Improved Estimates of the Industry Economic Accounts

Results of the 2018 Comprehensive Update

Taylor M. Grant and Blaire Thomson contributed to the analysis in this article.

On November 1st, the Bureau of Economic Analysis (BEA) released estimates of real gross domestic product (GDP) by industry for the second quarter of 2018, revised annual estimates for 1997 through 2017, and revised quarterly estimates for the first quarter of 2005 through the first quarter of 2018. The revised estimates reflect the results of the 2018 comprehensive update to the Industry Economic Accounts (IEAs). These accounts provide statistics on interactions among industries and the roles these industries play in the economy, including each industry’s contribution to GDP.

Additionally, the release includes the new 2012 supply-use tables (SUTs) as well as updated 2007 benchmark SUTs. The benchmark SUTs provide a detailed picture of the economy, showing relationships among approximately 405 industries and commodities. They serve as the statistical foundation for other BEA estimates, as they are fully integrated with the annual industry accounts and the National Income and Product Accounts (NIPAs). The full time series of annual SUTs, providing information on 71 industry categories, has been updated to be consistent with the benchmark tables.

The statistics show that 16 of 22 major industry groups contributed to an overall 4.2 percent increase in real GDP in the second quarter. Economic growth was led by information; real estate and rental and leasing; and professional, scientific, and technical services. Together, these industries contributed 2.08 percentage points to overall growth (chart 1).

Estimates of real gross output and real intermediate inputs are available for both annual and quarterly IEA statistics. Annual statistics are available for 1947 forward.1 In addition, annual statistics for 138 industries are now available as part of the underlying detail for the IEAs for 1997 forward. Quarterly statistics are now available at the expanded 71 industry publication level for the first quarter of 2005 forward.

Table 1 shows the percent changes in real value added in the second quarter of 2018 by major industry group.

  • Real GDP increased at a 4.2 percent annualized rate in the second quarter of 2018. Overall, 16 of 22 major industry groups contributed to GDP growth.
  • Information, the leading contributor to GDP growth, increased 13.4 percent in the second quarter after increasing 4.3 percent. The second-quarter increase was driven by a 30.3 percent increase in data processing, internet publishing, and other information services.
  • Real estate and rental and leasing, the second leading contributor to GDP growth, increased 5.3 percent after increasing 2.7 percent. This is the largest growth rate since the third quarter of 2013. This growth primarily reflected an 18.0 percent increase in other real estate.
  • Professional, scientific, and technical services increased 9.3 percent, the largest increase since the third quarter of 2014, after increasing 6.0 percent. The increase was led by a 9.2 percent increase in miscellaneous professional, scientific, and technical services.
  • Mining increased 11.7 percent after decreasing 18.0 percent. This growth was mainly driven by an 8.0 percent increase in oil and gas extraction.
  • Health care and social assistance increased 4.7 percent after increasing 4.6 percent, primarily reflecting a 7.0 percent increase in ambulatory health care services.
  • Durable-goods manufacturing increased 7.3 percent after decreasing 4.7 percent, the strongest growth rate since the third quarter of 2014. The increase primarily reflected an 18.4 percent increase in motor vehicles, bodies and trailers, and parts manufacturing.
  • Nondurable goods decreased 3.7 percent after increasing 12.1 percent. The second-quarter decrease was attributed to a 36.1 percent decrease in petroleum and coal products, which was partially offset by a 7.1 percent increase in chemical products.
  • Finance and insurance decreased 2.0 percent after decreasing 5.5 percent. The decrease was driven by a 15.0 percent decrease in securities, commodity contracts, and investments.

Table 2 shows each industry’s contribution to the real GDP growth in the second quarter of 2018.

