Current Employment Statistics Benchmark Article
Description
Benchmark Article
BLS Establishment Estimates Revised to Incorporate March 2010 Benchmarks
Introduction
Summary of the benchmark revisions
Revisions in the post-benchmark period
Corrected historical CES data for the other federal government series
Other changes to the CES published series
Why benchmarks differ from estimates
Benchmark revision effects for other data types
Methods
Introduction of quarterly net birth/death adjustments
Availability of revised data
Small domain model
Seasonal adjustment procedure
Special model adjustments
Seasonal adjustment of new all employee hours and earning series
Introduction
Jennifer M. Harris
Jennifer M. Harris is an economist in the Division of Current Employment Statistics, Office of Employment and Unemployment
Statistics, Bureau of Labor Statistics. Telephone: (202) 691-6555; e-mail: CESInfo@bls.gov
With the release of data for January 2011, the Bureau of Labor Statistics (BLS) introduced its annual revision of national estimates of
employment, hours, and earnings from the Current Employment Statistics (CES) monthly survey of nonfarm establishments. Each
year, the CES survey realigns its sample-based estimates to incorporate universe counts of employment—a process known as
1 benchmarking. Comprehensive counts of employment, or benchmarks, are derived primarily from unemployment insurance (UI) tax
reports that nearly all employers are required to file with State Workforce ...
BLS Establishment Estimates Revised to Incorporate March 2010 Benchmarks
Introduction
Summary of the benchmark revisions
Revisions in the post-benchmark period
Corrected historical CES data for the other federal government series
Other changes to the CES published series
Why benchmarks differ from estimates
Benchmark revision effects for other data types
Methods
Introduction of quarterly net birth/death adjustments
Availability of revised data
Small domain model
Seasonal adjustment procedure
Special model adjustments
Seasonal adjustment of new all employee hours and earning series
Introduction
Jennifer M. Harris
Jennifer M. Harris is an economist in the Division of Current Employment Statistics, Office of Employment and Unemployment
Statistics, Bureau of Labor Statistics. Telephone: (202) 691-6555; e-mail: CESInfo@bls.gov
With the release of data for January 2011, the Bureau of Labor Statistics (BLS) introduced its annual revision of national estimates of
employment, hours, and earnings from the Current Employment Statistics (CES) monthly survey of nonfarm establishments. Each
year, the CES survey realigns its sample-based estimates to incorporate universe counts of employment—a process known as
1 benchmarking. Comprehensive counts of employment, or benchmarks, are derived primarily from unemployment insurance (UI) tax
reports that nearly all employers are required to file with State Workforce ...
Sujets
Informations
| Publié par | Ulle |
| Nombre de lectures | 54 |
| Langue | English |
| Poids de l'ouvrage | 1 Mo |
Extrait
Benchmark Article
BLS Establishment Estimates Revised to Incorporate March 2010 Benchmarks
Introduction Summary of the benchmark revisions Revisions in the post-benchmark period Corrected historical CES data for the other federal government series Other changes to the CES published series Why benchmarks differ from estimates Benchmark revision effects for other data types Methods Introduction of quarterly net birth/death adjustments Availability of revised data Small domain model Seasonal adjustment procedure Special model adjustments Seasonal adjustment of new all employee hours and earning series
Introduction
Jennifer M. Harris
Jennifer M. Harris is an economist in the Division of Current Employment Statistics, Office of Employment and Unemployment Statistics, Bureau of Labor Statistics. Telephone: (202) 691-6555; e-mail: CESInfo@bls.gov
With the release of data for January 2011, the Bureau of Labor Statistics (BLS) introduced its annual revision of national estimates of employment, hours, and earnings from the Current Employment Statistics (CES) monthly survey of nonfarm establishments. Each year, the CES survey realigns its sample-based estimates to incorporate universe counts of employment — a process known as
benchmarking. Comprehensive counts of employment, or benchmarks, are derived primarily from unemployment insurance (UI) tax reports that nearly all employers are required to file with State Workforce Agencies.
Summary of the benchmark revisions
The March 2010 benchmark level for total nonfarm employment is 128,584,000; this figure is 378,000 below the sample-based estimate for March 2010, an adjustment of -0.3 percent. Table 1 shows the total nonfarm percentage benchmark revisions for the past ten years.
