Diversity Dilemma

arly this year, the Merit Systems Protection Board issued a stunning report on the federal hiring system. After an extensive review, MSPB recommended ending two special hiring programs designed to allow agencies to bypass standard procedures and quickly hire minority employees. The Outstanding Scholar and Bilingual/Bicultural programs, MSPB concluded, "conflict with the first statutory merit system principle . . . that hiring should be based on merit." In particular, the board found that the Outstanding Scholar program, under which agencies are allowed to make job offers to students based on their grade point averages, "has been severely misused in recent years."

Both the Office of Personnel Management and the Equal Employment Opportunity Commission quickly expressed opposition to the recommendation, voicing alarm that the Clinton administration's commitment to create a federal workforce that "looks like America" would be compromised by ending these targeted programs. Even as the administration oversaw substantial reductions of federal civilian employees, it used buyouts to encourage older white men to retire voluntarily, thus protecting the diversity gains the administration sought.

Barely a year before its report on the Outstanding Scholar and Bilingual/Bicultural programs, MSPB had reported that Hispanics remain underrepresented in the federal workforce. OPM Director Janice Lachance promoted flexible hiring programs to increase the number of Hispanics in government.

Such efforts to increase diversity in the federal workforce build on at least 30 years of bipartisan support for affirmative employment practices. Targeted efforts aimed at more extensive hiring of minorities and women have been used at least since the Johnson administration, with a strong impetus under President Nixon. In the waning weeks of the Carter administration, OPM signed a consent decree in federal court settling a lawsuit challenging the use of the Professional and Administrative Career Examination (PACE) for federal hiring. PACE was found to hurt the chances of African-American and Hispanic job applicants. The consent decree led to the creation of the special hiring programs, which were only supposed to be used temporarily until new tests were developed.

During the Reagan and Bush administrations, affirmative employment practices were ingrained in senior executives' performance evaluations. Federal agencies have long been recognized for providing greater opportunities than were generally open in the private sector. Indeed, MSPB concluded in its report, "the evidence suggests that representative African-American and Hispanic hiring now can be sustained without the non-merit based hiring authorities provided by the consent decree."

There is no visible opposition to the commitment to diversity in the federal workforce, and much concern about maintaining momentum in the diversity quest during a period of dramatic federal downsizing. But now MSPB openly differs from OPM and EEOC on the issue of whether special hiring programs-which often conflict with merit principles-are required anymore to ensure a diverse government workplace.

Downsizing's Effects

Federal agencies annually report to the EEOC on employment levels of white women, blacks, Hispanics, Asian-Pacific Islanders and American Indians. The federal workforce that President Clinton inherited, using data that agencies reported in 1993, included significant numbers of women and minorities. Blacks, Asian Americans and American Indians all were represented in government at levels slightly above their proportions of the nationwide civilian labor force. In aggregate, women and minorities constituted 56 percent of the federal workforce, compared with 57 percent in the overall civilian workforce. At 13 of 27 agencies analyzed for this article, minorities and women were employed in proportions greater than those in the civilian labor force. These numbers change slowly over the years, in large part because the federal workforce is very stable, with most change resulting from retirements and new hires.

Between 1993 and 1998, according to data that agencies reported to the EEOC, nearly 265,000 federal jobs were eliminated. Those reductions, however, did not reach proportionately across different segments of the workforce. Agencies made substantial cuts in the clerical ranks (cutting more than 81,000 positions, or about one-third of the clerical workforce) and among blue-collar employees (another 81,000 jobs, 72 percent of that segment). They also cut more than 50,000 professional positions.

Many clerical positions were eliminated as senior program analysts and managers lost secretaries while they gained personal computers. The blue-collar workforce was disproportionately affected by Defense Department base closures and increased contracting of commercial activities. Yet although the administration also targeted administrative positions (such as budget, accounting and personnel specialists) for cuts, this segment of the federal workforce experienced a reduction of less than 2 percent.

At the same time as agencies made these cuts, they also managed to achieve diversity gains. Minorities and women increased from 56 percent to 57 percent of the federal workforce between 1993 and 1998, so that these groups are now represented in the federal workforce in slightly greater proportion than in the civilian labor force. Between 1993 and 1998, 25 of 27 agencies showed increases in levels of employment of minorities and women. The two agencies showing declines were the EEOC, where women and minorities fell from 84 percent to 83.6 percent, and the Department of Health and Human Services, where the percentage dropped from 76 percent to 75 percent. By 1998, 16 of the 27 agencies in this analysis employed greater proportions of women and minorities than the civilian labor force.

The federal government may actually be doing better than these figures indicate, though, because comparisons with the civilian labor force can be misleading. The civilian workforce includes, for example, a substantial blue-collar segment that is a diminishing portion of the federal workforce. The overall American labor force also includes a robust agricultural sector that is largely omitted from the federal workforce. In assessing its own workforce, for example, the EEOC focuses on the professional and administrative categories that make up 80 percent of its workforce, according to Phil Calkins of the EEOC's federal sector office.

