The clerk's office is respectfully directed to amend the official caption of this… v. van Gorp et ano. UNITED STATES COURT OF APPEALS 1 FOR THE SECOND CIRCUIT 2 August…

* The clerk's office is respectfully directed to amend the
official caption of this case as shown above.
10-3297 (L)
Feldman v. van Gorp et ano.
UNITED STATES COURT OF APPEALS 1
FOR THE SECOND CIRCUIT 2
August Term, 2011 3
(Argued: January 30, 2012 Decided: September 5, 2012) 4
Docket Nos. 10-3297( Lead) 11-975 (Con) 5
------------------------------------- 6
United States of America ex rel. Daniel Feldman, 7
Plaintiff-Appellee, 8
- v - 9
Wilfred van Gorp and Cornell University Medical College, 10
Defendants-Appellants. *
11
------------------------------------- 12
Before: SACK, RAGGI, and CHIN, Circuit Judges. 13
Appeal from a judgment of the United States District 14
Court for the Southern District of New York (William H. Pauley 15
III, Judge) denying the defendants' motion for judgment as a 16
matter of law and their motion for a new trial following a jury 17
verdict partially in favor of the plaintiff on his claims brought 18
on behalf of the government pursuant to the False Claims Act, 31 19
U.S.C. § 3729 et seq., and awarding principally $855,714 in 20

-- 1 of 48 --

2
treble actual damages. We conclude that: 1) where the government 1
has provided funds for a specified good or service only to have 2
defendant substitute a non-conforming good or service, a court 3
may, upon a proper finding of False Claims Act liability, 4
calculate damages to be the full amount of the grant payments 5
made by the government after the material false statements were 6
made; 2) there was sufficient evidence from which a reasonable 7
jury could determine that the false statements at issue were 8
material to the government's funding decision; and 3) the 9
district court did not abuse its discretion in excluding evidence 10
of inaction on the part of the National Institutes of Health in 11
response to the plaintiff's complaint regarding the fellowship 12
program in which he had been enrolled. 13
Affirmed. 14
Appearances: TRACEY A. TISKA, R. Brian Black, Eva L. 15
Dietz, on the brief) Hogan Lovells US 16
LLP, New York, New York, for Defendant- 17
Appellant Cornell University. 18
Nina M. Beattie, Brune & Richard LLP, 19
New York, New York, for Defendant- 20
Appellant Wilfred van Gorp. 21
MICHAEL J. SALMANSON (Scott B. Goldshaw, 22
on the brief) Salmanson Goldshaw, P.C., 23
Philadelphia, Pennsylvania, for 24
Plaintiff-Appellee. 25
Jean-David Barnea, Rebecca C. Martin, 26
Sarah S. Normand, Assistant United 27
States Attorneys, of counsel, for Preet 28
Bharara, United States Attorney for the 29
Southern District of New York, for 30
Amicus Curiae, The United States of 31
America. 32

-- 2 of 48 --

3
SACK, Circuit Judge: 1
The defendants appeal from a judgment of the United 2
States District Court for the Southern District of New York 3
(William H. Pauley III, Judge) denying their motion for judgment 4
as a matter of law and their motion for a new trial following a 5
jury verdict partially in favor of the plaintiff on his claims 6
regarding the misuse of a research training grant brought on 7
behalf of the government pursuant to the False Claims Act, 31 8
U.S.C. § 3729 et seq., and awarding principally $855,714 in 9
treble actual damages. We conclude that: 1) where the government 10
has provided funds for a specified good or service only to have 11
defendant substitute a non-conforming good or service, a court 12
may, upon a proper finding of False Claims Act liability, 13
calculate damages to be the full amount of the grant payments 14
made by the government after the material false statements were 15
made; 2) there was sufficient evidence from which a reasonable 16
jury could determine that the false statements at issue were 17
material to the government's funding decision; and 3) the 18
district court did not abuse its discretion in excluding evidence 19
of inaction on the part of the National Institutes of Health in 20
response to the plaintiff's complaint regarding the fellowship 21
program in which he had been enrolled. 22

-- 3 of 48 --

4
BACKGROUND 1
In 1997, appellants Cornell University Medical College 2
("Cornell") and Dr. Wilfred van Gorp, a professor of psychiatry 3
at Cornell, applied for funding from the Ruth L. Kirschstein 4
National Research Service Award Institutional Research Training 5
Grant program, also known as the "T32" grant program, of the 6
National Institutes of Health ("NIH"). The T32 program funds 7
pre- and post-doctoral training programs in biomedical, 8
behavioral, and clinical research. T32 grants are meant to "help 9
ensure that a diverse and highly trained workforce is available 10
to assume leadership roles related to the Nation's biomedical and 11
behavioral research agenda." NIH Guide, "NIH National Research 12
Service Award Institutional Research Training Grants," at 1 (May 13
16, 1997), United States ex rel. Feldman v. Van Gorp, No. 10- 14
3297, Joint Appendix ("J.A.") 2437 (2d Cir. Jan. 26, 2012) ("NIH 15
Guide"). Positions funded through T32 grants may not be used for 16
study leading to clinically-oriented degrees, "except when those 17
studies are a part of a formal combined research degree program, 18
such as the M.D./Ph.D." Id. at 2, J.A. 2438. Instead, funded 19
programs must train their fellows "with the primary objective of 20
developing or extending their research skills and knowledge in 21
preparation for a research career." Id. 22
Institutions applying for T32 grants undergo a two- 23
tiered review process. It begins with a review of the proposal 24

-- 4 of 48 --

5
by a twenty-member "Initial Research Group" ("IRG"), also called 1
a "peer review committee." IRG members are independent experts 2
in scientific fields related to that of the grant application 3
under review; they are not NIH employees. Each member scores 4
applications based on his or her view of its scientific or 5
technical merit guided by specified criteria, including, among 6
other factors: the program director's and faculty's training 7
records, as determined by the success of former trainees; the 8
objective, design, and direction of the program; the caliber of 9
the faculty; the institutional training environment, including 10
the commitment of the institution to training and the resources 11
available to trainees; and the institution's proposed plans for 12
recruiting and selecting high-quality trainees. The scores are 13
then averaged to arrive at an IRG "priority score." Testimony of 14
Dr. Robert Bornstein at 1190-91, July 21, 2010 ("Bornstein 15
Testimony"), J.A. 1955. This score is included with the IRG 16
members' written comments in a summary statement, which is 17
transmitted to the NIH. 18
The "second tier" of review is performed by the 19
advisory council of the appropriate constituent organization of 20
the NIH, in this case the National Institute of Mental Health 21
("NIMH"). The advisory council ranks the applications by 22
priority score, and establishes a "pay line" at the point in the 23
list of applications where there is no more funding available; 24
only the applications above the "pay line" are recommended to the 25

-- 5 of 48 --

6
director of the funding institute as potential grant recipients. 1
"The role of the advisory council is not to second-guess the 2
scientific review of the IRG. Rather, [the council] reviews the 3
applications to ensure that they further the goals and interests 4
of the awarding institute. Thus, the IRG review and the 5
resulting high-priority score are keys to NIH funding." Id. at 6
1190, J.A. 1955-56. 7
Once an application has placed above the "pay line," 8
the advisory council makes recommendations based on the 9
scientific merit of the proposal, as judged by the IRG, and the 10
relevance of the proposal to the awarding institute's programs 11
and priorities. Funding is typically approved by the NIH for one 12
year, and recipient institutions are eligible for up to four 13
years of additional funding. 14
In order to renew a T32 grant, the recipient 15
institution (in this case Cornell) must submit an annual renewal 16
application and a progress report detailing the status of its 17
project. In contrast with initial grant applications, renewal 18
applications are reviewed solely by the NIH on a noncompetitive 19
basis. The NIH considers the progress made under the grant and 20
the grant's budget. By regulation, the annual progress report 21
must contain a "comparison of actual accomplishments with the 22
goals and objectives established for the period," and must 23
specify "[r]easons why established goals were not met," if indeed 24
they were not. 45 C.F.R. § 74.51(d)(1)-(2). 25

