Fair Lines America Foundation Inc. v. US Department of Commerce

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
____________________________________
)
FAIR LINES AMERICA )
FOUNDATION INC., )
)
Plaintiff, )
)
v. ) Civil Action No. 21-1361 (ABJ)
)
US DEPARTMENT )
OF COMMERCE, et al., )
)
Defendants. )
____________________________________)
MEMORANDUM OPINION
Fair Lines America Foundation, Inc. brought this Freedom of Information Act (“FOIA”)
lawsuit against the Bureau of the Census and Department of Commerce on May 18, 2021. See
Compl. [Dkt. # 1]. Both parties have filed motions for summary judgment. See Mem. of P. & A.
in Supp. of Defs.’ Mot for Summ. J. [Dkt. # 13-1] (“Defs.’ Mot”); Mem. of P. & A. in Supp. of
Pl.’s Combined Opp. to Defs.’ Mot. and Cross-Mot. for Summ. J. [Dkt. # 14-1] (“Pl.’s Mot.”).
The motions are fully briefed. See Defs.’ Reply in Supp. of Defs.’ Mot. and Opp. to Pl.’s Mot.
[Dkt. # 17] (“Defs.’ Reply”); Pl.’s Reply in Supp. of Pl.’s Mot. [Dkt. # 19] (“Pl.’s Reply”). For
the following reasons, plaintiff’s motion will be DENIED, and defendants’ motion will be
GRANTED.
BACKGROUND
Every ten years, defendants are tasked with conducting the census: a count of the
population of the United States. This duty is mandated by the United States Constitution, which
“requires an ‘actual Enumeration’ of the population every 10 years and vests Congress with the
1
authority to conduct that census ‘in such Manner as they shall by Law direct.’” Wisconsin v. City
of New York, 517 U.S. 1, 5 (1996), quoting U.S. Const. art. I, § 2, cl. 3.
Defendants conducted the most recent census in the year 2020. As the Supreme Court has
recognized, the census is a massive undertaking, but it is inevitably imperfect. “Although each
[census] was designed with the goal of accomplishing an ‘actual Enumeration’ of the population,
no census is recognized as having been wholly successful in achieving that goal.” Wisconsin,
517 U.S. at 6.
Despite consistent efforts to improve the quality of the count, errors persist.
Persons who should have been counted are not counted at all or are counted
at the wrong location; persons who should not have been counted (whether
because they died before or were born after the decennial census date,
because they were not a resident of the country, or because they did not
exist) are counted; and persons who should have been counted only once
are counted twice.
Id. One particular issue is that a lack of responses can – and inevitably does – lead to an
“undercount” of the population. Id. at 7. “The Census Bureau has recognized the undercount as
. . . [a] significant problem[], and in the past has devoted substantial effort toward achieving [its]
reduction.” Id. With this goal in mind, the Census Bureau does not simply rely on survey
responses to count the population; it has devised methods to improve the accuracy of its count and
to avoid undercounting.
One of these methods is that the Bureau “imputes” numbers based on other information it
has collected. The Supreme Court has explained that the Census Bureau utilizes
different kinds of “imputation” depending upon the nature of the missing or
confusing information. Where, for example, the missing or confused
information concerns the existence of a housing unit, the Bureau speaks of
“status imputation.” Where the missing or confused information concerns
whether a unit is vacant or occupied, the Bureau speaks of “occupancy
imputation.” And where the missing or confused information concerns the
number of people living in a unit, the Bureau refers to “household size
2
imputation.” In each case, however, the Bureau proceeds in a somewhat
similar way: It imputes the relevant information by inferring that the
address or unit about which it is uncertain has the same population
characteristics as those of a “nearby sample or ‘donor’” address or unit —
e.g., its “geographically closest neighbor of the same type (i.e., apartment
or single-family dwelling) that did not return a census questionnaire” by
mail.
Utah v. Evans, 536 U.S. 452, 458 (2002) (emphasis omitted).
One way defendants sought to obtain an accurate count in 2020 was a methodology called
“Group Quarters Count Imputation” (“GQCI”). Pl.’s Mot. at 1. Defendants describe
“group quarters” as
places where people live or stay in a group living arrangement that is owned
or managed by an organization providing housing and/or services for the
residents. Group quarters differ from typical household living arrangements
because the people living in them are usually not related to one another.