  • U.S. economic growth accelerated in the second quarter of 2018, increasing 4.2 percent after increasing 2.2 percent in the first quarter. Overall, 17 out of 22 major industry groups contributed to the acceleration in economic activity.
  • Mining was the leading contributor to the acceleration in real GDP growth, contributing 0.17 percentage point to GDP growth after subtracting 0.30 percentage point in the first quarter. The acceleration was driven by oil and gas extraction.
  • Information was the second-largest contributor to the acceleration in real GDP growth, contributing 0.70 percentage point after contributing 0.23 percentage point. The second-quarter acceleration was driven primarily by data processing, internet publishing, and other information services.
  • Utilities turned up, contributing 0.28 percentage point after subtracting 0.12 percentage point.
  • Real estate and rental and leasing accelerated, contributing 0.70 percentage point after contributing 0.36 percentage point. The acceleration was driven by an upturn in other real estate.
  • Nondurable goods, which turned down, was the largest offsetting industry group to GDP growth, subtracting 0.19 percentage point after contributing 0.58 percentage point. A downturn in petroleum and coal products drove this result.
  • Retail trade turned down, subtracting 0.06 percentage point to GDP growth after contributing 0.41 percentage point. This was led by a deceleration in other retail and a downturn in motor vehicle and parts dealers.

Comprehensive updates, which occur approximately every 5 years, differ from annual updates in the scope of improvements and in the number of years subject to revision. They provide opportunities for BEA to make changes to its economic accounts to increase the overall quality, relevance, and accuracy of the accounts, while better reflecting the evolving U.S. economy.

The 2018 comprehensive update introduced three major types of improvements: (1) integration with the 2012 benchmark SUTs and the 2018 comprehensive update of the NIPAs, (2) changes in definitions and classifications, which update the accounts to more accurately portray the dynamic U.S. economy and to better facilitate comparisons with economic data available from other countries, and (3) statistical changes, which update the accounts through the use of new and improved estimation methods and newly available and revised source data, including the Economic Census, which is used to benchmark the accounts. Combined, these improvements enable the accounts to continue to accurately measure the structure of the U.S. economy. Major additions and changes that were introduced in the 2018 comprehensive update include the following:

  • Incorporation of the results of the 2018 comprehensive update of the NIPAs.2
  • A shift in emphasis toward SUTs consistent with international recommendations from the 2008 System of National Accounts (SNA 2008) and away from the current make-use framework.3
  • Release of the new 2012 detailed benchmark SUTs.
  • Release of updated 2007 detailed benchmark SUTs consistent with the full time series of annual tables as well as the new 2012 benchmark tables.4
  • Updated industry and commodity definitions consistent with the 2012 North American Industry Classification System (NAICS).
  • Introduction of more detailed annual data on value added, gross output, and intermediate inputs at roughly the four-digit NAICS level of detail (138 industries) as part of the underlying detail for the IEAs. Currently, the most detailed annual data are published at roughly the three-digit NAICS level (71 industries).

The benchmark SUTs

A major highlight of the 2018 comprehensive update to the IEAs is the release of the 2012 benchmark SUTs. Benchmark tables provide a detailed picture of the economy, showing relationships among hundreds of industries and commodities. Benchmark tables also serve as the statistical foundation for other BEA estimates, including GDP. More specifically, the levels and commodity distributions of final-use categories are set as part of a reconciliation process between production-based data from the SUT framework and expenditure-based data from the NIPAs.

Beginning with this update, BEA's featured set of input-output tables will be presented in the supply-use framework, as recommended in the SNA 2008. With this change, U.S. data will be presented using valuations and a presentational format more comparable to international data. Although the supply-use format will be the featured set of input-output tables, BEA will continue to publish data in the current make-use format as supplementary tables.

Changes in definition

As part of the 2018 comprehensive update, several major changes in definition will be incorporated into both the IEAs and the NIPAs, reflecting ongoing work to further integrate these two sets of statistics. These changes include the following:

  • Reclassification of research and development (R&D) for software originals from own-account software (OAS) to own-account R&D
  • Recognition of capital services in own-account investment in software and R&D
  • Reclassification of “other” state and local personal current taxes as “other” taxes on production

In the supply table, industries that produce software investment saw a decrease in their secondary production of OAS and an offsetting increase in their production of own-account R&D, resulting in an overall increase in the domestic supply of R&D and a corresponding decrease in the domestic supply of software. For 1987–2001, the revisions for own-account software and R&D related to software originals offset one another. For 2002 forward, the revisions deviated from one another, reflecting slightly revised assumptions used in estimating own-account software. In 2012, own-account software was revised down $53.4 billion, while R&D was revised up $62.0 billion. The revisions each fall within private fixed investment (PFI), resulting in a net upward impact on total PFI and GDP.

