United States Court of Appeals
FOR THE DISTRICT OF COLUMBIA CIRCUIT
Argued October 3, 2014 Decided December 12, 2014
No. 14-5076
ASSOCIATED BUILDERS AND CONTRACTORS, INC.,
APPELLANT
v.
PATRICIA A. SHIU, ET AL.,
APPELLEES
Appeal from the United States District Court
for the District of Columbia
(No. 1:13-cv-01806)
Maurice Baskin argued the cause and filed the briefs for
appellant.
Stephanie R. Marcus, Attorney, U.S. Department of
Justice, argued the cause for appellees. With her on the brief
were Stuart F. Delery, Assistant Attorney General, Ronald C.
Machen, Jr., U.S. Attorney, and Marleigh D. Dover,
Attorney.
Daniel F. Goldstein was on the brief for amici curiae
American Association of People with Disabilities, et al. in
support of appellees.
2
Before: GARLAND, Chief Judge, TATEL, Circuit Judge,
and GINSBURG, Senior Circuit Judge.
Opinion for the Court filed by Circuit Judge TATEL.
TATEL, Circuit Judge: Section 503 of the Rehabilitation
Act of 1973 requires that certain government contractors
“take affirmative action to employ and advance in
employment qualified individuals with disabilities.” Until
recently, the Department of Labor’s implementing regulations
required government contractors to “invite” individuals
offered jobs to advise the contractor whether they believed
they were covered by the Act. Doubting that the existing
regulations were sufficiently advancing the employment of
qualified individuals with disabilities, the Department revised
the regulations to require contractors to extend this invitation
to job applicants, as well as to analyze the resulting data. The
revised regulations also adopt a “utilization goal” to serve as a
target for the employment of individuals with disabilities. In
this case, a trade group representing federal contractors
challenges these regulations, arguing that they exceed the
Department’s statutory authority and are arbitrary and
capricious. The district court rejected both challenges, as do
we.
I.
Congress enacted the Rehabilitation Act of 1973, 29
U.S.C § 701, et seq., “to empower individuals with disabilities
to maximize employment, economic self-sufficiency,
independence, and inclusion and integration into society,” as
well as “to ensure that the Federal Government plays a
leadership role in promoting the employment of individuals
with disabilities.” 29 U.S.C. § 701(b). Section 503 of the Act
provides that government contracts for more than $10,000
3
“shall contain a provision requiring that the party contracting
with the United States shall take affirmative action to employ
and advance in employment qualified individuals with
disabilities.” 29 U.S.C. § 793(a). The statute directs the
President to implement section 503 through regulations, id.,
and the President has delegated that authority to the Secretary
of Labor, who has in turn delegated it to the Office of Federal
Contract Compliance Programs (OFCCP). 41 C.F.R. § 60–
1.2.
The regulations in effect prior to the challenged
rulemaking required contractors to “prepare and maintain an
affirmative action program.” 41 C.F.R. § 60–741.40.
Specifically, the regulations required them to ensure that job
standards do not improperly exclude individuals with
disabilities, to publicize their affirmative-action plan, to
engage in steps to recruit qualified individuals with
disabilities, and to audit the effectiveness of the program. See
Superseded OFCCP Rule on Affirmative Action for Qualified
Individuals with Disabilities, 41 C.F.R. §§ 60–741.40 to –.47
(Effective Prior to Mar. 24, 2014). The regulations also
required contractors to “invite” individuals offered jobs to
inform the contractor if they believed they were covered by
the Act. Id.
By 2010, OFCCP had become concerned that the section
503 regulations were not sufficiently advancing the
employment of qualified individuals with disabilities. See
Affirmative Action and Nondiscrimination Obligations of
Contractors and Subcontractors; Evaluation of Affirmative
Action Provisions Under Section 503 of the Rehabilitation
Act, as Amended, 75 Fed. Reg. 43,116 (July 23, 2010).