-- 6 of 48 --

7
Recipient institutions must also "immediately notify" 1
NIH of "developments that have a significant impact" on the 2
research program, including "problems, delays, or adverse 3
conditions which materially impair the ability to meet the 4
objectives of the award." Id. § 74.51(f). This notification 5
must also include a "statement of the action taken or 6
contemplated, and any assistance needed to resolve the 7
situation." Id.; see also Draft OIG Compliance Program Guidance 8
for Recipients of PHS Research Awards, 70 Fed. Reg. 71312-01, 9
71320 (Nov. 28, 2005) ("Prompt voluntary reporting will 10
demonstrate the institution's good faith and willingness to work 11
with governmental authorities to correct and remedy the problem. 12
In addition, reporting such conduct may be considered a 13
mitigating factor by the responsible law enforcement or 14
regulatory office . . . ."). 15
Cornell's initial grant application at issue here 16
sought funding for a fellowship program entitled "Neuropsychology 17
of HIV/AIDS Fellowship." Van Gorp Grant Application at 1, J.A. 18
2254 (April 24, 1997) ("Grant Application"). The application 19
explained that the two-year fellowship would train as many as six 20
post-doctoral fellows at a time in "child and adult clinical and 21
research neuropsychology with a strong emphasis upon research 22
training with HIV/AIDS." Id. at 2, J.A. 2255. The training 23
program would, according to the application, build on the Cornell 24
faculty's extensive research into the neuropsychology of 25

-- 7 of 48 --

8
HIV/AIDS, which included projects examining distress levels in 1
HIV-AIDS patients during the course of their illness, the 2
relationship between the neuropsychology of HIV/AIDS and 3
patients' abilities to function at work or in school, and the 4
possibility of using neuropsychological testing to predict 5
whether AIDS patients will suffer from dementia. The 6
application further explained that van Gorp would serve as the 7
program director, and that he had a "long history of successful 8
research, training and mentoring of students in HIV[] related 9
work." Id. at 40, J.A. 2295. 10
The 123-page grant application outlined the 11
fellowship's curriculum in detail. Fellows would be required to 12
take "several formal, core didactic courses," and a number of 13
elective courses. Id. at 45, J.A. 2300. In the first year of 14
the fellowship, fellows would enroll in five core courses, some 15
of which "have been designed specifically for the HIV 16
Neuropsychology Fellowship." Id., J.A. 2300. These core courses 17
would be supplemented by a "large number of courses, lectures, 18
neuroscience educational programs, as well as other seminars in a 19
variety of sub-speciality areas." Id. The curriculum for the 20
second year, which included four core courses, would allow 21
fellows to "develop more independent research skills and devote 22
more time to their HIV research." Id. The fellows' progress 23
under the grant would be monitored monthly by a formal training 24
committee comprised of several faculty members, as "[o]ngoing 25

-- 8 of 48 --

9
evaluation of the curriculum, trainees and faculty is an integral 1
part of the training program." Id. at 48, J.A. 2303. 2
The Cornell grant application identified a list of 3
fourteen faculty members who would serve as "Key Personnel," 4
which the NIH defined as "individuals who contribute to the 5
scientific development or execution of the project in a 6
substantive way." NIH Grant Application Instructions at 26, J.A. 7
2612 (June 8, 1999). The application described in detail some of 8
these research projects. It also asserted, "Our faculty has a 9
solid track record in quality and productive research in brain- 10
behavior issues, including research in HIV/AIDS-related research 11
[sic]." Grant Application at 48, J.A. 2303. And the application 12
identified additional institutions which would serve as clinical 13
resources, including Cornell University, Memorial Sloan-Kettering 14
Cancer Center, St. Vincent's Hospital, and Gay Men's Health 15
Crisis Center. 16
In describing the fellowship program's commitment to 17
research training, the grant application explained that "the 18
majority of [the fellows'] clinical work will be with persons 19
with HIV infection." Grant Application at 44, J.A. 2299. 20
Fellows would "devote an average of 75% of their time to research 21
and an average of 25% [of their] time to clinical work with 22
persons with HIV/AIDS and other neuropsychiatric disorders." Id. 23
The IRG gave Cornell's grant application a high 24
priority score, and the NIH subsequently approved funding for two 25

-- 9 of 48 --

10
fellows for the fiscal year beginning September 30, 1997, with 1
the possibility of additional funding for up to four additional 2
years. Cornell submitted renewal applications in each of the 3
following four years, from fiscal year 1999 (July 1, 1998, 4
through June 30, 1999) to fiscal year 2002 (July 1, 2001, through 5
June 30, 2002), all of which the NIH approved. In the 6
accompanying annual progress reports, Cornell and van Gorp 7
indicated that there had been no material alterations to the 8
program as described in the original grant application. 9
In the renewal application for the second renewal year 10
(the third year overall), for example, Cornell and van Gorp wrote 11
that "[a]ll core and supporting faculty listed in our original 12
application are continuing. . . . There have been no alterations 13
in the courses or training program from that listed in the 14
original application, except for the addition of two [specified] 15
courses . . . ." 1999 Progress Report at 7, J.A. 2402 (January 16
19, 1999). The renewal application also explained that the 17
program had been relocated from Cornell's White Plains campus to 18
its New York City (Manhattan) campus in order to provide fellows 19
with "immediate access to subjects and patients who have 20
HIV/AIDS." Id. The renewal applications for the fourth and 21
fifth year stated that "[t]he core structure of our training 22
program has remained the same as in years past and to that 23
described in our initial application." 2000 Progress Report at 24
4, J.A. 2411 (January 24, 2000); 2001 Progress Report at 5, J.A. 25

-- 10 of 48 --

1 Because this suit is being brought by Feldman on behalf
of the United States, Feldman is technically the "plaintiff-
relator." See infra, note [3]. We nonetheless refer to him
simply as the "plaintiff."
11
2422 (January 22, 2001). The NIH approved each of these renewal 1
applications. 2
In September 1998, at about the time the first renewal- 3
year began, Daniel Feldman, the plaintiff, 1 was selected by 4
Cornell to participate in the fellowship program. He left the 5
program in December 1999, before the completion of his two-year 6
fellowship. Other fellows who participated in the program 7
included Elizabeth Ryan, Clifford Smith, Kimberly Walton Louis, 8
and Evan Drake. At trial, Feldman presented evidence that the 9
actual fellowship deviated in many ways from that described in 10
the Grant Application, and that Cornell and van Gorp failed to 11
inform NIH of these deviations. 12
Testimony presented at trial indicated that some of the 13
faculty members identified as "Key Personnel" in the initial 14
application did not in fact contribute in any substantive way to 15
the fellowship program. Van Gorp acknowledged that the 16
contributions to the program of two of these faculty members, Dr. 17
Tatsuyki Kakuma and Dr. Michael Giordano, were considerably 18
limited, if not entirely eliminated, by the fact that the two 19
doctors were not in physical proximity to the fellows during the 20
grant period. Many fellows, according to their testimony, had 21
little or no interaction with the remaining key personnel, and 22

-- 11 of 48 --

12
were unaware that these faculty members were or were supposed to 1
be available as resources. In addition, according to this 2
testimony, fellows were largely unaware of research opportunities 3
at medical centers other than Cornell. 4
There was also testimony in the district court to the 5
effect that Cornell and van Gorp failed to notify NIH that the 6
curriculum outlined in the initial grant application was never 7
implemented. Several core courses identified in the application 8
were not regularly conducted for fellows, and fellows were not 9
informed that these courses were a required component of the 10
program. Moreover, according to this testimony, fellows were 11
never evaluated or supervised by the training committee referred 12
to in the Grant Application. 13
Feldman also presented evidence that the research and 14
clinical training described in the initial grant application 15
differed significantly from the actual training received. NIH 16
rules provide that fellows in a T32 program "must devote their 17
time to the proposed research training and must confine clinical 18
duties to those that are an integral part of the research 19
training experience." NIH Guide at 3, J.A. 2439; T32 Training 20
Grant Announcement at 9, J.A. 2568 (June 16, 2006). And, in 21
accordance with these requirements, the grant application stated 22
that "the majority of [the fellows’] clinical work will be with 23
persons with HIV infection." Grant Application at 44, J.A. 2299. 24
Further, in explaining the training program's relocation from 25

