LCCR Coalition Letter to Amy A. Hobby Regarding CES WWS

Amy A. Hobby, BLS Clearance Officer
Division of Management Systems
Bureau of Labor Statistics
2 Massachusetts Ave., NE, Room 4080
Washington, DC 20212

Re: Comments on Proposed Discontinuation of Current Employment Statistics Women Worker Series

Dear Ms. Hobby:

The Current Employment Statistics Women Worker Series (CES WWS) is an important and unique source of information about how women are faring in the economy. Despite this fact, the Bureau of Labor Statistics (BLS) has proposed to eliminate the series, based on the assertions that 1) little of value is lost because the series has few users and other data sources are sufficient, and 2) continuing to require employers to report the number of jobs gained and lost by gender is an undue burden, especially in light of BLS’ desire to institute some new data elements. 69 Fed. Reg. 76793 (Dec. 22, 2004). The justifications offered for the discontinuation are unpersuasive, but even if they were, the Leadership Conference on Civil Rights (LCCR), many members of which are consumers of this information, believes such concerns are outweighed by the benefits of continuing to collect this valuable information. LCCR is a diverse coalition of more than 180 national organizations representing a broad constituency, including persons of color, women, children, labor unions, individuals with disabilities, older Americans, major religious groups, and gays and lesbians, with a shared commitment to advancing equal opportunity in employment and in ensuring that all Americans are treated fairly in the workplace. Thus, LCCR opposes the proposal to discontinue the collection of establishment data on jobs gained and lost by gender.

Discontinuation of the CES WWS Would Erase a Key Part of the Employment Picture

Despite significant gains by women in the workforce over the last few decades, women have not yet achieved equality with men in the workplace, and women and men continue to have different experiences with, and relationships to, employment. Women still earn only about three-fourths of what similarly situated men earn; women still take more time out of the workforce than men to care for family members and are more likely to work part-time; and the workforce is still highly sex-segregated, with women concentrated in different and lower-paying jobs and industries than men.[1] Because women and men often predominate in different industries, they can also have different experiences with job gains and losses over the course of economic downturns and recoveries.

The CES WWS enables researchers, advocates, policymakers, and the general public to gauge how women are faring in the economy. The WWS does not provide a complete picture, but it does provide a critical piece of the picture: a monthly indicator of the employment situation for nonsupervisory, nonagricultural workers, by industry. But for the WWS, for example, it would not have been possible either to determine that the latest recession triggered “”the only period of sustained job loss for women in the last forty years,””[2] or to identify the industries in which these job losses occurred. The elimination of the WWS thus necessarily would produce a less complete, and therefore possibly misleading, view of how female and male workers are affected by the business cycle, which in turn has implications for public policy. It is important to know whether and where women are losing jobs, and how that might differ from men’s experience, to help ensure both that economic trends that affect the sexes differently are identified and examined and that industries in which women have incurred disproportionate job losses are not overlooked in decisions about, for example, job training and displacement assistance.

While the Current Population Survey (CPS) is a rich source of employment and demographic information, it is no substitute for the information provided by the CES. The CPS collects household data from a sample size of only 60,000,[3] whereas the CES collects establishment data from actual business records submitted by between approximately 160,000[4]-283,000[5] employers, representing 400,000 workers.[6] The CPS provides information on the employment status of people, whereas the CES provides information on the presence or absence of jobs. The CES is “”widely viewed as a key measure of the health of the economy?”” and its estimates are “”some of the most timely and sensitive economic indicators published by the Federal Government.””[7] Despite the promise of a new CPS table on employed women by industry, as indicated in the revised and expanded January 31, 2005 version of BLS’ Notice,[8] the basic fact is that the CPS and the CES convey different, albeit related, information about women and employment.

BLS’ Justifications for Discontinuing the CES WWS Are Not Supported and Do Not Outweigh the Value of Maintaining the WWS

One justification offered by BLS to discontinue the WWS is that the “”data series does not have a large number of users””[9] and is “”little used.””[10] According to the revised and expanded Notice issued by BLS, “”use”” is based on the number of “”requests”” for the data, and the number of articles that resulted from an informal internet literature search. While it is unclear what constitutes a “”request,”” on both counts it appears that BLS is using an unduly narrow definition that fails to account for the myriad of ways in which this information is consumed. Does “”request”” only include use of the website’s extraction tool? The number of times the link to the relevant table has been clicked or downloaded? The number of special, custom requests by researchers or journalists for information not available on the website? There is no indication which of these or other uses were considered by BLS. Moreover, an informal internet search may turn up some articles, but it cannot detect how many researchers, journalists, advocates, policymakers, and members of the public read and relied on those articles, nor how many times they may have used that information. At the very least, BLS should be considering such other types of “”uses”” in assessing the value of the data.

The revised and expanded BLS Notice states that the Bureau is particularly interested in “”receiving examples of actual usage”” of the series. But the Bureau appears to consider only outside researchers and writers, and overlooks the type of extensive analysis conducted and published by the BLS itself, such as Women and Jobs in Recessions: 1969-92.[11] This article still appears on the DOL website,[12] indicating that it still has relevance, is of interest, and is read. If BLS stops collecting the underlying data, neither it nor anyone else will ever be able to produce or rely on a similar analysis.