Table 1. Percent differences between nonfarm employment benchmarks and estimates by industry supersector, March 2001-2010 (1) Industry 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Total - 0.1 - 0.2 - 0.1 0.2 - 0.1 0.6 -0.2 -0.1 -0.7 -0.3 Total private - .2 - .4 - .2 .2 - .2 .7 -.2 -.1 -.9 -.4 Mining & logging .9 .7 -.3 1.2 .4 -3.5 -3.0 Construction -.8 .6 .5 2.6 .1 .7 -2.9 -1.3 Manufacturing -1.1 -.4 -.3 -.1 -1.0 -.1 -.7 -1.0 Trade, trans ortation, & utilities .2 .2 -1.2 -.6 .3 .6 .5 Information -2.6 -1.0 -2.1 -.5 -1.8 .3 -1.5 -.4 Financial activities .2 .1 -.8 .4 -1.3 -.3 -.1 .4 Professional & business services -.7 -.2 -.4 1.3 .2 -.4 -.8 Education & health services .3 .2 .5 -.2 -.1 - 3 . Leisure & hospitality .5 1.2 .4 .3 -.8 -1.1 -.6 -.6 Other services 1.4 .5 -1.3 .5 .3 .2 -.8 .2 Government .3 1.0 .3 .1 -.2 .2 .1 .1 (1) Differences are based on comparisons of final published March estimates and benchmark levels, as originally published. (2) Less than 0.05 percent.
Table 2 shows the nonfarm employment benchmarks for March 2010, not seasonally adjusted, by industry. The majority of super sectors had downward revisions, with the exception of government, financial activities, professional and business services, and other services. The largest downward revision occurred in trade, transportation, and utilities with a revision of -143,000, or -0.6 percent. Within this sector, the revision is concentrated in wholesale trade, which revised by -124,500, or -2.3 percent. It is of note that within
trade, transportation, and utilities there were a series of large offsetting revisions made to utilities. Utilities as a whole was only revised downwards 3,400, or 0.6 percent, but electric power generation was revised by -67,800, or -39.9 percent. This revision was offset by a positive revision in electric power transmission and distribution, revised upwards by 65,600, or 28.7 percent. Manufacturing had a downward revision of 119,000, or 1.0 percent, and leisure and hospitality had a revision of -80,000, or -0.6 percent. The revision in manufacturing was concentrated in durable goods with a downward revision of 82,000, or 1.2 percent, while nondurable goods was revised downward by 37,000, or 0.8 percent. The largest downward revision within leisure and hospitality was food services and drinking places with a revision of 92,700, or 1.0 percent.
Table 2. Nonfarm employment benchmarks by industry for March 2010 (in thousands) Differences Industry Benchmark Estimate Amount Percent Total nonfarm 128,584 128,962 -378 -0.3 Total rivate 105,671 106,062 -391 -.4 Goods-producing 17,248 17,454 -206 -1.2 Service-providing 111,336 111,508 -172 -.2 Private service-providing 88,423 88,608 -185 -.2 Minin and lo in 668 688 -20 -3.0 Logging 48 46 2 4.2 Minin 621 642 -21 -3.4 Oil and gas extraction 156 161 -5 -3.2 Minin , exce t oil and as 194 205 -11 -5.7 Coal mining 79 81 -2 -2.5 Su ort activities for minin 271 276 -5 -1.8 Construction 5,213 5,280 -67 -1.3 Construction of buildin s 1,186 1,214 -28 -2.4 Heavy and civil engineering construction 735 721 14 1.9 Specialty trade contractors 3,292 3,345 -53 -1.6 Manufacturing 11,367 11,486 -119 -1.0 Durable oods 6,960 7,042 -82 -1.2 Wood products 33 3 339 -6 -1.8 Nonmetallic mineral products 358 369 -11 -3.1
Table 2. Nonfarm employment benchmarks by industry for March 2010 (in thousands) — continued Differences Industry Benchmark Estimate Amount Percent Primary metals 353 362 -9 -2.5 Fabricated metal roducts 1,253 1,273 -20 -1.6 Machinery 978 980 -2 -.2 Computer and electronic products 1,093 1,091 2 .2 Computer and peripheral equipment 159 158 1 .6 Communications equipment 116 119 -3 -2.6 Semiconductors and electronic components 366 362 4 1.1 Electronic instruments 406 405 1 .2 Electrical equipment and appliances 353 364 -11 -3.1 Trans ortation e ui ment 1,319 1,336 -17 -1.3 Furniture and related products 356 356 0 Miscellaneous manufacturing 564 572 -8 -1.4 Nondurable goods 