Group quarters include such places as college residence halls, residential
treatment centers, skilled nursing facilities, group homes, military barracks,
prisons, and worker dormitories.
Second Decl. of John M. Abowd, Ex. 1 to Defs.’ Mot. [Dkt. # 13-3] (“Abowd Decl.”) ¶ 63. 1 After
defendants’ “review indicated anomalies” in its count of group quarters, id. ¶ 64, they
used processing information from the group quarters enumeration records,
group quarters advance contact records, and administrative data to
determine whether records were double counted, appropriately counted, or
missing. The GQCI resolved the status of group quarters addresses for
frame eligibility (occupied or not; unoccupied group quarters are deleted
from the census frame) and, if occupied, the status of persons residing in
the group quarters—eliminating duplicates and imputing missing persons.
Id. ¶ 66.
“Due to its concerns about the Census Bureau’s lack of transparency in its methodology,
Plaintiff filed a FOIA Request seeking information to help inform the public about the impact of
1 John Abowd is “the Chief Scientist and Associate Director for Research and Methodology
at the United States Census Bureau.” Abowd Decl. ¶ 1.
3
GQCI on the 2020 Census, as well as the extent to which it was employed in the enumeration.”
Pl.’s Mot. at 2. The original FOIA request has been narrowed down throughout this litigation, and
now plaintiff “only seeks documents identifying the total population (number of individuals)
imputed statewide by the Census Bureau for group quarters for each U.S. state.” Id. (internal
quotation marks and citation omitted). In short, plaintiff wishes to know how many individuals
were imputed by the GQCI process in each state.
Defendants object to producing this information and argue that FOIA Exemption 3 applies.
Exemption 3 states that FOIA “does not apply to matters that are specifically exempted from
disclosure by statute . . . if that statute requires that the matters be withheld from the public in such
a manner as to leave no discretion on the issue.” 5 U.S.C. § 552(b)(3)(A)(i). 2 Defendants’ position
is that “the Census Act’s confidentiality provisions, 13 U.S.C. §§ 8(b) & 9,” Defs.’ Mot. at 1,
require that they not produce the requested information because its release would inevitably allow
third parties to “reconstruct[]” the larger dataset and then identify the private information of
individual people counted. Id. at 21.
STANDARD OF REVIEW
Summary judgment is appropriate “if the movant shows that there is no genuine dispute as
to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P.
56(a). The party seeking summary judgment “bears the initial responsibility of informing the
district court of the basis for its motion, and identifying those portions of the pleadings,
depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any,
which it believes demonstrate the absence of a genuine issue of material fact.” Celotex Corp. v.
2 There are other subsections of Exemption 3, but defendants rely on the portion of
Exemption 3 related to statutes allowing for no discretion. See Defs.’ Mot. at 1.
4
Catrett, 477 U.S. 317, 323 (1986) (internal quotation marks omitted). To defeat summary
judgment, the non-moving party must “designate specific facts showing that there is a genuine
issue for trial.” Id. at 324 (internal quotation marks omitted). When the court is presented with
cross-motions for summary judgment, it analyzes the underlying facts and inferences in each
party’s motion in the light most favorable to the non-moving party. See Anderson v. Liberty Lobby,
Inc., 477 U.S. 242, 247 (1986).
The mere existence of a factual dispute is insufficient to preclude summary
judgment. Anderson, 477 U.S. at 247–48. A dispute is “genuine” only if a reasonable fact-finder
could find for the non-moving party; a fact is “material” only if it is capable of affecting the
outcome of the litigation. Id. at 248; Laningham v. U.S. Navy, 813 F.2d 1236, 1241 (D.C. Cir.
1987). In the FOIA context, “the sufficiency of the agency’s identification or retrieval procedure”
must be “genuinely in issue” in order for summary judgment to be inappropriate. Weisberg v.
