A Solution in Search of a Problem? Discrimination,

Sally Coleman Selden
Lynchburg College

Theory to
Practice

A Solution in Search of a Problem? Discrimination,
Affirmative Action, and the New Public Service

Sally Coleman Selden is an associate
professor at Lynchburg College. Her current
research focuses on strategic human
resource management in state governments
and the impact of collaboration on
nonprofit organizational effectiveness.
She has published articles in the American

ne of the defining challenges of the ongoing
transition from government-centered to
multisectored models of new governance
in the 21st century is maintaining the values that one
cherishes in a democracy. One value that has
historically been pursued in the administrative state
is attaining a workforce that “looks like America.”
For some, this means having a public workforce that
reflects America’s diversity (passive representation),
with less concern about whether that representation
influences the substantive policy outputs of public
agencies (active representation). To others, looking
like America is important so that all groups—
especially historically underrepresented or excluded
groups—can actively promote and implement public
policies that reflect the needs, values, and aspirations
of the groups they represent.

Journal of Political Science, Administration
& Society, American Review of Public
Administration, Review of Public Personnel
Administration, Journal of Public
Administration Education, Public
Administration Review, and Journal of
Public Administration Research and Theory.
E-mail: selden@lynchburg.edu.

O

Passive representation is viewed as a worthwhile aim
of public policy because the public service has always
been a vehicle of social mobility for groups that
historically have been disadvantaged in the labor
market, and it implies a symbolic commitment to
equal access to power. Furthermore, the bureaucracy
is, and should be, more reflective of the economic
stature of most Americans than elected officials
because of its role in the policy process. Moreover, in a
global economy—and with a foreign policy premised
on human rights and dignity—lacking a diverse
workforce is embarrassing to the United States,
undermines its credibility, and hurts its business
success in dealing with other nations. Meanwhile,
active representation is valued because the operations
and policies of a demographically diverse agency will
look very different from what they would be if the
agency’s workforce were homogeneous.
To these ends, for over a half-century, the United
States has pursued a more diverse workforce, first
through an emphasis on equal opportunity and later
through affirmative action policies (see table 1 on the
PAR Web site). These efforts have come through an
amalgam of federal and state legislation, executive

orders, administrative rules, and judicial decisions.
With the exception of the abortion issue, this
morphing from equal employment opportunity to
an affirmative action (henceforth, AA) approach to
representation has ignited the most heated and
divisive controversies over social policy that this
nation has endured in the 20th century.
The battle has been led by both passionate advocates
and opponents (Holzer and Neumark 2000a; Jones
2005; Sabbagh 2003). To some proponents, moving
away from AA as the primary tool for bringing about
both passive and active representation in the United
States would violate the nation’s constitutional
covenant with its citizens. It is misguided, they argue,
to ignore the fact that stubborn remnants of race,
ethnicity, and gender remain in the hiring, promotion,
and retention decisions in the public, private, and
nonprofit sectors of this nation. Thus, relaxing AA
pressures would only undermine the progress made
over the past four decades in these areas. Moreover,
given the movement from government-centered to
multisectored models of governance that has taken
place in recent years, some worry that private and
nonprofit organizations may be less focused on
promoting diversity in their workforces, thereby
compromising the nation’s aspirations toward equal
opportunity for all Americans.
To opponents, however, AA is an equally divisive form
of reverse discrimination that is not needed precisely
because of the progress that has been made in the
antidiscrimination area. They are joined by others
who argue that AA, at least on the basis of race and
ethnicity, makes little sense given the changing racialethnic composition of today’s (and likely tomorrow’s)
population. To others, the continued migration of
workers from around the world into the United
States, coupled with the demographic projections of
the U.S. labor force by race and ethnicity, suggest we
will no longer be able to classify U.S. workers using
the five traditional racial and ethnic groups. Others in
this “beyond AA” camp argue that recruitment is not