4,407 4,444 -37 -.8 Food manufacturing 1,418 1,427 -9 -.6 Beverages and tobacco products 178 179 -1 -.6 Textile mills 119 123 -4 -3.4 Textile product mills 118 121 -3 -2.5 A arel 159 165 -6 -3.8 Leather and allied products 27 29 -2 -7.4 Pa er and a er roducts 393 394 -1 -.3 Printing and related support activities 487 494 -7 -1.4 Petroleum and coal products 110 110 0 Chemicals 785 781 4 .5 Plastics and rubber products 614 622 -8 -1.3 Trade, transportation, and utilities 24,278 24,421 -143 -.6 Wholesale trade 5,410 5,535 -125 -2.3 Electronic markets and agents and brokers 799 823 -24 -3.0 Retail trade 14,204 14,222 -18 -.1 Motor vehicle and parts dealers 1,599 1,608 -9 -.6 Automobile dealers 997 1,009 -12 -1.2
Table 2. Nonfarm employment benchmarks by industry for March 2010 (in thousands) — continued Differences Industry Benchmark Estimate Amount Percent Furniture and home furnishings stores 432 435 -3 -.7 Electronics and a liance stores 492 479 13 2.6 Building material and garden supply stores 1,132 1,154 -22 -1.9 Food and beverage stores 2,779 2,776 3 .1 Health and personal care stores 974 968 6 .6 Gasoline stations 804 809 -5 -.6 Clothing and clothing accessories stores 1,318 1,334 -16 -1.2 S ortin oods, hobb , book, and music stores 585 593 -8 -1.4 General merchandise stores 2,935 2,903 32 1.1 De artment stores 1,453 1,448 5 .3 Miscellaneous store retailers 747 754 -7 -.9 Nonstore retailers 408 410 -2 -.5 Transportation and warehousing 4,111 4,108 3 .1 Air transportation 462 452 10 2.2 Rail transportation 211 215 -4 -1.9 Water trans ortation 1 -1 -1.7 60 6 Truck transportation 1,208 1,201 7 .6 Transit and round assen er trans ortation 437 429 8 1.8 Pipeline transportation 42 39 3 7.1 Scenic and si htseein trans ortation 21 22 -1 -4.8 Support activities for transportation 532 534 -2 -.4 Couriers and messengers 514 517 -3 -.6 Warehousing and storage 625 639 -14 -2.2 Utilities 553 556 -3 -.5 Information 2,715 2,726 -11 -.4 Publishin industries, exce t Internet 762 762 0 Motion picture and sound recording industries 365 342 23 6.3 Broadcastin , exce t Internet 294 295 -1 .3 -Telecommunications 913 942 -29 -3.2 Data processing, hosting and related services 244 249 -5 -2.0
Table 2. Nonfarm employment benchmarks by industry for March 2010 (in thousands) — continued Differences Industry Benchmark Estimate Amount Percent Other information services 138 136 2 1.4 Financial activities 7,606 7,572 34 .4 Finance and insurance 5,695 5,656 39 .7 Monetary authorities - central bank 21 21 0 Credit intermediation and related activities 2,544 2,567 -23 -.9 Depositor credit intermediation 1,727 1,748 -21 -1.2 Commercial banking 1,304 1,310 -6 -.5 Securities, commodit contracts, investments 796 791 5 .6 Insurance carriers and related activities 2,249 2,194 55 2.4 Funds, trusts, and other financial vehicles 86 85 1 1.2 Real estate and rental and leasing 1,911 1,915 -4 -.2 Real estate 1,379 1,368 11 .8 Rental and leasing services 506 523 -17 -3.4 Lessors of nonfinancial intangible assets 25 25 0 Professional and business services 16,343 16,346 -3 Professional and technical services 7,468 7,469 -1 Legal services 1,109 1,101 8 .7 Accountin and bookkee in services 1,005 1,025 -20 -2.0 Architectural and engineering services 1,261 1,261 0 Com uter s stems desi n and related services 1,417 1,429 -12 -.8 Management and technical consulting services 982 974 8 .8 Management of companies and enterprises 1,850 1,815 35 1.9 Administrative and waste services 7,025 7,062 -37 -.5 Administrative and support services 6,681 6,722 -41 -.6 Employment services 2,517 2,571 54 -2.1 -Tem orar hel services 1,908 1,922 -14 -.7 Business support services 809 799 10 1.2 Services to buildin s and dwellin s 1,591 1,576 15 .9 Waste management and remediation services 344 340 4 1.2 Education and health services 19,599 19,592 7