OFCCP was especially worried that individuals with
disabilities had lower workforce participation rates and higher
unemployment rates than those without disabilities. Id. After
4
seeking public comment on how to strengthen the regulations,
id., OFCCP issued a Notice of Proposed Rulemaking in
December of 2011. See Affirmative Action and
Nondiscrimination Obligations of Contractors and
Subcontractors Regarding Individuals With Disabilities,
Notice of Proposed Rulemaking, 76 Fed. Reg. 77,056 (Dec. 9,
2011). In response to hundreds of comments on a variety of
issues, OFCCP made some modifications and issued the Final
Rule on September 24, 2013. See Affirmative Action and
Nondiscrimination Obligations of Contractors and
Subcontractors Regarding Individuals With Disabilities, Final
Rule, 78 Fed. Reg. 58,682, 58,685 (Sept. 24, 2013) (to be
codified at 41 C.F.R. pt. 60–741) (“Final Rule”).
The Final Rule makes several significant changes, two of
which are challenged here. First, it obligates contractors to
extend the invitation to self-identify to all job applicants and
to analyze the resulting data. This new requirement is
implemented by section 741.42(a) of the Final Rule, which
requires contractors to invite job applicants to indicate
whether they have a disability, 41 C.F.R. § 60–741.42(a), and
by section 741.44(k), which requires analysis of the data
collected, along with the number of job openings, the total
number of applicants, the number of applicants hired, and the
number of applicants hired who have disabilities, id. § 60–
741.44(k).
Second, section 741.45 of the Final Rule introduces a 7
percent “utilization goal” for the employment of individuals
with disabilities. For employers with 100 or fewer employees,
the goal applies to the employer’s entire workforce, while for
employers with more than 100 employees, the goal applies to
each job group within the workforce. The goal establishes “a
benchmark against which the contractor must measure the
representation of individuals” with disabilities. Id. § 60–
5
741.45. “The goal is not a rigid and inflexible quota which
must be met” but rather “is intended solely as a tool.” Final
Rule at 58,706.
To calculate the utilization goal, OFCCP used data from
the American Community Survey (ACS), a detailed view of
U.S. households produced by the Census Bureau. See Final
Rule at 58,703. OFCCP began by estimating that “5.7% of the
civilian labor force has a disability.” Id. at 58,704. (A member
of the civilian labor force is either presently working or
unemployed and looking for work.) Id. According to OFCCP,
this percentage would be higher absent discrimination on the
basis of disability. Id. at 58,704–06. OFCCP therefore
compared the percent of the civilian labor force with a
disability to the percent of the general population with a
disability who identify as having an occupation, from which it
derived what it called a “discouraged worker” effect of 1.7
percent. Id. Adding that figure to 5.7 percent, OFCCP arrived
at 7.4 percent, which it rounded down to 7 percent in order to
“avoid implying a false level of precision.” Id. at 58,705.
Taken together, these two requirements, OFCCP
explained, are “an important means by which the Government
can contribute to reducing the employment disparity between
those with and without disabilities.” Id. at 58,684. The new
provisions “are designed to bring more qualified individuals
with disabilities into the Federal contractor workforce and
provide them with an equal opportunity to advance in
employment.” Id. at 58,685.
Appellant, Associated Builders and Contractors, Inc.
(ABC), a “national trade association representing” members
from “construction and industry-related firms,” has many
members that are government contractors and therefore
subject to section 503. Appellant’s Br. 15. ABC sued in the
6
United States District Court for the District of Columbia,
challenging the Final Rule as both beyond OFCCP’s statutory
authority and arbitrary and capricious. Id. at 16. Rejecting
both arguments, the district court granted summary judgment
to OFCCP. Associated Builders & Contractors, Inc. v. Shiu,
No. 13–1806, 2014 WL 1100779 (D.D.C. Mar. 21, 2014). We
review the district court’s grant of summary judgment de
novo, “according no particular deference to the judgment of
the District Court.” Association of Private Sector Colleges
and Universities v. Duncan, 681 F.3d 427, 440–41 (D.C. Cir.
2012) (citation and internal quotation mark omitted).
II.
Because the Rehabilitation Act vests the executive branch
with rulemaking authority, we proceed under the familiar
two-step framework of Chevron, U.S.A., Inc. v. Natural
Resources Defense Council, 467 U.S. 837 (1984). In
accordance with that decision, we determine first “whether
Congress has directly spoken to the precise question at issue,”
and “if the statute is silent or ambiguous with respect to the
specific issue,” we ask whether the agency’s interpretation “is
based on a permissible construction of the statute.” Id. at 842–
43.