Discrimination, Affirmative Action, and the New Public Service 911

the problem (i.e., the recruitment “pipeline” is full);
rather, the problem is the retention of diverse workers.
Diversity management, therefore, is the key to growing, nurturing, and empowering a heterogeneous
workforce.
Has AA’s time passed, either politically, substantively,
or in impact? A complete review of the host of questions that must be answered to inform such an inquiry
would be impractical. Consequently, this article
focuses on what prior research tells us about four
important dimensions of this question that beg
understanding before an informed answer can be
given. First, what does prior research tell us about
where the public stands on AA? Second, what does it
tell us about where AA presently stands in terms of
state legislatures, the courts, and civil society? Third,
as a guide to arguing the merits and demerits of AA in
the future, what does prior research tell us about what
AA has accomplished so far in light of its passive and
active representation goals? Finally, in an era of new
governance, does a movement from governmentcentered to multisectored workforces compromise the
aims of AA if it is dismantled?
Reviewing a robust and still growing body of research
related to these questions, this article culls eight
general lessons related to AA that practitioners and
researchers should ponder as they work with or study
the future of AA in the new governance era. Perhaps
the most important lesson for society more broadly is
that AA is not a solution that is chasing a problem
from a bygone era. Today’s residual problems will
likely require redefinition and new solutions—a
refocus that will affect public, private, and nonprofit
organizations as the United States moves from a
government-centered to a multisectored model of
public service.
Divided It Falls?
Where do Americans stand politically on AA? Neither
practitioners nor scholars will be surprised to learn
that public opinion is highly divided on this issue. But
the real questions are how consistent are these polls,
whether citizens are more or less divided in light of the
broader discussions they have heard in the media over
the years, and whether opinion is fixed or has shifted
over time. Prior research indicates that the durability
of AA rests on a precarious and declining base of
public support in the United States, with racial gaps in
perceptions that are real, enduring, and divisive.
Lesson 1: Citizens remain conflicted about AA, support
for AA varies slightly across time and polls, and the
results vary across time and by race. As illustrated in
greater detail and with graphs in the longer version of
this article on the PAR Web site (see figures 1–3), polling
results indicate that support for AA varies slightly across
time and opinion polls (Jones 2005). A comparison of
912

Public Administration Review • November | December 2006

public opinion polls administered by CBS News and
the Gallup Organization demonstrate this fluctuation.
Compared to 2003 polling data, the results of a January
2006 CBS poll suggest that the American people are
growing impatient with AA programs (CBS News
2006). The level of support for continuing AA programs
has decreased substantially over the past three years. In
2003, the polling organization found that 53 percent
of people surveyed believed that AA programs should
be continued (CBS News 2003). In 2006, however,
only 36 percent of persons surveyed believed that AA
should be continued, a decrease of 17 percentage points
from 2003.
In contrast, respondents to Gallup’s annual survey of
minority rights and relations in 2005 were slightly
more supportive of AA programs for racial minorities
(50 percent) than opposed to such programs (42
percent). Moreover, between 2001 and 2005, the
percentage of people opposed to AA programs actually
decreased, albeit marginally, from 44 percent to 42
percent ( Jones 2005). Because the percentage of people who wanted to see a decrease in AA programs fell
from 37 percent in 1995 to 26 percent in 2003, Jack
Ludwig, director of research for Gallup Poll Social
Audits, argued that the public was warming to AA
(Ludwig 2003).
What has been consistent over the years, however, is
an enduring racial gap in perceptions of AA, with
minorities supporting AA by a margin of at least
20 percentage points in five annual CBS and Gallup
surveys (see figures 2 and 3 in the PAR Web version;
CBS News 2006; Jones 2005; Ludwig 2003). For
example, a 2003 Gallup Poll revealed that African
Americans and Hispanics were significantly more
supportive of AA programs than non-Hispanic whites.
In that survey, 70 percent of African American and
63 percent of Hispanic respondents favored AA,
compared to 44 percent of white respondents. More
in-depth exploration of this perceptual gap has led
some to conclude that the differences between African
Americans’ and whites’ perceptions of AA programs
“likely stem from the belief among a majority of
whites (59 percent) that blacks in this country have
equal job opportunities with whites, while only
23 percent of blacks agree. Roughly three in four
blacks believe that they do not have equal job opportunities in this country” ( Jones 2005, 2). Moreover,
the percentage of African Americans who believed
that the government should make every effort to help
blacks and other minorities increased from 59 percent
to 67 percent between 1997 and 2003 (see figure 3 in
the PAR Web version). The trend for white respondents in the CBS and Gallop polls differed slightly,
however, with the percentage supporting federal
government efforts increasing from 34 percent
in 1997 to 41 percent in 2001 but decreasing to
36 percent in 2003.