-- 12 of 48 --

2 The parties stipulated that of Ryan's 32 clinical
patients, two were HIV positive; of Smith's 35 clinical patients,
none were HIV positive; of Louis's 23 patients, none were HIV
positive; of Drake's 48 patients, none were HIV positive; and of
Feldman's 27 patients, one was HIV positive.
13
White Plains to Manhattan, the third-year renewal application 1
explained that "[f]ellows [would be] housed within a large, 2
medical/surgical setting with immediate access to subjects and 3
patients who have HIV/AIDS." 1999 Progress Report at 7, J.A. 4
2402. 5
But, as the plaintiff summarizes the trial testimony, 6
out of the 165 clinical cases that the fellows saw during their 7
fellowship, only three involved HIV-positive patients. 2 Pl.'s 8
Br. at 22. Several fellows testified that much of the research 9
that they performed under the grant program had no relation to 10
HIV or AIDS at all. For example, Clifford Smith testified that 11
the research projects he worked on under the T32 grant were 12
primarily related to epilepsy and aging, and did not involve an 13
HIV population. Out of the eight research projects that Evan 14
Drake worked on during his fellowship, he said, only one focused 15
specifically on HIV. Feldman similarly told the court that he 16
worked on only one HIV-focused project during his time as a 17
fellow. 18
In July 2001, after he had left the program, Feldman 19
submitted a letter to the NIH complaining about the program's 20
focus on clinical work rather than research, and the fellows' 21

-- 13 of 48 --

3
In a qui tam action, a private plaintiff,
known as a relator, brings suit on behalf of
the Government to recover a remedy for a harm
done to the Government. See United States ex
rel. Eisenstein v. City of New York, [556
U.S. 928, 932] (2009) (describing qui tam
actions under the False Claims Act, 31 U.S.C.
§ 3729 et seq.); see also Black's Law
Dictionary 1282 (8th ed. 2004) (defining "qui
tam action" as "[a]n action brought under a
statute that allows a private person to sue
for a penalty, part of which the government
or some specified public institution will
receive"). Qui tam plaintiffs, even if not
personally injured by a defendant's conduct,
possess constitutional standing to assert
claims on behalf of the Government as its
effective assignees. There is, however, no
common law right to bring a qui tam action;
rather, a particular statute must authorize a
private party to do so.
Woods v. Empire Health Choice, Inc., 574 F.3d 92, 97-98 (2d Cir.
2009) (footnote and some citations omitted; second brackets in
original).
Where the United States has elected not to proceed with the
action, as here, the relator is entitled personally to recover
14
limited access to HIV-positive patients. In March 2002, he 1
submitted another letter to the NIH, again complaining that the 2
fellowship program deviated from its description in the initial 3
grant application. In response, the NIH asked Cornell to conduct 4
an investigation of the complaint, which Cornell completed in 5
June 2003. Cornell then sent Feldman a letter informing him that 6
the investigation uncovered no wrongdoing. 7
On October 14, 2003, Feldman filed a qui tam complaint 8
pursuant to the False Claims Act ("FCA"), 31 U.S.C. § 3729 et 9
seq.,3 alleging that Cornell and van Gorp made false claims to 10

-- 14 of 48 --

between 25 and 30 percent of the proceeds of the action or
settlement, plus reasonable attorney's fees. See 31 U.S.C.
§ 3730(d)(2).
15
the United States in the Grant Application and in the four 1
renewal applications. Feldman alleged that statements made in 2
these applications were false because the fellowship's 3
curriculum, resources, faculty members, and training differed 4
significantly from that described in the application, and in the 5
subsequent renewal applications representing that no changes had 6
been made to the program. The complaint was unsealed in April 7
2007, after the United States declined to intervene in this 8
action. See Cook County v. United States ex rel. Chandler, 538 9
U.S. 119, 122 (2003) ("The relator must inform the Department of 10
Justice of her intentions and keep the pleadings under seal for 11
60 days while the Government decides whether to intervene and do 12
its own litigating." (citing 31 U.S.C. § 3730(b)(2)-(c))). 13
On January 9, 2009, after discovery had been completed, 14
Cornell and van Gorp moved for summary judgment. On December 7, 15
2009, the district court denied the motion, concluding that there 16
were genuine issues of material fact as to whether the defendants 17
made false statements in both the initial grant application and 18
the renewal applications, and whether those statements were 19
material to the funding decisions. United States ex rel. Feldman 20
v. Van Gorp ("Feldman I"), 674 F. Supp. 2d 475, 482-83 (S.D.N.Y. 21
2009). The district court also concluded that the plaintiff need 22

-- 15 of 48 --

16
not establish actual damages to the government as an element of 1
an FCA claim because that statute's provision of civil penalties 2
for false and fraudulent claims allowed courts to "find a 3
violation even in the absence of proof of damages to the United 4
States." Id. at 481. The court did not address, however, 5
whether Feldman's recovery would be limited to statutory damages. 6
On December 18, 2009, the defendants moved for 7
reconsideration of the summary judgment decision, arguing that 8
the district court had erred in failing to address the issue of 9
whether Feldman should be limited to statutory penalties because 10
he had not presented sufficient evidence of actual damages to the 11
United States. On May 3, 2010, the district court denied the 12
motion, explaining that although the damages to the United States 13
could not be calculated in the same way they would be in a 14
standard breach-of-contract action because no tangible benefit 15
had been received, the plaintiff would not be limited to 16
statutory damages. United States ex rel. Feldman v. Van Gorp 17
("Feldman II"), No. 03 Civ. 8135, 2010 WL 1948592, at *1-*2, 2010 18
U.S. Dist. LEXIS 47039, at *4-*6 (S.D.N.Y. May 3, 2010). The 19
court said that the "'benefit of the bargain' to the government 20
is providing funds to recipients who best fit its specified 21
criteria and that this benefit is lost when funds are diverted to 22
less eligible recipients." Id. at *2, 2010 U.S. Dist. LEXIS 23
47039, at *4-*5. Therefore, "if the fact-finder concludes that 24
the government would not have awarded the grant absent the false 25

-- 16 of 48 --

4 The district court similarly excluded evidence of
inaction on the part of the New York State Department of
Education and the American Psychological Association, but the
defendants do not challenge the exclusion of that evidence on
appeal.
17
claims, it may properly conclude that the measure of damages is 1
the total amount the government paid." Id., 2010 U.S. Dist. 2
LEXIS 47039, at *6. 3
Before trial, Feldman submitted a motion in limine to 4
exclude evidence including that of NIH's inaction towards Cornell 5
and van Gorp in response to Feldman's complaints about the 6
fellowship program. On July 8, 2010, the district court granted 7
Feldman's motion to exclude that evidence. The court concluded 8
that the evidence of NIH's inaction was irrelevant and therefore 9
inadmissible under Rule 402 because "no discovery was conducted 10
concerning the standards [NIH used] to determine the existence of 11
misconduct and whether those standards are at all similar to the 12
elements of an FCA claim." United States ex rel. Feldman v. van 13
Gorp ("Feldman III"), No. 03 Civ. 8135, 2010 WL 2911606, at *3, 14
2010 U.S. Dist. LEXIS 73633, at *7 (S.D.N.Y. July 8, 2010). 15
Moreover, the court concluded, even if "marginally relevant," the 16
evidence would have been excluded pursuant to Rule 403 because of 17
the possibility that it would confuse or mislead the jury. 4 Id. 18
The case was tried to a jury for eight days in July 19
2010, resulting in a partial verdict for Feldman. The jury found 20
the defendants not liable for false statements in the Grant 21

-- 17 of 48 --

18
Application and the first renewal application, but found 1
liability based on the renewal applications for the third, fourth 2
and fifth years of the grant, i.e., the second, third and fourth 3
renewal years. On August 3, 2010, the district court awarded 4
actual damages in treble the amount NIH paid for the last three 5
renewal years of the grant –- the trebling being provided for in 6
the FCA, 31 U.S.C. § 3729(a)(1) -- totaling $855,714. The 7
judgment also included statutory penalties of $32,000, for a 8
total of $887,714. The district court also awarded to the 9
plaintiff $602,898.63 in attorney's fees, $25,862.15 in costs, 10
and $3,121.47 in expenses. 11
On August 25, 2010, the defendants filed a motion for 12
judgment as a matter of law under Rule 50(b), or in the 13
alternative, for a new trial pursuant to Rule 59. The defendants 14
argued that there was insufficient evidence from which the jury 15
could properly have concluded that the false statements at issue 16
were material to the NIH's decisions to renew the T32 grant, and 17
that the court should grant judgment as a matter of law, or that 18
such a conclusion was against the weight of the evidence and 19
warranted a new trial. The defendants also argued that the 20
district court erred in determining as a matter of law that 21
damages were equal to the entire grant amounts for the years in 22
which liability was found rather than submitting that question to 23
the jury. 24

