5

INTRODUCTION
The Supreme Court’s decision in McDonnell v. United States1 upended the anti-corruption landscape by narrowing the
scope of honest services fraud 2 and Hobbs Act extortion 3 —two federal statutes commonly used to prosecute
corruption by state and local officials.4 In prosecutions under those statutes, jurors are often instructed that they
must find that a public official took bribes in exchange for an agreement to perform an “official act.”5 In McDonnell,
the Court held that merely setting up meetings with government officials or voicing support for causes are generally
not official acts. 6 Instead officials perform official acts only when they make a formal decision on a matter
themselves or use their position to exert pressure on another official to do so.7

With McDonnell limiting the scope of honest services fraud and Hobbs Act extortion, federal prosecutors might be
expected to turn to a third statute commonly used to prosecute state and local corruption: 18 U.S.C. § 666,
commonly known as federal program bribery. 8 The language of Section 666 is broader than the “official act”
language in McDonnell, making it a crime to give anything of value to “any person” with the intent to “influence” a
public official “in connection with any business, transaction, or series of transactions” of the entity involving
anything of value of $5,000 or more.9 Because McDonnell was ostensibly decided on statutory interpretation grounds,
and Section 666 is a different statute, prosecutors could theoretically bring charges under § 666 when public officials
accept bribes in exchange for setting up meetings and voicing support for causes—conduct the Court in McDonnell
found insufficient to constitute an official act for honest services fraud and Hobbs Act extortion purposes. This
possibility is bolstered by the fact that every court which has addressed the issue, including the Second and Sixth
Circuits, has found that McDonnell did not create an official act requirement for prosecutions under Section 666.10

But there are reasons to be skeptical. While the Court in McDonnell purported to decide the case on statutory
grounds, it noted that a broader interpretation of the statute might raise constitutional concerns by criminalizing
everyday politics and putting the federal government in the position of “setting standards of good government for

1 136 S. Ct. 2355 (2016).

2 18 U.S.C. § 1346.
3 Id. § 1951.
4 See Thomas M. DiBiagio, Federal Public Corruption Statutes Targeting State and Local Officials: Understanding the Core Legal Element and the

Government’s Burden of Proving Corrupt Intent After McDonnell, 7 U. DENV. CRIM. L. REV. 47, 47 (2017).
5 The term “official act” and its definition are borrowed from a different federal anti-bribery statute. See 18 U.S.C. § 201(a)(3) (defining an

“official act” as “any decision or action on any question, matter, cause, suit, proceeding or controversy, which may at any time be pending,
or which may by law be brought before any public official, in such official's official capacity, or in such official's place of trust or profit.”).
While the Hobbs Act and honest services fraud statutes themselves do not mention official acts, in prosecutions under those statutes,
courts routinely instruct jurors that they must find an agreement to commit an official act, apparently motivated by a desire to avoid the
constitutional vagueness concerns recognized by the Supreme Court in Skilling v. United States, 561 U.S. 358 (2010). In Skilling, the Court
cited constitutional concerns in limiting the reach of the honest services fraud statute to bribery and kickback schemes. And while the
Court in Skilling did not definitively hold that constitutional applications of the honest services fraud require a finding of an official act, it
noted that the statute’s prohibition on bribery and kickbacks should “draw content” from similar statutes, including § 201, which contains
an official act requirement. Id. at 412.
6 McDonnell, 136 S. Ct. at 2371.
7 Id.
8 See DiBiagio, supra note 5, at 47.
9 Section 666 makes it a crime to give or solicit bribes to “any person” with the intent to influence a public official of certain government

entities “in connection with any business, transaction, or series of transactions” of the entity involving anything of value of $5,000 or more.
18 U.S.C. § 666(a)(1)(B)–(2).
10 See United States v. Porter, 886 F.3d 562, 565–66 (6th Cir. 2018); United States v. Boyland, 862 F.3d 279, 291 (2d Cir. 2017); United States v.

Ferguson, No. 10-20403, 2018 WL 1071743, at *4 (E.D. Mich. Feb. 27, 2018); United States v. Percoco, 16-CR-776 (VEC), 2017 WL 6314146,
*2–4 (S.D.N.Y. Dec. 11, 2017).

1
local and state officials.”11 A broad construction of Section 666 that reaches conduct similar to that at issue in
McDonnell might raise the same issues.