A more insightful perspective can be gained by comparing agencies with the relevant professional, administrative, technical, clerical, other white-collar and blue-collar (PATCOB) segments of the civilian labor force. In measuring the diversity of an employer's workforce, analysts expect that the workforce will reflect the balance of people in the community with the relevant skills. Thus, for example, when the MSPB reviewed federal agencies' hiring of Hispanic Americans, it noted that many Hispanics working in the civilian labor force lack the citizenship required for most federal jobs. Many Hispanics also lack the educational requirements associated with white-collar occupations. Further, Hispanics constitute a much greater portion of the agricultural workforce than of the white-collar occupations that dominate the federal service. Awareness of these factors is important in understanding differences between the federal workforce and the national civilian labor force.

Comparing changes in the federal workforce's PATCOB categories between 1993 and 1998 provides several measures of the diversity gains the Clinton administration could claim. Black men and women and American Indian men and women were at least fully represented compared with the private sector in all six PATCOB employment categories in both 1993 and 1998. The proportion of Hispanic men in the federal workforce increased in five of the six employment categories. Hispanic women surpassed civilian labor force levels in the professional, technical and clerical sectors while also increasing their proportion in five of the six categories. Asian-Pacific Island men surpassed civilian labor force levels in the administrative area while remaining employed at levels above the civilian labor force levels for the professional, blue-collar and "other" categories.

Diversity gains have permeated every sector of the federal workforce except the clerical staff. As measured by the percentage of each workforce sector made up of women and minorities, professional employees increased from 44 percent women and minorities in 1993 to 48 percent in 1998, with increases almost as large in the administrative and technical workforces.

These data demonstrate that people who have relevant work skills are included in the federal workforce at levels at or above comparable sectors of the civilian workforce. Further, they confirm that diversity gains have been shared widely in the federal sector during the five-year period. "The government has a lot to feel good about in this area," says John Palguta, director of the MSPB's Office of Policy and Evaluation. "We still have areas where we could do better, but in many of these areas, we serve as a model to the private sector."

Plantations and Glass Ceilings

Data reported by agencies to the EEOC contradict many popular impressions about diversity--or the lack thereof-in the federal workforce. For example, at a 1997 House hearing, Oscar Eason, former president of Blacks in Government, alleged that white women were benefiting disproportionately from the employment opportunities that were opened through affirmative action. Eason said the federal workforce retained several "plantations" that continued to impede the employment objectives of blacks.

The data confirm part of Eason's premise, that the portion of federal jobs held by black men declined from 6.6 percent of the workforce in 1993 to 6.5 percent in 1998. During that span, however, black men gained in the federal professional and administrative workforces, and black women gained not only in the overall federal workforce, but also in the professional, administrative, technical and clerical categories. Eason's perspective is supportable only if declines in the blue-collar and "other" sectors of the federal workforce could be construed as harmful, or if gains by black women and other minority groups that result in black men becoming employed at lower rates could somehow be construed as racially motivated.

Instead, where reductions have adversely affected women and minorities, the data provide a clear indication that white women have been hit hardest. White women were not employed in the federal workforce at levels comparable to appropriate sectors of the civilian labor force for any of the occupational categories in either 1993 or 1998. Moreover, white women's portion of the federal clerical, blue-collar and "other" workforces declined during the Clinton years.

Another common assertion is that women and minorities are limited by a "glass ceiling" from ascending to the highest levels of the federal service. Agencies' reports to the EEOC, however, indicate that minorities have moved into the professional ranks of the federal workforce in large numbers. Indeed, the professional ranks are more diverse than any other sector of the federal labor force.

Black men are fully represented among professionals at 25 of the 27 agencies analyzed for this article, missing civilian labor force levels only at the departments of the Interior and Veterans Affairs. Black women are at least fully represented at 22 of the 27 agencies analyzed. Hispanic men and American Indian men are fully represented at 17 of the 27 agencies. Among professionals, white women are fully represented at only six of the 27 agencies.

Across the Agencies

A review of data from individual agencies shows the breadth of the recent effort to diversify the federal workforce. The Justice Department, for example, has created one of the most diverse workforces in government by most measures used in this analysis. From 1993 to 1998, Justice increased the portion of women and minorities in its workforce from 54 percent to 55 percent-slightly below the civilian labor force for the nation. Women and minorities increased from 43 percent to 50 percent of the professional positions at Justice and gained nearly 4 percent in the administrative workforce. These gains were offset by declines in the portion of women and minorities in the technical and clerical workforces. In effect, the data indicate that women and minorities gained substantially as Justice-which gained employees while other agencies were downsizing-restructured its workforce.

At the same time, the Defense Department, which bore the lion's share of the federal workforce reductions, also managed to increase its diversity. The overall portion of women and minorities in the civilian DoD workforce rose from 50 percent to 51 percent, with gains of 4 percent among professionals and 3 percent in the technical sector overcoming reductions in other categories.

The Environmental Protection Agency and the Federal Communications Commission also increased their diversity significantly. EPA's gains included increasing the proportion of American Indian employees, surpassing civilian labor force levels for both professionals and administrative personnel. Black women gained strongly in the agency's technical workforce, rising from 31 percent of such employees in 1993 to 48 percent in 1998. Black men are less than fully represented only within EPA's clerical workforce, but they still fall short of the portion of the agency workforce that would be expected based on the national civilian labor force. At the FCC, increases among black men and women in the technical and administrative workforces led the agency's overall percentage of women and minorities to rise from 60 percent to 63 percent.