Table 2. Nonfarm employment benchmarks by industry for March 2010 (in thousands) — continued Differences Industry Benchmark Estimate Amount Percent Educational services 3,285 3,283 2 .1 Health care and social assistance 16,314 16,309 5 Ambulatory health care services 5,924 5,873 51 .9 Offices of physicians 2,306 2,308 -2 -.1 Outpatient care centers 594 548 46 7.7 Home health care services 1,066 1,056 10 .9 Hospitals 4,666 4,697 -31 -.7 Nursin and residential care facilities 3,105 3,102 3 .1 Nursing care facilities 1,649 1,646 3 .2 Social assistance 2,620 2,637 -17 -.6 Child day care services 868 882 -14 1.6 -Leisure and hospitality 12,578 12,658 -80 -.6 Arts, entertainment, and recreation 1,740 1,737 3 .2 Performing arts and spectator sports 370 371 -1 -.3 Museums, historical sites, zoos, and parks 119 121 -2 -1.7 Amusements, amblin , and recreation 1,251 1,245 6 .5 Accommodation and food services 10,838 10,920 -82 -.8 Accommodation 1,679 1,668 11 .7 Food services and drinking places 9,159 9,252 -93 -1.0 Other services 5,304 5,293 11 .2 Repair and maintenance 1,128 1,139 -11 -1.0 Personal and laundry services 1,259 1,265 -6 -.5 Membership associations and organizations 2,917 2,890 27 .9 Government 22,913 22,900 13 .1 Federal 2,905 2,895 10 .3 Federal, exce t U.S. Postal Service 2,244 2,235 9 .4 U.S. Postal Service 661 661 0 State overnment 5,280 5,315 -35 -.7 State government education 2,507 2,540 -33 -1.3 State government, excluding education 2,774 2,775 -1
Table 2. Nonfarm employment benchmarks by industry for March 2010 (in thousands) — continued Differences Industry Benchmark Estimate Amount Percent Local government 14,728 14,690 38 .3 Local overnment education 8,432 8,379 53 .6 Local government, excluding education 6,296 6,312 -16 -.3 (1) Less than 0.05 percent. Table 3. Differences in seasonally adjusted levels and over-the-month changes, total nonfarm employment, January 2010-December 2010 (in thousands) Levels Over-the-month changes As reviousl As reviousl 2010 published As revised Difference published As revised Difference January 129,602 129,281 -321 14 -39 -53 Februar 129,641 129,246 -395 39 -35 -74 March 129,849 129,438 -411 208 192 -16 A ril 130,162 129,715 -447 313 277 -36 May 130,594 130,173 -421 432 458 26 June 130,419 129,981 -438 -175 -192 -17 July 130,353 129,932 -421 -66 -49 17 Au ust 130,352 129,873 -479 -1 -59 -58 September 130,328 129,844 -484 -24 -29 - 5 October 130,538 130,015 -523 210 171 -39 November 130,609 130,108 -501 71 93 22 December (p) 130,712 130,229 -483 103 121 18 (p) Preliminary
Construction was revised by -67,000, or -1.3 percent, while mining and logging was revised -20,000, or -3.0 percent. One of the largest revisions within construction was to building equipment contractors, which was revised downward by 32,700, or 2.1 percent. Information had a downward revision of 11,000, or 0.4 percent. Within the supersector the largest revision was in telecommunications with a revision of -29,000, or -3.2 percent.
Four sectors saw positive revisions. Financial activities was revised upward by 34,000, or 0.4 percent. The revision in financial activities was concentrated in insurance carriers and related activities which was revised upward by 55,000, or 2.5 percent. Government was revised upward by 13,000, or 0.1 percent, and other services had a revision of 11,000, or 0.2 percent. Education and health services had a negligible revision upward of 7,000, or less than 0.05 percent.
Revisions in the post-benchmark period Post-benchmark period estimates from April 2010 to December 2010 were calculated for each month based on new benchmark levels. Also, beginning in April, model-based estimates for the net birth/death employment were revised to incorporate information from the most recent year of universe employment counts. Text table A shows the net birth/death model figures for the supersectors over the post-benchmark period. From April 2010 to December 2010, the cumulative net birth/death model added 537,000, compared with 759,000 in the previously published April to December estimates.
