Introduction
Starting in 2025, the Bureau of Economic Analysis (BEA) made the decision to discontinue the publication of numerous data tables from its Regional Economic Accounts (REA) program. Several of these discontinued tables had been important components of the annual core IMPLAN data sets:
-
State-Level
- SAEMP25 (Total full-time and part-time employment by industry)
- SAEMP27 (Full-time and part-time wage and salary employment by industry.)
-
County-Level
- CAEMP25 (Total full-time and part-time employment by industry)
- CAINC35 (Personal current transfer receipts)
In this article, we’ll discuss the importance and use case for each of these data tables, and the alternative data sources and methods now employed by IMPLAN to preserve the richness of data that can be found in IMPLAN models.
The loss of the employment tables (SAEMP25, SAEMP27, and CAEMP25) represents the most substantial problem for producing data that feeds into IMPLAN models because these BEA employment tables represented:
- the singular source of proprietor counts, at any geographic level and any level of industry detail, with no comparable data set that can be used as a direct replacement; and
- the only complete count of all wage and salary employees, without any of the exclusions that affect other employment data sources like the Bureau of Labor Statistics’ (BLS) Census of Employment and Wages (CEW) data, which exclude many employees, as described here.
Wage and Salary Employment
The BLS CEW data represents the most important source of employment data for IMPLAN due to its higher level of industry detail and greater recency relative to the BEA REA data. However, the BEA REA employment data played an important role in adjusting, augmenting, or replacing the CEW data for a number of industries to account for various exclusions.
For industries for which we use adjustment ratios based on BEA REA values relative to BLS CEW values, the only change is that those ratios can now only be calculated for wage and salary income – we no longer have distinct wage-based ratios and employment-based ratios, so income-based ratios are applied to CEW employment as well as CEW wages.
For Military employment, the BLS CEW data do not provide any coverage at all, such that an adjustment approach is not possible, and another data source is needed. Therefore, we now use state-level employment data from the Department of Defense’s Defense Manpower Data Center, which are distributed to counties based on employee compensation values from BEA REA.
For Rail transportation employment, the BLS CEW data provide very little coverage, such that an adjustment approach is not appropriate, and another data source is needed. We now use wage and salary employment data from the U.S. Railroad Retirement Board.
Proprietor Employment
The BEA’s proprietor employment estimates were the sum of three types of proprietors: sole proprietors, limited partners, and general partners. Proprietors play a large role in the U.S. economy, making up 25% of all employment in the U.S. in recent years, and making up the majority of employment in a number of industries, including several crop and animal production industries; commercial fishing; commercial hunting and trapping; commercial sports; commercial real estate; independent artists, writers, and performers; oil and gas extraction, and investment brokerages, among others. This underscores the importance of including proprietors in economic impact models.
Farm and non-farm proprietor employment values in IMPLAN are estimated separately, using distinct data sources and methods for each. The series of steps described in Accounting for Non-Disclosures, Exclusions, and Undercoverage in Employment Data is largely based on the BEA’s published methodology for generating farm proprietor employment. This process involves several new data tables not previously used by IMPLAN, including the NASS total number of farms; ARMS estimates of corporate and non-corporate farms; NASS agricultural prices; IRS Statistics of Income for Sole-Proprietorships by 3-digit NAICS subsector or 4-digit NAICS industry group level (State); IRS Statistics of Income for Partnerships by 3-digit NAICS subsector or 4-digit NAICS industry group level (State); and several data points from the Census of Agriculture that had not previously been used.
The new process for estimating non-farm proprietors involves the use of two data tables from the Internal Revenue Service’s (IRS) Statistics of Income (SOI) program: Schedule C data cover sole proprietorships, while K1 data cover limited partnerships. Because the IRS data exclude general partnerships, they are not used directly but rather are used to calculate growth rates that can be applied to previous-year full proprietor counts from IMPLAN. They also serve as minimums for non-farm proprietor employment.
The BEA proprietor income tables, which still exist, reveal cases where new proprietor employment should exist in a given industry and geography or where previously-existent proprietor employment should no longer exist in a given industry and geography.
As with all IMPLAN data, various geographic and industry hierarchy controls are imposed, and various limits and quality control checks are put into effect.
Early testing suggests that the methods discussed herein (and in further detail in the Data Sources and Estimation Methods section of IMPLAN’s support site) yield results similar to those that would have been provided by the BEA. Comparing historical data provided for proprietors with calculated proprietor counts using these methods allowed IMPLAN to refine its methodology with accuracy in mind. In almost all cases, estimates by state and by industry were within 10% of expected values.
Personal Current Transfer Receipts
Various data points from the discontinued BEA table CAINC35 were used to distribute various U.S. transfer payments received by households to states and counties. Including business transfers to households and government social benefits payments to households. IMPLAN now employs projections and alternative distributor data from BEA table CAINC30 (Economic profile).
Other Difficulties with BEA Data
As noted in the State and Local section on the BEA’s Discontinued Data Tables webpage, the BEA will not implement Connecticut’s transition to planning regions that took place in 2024 and will continue to report Connecticut county-level data at the outdated county level. This is in contrast to all other county-level data sources used by IMPLAN, thereby requiring the use of other data sources and techniques to distribute BEA state-level data for Connecticut to the new planning regions.
Conclusions
While the discontinuation of the BEA data tables discussed in this article have presented challenges for IMPLAN’s data production team, IMPLAN remains committed to providing the high quality regional economic data and modeling capabilities our users are accustomed to.
In accordance with our core value of stewardship, we will continue to test the quality of the IMPLAN data on an ongoing basis, each step along the data production process, and make refinements as warranted. Please contact us with any questions or comments.
Written October, 21, 2025