DOJ, 627 F.2d 365, 371 n.54 (D.C. Cir. 1980), quoting Founding Church of Scientology v. NSA,
610 F.2d 824, 836 (D.C. Cir. 1979) (internal quotation marks omitted).
“[S]ummary judgment may be granted on the basis of agency affidavits” in FOIA cases,
when those affidavits “contain reasonable specificity of detail rather than merely conclusory
statements,” and when “they are not called into question by contradictory evidence in the record
or by evidence of agency bad faith.” Jud. Watch, Inc. v. U.S. Secret Serv., 726 F.3d 208, 215 (D.C.
Cir. 2013), quoting Consumer Fed’n of Am. v. Dep’t of Agric., 455 F.3d 283, 287
(D.C. Cir. 2006). See also Chambers v. Dep’t of Interior, 568 F.3d 998, 1003 (D.C. Cir. 2009)
(“In a suit seeking agency documents—whether under the Privacy Act or FOIA—‘[a]t the
summary judgment stage . . . the court may rely on a reasonably detailed affidavit.’”). A plaintiff
cannot rebut the good faith presumption afforded to an agency’s supporting affidavits through
5
“purely speculative claims about the existence and discoverability of other documents.” SafeCard
Servs., Inc. v. SEC, 926 F.2d 1197, 1200 (D.C. Cir. 1991), quoting Ground Saucer Watch, Inc. v.
CIA, 692 F.2d 770, 771 (D.C. Cir. 1981).
In a FOIA case, the burden rests with “the agency to sustain its action,” Military Audit
Project v. Casey, 656 F.2d 724, 738 (D.C. Cir. 1981), quoting 5 U.S.C. § 552(a)(4)(B), and “the
vast majority of FOIA cases can be resolved on summary judgment.” Brayton v. Off. of U.S. Trade
Rep., 641 F.3d 521, 527 (D.C. Cir. 2011). See also Wheeler v. CIA, 271 F. Supp. 2d 132, 136
(D.D.C. 2003) (the agency “bears the burden of sustaining its decision to claim an exemption from
disclosure”), citing 5 U.S.C. § 552a(g)(3)(A). The district court reviews agency decisions de
novo. 5 U.S.C. § 552(a)(4)(B).
ANALYSIS
I. General FOIA Principles
FOIA requires federal agencies to disclose records upon request. See
5 U.S.C. § 552(a)(3)(A). The only exception is for records that fall into one, or more, of the nine
narrowly construed exemptions. See id. § 552(b); see also FBI v. Abramson, 456 U.S. 615, 630
(1982). “[W]hen an agency seeks to withhold information, it must provide a relatively detailed
justification” for the withholding, Morley v. CIA, 508 F.3d 1108, 1122 (D.C. Cir. 2007) (citation
and internal quotation marks omitted), through a Vaughn index, an affidavit, or by other means.
See Gallant v. NLRB, 26 F.3d 168, 172–73 (D.C. Cir. 1994), citing Vaughn v. United States, 936
F.2d 862, 867 (6th Cir. 1991). The general rule in FOIA cases is that:
If an agency’s affidavit describes the justifications for withholding the
information with specific detail, demonstrates that the information withheld
logically falls within the claimed exemption, and is not contradicted by
contrary evidence in the record or by evidence of the agency’s bad faith,
then summary judgment is warranted on the basis of the affidavit alone.
6
ACLU v. U.S. Dep’t of Def., 628 F.3d 612, 619 (D.C. Cir. 2011).
After asserting and explaining the use of particular exemptions, an agency must release
“[a]ny reasonably segregable portion of a record,” 5 U.S.C. § 552(b), unless the non-exempt
portions are “inextricably intertwined with exempt portions” of the record. Johnson v. Exec. Off.
for U.S. Att’ys, 310 F.3d 771, 776 (D.C. Cir. 2002), quoting Mead Data Cent., Inc. v. U.S. Dep’t
of Air Force, 566 F.2d 242, 260 (D.C. Cir. 1977). “In order to demonstrate that all reasonably
segregable material has been released, the agency must provide a ‘detailed justification’ for its
non-segregability,” although “the agency is not required to provide so much detail that the exempt
material would be effectively be disclosed.” Johnson, 310 F.3d at 776, quoting Mead Data, 566
F.2d at 261.