Text Table A. Net Birth/Death Estimates, Post-Benchmark 2010 (in thousands)
2010 April May June Jul August Se tember October November December Cumulative Total
0 1 2 1 1 1 1 0 0 7
8 36 21 -18 5 4 0 -16 -19 21
9 -7 5 -11 3 0 3 -1 0 -7
-1 21 12 -19 11 6 16 -2 6 50
2 3 1 -3 4 0 4 1 2 14
2 -6 5 -11 5 -1 11 0 11 24
50 18 15 1 -19 -9 42 2 -2 134
20 15 -7 2 19 11 45 7 2 114
66 78 72 29 21 -36 -44 -24 4 166
7 7 5 -7 3 -1 -1 -1 2 14
141 192 131 -38 91 -25 71 -32 6 537
Revisions to November and December also reflect incorporation of the annual CES sample update and the routine inclusion of additional sample units not available for the respective months' preliminary estimates. Corrected historical CES data for the other federal government series With the 2010 benchmark revision BLS corrected historical CES employment levels for all employees and women employees for the other federal government series (91-999900). CES data series on hours and earnings for all employees and for production and nonsupervisory employees were not impacted. The corrections affected April 2009 to April 2010 data, and reflect corrections to initial counts for Census temporary and intermittent workers for Census 2010. These corrections were applied before any benchmark processing was done. The corrected March 2010 value was used to calculate the benchmark revision for this industry, and the corrected values from April 2009 to February 2010 were then subject to the normal wedging process. (For more information on the wedging process please reference the section on benchmark methods .) The corrected values for March and April 2010 were used to calculate a sample link for April 2010, which was applied to the new benchmark level during the post-benchmark processing to generate the April 2010 post-benchmark estimate. These corrections did not impact employment levels after April 2010; post-benchmark estimates for May through October 2010 were calculated using a sample link generated from previously published values. Exhibit 1 shows the previously published employment level, the corrected value applied before benchmark processing, and the difference between the two for all employees. Exhibit 2 shows the same for women employees.
Exhibit 1. Corrections to all employees for other federal government (91-999900) Previousl ublished Corrected value a lied before Month employment benchmark processing Apr-09 1,377.2 1,373.4 May-09 1,330.4 1,324.2 Jun-09 1,289.5 1,292.3 Jul-09 1,298.8 1,301.6 Aug-09 1,296.3 1,300.3 Sep-09 1,287.9 1,292.3 Oct-09 1,296.9 1,301.4 Nov-09 1,285.3 1,288.8 Dec-09 1,287.3 1,291.0 Jan-10 1,289.4 1,291.9
Difference -3.8 -6.2 2.8 2.8 4.0 4.4 4.5 3.5 3.7 2.5
Exhibit 1. Corrections to all employees for other federal government (91-999900) — continued Previously published Corrected value applied before Month employment benchmark processing Difference Feb-10 1,312.9 1,314.5 1.6 Mar-10 1,367.3 1,371.6 4.3 Apr-10 1,452.5 1,443.7 -8.8 Exhibit 2. Corrections to women employees for other federal government (91-999900) Corrected value a lied before Month Previously published employment benchmark processing A r-09 654.3 652.5 May-09 626.8 623.9 Jun-09 603.9 605.2 Jul-09 606.3 607.6 Aug-09 604.9 606.8 Sep-09 602.1 604.2 Oct-09 607.9 610.0 Nov-09 604.0 605.7 Dec-09 605.0 606.7 Jan-10 608.2 609.4 Feb-10 623.0 623.7 Mar-10 650.6 652.6 Apr-10 691.3 686.9
Difference -1.8 -2.9 1.3 1.3 1.9 2.1 2.1 1.7 1.7 1.2 0.7 2.0 -4.4
Other changes to the CES published series All CES series are evaluated annually for sample size, coverage, and response rates. The following series changes result from a re-evaluation of the sample and universe coverage for NAICS industries. Some small industries no longer have sufficient sample to be estimated and published separately and have been combined with other similar industries for estimation and publication purposes, as shown below. Most of the collapsed and deleted series are in the manufacturing sector where employment has been declining over a number of years. Historical data for the series with changed scope were reconstructed to provide consistent time series.
© 2010-2024 YouScribe