For ABC “to prevail under Chevron step one, [it] must do
more than offer a reasonable or, even the best, interpretation;
it must show that the statute unambiguously forecloses”
OFCCP’s interpretation. Village of Barrington, Illinois v.
Surface Transportation Board, 636 F.3d 650, 661 (D.C. Cir.
2011). ABC argues that the word “qualified” as used in
section 503—“take affirmative action to employ and advance
in employment qualified individuals with
disabilities”—expressly limits affirmative action to
individuals already offered jobs. But that word does no such
7
thing. It does not modify “affirmative action,” nor does
anything in section 503 limit “affirmative action” to those
offered jobs. Rather, the word “qualified” describes the
statute’s beneficiaries—“qualified individuals with
disabilities.” In fact, the provisions of the final rule ABC
challenges are all expressly designed to promote the
“employ[ment] and advance[ment] in employment [of]
qualified individuals.” 29 U.S.C. § 793 (emphasis added).
Undaunted by the statute’s plain language, ABC invokes
other evidence to make its case. Observing that “Congress
repeatedly amended the Act without expressing any
disapproval of OFCCP’s implementation of ” the statute, it
argues that “[b]oth the Supreme Court and this Court have
repeatedly held that Congressional re-enactment of a statute
without pertinent change to an agency’s longstanding
interpretation of it is persuasive evidence that the
interpretation is the one intended by Congress.” Appellant’s
Br. 28 (citation and internal quotation marks omitted).
Although this is certainly true in principle, in this case
OFCCP never issued a limiting “interpretation” that Congress
could have endorsed via silence. Although the previous
regulations included neither a pre-job-offer data-collection
requirement nor a utilization goal, OFCCP never said it
lacked authority to include such requirements or that it would
not do so in the future. In other words, although OFFCP did
not make use of its full panoply of powers with the earlier
regulations, “powers . . . are not lost by being allowed to lie
dormant.” Altman v. SEC, 666 F.3d 1322, 1327 (D.C. Cir.
2011) (quoting United States v. Morton Salt Co., 338 U.S.
632, 647 (1950)). Indeed, were ABC correct, agencies would
be unable to strengthen regulations implementing statutes that
Congress has amended. This is simply not how administrative
law works.
8
ABC grounds its next argument in the Vietnam Era
Veterans Readjustment Assistance Act (VEVRAA), also
administered by OFCCP, which expressly requires contractors
to report data on the veteran status of new hires. According to
ABC, this “expression of Congressional intent to delegate
authority to an agency to engage in an activity . . . in very
similar legislation to the statute at issue, combined with the
Congressional failure to include such authorization in the
challenged statute itself, [is] compelling evidence as to
Congressional intent.” Appellant’s Br. 30. But that situation
differs from the one we face here. Nothing in VEVRAA's
original language called for data reporting, but OFCCP
required it nonetheless. Only after OFCCP discontinued the
requirement did Congress amend VEVRAA with language
“motivated by Congress’s desire to restore OFCCP’s prior
practice of requiring similar reports by regulation.”
Associated Builders & Contractors, 2014 WL 1100779 at *8.
The VEVRAA amendment thus tells us nothing about the
issue in this case.
In a footnote, ABC also offers a Chevron step two
argument, but it is the same as its step one argument. It fails
for the same reason.
III.
Turning to ABC’s arbitrary and capricious challenge, we
must first consider the association’s argument that these
regulations are subject to heightened review under FCC v.
Fox Television Stations, Inc., 556 U.S. 502 (2009). There, the
Supreme Court held that when a change in agency policy
“rests upon factual findings that contradict those which
underlay its prior policy . . . a reasoned explanation is needed
for disregarding facts and circumstances that underlay” the
prior policy. Fox, 556 U.S. at 515–16. According to ABC,
9
OFCCP has found “that the ACS survey data was somehow
sufficient” to set a utilization goal, and this conflicts with its
earlier finding that it had insufficient data to set such a goal.