In 2016, Congress added language to FOIA, mandating that “an agency shall withhold
information under this section only if the agency reasonably foresees that disclosure would harm
an interest protected by [the] exemption . . . or disclosure is prohibited by law.”
5 U.S.C. § 552(a)(8)(A)(i). As the D.C. Circuit has explained, “Congress added the distinct
foreseeable harm requirement to foreclose the withholding of material unless the agency can
‘articulate both the nature of the harm [from release] and the link between the specified harm and
specific information contained in the material withheld.’” Reps. Comm. for Freedom of the Press
v. Fed. Bureau of Investigation, 3 F.4th 350, 369 (D.C. Cir. 2021) (“Reporters Committee v. FBI”),
quoting H.R. Rep. No. 114-391, at 9 (2016). “In that way, the foreseeable harm
requirement imposes an independent and meaningful burden on agencies.” Id. (brackets and
internal quotation marks omitted).
7
II. Exemption 3
Under FOIA Exemption 3, agencies need not produces matters
“specifically exempted from disclosure by statute,” provided that such statute leaves no discretion
on disclosure. 5 U.S.C. § 552(b)(3)(a)(i). “Under that exemption, the [agency] need only show
that the statute claimed is one of exemption as contemplated by Exemption 3 and that the withheld
material falls within the statute.” Larson v. Dep’t of State, 565 F.3d 857, 865 (D.C. Cir. 2009).
The Supreme Court has already made clear that the portions of the Census Act defendants
cite are an appropriate basis for the denial of a FOIA request under Exemption 3. See Baldrige v.
Shapiro, 455 U.S. 345, 355 (1982) (“Sections 8(b) and 9(a) explicitly provide for the nondisclosure
of certain census data. No discretion is provided to the Census Bureau on whether or not to
disclose the information referred to in §§ 8(b) and 9(a). Sections 8(b) and 9(a) of the Census Act
therefore qualify as withholding statutes under Exemption 3.”). So the only question here is
whether the withheld material falls within those confidentiality provisions.
13 U.S.C. § 8(b) states that “the Secretary may furnish copies of tabulations and other
statistical materials which do not disclose the information reported by, or on behalf of, any
particular respondent.” 13 U.S.C. § 9(a)(2) prohibits the agency from making “any publication
whereby the data furnished by any particular establishment or individual under this title can be
identified.”
Plaintiff points out that its FOIA request, as narrowed, does not ask for the “information
reported” or “data furnished” by any particular individual or respondent. Instead, plaintiff is
requesting aggregated data: “the total population (number of individuals) imputed statewide by
the Census Bureau for group quarters for each U.S. state.” Pl.’s Mot. at 2 (internal quotation marks
omitted). And defendants acknowledge that this request, on its face, would not require them to
8
disclose the information or data reported by any particular individual or respondent. See Defs.’
Mot. at 2 (“The forced publication of any portion of the withheld data in this case without privacy
protections may not, by itself, directly reveal the data disclosed by any particular respondent.”).
The issue, according to defendants, is the impact that this disclosure would have on the
strength of the larger dataset. They present a lengthy description of their plan to protect “the
underlying raw data” from being “reconstruct[ed]” “through a series of mathematical algorithms.”
Defs.’ Mot. at 6. Central to this process of protecting the census dataset is the allocation of “precise
amounts of statistical noise” while maintaining some “invariants.” Id. at 8–9 (internal quotation
marks omitted). Noise infusion “introduces controlled amounts of error or ‘noise’ into the data.”
Id. at 3 n.4. Meanwhile, “[i]nvariants are data held constant.” Id. at 9. In other words, when
releasing information related to the 2020 Census, the Census Bureau publicizes some numbers as
they were actually reported 3 to the agency. For example, “the state level population totals” are
invariant, so each reported population of an individual state is the Census Bureau’s actual count
of each state’s population. Id. Meanwhile, the agency publicizes other numbers slightly
differently than how they were actually reported – these numbers have been infused with noise.
For example, if this courthouse were to conduct a census, and it was determined that there were
sixty-five law clerks working for the district court judges, that number might be reported as sixty-
four, or sixty-six.