-- 18 of 48 --

19
The district court denied this motion on December 9, 1
2010. United States ex rel. Feldman v. van Gorp ("Feldman IV"), 2
No. 03 Civ. 8135, 2010 WL 5094402, at *5, 2010 U.S. Dist. LEXIS 3
130358, at *14-*15 (S.D.N.Y. Dec. 9, 2010). The court concluded 4
that Feldman had presented sufficient evidence for the jury to 5
conclude that the false statements were material to the NIH's 6
funding decisions, noting that NIH's guidelines and instructions 7
on the renewal applications unambiguously stated that it should 8
be notified of any changes made to the grant program. Id. at *2- 9
*5, 2010 U.S. Dist. LEXIS 130358, at *4-*14. The district court 10
also relied on its opinion in Feldman III to deny the motion for 11
a jury trial on damages. Id. at *5, 2010 U.S. Dist. LEXIS 12
130358, at *13-*15. 13
The defendants appeal. 14
DISCUSSION 15
The defendants contend that: (1) the district court 16
erred in its methodology for determining damages and in 17
determining the amount of those damages, as a matter of law; (2) 18
the jury did not have sufficient evidence from which to conclude 19
that the false statements at issue were material to the funding 20
decision; and (3) the district court erred in excluding evidence 21
of NIH's "inaction" in response to Feldman's complaint. 22
I. Damages 23
The False Claims Act prohibits a person from "knowingly 24
present[ing], or caus[ing] to be presented, [to an officer or 25

-- 19 of 48 --

5 In 2009, Congress amended the False Claims Act to add a
specific requirement that to be actionable a false statement must
be material. 31 U.S.C. § 3729(a)(1)(B). It purports to apply
prospectively and therefore would not apply to this case. See
Feldman I, 674 F. Supp. 2d at 480. Never prior to that enactment
and absent its materiality provision did we explicitly require a
showing of materiality in FCA cases, although six of the seven
circuits to address the issue did. See id. (citing decisions).
We need not decide here whether a showing of materiality was
required because, assuming that it was, the requirement has been
met, as we explain in Part II, below.
20
employee of the United States Government,] a false or fraudulent 1
claim for payment or approval." 31 U.S.C. § 3729(a)(1)(A). 2
Liability under the Act also requires a showing of materiality. 5
3
Under the Act as currently in force, "the term 'material' means 4
having a natural tendency to influence, or be capable of 5
influencing, the payment or receipt of money or property." 6
Id. § 3729(b)(4); see also Neder v. United States, 527 U.S. 1, 16 7
(1999) ("In general, a false statement is material if it has a 8
natural tendency to influence, or [is] capable of influencing, 9
the decision of the decisionmaking body to which it was 10
addressed." (brackets in original; internal quotation marks 11
omitted) (criminal fraud case)). 12
The FCA provides for damages equal to "3 times the 13
amount of damages which the Government sustains because of the 14
act of that person," in addition to a "civil penalty." 31 U.S.C. 15
§ 3729(a)(1). The Act does not specify how damages are to be 16
calculated, but the Supreme Court has recognized that the purpose 17
of damages, even as multiplied, under the Act is to make the 18

-- 20 of 48 --

21
government "completely whole" for money taken from it by fraud. 1
United States ex. rel. Marcus v. Hess, 317 U.S. 537, 551-52 2
(1943), superseded by statute as recognized by United States ex 3
rel. Kirk v. Schindler Elevator Corp., 601 F.3d 94 (2d Cir. 2010) 4
("We think the chief purpose of the statutes here [predecessors 5
of the current False Claims Act, providing for double rather than 6
treble damages] was to provide for restitution to the government 7
of money taken from it by fraud, and that the device of double 8
damages plus a specific sum was chosen to make sure that the 9
government would be made completely whole."). Because the 10
district court here determined that damages could be established 11
as a matter of law, we review that conclusion de novo. See 12
Bessemer Trust Co., N.A. v. Branin, 618 F.3d 76, 85 (2d Cir. 13
2010) (stating that where the district court has determined 14
damages, we review its application of legal principles de novo 15
and its factual findings for clear error). 16
The question of how damages should be measured in an 17
FCA case where "contracts entered into between the government and 18
the Defendants did not produce a tangible benefit to the 19
[government]," United States ex. rel. Longhi v. United States, 20
575 F.3d 458, 473 (5th Cir. 2009), is one of first impression in 21
this Court. The defendants argue both that the district court 22
erred in concluding that application of the standard benefit-of- 23
the-bargain calculation as a methodology for determining damages 24
was inappropriate in this case, and that it erred in deciding the 25

-- 21 of 48 --

22
amount of damages as a matter of law based on the jury's verdict, 1
rather than allowing the jury to assess the amount of damages 2
due. 3
A. Proper Measure of Damages 4
In most FCA cases, damages are measured as they would 5
be in a run-of-the-mine breach-of-contract case –- using a 6
"benefit-of-the-bargain" calculation in which a determination is 7
made of the difference between the value that the government 8
received and the amount that it paid. See United States v. 9
Foster Wheeler Corp., 447 F.2d 100, 102 (2d Cir. 1971) 10
(collecting cases); cf. Terwilliger v. Terwilliger, 206 F.3d 240, 11
248 (2d Cir. 2000) ("[S]o far as possible, [New York contract] 12
law attempts to secure to the injured party the benefit of his 13
bargain, subject to the limitations that the injury -- whether it 14
be losses suffered or gains prevented -- was foreseeable, and 15
that the amount of damages claimed be measurable with a 16
reasonable degree of certainty and, of course, adequately 17
proven." (internal quotation marks omitted)). This method of 18
calculation is employed, for example, when the government has 19
paid for goods or services that return a tangible benefit to the 20
government. 21
There are generally two ways of determining damages in 22
such cases. First, if the non-conforming goods or services have 23
an ascertainable market value, then damages are measured 24
according to the "'difference between the market value of the 25

-- 22 of 48 --

23
product [the government] received and retained and the market 1
value that the product would have had if it had been of the 2
specified quality.'" United States v. Science Application Int'l 3
Corp., 626 F.3d 1257, 1279 (D.C. Cir. 2010) (quoting United 4
States v. Bornstein, 423 U.S. 303, 316 n.13 (1976)) (alterations 5
omitted). If the non-conforming goods' or services' market value 6
is not ascertainable, then the fact-finder determines the amount 7
of damages by calculating the difference between "the amount the 8
government actually paid minus the value of the goods or services 9
the government received or used," as judged by the fact-finder. 10
Id. 11
The defendants contend that a "benefit-of-the-bargain" 12
calculation was appropriate in this case, and that the district 13
court erred by awarding the government the full amount of the 14
grant for the years for which the violations were found rather 15
than the difference between the value of the training promised 16
and that actually delivered. The plaintiff argues, to the 17
contrary, that a different measure of damages is appropriate in 18
cases such as this, where "the defendant fraudulently sought 19
payments for participating in programs designed to benefit third- 20
parties rather than the government itself" and the government 21
received nothing of tangible value from the defendant. Id.; see 22
also Longhi, 575 F.3d at 473 ("[W]here there is no tangible 23
benefit to the government and the intangible benefit is 24
impossible to calculate, it is appropriate to value damages in 25

-- 23 of 48 --

24
the amount the government actually paid to the Defendants."). 1
This approach rests on the notion that the government receives 2
nothing of measurable value when the third-party to whom the 3
benefits of a governmental grant flow uses the grant for 4
activities other than those for which funding was approved. In 5
other words, when a third-party successfully uses a false claim 6
regarding how a grant will be used in order to obtain the grant, 7
the government has entirely lost its opportunity to award the 8
grant money to a recipient who would have used the money as the 9
government intended. 10
The plaintiff and the United States, as amicus curiae, 11
argue that this is such a case: The government received no 12
tangible benefit from the T32 grant -- students and others may 13
have, but not the government. The grant represented an attempt 14
to, but did not thereby, promote "child and adult clinical and 15
research neuropsychology with a strong emphasis upon research 16
training with HIV/AIDS." Grant Application at 2, J.A. 2255. The 17
plaintiff argues that the government is therefore entitled to 18
damages equal to the full amount of grants awarded to the 19
defendants based on their false statements. 20
We conclude that the measure of damages advocated by 21
the plaintiff and the United States is correct. 22
Although we have not addressed this question, several 23
of our sister circuits have done so in decisions that support the 24
conclusion we now reach. See Science Application, 626 F.3d at 25