Appellant’s Br. 37. This is inaccurate. Prior to the challenged
rulemaking, OFCCP never found that setting a utilization goal
was infeasible; indeed, nothing in the administrative record
suggests that it even considered setting such a goal. In other
words, no prior factual finding conflicts with the finding
underlying the challenged Rule, i.e., that the ACS provides a
feasible basis for calculating a utilization goal. Given this, we
shall proceed in accordance with the normal arbitrary and
capricious standard. See 5 U.S.C. § 706(2)(A). “The scope of
review . . . is narrow and a court is not to substitute its
judgment for that of the agency. Nevertheless, the agency
must examine the relevant data and articulate a satisfactory
explanation for its action including a rational connection
between the facts found and the choice made.” Motor Vehicle
Manufacturers Association of the United States, Inc. v. State
Farm Mutual Automobile Insurance Co., 463 U.S. 29, 43
(1983) (citation and internal quotation marks omitted).
ABC advances several arbitrary and capricious
challenges. For purposes of our analysis, we have grouped
them into four categories.
ABC first argues that OFCCP has failed to explain the
need for the Final Rule. Specifically, “OFCCP does not claim
that the lack of improvement [in the employment of
individuals with disabilities] exists among government
contractors . . . but only that a continuing disparity exists in
the workforce population as a whole.” Appellant’s Br. 35. But
OFCCP had no obligation to make such a particularized
finding. Rather, it was permitted to infer the existence of
employment barriers from its analysis of the workforce as a
whole without “a finding in each case that the status quo is
10
discriminatory,” Allen v. Heckler, 780 F.2d 64, 68 (D.C. Cir.
1985).
Next, ABC challenges the requirement that contractors
collect data from all job applicants instead of from just those
offered jobs. According to ABC, although “newly hired
employees are presumably qualified for the positions to which
they have been hired,” the “new data collection on mere job
applicants is meaningless, because there is no way to tell
whether the applicants measured are qualified or not.”
Appellant’s Br. 33. If this argument sounds familiar, that’s
because it reprises the statutory argument we have already
rejected. As explained above, supra at pp. 5–6, the word
“qualified” refers to the beneficiaries of affirmative action; it
does not limit the kind of affirmative action OFCCP can
require. In a related argument, ABC contends that OFCCP
failed to explain how the new data collection “will enable
anyone to better monitor or evaluate contractors’ hiring of
qualified individuals with disabilities.” Appellant’s Br. 33–34.
But doing just that, OFCCP explained that “[m]aintaining this
information will provide meaningful data to assist the
contractor in evaluating and tailoring its recruitment and
outreach efforts.” Final Rule at 58,701. Absent such data, it is
“nearly impossible for the contractor and OFCCP to perform
even rudimentary evaluations of the availability of individuals
with disabilities in the workforce, or to make any sort of
objective, data-based assessments of how effective contractor
outreach and recruitment efforts have been in attracting
individuals with disabilities as candidates.” Id. OFCCP has
more than satisfied its obligation to provide a reasoned
explanation and to draw a connection between the problem
(the low workforce participation of individuals with
disabilities) and the regulatory solution (more refined data
collection).
11
ABC next challenges the utilization goal, pointing out
that the ACS does “not use the same definition of disabilities
as the new Rule,” does not break down the data by industry or
geography, and “could not possibly have surveyed whether
the disabled workers in question were ‘qualified’ for jobs in
different industries in any particular percentages.” Appellant’s
Br. 35–36. Of course, OFCCP knew all of this. As to ABC’s
first point, OFCCP acknowledged that “[t]he definition of
disability used by the ACS . . . is clearly not as broad as that
of the Rehabilitation Act,” Final Rule at 58,703, and, if
anything, this difference would result in an underestimate of
the size of the population with disabilities. OFCCP also
explained that its decision to set a single national goal rested
on the fact that since “the ACS disability data is based on
sampling, and because the percentage of that sample who
identify as having a disability is [small], it cannot be broken
down into as many job titles, or as many geographic areas as
the data for race and gender.” Id. at 58,704. What’s more,
based on the geographic data that OFCCP did have, it
observed that there was an “almost uniform distribution” of
individuals with disabilities and explained that “[t]his general
uniformity is consistent with the use of a single national
goal.” Id. at 58,704 n. 24.