Few agencies' workforces can be described as less diverse than at the beginning of the Clinton administration, and in most cases they are probably simply running up against the mathematical limits of increasing their diversity. The percentage of women and minorities at the EEOC, for example, dropped slightly from 1993 to 1998 but still stands above 83 percent.

EEOC's Calkins attributed the agency's workforce characteristics to such factors as its applicant pool, made up of people who are likely to have the most intense commitment to its mission. He says that when he makes recommendations to the agency's leadership about its diversity profile, he suggests hiring among Asian attorneys or American Indians if civilian labor force levels could be surpassed with only a few hires in a small agency. He says he is not concerned about the relatively small numbers of white men at the agency.

Diversity and Discrimination

A review of diversity profiles of different agencies and employment categories confirms that there were very real gains in the diversity of the federal workforce during the period from 1993 to 1998. However, there do not appear to be systematic links between the diversity data and the number of discrimination complaints in the federal sector. Rates of filing such complaints appear unrelated to the level of diversity at an agency, the rate of change in diversity, or any other theoretical relationship. Achieving diversity is apparently something very different from eliminating discrimination.

The levels of women and minorities employed by different agencies, and in different employment classifications, in some cases approach mathematical levels that would make additional gains difficult to achieve. This results in odd tensions within some agency workforces. The Social Security Administration, for example, has a workforce that was 71 percent women in 1998. In that environment, both black males (organized as Black Males for Justice) and white males (grouped as Government Men for Equality Now, or G-MEN) have filed class action lawsuits claiming adverse effects from the climate at the agency. The data indicate, however, that black men are fully represented in each of the employment categories examined in this analysis, compared with the private sector.

Human resources officials from several agencies say that a perception of discrimination in the federal workforce persists despite the data. Officials speak of the "overrepresentation of white men" even when women and minorities make up more than 65 percent of the workforce in some agencies. Despite clear data regarding the underrepresentation of white women, they are not beneficiaries of targeted hiring efforts in the government.

As the overall federal workforce becomes more diversified, the EEOC is shifting its focus to the portions of women and minorities in the GS-15 and Senior Executive Service ranks of agencies. As the MSPB's Palguta observes, however, this situation will require a long-term effort to address. Most agencies have very low quit rates, so opportunities to change the demographic characteristics of the workforce occur primarily when older employees retire and are replaced by new hires, usually at the GS-5 to GS-7 levels. Most promotional opportunities occur within the agencies, with very few external hires, so moving today's GS-7 workforce to the Senior Executive Service is a 15- to 20-year process. As Palguta observes, reshaping the demographics of the workforce at this level is not an objective that can be achieved through a focus on hiring of employees.

The EEOC's Calkins says his agency is increasing attention to factors such as training, detail assignments, temporary promotions and other developmental efforts to expand the pool of women and minorities applying for upper-level jobs. Similar efforts by other agencies in recent years may already be paying off. Data that agencies report to EEOC indicate that at the GS-7 level, the federal workforce remained steady at 33 percent male from 1993 to 1998. Men surpass the 50 percent level among federal employees at the GS-10 level, rising at each grade through senior executive. During the five-year interval studied, however, the portion of men in each of these upper grade levels decreased. The SES, for example, dropped from 87 percent male in 1993 to 78 percent in 1998. The relatively low portions of men in junior federal positions, and the high portions of women in many agencies' workforces, could combine to eventually produce a federal executive corps that has more minorities and women.

Shifting Objectives

Although data show the Clinton administration has succeeded in making the federal workforce more diverse, agency officials continue to speak of "doing better" or "improving." However, none of these agencies have clearly defined their goals.

Palguta notes that few agencies have engaged in open discussion of their diversity profiles, and in the absence of open discussion, the information vacuum is filled with conjecture about "doing more" in pursuit of an elusive objective. If the nation's workforce is changing as a result of immigration, different birth rates in different segments of the population, or other long-term demographic factors, the federal workforce will change over time, within the context of merit hiring and promotion. EEOC provides guidance to agencies about diversifying their workforces, and Calkins says agencies can diversify through a variety of approaches.

Nevertheless, a perception that "something is wrong" pervades the federal workforce. After reviewing the data, it seems clear that the absence of a defined objective contributes to the uncertain context of this discussion. Rather than establish targets, however, most federal human resources officials seem inclined to continue the pursuit of an uncertain or changing objective. This will inevitably feed future frustration-and more disagreements among agencies such as MSPB, OPM and the EEOC-as long as policy remains an elusive objective operating in conflict with merit system principles. Under current practices, however, it appears that the merit principle is losing a war of attrition against the diversity commitment. Where these principles are in conflict, we cannot measure progress without comparable measures of workforce quality.

Edward J. Lynch has decades of federal experience at seven agencies and on Capitol Hill. He writes a column on civil service issues each Monday on GovExec.com.