-- 24 of 48 --

6 District courts within this Circuit have also employed
this methodology. See United States v. Karron, 750 F. Supp. 2d
480, 493 (S.D.N.Y. 2011), appeal filed, No. 11-1924 (concluding
that the defendant was liable for the full amount of a
government-funded research grant because he "cannot establish
that the Government received any ascertainable benefit from its
relationship with CASI. Even assuming that CASI in fact met
various milestones and provided reports to the Government, such
actions yielded no tangible benefit to the Government."); United
States ex rel. Antidiscrimination Ctr. of Metro N.Y., Inc. v.
Westchester County, No. 06 Civ. 2860, 2009 WL 1108517, at *3,
2009 U.S. Dist. LEXIS 35041, at *9 (S.D.N.Y. Apr. 24, 2009)
("Westchester has identified no tangible asset or structure it
provided to the United States such that this theory would be
applicable; it did not have a contract with the government to
build any sort of facility for the government's use or to provide
it with goods.").
25
1279 (D.C. Cir.); Longhi, 575 F.3d at 473 (5th Cir.); United 1
States v. Rogan, 517 F.3d 449, 453 (7th Cir. 2008) ("The 2
government offers a subsidy . . . with conditions. When the 3
conditions are not satisfied, nothing is due."); United States v. 4
Mackby, 339 F.3d 1013, 1018-19 (9th Cir. 2003) ("Had Mackby been 5
truthful, the government would have known that he was entitled to 6
nothing . . . ."). 6
7
The defendants point out, however, that other courts 8
have applied the "benefit-of-the-bargain" calculation in cases 9
they assert are similar to this one. They argue that because 10
"[t]he ultimate beneficiary of all government grants or contracts 11
is the public regardless of who receives the 'direct' benefit," 12
the flow of benefits to a third-party should not be determinative 13
of the damages measure. Defs.' Reply Br. at 5. 14

-- 25 of 48 --

26
In support of this theory, the defendants cite United 1
States v. Hibbs, 568 F.2d 347 (3d Cir. 1977). There, the Third 2
Circuit applied a benefit-of-the-bargain calculation in an FCA 3
case involving the defendants' fraudulent statements to the 4
Federal Housing Administration regarding the condition of various 5
residential properties. Relying on these representations, the 6
agency insured mortgages on several properties, and the agency 7
was required to pay these mortgages when the purchasers 8
defaulted. Id. at 349. 9
The government argued that its damages were the total 10
amount of the mortgage debt it had assumed, insisting that "had 11
[the defendant] not furnished the false certification, it would 12
not have insured the mortgage[s] and therefore would not have 13
been called upon to make any payment." Id. at 351. 14
The Third Circuit rejected this argument. 15
The government's actual damage was the 16
decrease in worth of the security that was 17
certified as being available, measured by the 18
difference in value between the houses as 19
falsely represented, and as they actually 20
were. Since the government was given 21
security which was less than what it was 22
represented to be, the damages are 23
essentially similar to those sustained when a 24
defective article is purchased in a 25
fraudulent transaction. In those instances, 26
decisional law sets the damages as the 27
difference in cost between that contracted 28
for and that received. 29
Id. 30

-- 26 of 48 --

27
Similarly, in Coleman v. Hernandez, 490 F. Supp. 2d 278 1
(D. Conn. 2007), a case involving the so-called "Housing Choice 2
Voucher Program" or "Section 8," under which the government 3
provides housing subsidies to qualifying individuals, the 4
district court declined to award the plaintiff the full amount 5
that the government paid to subsidize her rent, even though her 6
landlord had allegedly made false statements to the government by 7
overcharging the plaintiff for rent. Id. at 280-83. The Coleman 8
court acknowledged that in other FCA cases, courts had awarded 9
damages equal to the full amount of the government's payment. 10
Id. at 281-82. But the court decided that in the case before it, 11
the awardable damages were equal to the difference between the 12
market rent, and the amount that the landlord charged the 13
government including the additional, improper payments it had 14
received, i.e., the amount of the overcharge. Id. at 282. The 15
government was then made whole, receiving the full benefit of its 16
bargain –- trebled by statute. 17
The defendants also look to Medicaid and Medicare FCA 18
cases for support. They contend that adopting the plaintiff's 19
theory of damages, all such cases would result in damages equal 20
to the full amount the government paid in reimbursements to 21
physicians because "the direct benefit always goes to patients." 22
Defs.' Reply Br. at 5. 23
This is not, however, the methodology generally 24
employed by courts evaluating FCA claims based on Medicaid or 25

-- 27 of 48 --

28
Medicare fraud. In United States ex. rel. Tyson v. Amerigroup 1
Illinois, Inc., 488 F. Supp. 2d 719 (N.D. Ill. 2007), the court 2
awarded damages based on the difference between the amount of 3
Medicare payments that the defendant should have received, and 4
the amount that it had actually charged the government. Id. at 5
739. Similarly, in United States ex. rel. Doe v. DeGregorio, 510 6
F. Supp. 2d 877 (M.D. Fla. 2007), the court also held that 7
damages were the "the amount of money the government paid out by 8
reason of the false claims over and above what it would have paid 9
out if the claims had not been false." Id. at 890. 10
In short, in each of the cases cited by the 11
defendants, the government paid for a contracted service with a 12
tangible benefit -- whether it be medical care, security on 13
mortgages, or subsidized housing -- but paid too much. The 14
government in these cases got what it bargained for, but it did 15
not get all that it bargained for. Thus, courts treated the 16
difference between what the government bargained for and what it 17
actually received as the measure of damages. Here, by contrast, 18
the government bargained for something qualitatively, but not 19
quantifiably, different from what it received. 20
This approach comports with the one we discussed in 21
making a sentencing calculation of loss in United States v. 22
Canova, 412 F.3d 331, 352 (2005) (rejecting argument that 23
abbreviated medical tests performed by the defendant were as 24
clinically sound as full tests required by Medicare so that the 25

-- 28 of 48 --

29
government sustained no loss). There, we explained that it was 1
not a court's task to second-guess a victim's judgment as to the 2
necessity of specifications demanded and paid for. See id. 3
("Whether the testing time on a pacemaker, the number of rivets 4
on an airplane wing, or the coats of paint on a refurbished 5
building is a matter of necessity or whim, the fact remains that 6
the victim has been induced to pay for something that it wanted 7
and was promised but did not get, thereby incurring some measure 8
of pecuniary 'loss.'") To be sure, Canova recognized that "a 9
victim's loss in a substitute goods or services case" does not 10
"necessarily equal[] the full contract price paid." Id. at 353. 11
But this was not because a defendant had the right to an offset 12
for the value of the substituted good or service. Rather, the 13
proper focus of any loss calculation was on "the 'reasonably 14
foreseeable costs of making substitute transactions and handling 15
or disposing of the product delivered or retrofitting the product 16
so that it can be used for its intended purpose,' plus the 17
'reasonably foreseeable cost of rectifying the actual or 18
potential disruption to [the victim's] operations caused by the 19
product substitution." Id. (quoting U.S.S.G. § 2f1.1, cmt 20
n.8(c)). Canova emphasized that a court calculating loss cannot 21
simply "rewrit[e] the parties' contract to excise specifications 22
paid for but not received and, thereby, conclud[e] that the 23
victim sustained no [or a reduced] loss." Id. 24