This paper argues that this concern is well-founded. A broad reading of Section 666 raises many of the same
constitutional concerns that the Supreme Court flagged in McDonnell and is hard to square with the narrow
approach that the Court has taken to public corruption statutes in recent years.12 Part I of this paper analyzes the
Court’s opinion in McDonnell and argues that constitutional concerns played a bigger role in the decision than most
courts and commentators have been willing to acknowledge. Part II examines the differences between Section 666
and the statute at issue in McDonnell, exploring whether those differences are enough to save Section 666 from
constitutional problems and ultimately concluding that they are not. Part III looks to the implications of that
conclusion, arguing that prosecutors bringing a § 666 charge should request an official act instruction when they
think they can prove it and should reserve prosecutions without an official act instruction for egregious corruption
cases where they nevertheless are not confident than they can prove an agreement to perform official act.

I. MCDONNELL V. UNITED STATES

A. Background
In McDonnell, the Supreme Court confronted the case of former Virginia governor Bob McDonnell, who accepted
more than $175,000 in gifts and loans from the CEO of a nutritional supplement company, Star Scientific, in
exchange for helping to promote one of the company’s products—a supplement called Anatabloc.13 To convict, the
jury was instructed that it had to find that McDonnell accepted the gifts and loans in exchange for agreeing to
perform an “official act,” as defined in a separate federal bribery statute.14 The government alleged that McDonnell
had agreed to take five “official acts”:

1. Arranging meetings between the company’s CEO and other Virginia government officials to
discuss and promote Anatabloc.

2. Hosting and attending events at the governor’s mansion designed to encourage Virginia university
researchers to initiate studies of the active ingredient in Anatabloc and promote the product to
doctors.

3. Contacting other officials in the governor’s office as part of an effort to encourage the universities
to begin the aforementioned studies.

4. Allowing the CEO to invite individuals important to Star Scientific’s business to exclusive events
at the governor’s mansion.

5. Recommending that Virginia government officials meet with representatives of Star Scientific to
discuss ways that the company’s products could lower healthcare costs.15

A jury convicted McDonnell of honest services fraud and Hobbs Act extortion,16 but the Supreme Court reversed.
While noting that there was “no doubt” that McDonnell’s conduct was “distasteful,”17 the Court held unanimously

11 McDonnell, 136 S. Ct. 2373.

12 See id.; Skilling v. United States, 561 U.S. 358 (2010); United States v. Sun-Diamond Growers, 526 U.S. 398 (1999).
13 McDonnell, 136 S. Ct. at 2361.
14 Id. at 2365. See supra note 5 and accompanying text.
15 McDonnell, 136 S. Ct. at 2365–66.
16 Id.
17 Id. at 2375.

2
that arranging meetings, hosting events, and contacting other officials to voice support for particular matters—
without more—did not qualify as “official acts” under the statute.18

B. The Basis for the Decision

The Court wrote that it had taken into account “the text of the statute, the precedent of this court, and . . .
constitutional concerns” in reaching its decision.19 But this vague summary failed to explain the extent to which
constitutional concerns drove the decision, leaving its effect on different statutes, like Section 666, unclear.

1. Statutory Analysis
The bulk of the Court’s opinion was devoted to statutory interpretation. An “official act” is defined by the relevant
statute as “any decision or action on any question, matter, cause, suit, proceeding or controversy, which may at any
time be pending, or which may by law be brought before any public official, in such official's official capacity.”20
The Court first examined whether arranging a meeting or contacting another official to voice support for an
initiative could be a “question, matter, cause, suit, proceeding or controversy.” 21 It found that they could not
because the statutory language “connote[s] a formal exercise of government power” like a lawsuit, hearing, or
administrative decision.22

The Court then turned to whether arranging meetings or voicing support could be “decision[s] or action[s]” on the
matters of convincing Virginia government entities to fund research of Anatabloc and include the supplement in
the state employee health insurance plan. While hosting meetings and events and speaking with interested parties
may be related to a particular matter, the Court concluded they do not qualify as “decisions” or “actions” on it.23
Instead, an official must do something more concrete like personally approving a research study—or at the very
least, making a decision on a qualifying step, such as narrowing the scope of potential research topics.24

2. Constitutional Analysis

Having decided that McDonnell’s conduct did not fall within the text of the official act statute, the Court could
have stopped there. But it did not, instead proceeding to explain that the government’s broad interpretation of the
statute raised “significant constitutional concerns.”25

The Court pointed to three separate constitutional issues. First, a reading of the statute that made nearly everything
a public official accepts a quid and nearly everything he or she does a quo would risk criminalizing everyday
politics.26 The Court expressed “substantial” concern that “officials might wonder whether they could respond to
even the most commonplace requests for assistance, and citizens with legitimate concerns might shrink from
participating in democratic discourse.”27 Second, the Court suggested that a broad interpretation of “official act”
could make the term so indefinite as to raise constitutional vagueness issues.28 Finally, the Court wrote that giving

18 Id. at 2371.
19 Id. at 2367.
20 18 U.S.C. § 201(a)(3).
21 McDonnell, 136 S. Ct. at 2368–69.
22 Id.
23 Id. at 2370 (citing United States v. Sun-Diamond Growers of Cal., 526 U.S. 398, 407–08 (1999)).
24 Id. The official does not have to actually make the decision himself—it is enough if he uses his position to exert pressure on another

official to take official action or provides advice, knowing that it will form the basis of an official act by another. Id.
25 Id. at 2372.
26 Id.
27 Id.
28 Id. at 2373.