With respect to ABC’s complaint that the ACS is
incapable of measuring the number of qualified individuals
with disabilities in particular industries, we are unsure how
the survey could do that since job qualifications vary from
position to position and industry to industry. ABC believes
this means that the use of any survey data is inappropriate,
since it could include individuals who are unqualified for
particular jobs. Of course, there may be fewer individuals
with disabilities who are qualified to perform certain jobs, just
as there are fewer individuals without disabilities who are
qualified to fill some positions because skills are unevenly
12
distributed across the labor force. As mentioned above,
OFCCP determined that, based upon the ACS data, 5.7
percent of the civilian labor force has a disability. See supra
p. 5. OFCCP also determined that an additional 1.7 percent of
the population has a disability and an occupation, but is not
presently seeking employment. See supra p. 5. It reasoned
that many people who are working, actively looking for work,
or identify as having an occupation are qualified to perform at
least some jobs that might be offered by a federal contractor.
Final Rule at 58,705–06. Although both ABC and OFCCP
might prefer a utilization goal that accounts for variations in
the number of qualified individuals with a disability by
industry or job type, the agency adequately explained why the
best available data did not allow it to create a tailored goal
and why the uniform goal advances its regulatory objective.
See WorldCom, Inc. v. FCC, 238 F.3d 449, 461–62 (D.C. Cir.
2001) (“[T]he [agency] is not required to identify the optimal
threshold with pinpoint precision. It is only required to
identify the standard and explain its relationship to the
underlying regulatory concerns.”).
ABC also challenges the way in which OFCCP
calculated the utilization goal. Specifically, it objects to the
“discouraged worker effect” because, it says, OFCCP
“rejected without any evidence the likelihood that a
significant number of such workers were unable to work
because of the disqualifying nature of their disabilities.”
Appellant’s Br. 38. But OFCCP knew that the
underemployment of individuals with disabilities could have
different causes and concluded “that at least a portion of this
gap is due to discrimination.” Final Rule at 58,706.
Furthermore, OFCCP recognized that “[w]hile not perfect, the
goal will provide a yardstick against which contractors will be
able to measure the effectiveness of their equal employment
opportunity efforts.” Id.
13
Finally, ABC argues that, for several reasons, OFCCP
should have exempted the construction industry from the
Final Rule. Explaining that the industry is “uniquely
hazardous and physical compared to other industries,” ABC
insists that “[i]n this environment, decisions to hire and/or
employ disabled individuals must be made on a case by case
basis, without regard to statistics, in order to determine the
ability of each individual to perform the essential functions of
particular construction jobs.” Appellant’s Br. 40. ABC
believes the construction industry will find it especially
difficult to comply with the Final Rule because the fluid and
transitory nature of its workforce makes it hard to perform
utilization-goal analysis on a job-group basis. Construction
contractors, ABC also tells us, have “no experience” with
job-group analysis because under Executive Order 11246,
which requires affirmative action in the hiring of women and
minorities, construction contractors are required to perform
utilization-goal analysis only on an employer-wide basis. Id.
at 39–40.
None of these arguments demonstrates that OFCCP acted
arbitrarily and capriciously by failing to exempt the
construction industry from the Final Rule. For one thing, the
Final Rule does not prohibit employers from making
case-by-case hiring decisions based on the qualifications of
each individual. As OFCCP emphasized, nothing in the Final
Rule “require[s] a contractor to hire an individual who cannot
perform the essential functions of [a] job.” Final Rule at
58,707. ABC, moreover, never explains how the fluidity of
the construction industry workforce makes job-group analysis
so burdensome as to require an industry exemption, especially
given that the Final Rule exempts small contractors from the
job-group requirement. Id. at 58,709. And ABC’s final
argument proves too much, as the “no experience” claim
14
would doom virtually any regulation that imposes new
obligations on regulated entities.
We end as we began by emphasizing that our review of
an agency’s exercise of its rulemaking authority is narrow.
Judicial review exists to ensure that agency actions are the
“product of reasoned decisionmaking.” Fox v. Clinton, 684
F.3d 67, 75 (D.C. Cir. 2012). Here, ABC points to nothing in
the rulemaking that suggests OFCCP flunked this highly
deferential standard.
IV.
For the foregoing reasons, we affirm.
So ordered.