The goal of inserting deliberate amounts of noise is “preserving the overall statistical
validity of the resulting data while introducing enough uncertainty into the data that attackers
would not have any reasonable degree of certainty that they had isolated data for any particular
3 More precisely, the agency reports these numbers as it has calculated them after engaging
in other techniques to increase the accuracy of their count, such as imputation.
9
respondent.” Defs.’ Mot. at 3 n.4, citing Abowd Decl. ¶ 24. Defendants argue that this process
“protects privacy while maintaining the overall statistical validity of the data.” Id. at 8.
This brings us to how releasing the requested information would affect this effort.
Defendants note that “the group quarters population totals Fair Lines seeks” were not set as
invariant for the 2020 Census. Defs.’ Mot at 9. They argue that “[t]he inclusion of additional, as-
yet unaccounted for invariants—e.g., in the form of the forced disclosure of unobscured data—
would undermine the sensitive balance the Census Bureau has drawn,” and increase the likelihood
that third parties could circumvent defendants’ efforts and reconstruct the dataset, discovering
private information about particular respondents. Id. at 10, citing Abowd Decl. ¶ 60. In essence,
the more data that is publicized as invariant and without the introduction of noise to protect
anonymity, the easier it would be for someone to “reverse-engineer releases of aggregated data to
identify individual data,” especially in this modern era with the extreme amount of “computing
power that exists today.” Id. at 21.
It is important to emphasize in the wake of this technical discussion that the question to be
determined is not whether an agency is following best practices to maintain privacy; the steps
taken by the agency to maintain confidentiality are relevant only insofar as they relate to whether
the requested information is “specifically exempted from disclosure by statute . . . in such a manner
as to leave no discretion on the issue.” 5 U.S.C. § 552(b)(3)(a)(i). In other words, the agency’s
privacy practices provide context for the legal question, but the direct question here is: does the
Census Act prohibit disclosure of the requested information?
The first provision, 13 U.S.C. § 8(b), clearly does not prohibit this kind of disclosure. This
conclusion is mandated by the structure of the sentence. 13 U.S.C. § 8(b) states that “the Secretary
may furnish copies of tabulations and other statistical materials which do not disclose the
10
information reported by, or on behalf of, any particular respondent.” (Emphasis added). The word
“which” is a relative pronoun that refers to the earlier described “copies of tabulations and other
statistical materials.” See Genus Lifesciences, Inc. v. Azar, 486 F. Supp. 3d 450, 460 (D.D.C.
2020) (“[A] pronoun, relative pronoun, or demonstrative adjective generally refers to the nearest
reasonable antecedent.”) (citation and quotation marks omitted). So the most natural reading of
the sentence is that disclosure is prohibited when the “copies of tabulations and other statistical
materials” themselves would “disclose the information reported by, or on behalf of, any particular
respondent.” See Ardestani v. I.N.S., 502 U.S. 129, 135 (1991) (“The starting point in statutory
interpretation is the language of the statute itself”; courts must focus on “the most natural reading”
and “plain and ordinary meaning” of each word) (quotation marks, brackets, and citations omitted).
Plaintiff has not asked for “raw census data,” Baldrige, 455 U.S. at 358, submitted by any
individual; the narrowed request calls for an aggregated count of “imputed” individuals – just the
total number without any further elaboration or identification. The request is therefore outside of
the 13 U.S.C. § 8(b) in two different ways. First, the furnished information would not include
information that has been “reported.” 13 U.S.C. § 8(b). The GQCI numbers are not reported by
anyone; they are imputed, and therefore the source of the information is the agency itself. Second,
the request does not ask defendants to “disclose” information that stems from “any particular
respondent.” 13 U.S.C. § 8(b). Again, the nature of imputation is that the numbers are not tethered
to, let alone submitted by, any particular respondent. Because the question under 8(b) is whether
the furnished information would itself disclose reported information – and it clearly would not –
it is irrelevant whether disclosure could be part of the longer chain of events that defendants
describe, in which such information could eventually be discovered by an unrelated third party.