Affirmative Action on the Defensive: The
View from the Courts, Legislatures, and
Governors’ Offices
Against this leitmotif of eroding support for AA
among large segments of the American public, and
perhaps helping to coalesce it, recent years have seen a
sustained attack on AA in Washington, in the states,
and in the courts. In the process, AA has seen limitations placed on its scope, with narrow tailoring of its
application—part of the test that public managers
must anticipate as they work in this area. Collectively,
these developments do not bode well for the future of
AA as we know it.
Lesson 2: A narrowing of AA by the courts has been
under way since the late 1980s, though with some
protections added in higher education. At the federal
level, the seeds of President Ronald Reagan’s efforts to
appoint federal judges “opposed to regulation in general and AA in particular” (Kelly and Dobbin 2001,
95) blossomed after Reagan left office (Kellough 1989;
Naff 2004). Before then—although many of the
decisions were split—AA typically received (with few
exceptions) the support of the U.S. Supreme Court
(Kellough 1989; Naff 2004). For example, the Court
ruled that colleges and universities were permitted to
use race or ethnicity as a flexible factor in its admissions decisions. Likewise, public and private employers could voluntarily employ AA programs to address
racial, ethnic, or gender imbalances in their workforces. Moreover, the U.S. Congress could require AA
programs from employers receiving federal funding.
Since then, however, AA has largely been on the
defensive in the federal courts. The Supreme Court’s
first major decision in this regard challenged the legal
standing of AA programs at the state and local levels
by requiring programs to pass a more demanding and
stringent level of scrutiny by the courts. More precisely,
in Richmond v. J. A. Croson Co. (488 U.S. 469 [1989]),
the Court ruled that the city of Richmond’s 30 percent
set-aside program for minority-owned construction
firms was “an unyielding racial quota.” The Court
established that AA programs at the state and local
levels should be subject to strict scrutiny by the courts.
In Metro Broadcasting, Inc. v. FCC (497 U.S. 547
[1990]), Justice William J. Brennan, in his majority
decision, held that the minority preference policies
of the Federal Communications Commission did
not violate the equal protection clause of the Fifth
Amendment. Moreover, unlike the Croson decision,
which focused on a local government’s use of preferences, the Court even found that AA plans adopted
by Congress are not subject to strict scrutiny. But
the justices’ support for legislatively mandated AA
programs was temporary. In Adarand Constructors, Inc.
v. Peña (515 U.S. 200 [1995]), the Supreme Court
actually struck down its Metro Broadcasting decision

and ruled that the strict standards established in
Croson were the proper standard for examining federal
set-aside programs. Although the Court found that
federal AA programs need to serve “a compelling
governmental interest,” the majority decision affirmed
that the use of AA may be justified in particular
instances, such as when systematic discrimination
exists. Importantly, the aforementioned rulings do
not preclude local, state, or federal governments from
using AA programs. However, the decisions do “raise
the standards for their legal justification” (Holzer and
Neumark 2000a, 491).
In the aftermath of these decisions, the Clinton
administration responded quickly, recognizing that
“federal HRM affirmative action for minorities had
become constitutionally questionable” (Naylor and
Rosenbloom 2004, 151). Indeed, although the Adarand decision pertained only to government contracting, the administration instructed all federal agencies
and departments to apply the decision to both contracting and employment (Naff 2004; Naylor and
Rosenbloom 2004). Moreover, and at about the same
time, two important circuit court decisions called AA
into question.
In Podberesky v. Kirwan (38 F.3d 147 [1994]), the
U.S. Court of Appeals for the Fourth Circuit ruled
that the University of ’s Banneker scholarship program, which was limited to African Americans,
was unconstitutional. Two years later, the Fifth Circuit
Court of Appeals’ decision in Hopwood v. University of
Texas (78 F.3d 932 [1996]) prohibited consideration
of race or ethnicity in admissions decisions, even for
the express purpose of creating a diverse student body.
According to the court, diversity in higher education
did not represent a compelling state interest (AAUP
2006). The Fifth Circuit Court of Appeals held that
although addressing past discrimination was a compelling state interest, discrimination in the system in
general was not. The Supreme Court let the Hopwood
decision stand without review. More recently,
however, the Supreme Court ruled more positively
on the application of AA to higher education in
Gratz v. Bollinger (539 U.S. 244 [2003]) and Grutter
v. Bollinger (539 U.S. 306 [2003]). It has done so,
however, in ways that are decidedly more constraining
than during the pre-Reagan era.
Lesson 3: Grassroots legislative actions, ballot initiatives, and executive orders threaten AA, but state
legislators and governors are not without the tools to
hold its demise at bay. Prompted especially by the
Fifth Circuit Court’s decision, grassroots anti-AA
movements began growing in some states and were
quickly joined by legislative and gubernatorial initiatives. Some residents and elected officials, for example,
have taken action at the state level to curb AA through
legislation, ballot initiatives, and executive orders

