data. We have methods to estimate wage and salary employment in cases where BEA REA show EC for the geography but the RRB data do not show any employment.
GOVERNMENT
ADMINISTRATIVE GOVERNMENT VERSUS GOVERNMENT ENTERPRISES
IMPLAN data includes several types of government activity. "Administrative" or "general" government is considered an Institution in IMPLAN. There are payroll-only Industries that are purchased exclusively by administrative government spending patterns. In contrast, government enterprises are Industries that have profiles similar to those of Industries. The Bureau of Economic Analysis (BEA) defines government enterprises as "Government agencies that cover a substantial portion of their operating costs by selling goods and services to the public and that maintain their own separate accounts." The BEA Benchmark I-O tables treat government enterprises as industries and IMPLAN follows this convention. IMPLAN has 8 government enterprise Industries. All remaining government agencies are part of "general" or "administrative" government.
ASSIGNING NAICS-BASED ESTABLISHMENTS TO IMPLAN CODES
Most data sources other than the BEA Benchmark table do not distinguish government enterprises from administrative government establishments. The BLS' CEW program, for example, reports employment and wages for government establishments by NAICS code and by ownership type (federal, state, or local). We assign government establishments to enterprise or administrative payroll Industries according to the BEA Benchmark table. If a government enterprise makes a particular commodity, then any government establishment that fits the NAICS code of that commodity is classified as a government enterprise. For example, in the latest (2012) BEA Benchmark table, "Electric power generation, transmission, and distribution," which corresponds to NAICS 2211, is made by private businesses in the electrical Industries but also by "State and local government electric utilities." Accordingly, any government establishment owned by state or local government under NAICS 2211 is classified in IMPLAN Industry 512 or 515, respectively. If the BEA Benchmark table does not show a government enterprise Industry making a Commodity, then any government establishment in that Commodity's corresponding NAICS code is classified as administrative government. Tables 1 through 3 show IMPLAN Industry assignments for establishments by ownership type and NAICS code. Note that military employment is not covered by any of our NAICS-based sources and is estimated from military employment data reported by the BEA.
Table 1: Federal Government
Table 2: State Government
Table 3: Local Government
EMPLOYMENT AND LABOR INCOME
The projected BEA REA data serve as our controls for Employee Compensation for 4 aggregate groupings of IMPLAN government sectors: Federal Government Non-Military, Federal Government Military, State Government, and Local Government. Federal Government Military is the only one of the four that maps one-to-one with an IMPLAN sector; the rest are distributed to multiple IMPLAN sectors based on the government-level BLS CEW data.
Because the BEA REA data no longer include any information on Employment, we rely on BLS CEW data (adjusted for undercoverage of administrative government) for most IMPLAN government sectors. However, because the BLS CEW data do not cover military at all, the adjustment method does not work (there are no CEW values to adjust); therefore, we turn to state-level employment data from the Department of Defense’s Defense Manpower Data Center, which are distributed to counties based on Employee Compensation from BEA.
RELATED ARTICLES
Employment and Labor Income Data Sources
Accounting for Non-Disclosures, Exclusions, and Undercoverage in Employment Data
1ERS and NASS occasionally change their reporting for aggregate categories between a) including the sum of non-disclosed values for subcategories in their aggregate category values and b) making the aggregate value simply the sum of disclosed children. We adjust our processes accordingly.
Written April 18, 2024
Updated October 21, 2025