-- 29 of 48 --

30
Canova's reasoning supports the challenged loss 1
calculation. As a result of the fraudulent renewals, the 2
government was paying for a program that was not at all as 3
specified. By contrast to the Medicare cases cited by 4
defendants, the government did not receive less than it bargained 5
for; it did not get the "neuropsychology with a strong emphasis 6
upon research training with HIV/AIDS" program it bargained for at 7
all. Further, nothing in the record indicates that it could now 8
secure such a program at any lesser cost. We therefore conclude 9
that the appropriate measure of damages in this case is the full 10
amount the government paid based on materially false statements. 11
B. Fraudulent Inducement 12
The defendants acknowledge that courts have applied the 13
plaintiff's theory of damages in cases including Mackby, Rogan, 14
and Longhi, but argue that those cases are distinguishable from 15
this one because the defendants in each of those cases obtained 16
funds through fraudulent inducement -- and that any such theory 17
would fail here because no liability was found with respect to 18
the Grant Application. "In a fraudulent inducement case, [it is] 19
the false statements [that] allow the defendant to obtain the 20
funding in the first place." Defs.' Reply Br. at 9. 21
According to the defendants, because a defendant in a 22
fraudulent inducement case would not be eligible for any funding 23
received after the initial false claim, a court in such a case 24
could properly conclude that the defendant is liable for the 25

-- 30 of 48 --

31
entire amount that the government paid. But "[h]ere, the jury 1
expressly found that the initial Application contained no false 2
statements, and there was no false certification ever at issue." 3
Id. The defendants argue that Mackby, Rogan, and Longhi 4
therefore do not support the damages theory employed by the 5
district court. 6
We see no principled distinction, however, between 7
fraudulently inducing payment initially, thereby requiring all 8
payments produced from that initial fraud to be returned to the 9
government (trebled and with certain fees and costs added as 10
provided by statute), and requiring payments based on false 11
statements to be returned to the government when those false 12
statements were made after an initial contractual relationship 13
based on truthful statements had been established. Although it 14
may be true that under a fraudulent inducement theory, 15
"subsequent claims for payment made under the contract [that] 16
were not literally false, [because] they derived from the 17
original fraudulent misrepresentation, [are also] . . . 18
actionable false claims," Longhi, 575 F.3d at 468 (second 19
brackets in original; internal quotation marks omitted), this 20
proposition simply speaks to the time period for which FCA 21
liability may be found. It does not suggest that without 22
fraudulent inducement, no subsequent false statements can result 23
in FCA liability. 24

-- 31 of 48 --

32
If the government made payment based on a false 1
statement, then that is enough for liability in an FCA case, 2
regardless of whether that false statement comes at the beginning 3
of a contractual relationship or later. The only difference 4
would be that liability begins when the false statement is made 5
and relied upon, rather than at the beginning of the contractual 6
relationship, as it would be in a fraudulent inducement case. 7
Here, the jury found that materially false statements had been 8
made by the defendants in years 3, 4, and 5 of the grant, and the 9
court properly awarded damages based on that finding. 10
C. Damages as a Matter of Law 11
The defendants argue that the calculation of damages 12
should have been decided as a question of fact by a jury, not as 13
a matter of law by the district court. Indeed, in FCA cases, the 14
jury ordinarily does determine the amount of damages to be 15
imposed upon the defendant. See Chandler, 538 U.S. at 132. We 16
conclude, however, that here, where the question is not the 17
benefit of the bargain between the plaintiff and the defendants, 18
and the amount of each payment for which liability has been 19
assessed is not in dispute, no further finding of fact as to the 20
amount of the damages was necessary. 21
As the government correctly observes in its amicus 22
brief, awarding damages in this manner is not novel. And often, 23
the amount of damages in such cases has been determined as a 24
matter of law in the course of the court's grant of summary 25

-- 32 of 48 --

33
judgment to the plaintiff. See, e.g., Longhi, 575 F.3d at 461 1
(affirming summary judgment and damages award); United States v. 2
TDC Mgmt. Corp., 288 F.3d 421, 428 (D.C. Cir. 2002) (agreeing 3
that the district court could properly decide the damages award 4
where the government received no benefit from the transaction). 5
United States ex rel. Antidiscrimination Center of 6
Metro New York, Inc. v. Westchester County, No. 06 Civ. 2860, 7
2009 WL 1108517, 2009 U.S. Dist. LEXIS 35041 (S.D.N.Y. Apr. 24, 8
2009), is illustrative. There the federal government paid 9
approximately $52 million as part of a federal grant to 10
Westchester County for the purposes of housing and community 11
development. Id. at *2-*4, 2009 U.S. Dist. LEXIS 35041, at *5- 12
*11. The grant required the county to certify that it would 13
"conduct an analysis of impediments . . . to fair housing choice, 14
including those impediments imposed by racial discrimination and 15
segregation, to take appropriate actions to overcome the effects 16
of any identified impediments, and to maintain records reflecting 17
the analysis and actions." Id. at *1, 2009 U.S. Dist. LEXIS 18
35041, at *2-*3. The court granted summary judgment for the 19
plaintiff after finding that Westchester County had not conducted 20
the analysis as promised. The court agreed with the plaintiff's 21
contention that damages should be the full amount the government 22
paid, and rejected the county's argument that the damages 23
question should be submitted to the jury. There, as here, "the 24
United States did not get what it paid for," and there was no 25

-- 33 of 48 --

34
role for the jury because "Westchester's damages cannot be 1
reduced by reference to the alleged 'benefit' it provided to 2
HUD." Id. at *3, 2009 U.S. Dist. LEXIS 35041, at *9. 3
We conclude that in the case before us, inasmuch as the 4
damages equal the full amount that the government paid and that 5
amount is not in dispute, they were properly determined by the 6
district court as a matter of law. 7
D. Sufficiency of the Evidence 8
Finally, the defendants contend that the plaintiff did 9
not submit sufficient evidence to the jury to establish by a 10
preponderance of the evidence that the government suffered 11
damages equal to the full amount of the T32 grant. The 12
defendants argue that "to prove that the amount of damages was 13
the entire amount of the grant, a relator would be required to 14
prove that the government received no value -- at all -- through 15
the grant work it funded." Defs.' Br. at 37. 16
The defendants support this contention by citing 17
benefit-of-the-bargain cases. The defendants' argument is 18
therefore unavailing. Unlike a benefit-of-the-bargain case, no 19
specific amount of damages must be proved because, as we have 20
explained at length, damages in this case equal the entire amount 21
of the grant that was lost as a result of the fraud. 22
II. Materiality 23
The defendants assert that the false statements to the 24
government that are at issue were not material to the 25

-- 34 of 48 --

7 For the reasons referred to in note [5], supra, we assume
that materiality is required by the pre-2009 version of the FCA,
although we need not decide that issue on this appeal.
35
transactions in question. The district court therefore erred, 1
they say, in denying the defendants' motion for judgment as a 2
matter of law and for a new trial. 7
3
We conclude that the jury had sufficient evidence from 4
which to conclude, as it did, that the defendants' false 5
statements materially influenced NIH's decisions to renew the T32 6
grant. 7
A motion for a new trial will ordinarily be granted "so 8
long as the district court determines that, in its independent 9
judgment, the jury has reached a seriously erroneous result or 10
[its] verdict is a miscarriage of justice." Nimely v. City of 11
New York, 414 F.3d 381, 392 (2d Cir. 2005) (internal quotation 12
marks omitted). We review the district court's denial of a 13
motion for a new trial for abuse of discretion. Id. 14
A motion for judgment as a matter of law may be granted 15
only "[i]f a party has been fully heard on an issue during a jury 16
trial and the court finds that a reasonable jury would not have a 17
legally sufficient evidentiary basis to find for the party on 18
that issue." Fed. R. Civ. P. 50(a)(1). "A court evaluating such 19
a motion cannot assess the weight of conflicting evidence, pass 20
on the credibility of witnesses, or substitute its judgment for 21
that of the jury." Black v. Finantra Capital, Inc., 418 F.3d 22

-- 35 of 48 --

36
203, 209 (2d Cir. 2005) (internal quotation marks omitted). 1
Because such a judgment is made as a matter of law, we review it 2
de novo. We must "consider the evidence in the light most 3
favorable to the party against whom the motion was made and . . . 4
give that party the benefit of all reasonable inferences that the 5
jury might have drawn in his favor from the evidence." Id. at 6
208-09 (internal quotation marks omitted). 7
The district court concluded that the plaintiff had 8
"presented significant documentary evidence to support a finding 9
of materiality." Feldman IV, 2010 WL 5094402, at *2, 2010 U.S. 10
Dist. LEXIS 130358, at *5. 11
First, the parties stipulated that in order for a 12
grantee to receive additional funding after the initial grant 13
year, the "grantee must submit a noncompetitive renewal 14
application . . . includ[ing] a progress report which NIH expects 15
will provide information about the trainees['] activities during 16
the previous funding period." Id. Second, the renewal 17
instructions for the T32 grant contain a statement explaining 18
that "'Progress Reports provide information to awarding component 19
staff that is essential in the assessment of changes in scope or 20
research objectives . . . from those actually funded. They are 21
also an important information source for the awarding component 22
staff in preparing annual reports, in planning programs, and in 23
communicating scientific accomplishments to the public and to 24
Congress.'" Id., 2010 U.S. Dist. LEXIS 130358, at *6 (quoting 25