The incorporation of capital services into own-account investment in software and R&D industries that produce these commodities resulted in an overall upward revision in production value of about $11.2 billion in 2012. Between 2012 and 2017, the revisions ranged from $11.2 to $14.4 billion. The resulting increase is absorbed in the use table by corresponding increases in nonresidential private fixed investment in intellectual property products, which resulted in upward revisions to both total PFI and GDP.

Finally, the reclassification of selected state and local payroll taxes from personal current taxes to taxes on production and imports led to offsetting $1.6 billion revisions in the use table in 2012, as this value was reassigned from gross operating surplus to other taxes on production. This change left total value added by industry unaffected.

Changes in classification

IEA statistics released as part of the 2018 comprehensive update are classified and presented on a 2012 NAICS basis; previously, the statistics were classified and presented on a 2007 NAICS basis. Overall, changes stemming from using the 2012 NAICS are small; the manufacturing, wholesale, and retail trade sectors have several changes, but none involved reclassification to another sector.5

With the release of the 2018 IEA comprehensive update, BEA published 405 industries in the 2012 benchmark SUTs, compared with 388 industries in the 2007 benchmark SUTs. Wholesale and retail trade were among the largest expansions. To provide a more detailed assignment of trade margins within the supply-use framework, secondary production of retail and wholesale commodities was broken-out to reflect the specific type of retail or wholesale commodity. This reallocation represents a significant improvement in the specificity and granularity of the trade margin data. For wholesale trade, BEA published 11 industries in 2012; in 2007, BEA only published a total wholesale trade aggregate. For retail trade, BEA published 9 industries in 2012, compared with 4 industries in 2007 (table 3).

Table 3. Effects of Expanding Industry Detail in the Industry Economic Accounts
2007 BEA codes Previous
publication level
2012 BEA codes New
publication level
HS Real estate HSO Owner-occupied housing services
HST Tenant-occupied housing services
524100 Insurance carriers 524113 Direct life insurance carriers
5241X Insurance carriers, except direct life
524200 Insurance agencies, brokerages, and related activities 524200 Insurance agencies, brokerages, and related activities
GSLG General state and local government services GSLGE State and local government educational services
GSLGH State and local government hospitals and health services
GSLGO State and local government other services
420000 Wholesale trade 423100 Motor vehicle and motor vehicle parts and supplies
423400 Professional and commercial equipment and supplies
423600 Household appliances and electrical and electronic goods
423800 Machinery, equipment, and supplies
423A00 Other durable goods merchant wholesalers
424200 Drugs and druggists’ sundries
424400 Grocery and related product wholesalers
424700 Petroleum and petroleum products
424A00 Other nondurable goods merchant wholesalers
425000 Wholesale electronic markets and agents and brokers
4200ID Customs duties
4A0000 Other retail 444000 Building material and garden equipment and supplies dealers
446000 Health and personal care stores
447000 Gasoline stations
448000 Clothing and clothing accessories stores
454000 Nonstore retailers
4B0000 All other retail

Statistical improvements

Statistical improvements are changes in estimation procedures to incorporate new and improved methods as well as newly available and revised source data. Several notable improvements in statistical methods were introduced with the release of the 2018 comprehensive update. These improvements include the following:

  • Reclassification of taxes
  • Introduction of new digital media product lines to the information sector

The updates in taxes improved the distinction between “taxes on products” (TOP) and “other taxes on production” (OTOP), which is more consistent with the recommendations of the SNA 2008 and enhances the compatibility of the NIPA measures of taxes on production and imports with the measures in the industry accounts. The changes had no aggregate effect on the estimates, as they represent reassignment of value rather than the introduction of new value to the SUTs. However, the updated distribution of taxes across different categories significantly enhances the accuracy of TOPs by commodity and OTOPs by industry.