Moreover, even assuming the number of “”users”” was relatively low, it is unfair to indict the WWS for a lack of use when the data are not readily visible. The WWS results are released one month after the rest of the job figures; by the time they appear, therefore, they are considered “”old news”” by journalists and are not typically included in news stories about the monthly jobs report. “”[T]he first published estimates are the most widely anticipated and analyzed.””[13] Detached from the rest of those monthly jobs estimates, it is not surprising that the WWS figures receive little attention or visibility in the news. In addition, the data are not easy to find or access from BLS. The links to the appropriate tables are not readily apparent on the BLS website, and the data do not appear among the listed publications and releases. Finally, even if there have been relatively few “”requests”” for the WWS from the public, that is evidence only that too many requesters are glossing over possible differences between women and men, not evidence that the information is not needed.

BLS’ second justification for discontinuing the Series is that it “”imposes a significant reporting burden””[14] on employers and that it wishes to relieve employers of some data submission duties at the same time it is imposing new ones. Yet, on the very forms used to collect the gender information from employers, BLS estimates that it takes an average of only seven minutes to fill out the entire form each month.[15] To be clear, the WWS is not a lengthy questionnaire. The “”Series”” is based on one item: “”Enter the number of employees from Column 1 who are women.””[16] BLS has not provided evidence of the actual, demonstrated burden of answering this one question; nor has it offered any evidence of the percentage of employers that are already collecting this information. But certainly, those with 100 or more employees already capture and report this and more extensive information on their EEO-1 forms.[17] Federal contractors must also track the sex of their employees as part of their various reporting obligations. Smaller employers presumably have little trouble identifying and reporting the sex of their employees. LCCR sees no reason to believe that it is burdensome for employers to continue reporting this information.

The Leadership Conference on Civil Rights recognizes that BLS wishes to add new data elements to the CES, and it commends BLS for its decision to add supervisory workers to its dataset.[18] However, LCCR is not persuaded that the “”burden”” of continuing the WWS – assuming any such “”burden”” exists − outweighs the benefits of its retention. LCCR therefore urges BLS not to stop asking a question that it already asks and that business establishments already answer.

LCCR appreciates the opportunity to submit these comments and urges BLS to continue the Women Worker Series.

Sincerely,

AFL-CIO
American Association of University Women
American Civil Liberties Union
American-Arab Anti-Discrimination Committee
Church Women United
Communications Workers of America
Disability Rights Education and Defense Fund
Feminist Majority
Lawyers’ Committee for Civil Rights Under Law
National Association of Social Workers
National Bar Association
National Council of Jewish Women
National Gay and Lesbian Task Force
National Organization for Women
National Partnership for Women and Families
National Women’s Law Center
Parents, Families, and Friends of Lesbians and Gays (PFLAG)
Robinson, Curly and Clayton
Unitarian Universalist Association of Congregations
Women Employed

Footnotes

[1] Bureau of Labor Statistics, U.S. Dept. of Labor, Women in the Labor Force: A Databook, 14, 25-34, 38-39, 48-50 (Feb. 2004), available at http://www.bls.gov/cps/wlf-databook.pdf (last visited Feb. 1, 2005).
[2] Heidi Hartmann, Vicky Lovell, & Misha Werschkul, Women and the Economy: Recent Trends in Job Loss, Labor Force Participation, and Wages (Institute for Women’s Policy Research, Oct. 2004), available at http://www.iwpr.org/pdf/B245.pdf (last visited Feb. 1, 2005).
[3] Bureau of Labor Statistics, U.S. Dept. of Labor, Labor Force Statistics from the Current Population Survey: Overview, available at http://www.bls.gov/cps/cps_over.htm#available (last visited Feb. 3, 2005).
[4] Bureau of Labor Statistics, U.S. Dept. of Labor, Planned Changes to the Current Employment Survey, available at http://www.bls.gov/ces/cesww.htm (modified Jan. 31, 2005) [hereinafter Revised Notice].
[5] Proposed Collection, Comment Request, 69 Fed. Reg. 76793 (proposed Dec. 22, 2004) (hereinafter Fed. Reg. Notice).
[6] Revised Notice, supra note 4, at 1.
[7] Chris Manning, Concurrent Seasonal Adjustment for National CES Survey, 126 Monthly Labor Review 39 (Oct. 2003) [hereinafter Manning].
[8] Revised Notice, supra note 4, at 1.
[9] Fed. Reg. Notice, supra note 5, at 76794.
[10] Revised Notice, supra note 4.
[11] William Goodman, Stephen Antczak, and Laura Freeman, Women and Jobs in Recessions: 1969-92, 116 Monthly Labor Review 26 (July 1993), available at http://www.bls.gov/opub/mlr/1993/07/art3full.pdf (last visited Feb. 3, 2005).
[12] U.S. Dept. of Labor, Find It! By Audience − Women, available at http://www.dol.gov/dol/audience/aud-women.htm (last visited Feb. 1, 2005).
[13] Manning, supra note 7, at 39.
[14] Revised Notice, supra note 4, at 1.
[15] See, e.g., Bureau of Labor Statistics, U.S. Dept. of Labor, Bureau of Labor Statistics Report on Current Employment Statistics – Service-Providing 1, Form # BLS-790 E, available at http://www.bls.gov/ces/bls790e.pdf (last visited Feb. 3, 2005).
[16] Id.
[17] See 29 C.F.R. §§ 1602.7-1602.14 (2004); U.S. Equal Employment Opportunity Commission & Office of Federal Contract Compliance Programs, Equal Employment Opportunity Employer Information Report EEO-1, at 2, available at http://www.eeoc.gov/eeo1survey/eeo1.pdf (last visited Feb. 7, 2005).
[18] Although LCCR welcomes this addition, it has concerns about BLS’ apparent plans to aggregate supervisory with nonsupervisory workers, for reasons analogous to the ones raised here: the economy may affect supervisory and nonsupervisory employees differently, and it is important not to lose information about how the average (nonsupervisory) worker is faring.