-- 36 of 48 --

37
NIH Grant Continuation Instructions at 7, J.A. 2462). Third, the 1
renewal instructions direct grantees to "highlight progress in 2
implementation and developments or changes that have occurred. 3
Note any difficulties encountered by the program. Describe 4
changes in the program for the next budget period, including 5
changes in training faculty and significant changes in available 6
space and/or facilities." Id. (internal quotation marks and 7
brackets omitted). The instructions also ask for "'information 8
describing which, if any, faculty and/or mentors have left the 9
program.'" Id. at *3, 2010 U.S. Dist. LEXIS 130358, at *6-*7 10
(quoting T32 Program Announcement PA-06-648 at 22, J.A. 2581 11
(June 16, 2006)). 12
The district court rejected the defendants' argument 13
that the jury was required to accept Dr. Robert Bornstein's 14
unrebutted testimony on the issue of materiality. Id., 2010 U.S. 15
Dist. LEXIS 130358, at *7. Bornstein was a member of the IRG 16
that reviewed the defendants' initial grant application. At 17
trial, he testified as to the factors he considered material to 18
his analysis of a grant application. He asserted that although 19
he reviewed the application, he did not expect that every faculty 20
member identified in the initial grant application would be 21
involved with the fellowship program. He also testified that he 22
did not expect the fellowship program to follow the exact 23
curriculum outlined in the initial application. The defendants 24

-- 37 of 48 --

38
argued that this testimony established that not all false 1
statements in the renewal applications were material. 2
The district court rejected this argument because 3
Bornstein never reviewed the renewal applications, nor did he 4
have an independent recollection of reviewing the initial grant 5
application. Id., 2010 U.S. Dist. LEXIS 130358, at *7-*8. The 6
court also concluded that "[t]he absence of testimony by a 7
government official supporting a finding of materiality does not 8
mean that the jury was required to accept Bornstein's testimony." 9
Id., 2010 U.S. Dist. LEXIS 130358, at *7. "[T]he jury was well 10
within its bounds to credit NIH's unambiguous guidelines and 11
instructions over Bornstein's conclusory testimony that little in 12
the Grant Application really would have mattered to him had he 13
remembered reviewing it at all." Id., 2010 U.S. Dist. LEXIS 14
130358, at *8. 15
On appeal, the defendants do not dispute that the 16
renewal applications contained NIH's instructions and guidelines. 17
They contend instead that "none of these statements, taken 18
individually or together, establish what information was material 19
to NIH's funding decisions on renewals," Defs.' Br. at 47, "the 20
Renewal Instructions and the Program Announcement are silent as 21
to what information matters to NIH for purposes of its funding 22
decision." Id. at 52. The defendants argue in substance that 23
there is no evidence from which the jury could have decided that 24

-- 38 of 48 --

8 The defendants also argue, however, that the district
court erred in interpreting NIH's guidelines as "unambiguous" --
in other words, that to the extent the plaintiff did point to
evidence of materiality, that evidence was insufficient to
support a jury verdict. Defs. Br. at 52. The defendants note
that the renewal application's instructions do not specify what
information needs to be included in a progress report, only that
the report should include "difficulties" with or "changes" to a
grant program. Id. at 53. The instructions do not explicitly
state that grantees must report all changes. Because the NIH
guidelines are "necessarily ambiguous," defendants argue that the
court cannot rely upon these guidelines as a "legal standard for
materiality." Id.
But the district court never relied on these guidelines, nor
instructed the jury to rely on these guidelines, as a "standard
for materiality." The guidelines served instead as evidence that
the jury was permitted to rely upon in evaluating what was
material to the government in its monitoring of grants.
Therefore, we agree with the district court that they provided
sufficient evidence from which the jury could reach a conclusion
as to materiality. To the extent that these guidelines are
ambiguous, it was the jury's function to resolve any disputes
about their meaning.
39
the statements it found to be false materially influenced NIH's 1
decision to renew the T32 grant. 8
2
This argument, however, misapprehends the focus of the 3
materiality analysis. In Rogan, the defendant hospital admitted 4
patients through illegal referrals in violation of the Anti- 5
Kickback Act, 42 U.S.C. § 1320a-7b. 517 F.3d at 452. Because of 6
the violation, the defendant was ineligible to receive Medicare 7
payments. The defendant did not deny that it had violated the 8
Act, but instead argued that its failure to disclose information 9
regarding the illegal referrals was immaterial to the 10
government's decision to approve the hospital's Medicare claims, 11
because materiality could only be established if a government 12

-- 39 of 48 --

40
employee involved in the decision making process testified that 1
the government would have terminated payments. Id. 2
The court rejected this view of materiality, explaining 3
that a "statement or omission is 'capable of influencing' a 4
decision even if those who make the decision are negligent and 5
fail to appreciate the statement's significance." Id. As the 6
court stated, "[t]he question is not remotely whether [the 7
applicant] was sure to be caught . . . but whether the omission 8
could have influenced the agency's decision." Id. 9
In short, even if a program officer does not 10
subjectively consider a statement to be material, it can be found 11
to be material from an objective standpoint because it is 12
"capable of influencing" the program officer. Id. As the 13
plaintiff in this case argues, materiality is "determined not by 14
what a program officer at NIH declares material, but rather [is] 15
based on the agency's own rules and regulations." Pl.'s Br. at 16
48. 17
The Rogan court discussed the purpose of laws 18
prohibiting fraud: 19
Another way to see this is to recognize that 20
laws against fraud protect the gullible and 21
the careless -- perhaps especially the 22
gullible and the careless -- and could not 23
serve that function if proof of materiality 24
depended on establishing that the recipient 25
of the statement would have protected his own 26
interests. The United States is entitled to 27
guard the public fisc against schemes 28
designed to take advantage of overworked, 29
harried, or inattentive disbursing officers; 30

-- 40 of 48 --

41
the False Claims Act does this by insisting 1
that persons who send bills to the Treasury 2
tell the truth. 3
517 F.3d at 452 (citation omitted). 4
We agree with the plaintiff that the test for 5
materiality is an objective one. It does not require evidence 6
that a program officer relied upon the specific falsehoods proven 7
to have been false in each case in order for them to be material. 8
The fact-finder must determine only whether the proven falsehoods 9
have a "natural tendency to influence, or be capable of 10
influencing, the payment or receipt of money or property." 31 11
U.S.C. § 3729(b)(4). 12
To decide otherwise –- that materiality must be 13
established in each case based on the testimony of a 14
decisionmaker –- would subvert the remedial purpose of the FCA. 15
The resolution of each case would depend on whether such a 16
decisionmaker could be identified and located, and whether that 17
particular person would have treated the claims as material, 18
regardless of whether they were one of several individuals 19
charged with evaluating the claims at issue. 20
The defendants' contention would also render the 21
language of the statute superfluous. If no one other than an 22
actual decisionmaker could determine whether a statement had a 23
"natural tendency to influence" payment, the statute could have 24
provided that a statement is "material" if it actually influenced 25
a decision maker who was aware of the statement. 26