The incorporation of new digital media products to information sector output estimates reflects new 2012 Economic Census revenue data that captures downloads of electronic media for permanent ownership. In addition, the accounts now capture retail trade sales statistics for “video content downloads,” the retail trade industry's secondary production of digital downloads. Although this new output represents a small upward revision (less than $1 billion) in 2012, BEA expects digital media content to show significant growth moving forward. With the rising importance of the digital economy, incorporating this improvement is a key step toward capturing this dynamic and rapidly changing sector of the economy with improved accuracy.6

Source data

The primary data source for the benchmark SUTs is the Economic Census, which the Census Bureau conducts every 5 years. The Economic Census is the preferred data source because it provides the most comprehensive data available in terms of industry coverage and captures activity in the relevant economic units for those industries. The Economic Census collects data at the level of the smallest operating unit, the “establishment,” and provides most of the essential data required for the tables, including inventories, receipts and expenses of business establishments and of government, sales by detailed industry and product line, final industry and product shipments, input costs by general category, and trade margins.

For 2014 through 2017, the updated estimates reflect the incorporation of newly available and revised source data, which are regularly included in the annual updates, and which became available after the annual update in November of 2017. These data include the following:

  • Annual Survey of State and Local Governments for fiscal year 2015 (revised) and 2016 (new)
  • Annual Survey of Manufactures for 2015 (revised) and 2016 (new)
  • Annual Survey of Wholesale Trade for 2015 (revised) and 2016 (new)
  • Annual Retail Trade Survey for 2015 (revised) and 2016 (new)
  • Service Annual Survey for 2016 (revised) and 2017 (new)
  • Value of Construction Put in Place Survey for 2015 and 2016 (revised) and 2017 (preliminary)
  • Office of Management and Budget federal government budget data for fiscal Year 2017 (revised) and 2018 (new)
  • BEA data for 2015–2017 (revised) from the International Transactions Accounts
  • Bureau of Labor Statistics (BLS) Quarterly Census of Employment and Wages for 2014–2016 (revised)
  • Internal Revenue Service (IRS) tabulations of corporate tax returns for 2015 (revised) and for 2016 (new)
  • IRS tabulations of sole proprietorship and partnership tax returns for 2016 (new)
  • U.S. Department of Agriculture farm statistics for 2015–2017 (revised)

Principal sources of data used to construct current-dollar and chained-dollar estimates for benchmark and nonbenchmark years can be found in table A and table B at the end of this article. Principal sources of data used to construct the quarterly estimates can be found in table C, also at the end of this article.

Gross output

Starting in 2010, data from the Census Bureau Service Annual Survey (SAS) replaced Bureau of Transportation Statistics (BTS) data as the annual indicator for the air transportation industry. The SAS data provided break-outs of domestic and international freight and passenger transportation that assisted in the construction of improved supply-use tables. The Economic Census was used to establish the 2007 and 2012 pillar estimates for the time series, and the 2007 level was backcast to 1997 using data from BTS. BTS data was also used as the source indicator up to 2010, when it was replaced with SAS data to complete the time series.

For IEA quarterly statistics, the Census Bureau Quarterly Services Survey (QSS) data are used as output indictors in place of existing NIPA personal consumption expenditures (PCE) estimates for educational services and accommodation services. NIPA PCE indicators capture personal consumption and do not include business intermediate purchases, while the QSS data captures both. Output estimates for the educational services industry and the accommodation services industry were improved by the replacement because the QSS data more accurately reflect activities that are conceptually included in those industries.

Prices

BEA continually strives to improve the price indexes used throughout the industry accounts and consults regularly with colleagues at BEA, BLS, and the Federal Reserve Board. BEA has introduced several improved indexes and extended improvements introduced in prior updates.

Software, medical equipment, and communications equipment. Software, medical equipment, and communications equipment typically experience rapid innovation and are associated with state-of-the-art technologies. Such products present challenges when using standard matched-model techniques to construct quality-adjusted price indexes. As part of the 2018 comprehensive update, BEA introduced several improved indexes and extended improvements that were introduced in prior updates to previous years to improve the deflation of output measures of software, medical equipment, and communications equipment.7

Amusement, gambling, and recreation. Three new BLS producer price indexes became available within “amusement, gambling, and recreation” and were incorporated into the annual and quarterly time series, replacing more aggregated PCE deflators from the NIPAs.