Discrimination, Affirmative Action, and the New Public Service 913

(Americans for a Fair Chance 2005). Two states,
California (Proposition 209) and Washington (I-200),
passed citizen initiatives in 1996 and 1998, respectively, prohibiting state and local governments from
discriminating against or granting preferential treatment to any individual or group based on race, sex,
color, ethnicity, or national origin. Again, however,
racial and ethnic disparities in voting patterns were
readily apparent. Overall, California voters approved
Proposition 209 by a 10-point margin (55 percent
supported the measure and 45 percent opposed it).
But polling results suggest that 61 percent of the
white voters supported Proposition 209, compared to
only 25 percent of African American and 28 percent
of Hispanic voters (Hardy-Fanta 2000).
Though citizen initiatives have been successful in only
two states, ballot initiative campaigns led by Ward
Connerly, a former member of the California Board
of Regents and the leader of Proposition 209, may be
forthcoming in Arizona, Colorado, Missouri, and
New Hampshire (Americans for a Fair Chance 2005).
Moreover, even when citizen initiatives have failed or
have not reached the ballot, some governors have
taken unilateral action to end, limit, or redefine AA.
In Florida, for example, Governor Jeb Bush ended AA
in the state’s employment, contracting, and education
systems by issuing Executive Order 99-201, titled the
“One Florida Initiative” (U.S. Commission on Civil
Rights 2000). Still, since 2000, Florida is the only
state that has eliminated the use of AA through
executive order.
By the same token, however, legislative supporters of
AA can work with sympathetic governors to limit the
impact of anti-AA campaigns and court decisions. In
response to the dismantling of AA in higher education
admissions by the courts, for example, legislatures in
California, Florida, and Texas adopted a new tool of
public action, “percentage plans,” to address equal
educational opportunity (see table 3 in the Web
version of PAR). Though these plans differ somewhat,
they all grant top-graduating high school students
admission to the state university system.
Adopted in response to the Hopwood decision, for
example, Texas’s “Ten Percent Plan” allows students
who graduate in the top 10 percent of their class
admission to the University of Texas system. Similarly,
but more narrowly, California adopted a plan to
guarantee the top 4 percent of students graduating
from California’s public schools admission to the
University of California system (AAUP 2006). Meanwhile, Florida’s “Talented 20 Program” allows the top
20 percent of graduating seniors from Florida high
schools admission to one of the University of Florida’s
public institutions (AAUP 2006). Likewise, Governor
Ruth Ann Minner (D) issued an executive order in
2001 supporting the use of AA programs in Delaware,
914

Public Administration Review • November | December 2006

and Oregon governor Ted Kulongoski (D) made his
support of AA clear in both his 2005 State of the State
address and by executive order.
Has Affirmative Action Reached Its Goals?
After more than 30 years of implementation, what has
AA actually accomplished? Although the majority
of the empirical research demonstrates the positive
impact of AA policies on educational and employment opportunities for minorities and women (Chay
1998; GAO 1991; Goldstein and Smith 1976; Holzer
and Neumark 2000a, 2000b; Kellough 1990a, 1990b;
Leonard 1990; Naff 2001; Naylor and Rosenbloom
2004; Rodgers and Spriggs 1996), some studies have
raised questions about its true benefits (Bowen and
Bok 1998; Datcher, Garman, and Garman 1993,
1995; Davidson and Lewis 1997; Kane 1998; Sander
2004; Vars and Bowen 1998). Overall, one finds
research support for progress on the critical pipeline
dimension of educational opportunity. However, there
is decidedly less significant advancement when it
comes to other major goals such as hiring, equal pay,
and the elimination of discrimination once minorities
and women are hired. Therefore, it seems that future
progress still requires the kind of “pressure of court
decisions, legislation, executive action, and the power
of examples in the public and private sector” that was
so essential to the progress made since the 1960s
(Clinton 1995).
Lesson 4: Substantial and important progress is being
made in the critical area of educational opportunities,
but the retention and academic performance measures
of some minority students are disappointing and suggest that additional or different efforts are needed. The
impact of affirmative action on educational opportunity has been researched extensively, and the impact is
clear: More women and minorities are represented on
college campuses and in professional programs today
than prior to the use of AA (Stephanopoulos and
Edley 1995). Studies examining the link between AA
and student performance, however, are less compelling for minorities; on average, minority students do
not perform as well as white students. On the other
hand, women perform better academically than men
(Berkner, Cuccaro-Alamin, and McCormick 1996;
Bierman 2006; Clune, Nuñez, and Choy 2001;
NCES 1999).
Between 1960 and 2003, men enrolling in college
after graduating from high school increased by 7.2
percentage points, from 54 percent to 61.2 percent.
During this same period, the percentage of women
going to college after high school grew more rapidly.
In 1960, about 38 percent of women graduating
from high school went to college, compared to 66.5
percent of women in 2003 (NCES 2004a). According
to U.S. Department of Education data (NCES
2004b), between 1974 and 2003, a larger percentage

