BENEFITS OF IMPLAN TIME SERIES DATA
USE OF REVISED RAW DATA
Some government data sources are revised multiple times after the annual IMPLAN data process has been completed and the data have been released to IMPLAN users. We use the revised raw data when estimating the IMPLAN Time Series Data.
USE OF NON-LAGGED RAW DATA
When producing the annual IMPLAN data sets, a number of our raw data sources are lagged relative to the IMPLAN data year. This is not the case when going back and re-estimating past years’ data, so no projections are needed – we are able to use the raw data that matches the year of the resulting IMPLAN data.
CONSISTENT ESTIMATION METHODOLOGIES ACROSS YEARS
Re-estimating past years’ IMPLAN data allows us to incorporate all data and methodology improvements and changes that have been adopted over the past 5 years since the previous set of Time Series Data was released. The more significant of these improvements are detailed in a later section of this document. All other methodological changes that have been incorporated can be found in the U.S. annual data release notes.
CONSISTENT SECTORING SCHEME AND HOUSEHOLD INCOME CATEGORIES ACROSS YEARS
All years of the Time Series Data are in the latest IMPLAN sectoring scheme, which changes every 5 years. Similarly, all years of the Time Series Data will use the latest Household Income Groups available in IMPLAN. This makes the IMPLAN data suitable for statistical analysis purposes.
REVISED RAW DATA
Revised raw data that were incorporated into the latest Time Series Data estimation process include:
- The Census Bureau’s Annual Survey of Manufactures (ASM) data, used for Output and Inventory values for the manufacturing sectors.
- All data sets from the Bureau of Economic Analysis (BEA), including:
- BEA Regional Economic Accounts (REA) data, which are used for estimating Employee Compensation (EC), Proprietor Employment, and Proprietor Income (PI), as well for adjusting for undercoverage of some sectors by the Bureau of Labor Statistics’ (BLS) Census of Employment and Wages (CEW) data series, as discussed in this article.
- BEA Industry Output Series, used to estimate Output for most service sectors.
- BEA National Income and Product Accounts (NIPA) tables. The NIPA tables are at the national level only and do not contain any Industry or commodity detail. They serve as overall control totals (e.g., total proprietor employment, total GDP, total government spending, etc.) for nearly all IMPLAN data points, which do have geographic, industry, and commodity detail.
- BEA State GDP Series, which we use for Other Property Income (OPI), Taxes on Production and Imports (TOPI), and Subsidies by industry and state.
- BEA Personal Consumption Expenditure data, which are used to estimate household spending on goods and services.
SOME NOTABLE METHODOLOGICAL CHANGES INCORPORATED INTO THE LATEST TIME SERIES DATA
NO LONGER ESTIMATING NON-DISCLOSED COUNTY BUSINESS PATTERNS (CBP) VALUES
Beginning with their 2017 data, the Census Bureau has adopted a disclosure policy under which they no longer provide establishment counts for cases in which the establishment count is less than three, as the number of establishments is now considered sensitive. This omission from their tables of all records with fewer than three establishments was set in place to protect the confidentiality of businesses.
This practice makes a missing record of 1 or 2 establishments (an existing industry) indistinguishable from a missing record of 0 establishments (a non-existent industry). The omission of records with fewer than 3 establishments, in addition to making it impossible to estimate the employment and income values for those records, also makes it impossible to obtain high quality estimates for the non-disclosed records in the table (i.e., those with 3 or more establishments) since we are not able to roll up across NAICS levels without having values for sibling sectors.
Therefore, we no longer generate estimates for non-disclosed County Business Patterns (CBP) data; rather, we only use the disclosed values. For more information on our processes for disclosing BLS CEW data, see this article.
NO LONGER USING CBP BY LEGAL FORM OF ORGANIZATION DATA
Due to the Census Bureau’s new practice of omitting records with fewer than 3 establishments (described in more detail above) and the fact that state-level data by industry and legal form of organization often have records with fewer than 3 establishments, these data are not sufficiently reliable in their current form. Therefore, we have returned to the previous method of relying solely on BLS CEW data for the distribution of BEA REA proprietor data among the IMPLAN sectors that map to each more aggregate BEA REA sector.
INCORPORATION OF THE LATEST BEA BENCHMARK I-O TABLES
Every 5 years, the BEA releases a comprehensive revision and update to its Benchmark Input-Output (I-O) tables. The release of new Benchmark tables provides many important updates to the base data used for many data elements in IMPLAN’s annual U.S.-based models, as described briefly in this article and this webinar recording.
COMPARING THE 528 INDUSTRY SET DATA TO THE 546 INDUSTRY SET DATA
In the production of Time Series Data 2001-2022 in the 528 scheme, the Gravity Model produced an increased number of trade records and increased amount of domestic trade overall between states and counties. No changes have been made to our Gravity Model process compared to the 2023 528 scheme data and the 546 scheme data. IMPLAN's Data Team produced an additional process to ensure the validity of the trade data that corroborated the Gravity Model’s outputs. Increased trade between Regions leads to increased leakages in a single region analysis, causing a reduction in results compared to our existing data sets.
Written July 17, 2025