Motion picture and sound recording. Intermediate business purchases of licensing make up a substantial portion of motion picture and sound recording output. A BLS producer price index was incorporated to improve deflation of this output, replacing more aggregated PCE deflators from the NIPAs that largely capture personal consumption.

The percent change in real GDP growth for the first quarter of 2018 was revised up 0.2 percentage point to 2.2 percent from the previously published 2.0 percent (table 4). This revision reflected an upward revision to the production of private industries (both goods and services). The direction of growth in real value added was revised for 5 of 22 major industry groups.

  • Nondurable-goods manufacturing was revised up 8.4 percentage points, driving the upward revision within private goods-producing industries; value-added growth in this industry group was revised up to 12.1 percent from 3.8 percent.
  • Professional, scientific, and technical services was the second leading driver of the upward revision to real value added and the primary driver of the upward revision within private services-producing industries. Value-added growth in this industry group was revised up 4.2 percentage points to 6.0 percent from 1.8 percent.
  • The upward revision to real GDP growth was offset by a downward revision of 7.8 percentage points to finance and insurance, which was revised down to −5.5 percent from 2.2 percent.

Quarterly statistics for 2005–2017 were benchmarked to the corresponding annual estimates, and revisions to these quarters typically follow the revisions to the annual data. Updated quarterly source data and revised seasonal factors are also incorporated in the revisions to the quarterly estimates.

2017

Real GDP growth was revised down 0.1 percentage point in 2017 to 2.2 percent from the previously published 2.3 percent. Value-added growth in private services-producing industries was revised downward to 2.4 percent from 2.6 percent. Growth in private goods-producing industries was revised down 0.4 percentage point to 1.2 percent. The direction of growth was unrevised for 17 of 22 major industry groups (chart 2).

  • Finance and insurance was the leading contributor to the downward revision to private services-producing industries. Value-added growth in this industry was revised down to 0.4 percent from 1.5 percent, reflecting downward revisions to Federal Reserve banks, credit intermediation, and related activities and to insurance carriers and related activities.
  • Mining drove the downward revision to private goods-producing industries. Growth in this industry was revised down 16.4 percentage points to a decrease of 7.9 percent from the previously published increase of 8.5 percent, led by a downward revision to oil and gas extraction.
  • Growth in information was revised to 6.4 percent from 4.0 percent, partially offsetting the downward revision to real GDP growth. All four industries within information were revised upward, most significantly data processing, internet publishing, and other information services.
Chart X. Title

[Click chart to expand]

2016

Real GDP growth was revised up 0.1 percentage point in 2016 to 1.6 percent from the previously published 1.5 percent. Value-added growth in private services-producing industries was revised up to 2.3 percent from 1.9 percent. Growth in private goods-producing industries was revised down 1.3 percentage points from 0.3 percent to −1.0 percent. The direction of growth was revised for 5 of 22 major industry groups.

  • Information was the leading contributor to the upward revision to private services-producing industries. Value-added growth in this industry was revised up to 13.4 percent from 5.5 percent, primarily reflecting an upward revision to data processing, internet publishing, and other information services.
  • Nondurable-goods manufacturing drove the downward revision to private goods-producing industries. Growth in this industry was revised down 3.2 percentage points to a decrease of 2.0 percent from the previously published increase of 1.2 percent, reflecting a downward revision in petroleum and coal products growth.
  • Wholesale trade growth was revised downward to −1.5 percent from 0.2 percent, partially offsetting the upward revision to real GDP growth and growth in private services.

2015

Real GDP growth was unrevised in 2015 at 2.9 percent. Value-added growth in private services-producing industries also went unrevised. Private goods-producing industries growth was revised up 0.1 percentage point from 2.8 percent to 2.9 percent. Government was revised downward from 0.2 percent to flat growth. The direction of growth was unchanged for 19 of 22 major industry groups.