of graduating high school students across racial and
ethnic groups went to college. Specifically, although
white enrollment rose from 47.8 percent to 66.2
percent of the available pool during that period,
African American enrollment rose from 32.5 percent
to 57.5 percent. Hispanic enrollment improved only
marginally, from 54.1 percent to 58.6 percent. Thus,
the news—though positive in terms of AA goals
on this dimension—is not all good. For example,
over a 30-year period, there was a slightly higher rate
of growth of African Americans going to college
(25.0 percent) than for white students (18.4 percent).
But the rate of growth for Hispanics going to college
during this same period was significantly lower
(4.5 percent) than the growth rate for whites and
African Americans. Still, these data indicate that
minority students constitute a much higher percentage of college and university students today. In 2002,
29.4 percent of undergraduate students were
minorities, compared to 15.4 percent in 1976
(NCES 2004c).
Prior research also finds that women and minorities
have increased their representation in graduate and
professional schools during the last 25 years (NCES
2004c, 2004d). In 1969, women represented 38
percent of graduate school enrollments, compared to
58 percent of graduate students in 2003 (NCES
2004e). Between 1976 and 1977, whites received
91.7 percent of first professional degrees awarded;
African Americans received 4 percent, Hispanics
1.7 percent, Asians 1.6 percent, and Native Americans
0.3 percent. By 2002–03, however, minorities were
receiving a larger share of first professional degrees
conferred: African Americans received 7.1 percent,
Hispanics received 5.1 percent, Asians received
12.1 percent, and Native Americans received 0.7
percent (NCES 2004d). The progress of women in
professional programs has been even more impressive.
Whereas women received 19 percent of first professional degrees issued in 1976–77, they received
48 percent of first professional degrees awarded in
2002–03 (NCES 2004d). Likewise, in 1955–56,
women received only 1.1 percent, 5.1 percent, and
3.9 percent of dental, medical, and law degrees
conferred, respectively. In 2002–03, however, women
received 38.9 percent, 45.3 percent, and 49 percent
of dental, medical, and law degrees awarded, respectively, in these fields.
Prior research also reveals, however, a controversy
over what these otherwise impressive gains mean in
terms of academic performance (e.g., Bowen and Bok
1998; Datcher, Garman, and Garman 1993, 1995;
Davidson and Lewis 1997; Kane 1998; Sander 2004;
Vars and Bowen 1998). For example, a recent article
by Sander (2004) raises and validates a concern of
many AA critics—that is, AA results in minority
applicants who are less qualified being accepted to

more elite universities. Assessing grade point averages
and graduation rates, Sander (2004) argues that many
African American students struggle in law school
academically and fail at higher rates than they would
have if preferences had not been awarded.
Without AA, Sander contends that A…

Explanation & Answer

Our website has a team of professional writers who can help you write any of your homework. They will write your papers from scratch. We also have a team of editors just to make sure all papers are of HIGH QUALITY & PLAGIARISM FREE. To make an Order you only need to click Order Now and we will direct you to our Order Page at Litessays. Then fill Our Order Form with all your assignment instructions. Select your deadline and pay for your paper. You will get it few hours before your set deadline.