-- 41 of 48 --

42
Our conclusion finds support in other areas of the law. 1
In TSC Indus., Inc. v. Northway, Inc., 426 U.S. 438 (1976), for 2
example, the Supreme Court addressed the meaning of "materiality" 3
in the context of a suit brought under the federal securities 4
laws. The Court determined that a fact is "material" if there is 5
a "substantial likelihood that a reasonable shareholder would 6
consider it important in deciding how to vote." Id. at 448. 7
As an abstract proposition, the most 8
desirable role for a court in a suit of this 9
sort . . . would perhaps be to determine 10
whether in fact the proposal would have been 11
favored by the shareholders and consummated 12
in the absence of any misstatement or 13
omission. But as we [have] recognized . . . 14
such matters are not subject to determination 15
with certainty. Doubts as to the critical 16
nature of information misstated or omitted 17
will be commonplace. And particularly in 18
view of the prophylactic purpose of the Rule 19
and the fact that the content of the proxy 20
statement is within management's control, it 21
is appropriate that these doubts be resolved 22
in favor of those the statute is designed to 23
protect. 24
Id. 25
The same reasoning applies here. Like the securities 26
laws at issue in TSC Industries, this objective approach ensures 27
that the FCA serves as a robust prophylactic against fraud by 28
putting the question of materiality to the jury, rather than 29
attempting to trace it back to the state of mind of the 30
decisionmaker. 31
In Bustamante v. First Federal Savings & Loan 32
Association of San Antonio, 619 F.2d 360 (5th Cir. 1980), the 33

-- 42 of 48 --

43
plaintiffs alleged that the defendants violated the Truth-in- 1
Lending Act in a loan transaction. The court noted that 2
when a security interest [with an exception 3
not relevant here] is acquired in real 4
property which is the residence of the person 5
to whom credit is extended, the borrower has 6
a right of rescission within three business 7
days of either consummation of the 8
transaction or "the delivery of the 9
disclosures required under this section and 10
all other material disclosures required under 11
this part, whichever is later . . . ." 12
Id. at 362. Here again, the court applied an objective rather 13
than a subjective materiality standard. "[T]o apply a subjective 14
standard to the test for materiality would misperceive the 15
remedial purpose of the Act." Id. at 364. The court concluded 16
that if materiality could be established by a subjective 17
determination of whether or not particular information would 18
affect a credit shopper's decision to utilize the credit, 19
unsophisticated or uneducated consumers would not be sufficiently 20
protected. Id. 21
Having concluded that the test of materiality in the 22
case before us is objective -- asking what would have influenced 23
the judgment of a reasonable reviewing official -- rather than 24
subjective -- asking whether it influenced the judgment of a 25
reviewer of a proposal in the case at hand -- we agree with the 26
district court that a reasonable jury could have found the 27
defendants' statements to be material to the renewal decisions in 28
the third, fourth, and fifth years of the grant. Based on the 29

-- 43 of 48 --

44
stipulations regarding criteria relevant to funding and the 1
testimony at trial, the jury had an ample basis for understanding 2
the grant process based upon which it could determine whether 3
statements that were made or omitted concerning changes to 4
curriculum, personnel and clinical opportunities in the renewal 5
applications had a "natural tendency" to influence NIH's funding 6
decisions. The instructions regarding the grant application and 7
renewal process provided the jury with a clear understanding of 8
what information the NIH considers in evaluating progress 9
reports, such as changes or developments to the program. 10
The defendants did not inform NIH that not all faculty 11
members identified in the initial grant were "key personnel" in 12
the program. The defendants also failed to inform NIH that 13
several of the core courses listed in the proposed curriculum 14
were never implemented, and that fellows were never evaluated by 15
a training committee. NIH was not informed that the fellows did 16
not have access to research and clinical resources described in 17
the initial grant application. NIH was also not aware that the 18
fellows had very limited access to HIV positive patients in their 19
research. In addition, many of the fellows spent much of their 20
time working on projects unrelated to HIV, such as research into 21
aging and epilepsy, which was not reported to the NIH. We 22
conclude that these facts were more than sufficient to allow a 23
reasonable jury to conclude that had the facts been disclosed 24
they would have had a natural tendency to influence, or would 25

-- 44 of 48 --

45
have been capable of influencing, the decision to renew the grant 1
and pay money to the defendants pursuant to it. 2
We therefore also conclude that the district court did 3
not abuse its discretion in denying the motion for a new trial -- 4
the jury's verdict was not "seriously erroneous" or "a 5
miscarriage of justice." Nimely, 414 F.3d at 392 (internal 6
quotation marks omitted). 7
III. Exclusion of Evidence 8
Demonstrating NIH's Inaction 9
The defendants argue that the district court abused its 10
discretion by excluding evidence of NIH's alleged failure to take 11
remedial action in response to the plaintiff's complaints, and 12
that a new trial is therefore warranted. We review a district 13
court's decision to exclude evidence for abuse of discretion. 14
Schering Corp. v. Pfizer Inc., 189 F.3d 218, 224 (2d Cir. 1999). 15
"We [also] review a district court's denial of a motion for a new 16
trial for abuse of discretion." United States v. Brunshtein, 344 17
F.3d 91, 101 (2d Cir. 2003), cert. denied, 543 U.S. 823 (2004). 18
The defendants contend that they should have been 19
permitted to elicit evidence of NIH's relative inaction in 20
response to complaints because it is relevant as to whether or 21
not their statements in the renewal applications were false and 22
material. Feldman told NIH about the defendants' fraudulent 23
claims and, according to the defendants, the agency saw no 24
validity to the complaints as evidenced by its failure to take 25

-- 45 of 48 --

46
action beyond asking Cornell itself to investigate the 1
complaints. The defendants argue that they should have been able 2
to present this evidence to the jury in an effort to persuade it 3
that the statements had not misled the agency. If this evidence 4
was presented, they say, the plaintiff "could then have put on 5
any rebuttal evidence about why the jury should find the 6
statements were false and material despite NIH's lack of reaction 7
when presented with those allegations." Defs.' Br. at 61. 8
Federal Rule of Evidence 402, provides, inter alia, 9
that "[i]rrelevant evidence is not admissible." The district 10
court reasoned that the evidence in question was irrelevant 11
because the NIH's failure to act in response to Feldman's 12
complaints did not speak to the seriousness of those complaints 13
or the likelihood that false claims had been made. The jury did 14
not have before it the standard that NIH used to determine 15
whether or not action was warranted in response to a funding 16
complaint. "[N]o discovery was conducted concerning the 17
standards these agencies employ to determine the existence of 18
misconduct and whether those standards are at all similar to the 19
elements of an FCA claim." Feldman III, 2010 WL 2911606, at *3, 20
2010 U.S. Dist. LEXIS 73633, at *7. "Specifically, as to [the 21
plaintiff's] deposition testimony on the NIH decision, [he] does 22
not, and indeed cannot, speak to the standards NIH used to judge 23
the merits of his claims." Id. Without evidence as to what the 24
standards of the agency were for beginning an investigation, the 25

-- 46 of 48 --

9 The defendants point to United States v. Southland
Management Corp., 326 F.3d 669 (5th Cir. 2003) (en banc), where
the court considered the relevance of the course of conduct
between a landlord receiving Section 8 funds and HUD. The court
concluded that the communication between HUD and the landlord
demonstrated that "HUD was willing to work with the Owners" on
remedying maintenance problems, and that "HUD seemed to recognize
that the property's noncompliance was at least partially
explained by a lack of funds and nearby criminal activity." Id.
at 677. Based in part on this pattern of honest and open
communication, the court concluded that there could be no FCA
liability. Unlike in Southland Management, there is no
indication here that the defendants communicated compliance
issues to the government or sought its help in addressing them.
Where the government acts in response to potential false claims,
its activity may reveal something about its understanding as to
whether those claims were deliberately false or the result of
extrinsic factors, as in Southland Management. But where, as
here, there is no evidence of government action, nothing relevant
can be ascertained without knowing for which of many possible
reasons it did not act.
The defendants also cite United States ex rel.
Kreindler & Kreindler v. United Technologies Corp., 985 F.2d 1148
(2d Cir. 1993), in which we stated that "government knowledge may
be relevant to a defendant's liability." Id. at 1157. Indeed it
"may be," but is not here where the significance of the knowledge
and the responsibilities of the recipients have not been
established.
47
jury could not determine whether the complaints made by Feldman 1
should have instigated one. 9 Id., 2010 U.S. Dist. LEXIS 73633, 2
at *7-*8. 3
The defendants further argue that to the extent that 4
the district court excluded evidence of NIH's inaction pursuant 5
to Rule 403, it did so in error. While ultimately we would be 6
inclined to agree with the district court, we need go no further 7
in our analysis because the evidence was properly excluded under 8

-- 47 of 48 --

48
Rule 402 in any event. This conclusion was not an abuse of 1
discretion. 2
CONCLUSION 3
For the foregoing reasons, we affirm the judgment of 4
the district court. 5

-- 48 of 48 --