  • Mining drove the upward revision to private goods-producing industries. Growth in this industry was revised up 11.8 percentage points to 17.5 percent from the previously published increase of 5.8 percent, driven by an upward revision to growth of oil and gas extraction.
  • State and local government growth was revised downward 0.4 percentage point, changing the previously published growth rate of 0.4 percent to a slight decline of −0.1 percent. This revision drove the downward revision in the government sector growth.

2014

Real GDP growth was revised down 0.1 percentage point in 2014 to 2.5 percent from the previously published 2.6 percent. Private services-producing industries, which led the growth in real GDP with growth of 2.7 percent, was revised down from 2.8 percent. Growth in private goods-producing industries was revised downward 0.3 percentage point from 2.4 percent to 2.1 percent. The direction of growth was unchanged for 20 of 22 major industry groups.

  • Real estate and rental and leasing drove the downward revision to private services-producing industries. Value-added growth in this industry was revised down to 1.9 percent from 2.5 percent, reflecting a downward revision in growth of other real estate.
  • Durable-goods manufacturing drove the downward revision to private goods-producing industries. Growth in this industry was revised down 0.7 percentage point to 0.8 percent from the previously published increase of 1.5 percent. Many durable-goods industries saw downward revisions to real value-added growth, led by fabricated metal products and motor vehicles, bodies and trailers, and parts.
  • Growth in information was revised up to 2.1 percent from 0.2 percent, partially offsetting the downward revision to real GDP growth. The revision was mainly driven by data processing, internet publishing, and other information services.

2013

Real GDP growth was revised up 0.1 percentage point in 2013 to 1.8 percent from the previously published 1.7 percent. Private services-producing industries led the growth in real GDP with an unrevised growth rate of 1.3 percent. Growth in private goods-producing industries was revised up 0.2 percentage point to 3.9 percent. The direction of growth was unchanged for 21 of 22 major industry groups.

  • Durable-goods manufacturing was the leading contributor to the upward revision to private goods-producing industries. Growth in this industry was revised up 1.3 percentage points to 2.5 percent from the previously published increase of 1.2 percent, primarily reflecting an upward revision in growth to computer and electronic products.
  • Mining-sector growth was revised to 3.9 percent from 6.7 percent, partially offsetting the upward revision to real GDP growth and growth in private goods, reflecting a downward revision in growth to oil and gas extraction.

2007–2012

The average annual growth rate for real GDP was revised up 0.1 percentage point, primarily reflecting an upward revision to growth in private services-producing industries.

  • Average annual growth in real value added for professional, scientific, and technical services was revised up, primarily reflecting upward revisions to computer-systems design and related services and miscellaneous professional, scientific, and technical services.

2002–2007

The average annual growth rate for real GDP was revised up slightly, primarily reflecting an upward revision to growth in private goods-producing industries.

  • Average annual growth in real value added for mining was revised up, reflecting upward revisions to oil and gas extraction and support activities for mining.

1997–2002

The growth in private goods-producing industries was revised up slightly, but the average annual growth rate for real GDP was unrevised.

 


  1. Statistics for 1947–1996 have not yet been updated to reflect the 2018 comprehensive update.
  2. For more information, see Jason W. Chute, Stephanie H. McCulla, and Shelly Smith, “Preview of the 2018 Comprehensive Update of the National Income and Product Accounts,” Survey of Current Business 98 (April 2018).
  3. For additional details on the supply-use framework and how it compares to BEA's current make-use tables, see Jeffrey A. Young, Thomas F. Howells III, Erich H. Strassner, and David B. Wasshausen, “Supply-Use Tables for the United States,” Survey 95 (September 2015).
  4. For more information, see Erich H. Strassner and David B. Wasshausen, “Preview of the 2013 Comprehensive Revision of the Industry Economic Accounts,” Survey 93 (June 2013).
  5. Concordances between 2007 NAICS and 2012 NAICS are available through the Census Bureau website.
  6. Background information on BEA's efforts related to the digital economy are available online.
  7. For more information, see Jason W. Chute, Stephanie H. McCulla, and Shelly Smith, “Preview of the 2018 Comprehensive Update of the National Income and Product Accounts,” Survey of Current Business 98 (April 2018).