11
13 U.S.C. § 9(b) is written differently, though. The text of § 9(b) prohibits “any publication
whereby the data furnished by any particular establishment or individual under this title can be
identified.” The interpretation of this sentence hinges on the word “whereby.” Words in statutes
are interpreted according to their “ordinary, contemporary, common meaning.” Food Mktg. Inst.
v. Argus Leader Media, 139 S. Ct. 2356, 2363 (2019) (interpreting the word “confidential” in
FOIA’s Exemption 4 by asking what that term’s ordinary, contemporary, common meaning was
when Congress enacted the statute in 1966); see also New Prime Inc. v. Oliveira, 139 S. Ct. 532,
539 (2019) (“It’s a fundamental cannon of statutory construction that words generally should be
interpreted as taking their ordinary meaning at the time Congress enacted the statute.”) (quotation
marks, brackets, and ellipses omitted). Contemporary meaning is determined by referencing
dictionaries published around the time of enactment. See, e.g., Food Mktg., 139 S. Ct. at 2363
(referencing dictionaries from 1963, 1961, and 1968 to interpret a 1966 statute); New Prime, 139
S. Ct. at 539 (distinguishing modern definitions of the term “contract of employment” from
definitions contemporary with enactment). This is the practice even with adverbs like “whereby”;
for example, in Begay v. United States, 553 U.S. 137 (2008), the Supreme Court focused on the
meaning of the term “otherwise” in the Armed Career Criminal Act of 1984 and cited to Webster’s
Third from 1961. Id. at 144.
The relevant provision in 13 U.S.C. § 9(a)(2) was first enacted in 1954. The text of this
provision has not been changed by subsequent amendments. See H.R. 9729, 83d Cong. § 9(a)(2)
12
(1954). Webster’s New International Dictionary (2d ed. unabr. 1953) defines “whereby” as “by
or through which; by the help of which; in accordance with which.” 4
This definition allows for some level of attenuation in the chain of causation. “By the help
of which” and “by or through which,” in particular, encompass situations in which an initial event
sets a process in motion and an effect subsequently occurs. In that circumstance, the initial event
helped bring about the later effect. Applied to this situation, the adverb “whereby’ means that
publication is prohibited if the publication could help “the data furnished by any particular
establishment or individual . . . be identified,” or through publication, “the data furnished by any
particular establishment or individual . . . can be identified.” 13 U.S.C. § 9(b).
The situation defendants describe – in which release would contribute to the ability of a
third party to reconstruct the dataset – falls within the category of a situation “whereby,” or
“through which,” “data furnished by any particular establishment or individual” could be
identified. It would be through the publication of additional invariant data that the third party
could more easily reconstruct data furnished by a particular individual.
This conclusion brings us to plaintiff’s other argument: that regardless of how the statute
is interpreted, the hypothetical attack imagined by defendants is not just far-fetched; it may even
be impossible. See Pl.’s Mot. at 4 (criticizing defendants’ “conjecture about a chain of unlikely (if
not impossible events.”). Plaintiff has supplied the analysis of Dr. Steven Ruggles, a professor of
history and population studies, who concluded that “[t]here is no possible means by which the
number of imputed cases could be used in combination with other statistics to allow for
4 This is the first definition; the dictionary provides two additional definitions, but neither is
relevant (the second definition is for interrogative use and the third definition is labeled as “obscure
except dialect”).
13
identification of an individual.” See Expert Report of Dr. Steven Ruggles, Appendix A to Ex. 2 to
Pl.’s Mot. [Dkt. # 14-4] at 5 (“Ruggles Report”); see also Pl.’s Reply at 7 (defendants “stack
speculation and conjecture about a hypothetical attack by a supercomputer”).
Plaintiff asks the Court to accept its expert’s opinion on the likelihood of harm the
withholding is designed to prevent. 5 It insists that defendants’ efforts regarding invariants and
statistical noise are unnecessary. See Ruggles Report at 7. But the question when examining the
sufficiency of an agency’s reasoning in a FOIA affidavit is not whether the plaintiff has an equally
or even more compelling theory, but rather, whether the agency’s justification appears logical or
plausible. 6 See ACLU, 628 F.3d at 619 (“Ultimately, an agency’s justification for invoking
a FOIA exemption is sufficient if it appears ‘logical’ or ‘plausible.’”), quoting Larson, 565 F.3d
at 862. The agency has explained its thought process in an extremely detailed affidavit that evinces
a good faith belief that the risk is real. See generally Abowd Decl. This belief is supported by
5 It is worth noting that plaintiff’s expert, while clearly qualified to discuss the technical
portions of this case, strays well beyond that and makes extensive legal arguments in his report,
despite not being an attorney himself. See, e.g., Ruggles Report at 3 (arguing that “the counts of
imputed group quarters cases are not ‘data furnished by any particular establishment or
individual’”), 5 (explaining that a particular regulation “does not apply in this case”). This
speculation did not contribute to the Court’s understanding of the issues involved.
6 While the Court must review the agency’s reasoning to confirm that its account is plausible,
the level of technical expertise involved in weighing the risk of reconstruction is substantial, which
is precisely why defendants’ explanations – so long as they are logical and made in good faith –
are entitled to a degree of deference. As another court observed in a non-FOIA context, “disputes
regarding the Bureau’s use of statistics in taking the census ‘are best resolved not by the courts but
by the Bureau itself, whose experience with prior censuses and expertise in the collection and
analysis of statistical information render it especially qualified to make the appropriate decisions.’”
City of Los Angeles v. U.S. Dep’t of Com., 307 F.3d 859, 876 (9th Cir. 2002), quoting City of
Philadelphia v. Klutznick, 503 F. Supp. 663, 676 (E.D.Pa. 1980).
14
historical examples 7 in addition to theoretical reasoning. See Defs.’ Mot. at 6–7. Given the
agency’s statutory responsibility to protect its own dataset, its greater level of familiarity with all
of the numbers involved (including those that have been infused with noise for publication), and
its lengthy deliberation about these very risks over the last ten years – including reconstruction
simulations and “regular outreach with dozens of stakeholder groups,” Defs.’ Mot. at 7–8, – the
agency’s good faith determination that the risk is real suffices. 8 So while the Court understands
the plaintiff’s skepticism, and it is inclined to agrees that even defendants’ submissions suggest
that the risk is remote, it finds that the agency has carried its burden and is permitted to invoke
FOIA Exemption 3 in declining to release the GQCI numbers. 9
7 Plaintiff’s expert emphasizes that “there is not a single documented case of anyone outside
the Census Bureau revealing the responses of a particular identified person using data from the
decennial census.” Ruggles Report at 7 (emphasis removed). But defendants offer examples from
other publicly available datasets, and the absence of a previous successful reconstruction does not
mean that reconstruction cannot occur in the future.
8 Moreover, defendants need not show a “parade of horribles,” as plaintiff puts it. Pl.’s Reply
at 11. The statute forbids disclosure if “the data furnished by any particular establishment or
individual . . . can be identified.” 13 U.S.C. § 9(a)(2) (emphasis added). So even if a full
reconstruction remains impossible, the burden is only on the agency to establish that “any”
individual or establishment, i.e., a single person, could be identified.
9 One final issue to address is that there is no dispute regarding the adequacy of the search.
Plaintiff’s complaint arguably initially challenged the adequacy of defendants’ search. See Compl.
at 10 (plaintiff’s prayer for relief, asking this Court to “Order Defendants to expeditiously conduct
an adequate search for all records responsive to Plaintiff’s Request in accordance with 5 U.S.C. §
552(a)(3)(C) at no cost to Plaintiff”). However, defendants moved for summary judgment on this
issue, making specific arguments about the adequacy of their search, see Defs.’ Mot. at 15–18, and
plaintiff failed to respond, therefore waiving its right to argue that the search was inadequate. See
generally Pl.’s Mot.; Pl.’s Reply. As a result, defendants are entitled to summary judgment as to
this requested form of relief. See United Mine Workers of Am. 1974 Pension v. Pittston Co., 984
F.2d 469, 478 (D.C. Cir. 1993) (failure to raise an argument during summary judgment briefing
constitutes a waiver).
15
CONCLUSION
Defendants’ motion for summary judgment will be GRANTED and plaintiff’s motion will
be DENIED.
A separate order will issue.
AMY BERMAN JACKSON
United States District Judge
DATE: August 2, 2022
16