3
the statute a broad reach could raise federalism concerns by putting the federal government in the position of
“setting standards of good government for state and local officials.”29

But while the Court devoted a section of its opinion to discussing the potential constitutional issues lurking in the
government’s view of the statute, it failed to make clear the extent to which they drove the statutory analysis. It did
not invoke the canon of constitutional avoidance, under which statutes susceptible to multiple interpretations are
construed to avoid constitutional issues,30 nor did it otherwise expressly state that the government’s reading of the
statute would render it unconstitutional. However, it seems unlikely that all nine justices would sign on to the
constitutional discussion if it was merely gratuitous dicta. Its inclusion at least strongly suggests that the Court
believed that a narrow interpretation of “official act” was constitutionally required.

II. FEDERAL PROGRAM BRIBERY (SECTION 666)

A. Background
Whether McDonnell should be read as a statutory or constitutional decision has important implications for 18
U.S.C. § 666, the federal program bribery statute. Section 666 differs from the honest services fraud statute in ways
that could limit McDonnell’s effect on it.

1. Textual Differences
The text of Section 666 contains broader language than the official act statute at issue in McDonnell. As discussed
above, to be convicted of honest services fraud, a public official must accept benefits in exchange for an agreement
to make a “decision or action on any question, matter, cause, suit, proceeding or controversy” pending before him
or her—a phrase the Court in McDonnell construed to be limited to formal decisions or actions on discrete matters.31

But the text of Section § 666 is not so limited. It prohibits officials from accepting “anything of value from any
person intending to be influenced or rewarded in connection with any business, transaction, or series of transactions
of such organization, government, or agency involving anything of value of $5,000 or more.”32 This broad language
would seem to permit prosecuting public officials for setting up meetings in exchange for bribes, so long as the
meetings were “in connection with” business involving the government entity—in other words, it would allow
prosecution for conduct that cannot be prosecuted as honest services fraud after McDonnell.33

2. Constitutional Differences
Congress enacted Section 666 pursuant to its authority under the Spending Clause to “appropriate federal moneys
to promote the general welfare” and its corresponding authority under the Necessary and Proper Clause to ensure
that the money it appropriates is spent for proper purposes and not “frittered away in graft.” 34 In contrast, the
honest services fraud statute is grounded in different constitutional provisions. Honest services wire fraud was
enacted pursuant to Congress’s power to regulate interstate and foreign commerce and therefore requires a showing

29 Id.
30
See, e.g., M Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 575 (1988)
31 McDonnell, 136 S. Ct. at 2370.
32 18 U.S.C. § 666(a)(1)(B).
33 Section 666 does have some limitations that honest services fraud does not. To fall within the statute, the public official must be an agent

of an organization that receives more than $10,000 per year in funds from a federal program. Id. § 666(b). And the bribe must be given in
connection with something worth $5,000 or more. Id. But prosecutors in the most serious cases have little trouble satisfying these
requirements. The governments of every state and myriad localities receive more than $10,000 from the federal government, and as long as
the funds given to a government entity exceed the threshold, there does not have to be an actual nexus between the bribe and the federal
funds. § 666(a)(1)(B). Sabri v. United States, 541 U.S. 600, 604–05 (2004).
34 Id. at 605 (citing U.S. CONST. art I, § 8, cl. 8, 18).

4
that the fraudulent communication crossed state or national borders.35 Honest services mail fraud was created as an
exercise of both Congress’s commerce power as well as its independent authority to establish and regulate post
offices.36

The Court has generally held that federalism cannot serve as an external constraint on Congress’s independent
authority under the Spending Clause to impose conditions on government grants—as long as the conditions are not
unduly coercive.37 And the Court has already held that Section 666 is not unduly coercive.38 So it is possible that
Section 666’s basis in Congress’s spending power renders the federalism concerns that the Court voiced in
McDonnell inapplicable.

Further, while the honest services fraud statute criminalizes all bribes that meet the McDonnell standard, no matter
how small, Section 666 applies only to bribes given in connection with business “involving anything of value of
$5,000 or more.” 39 By limiting criminal liability to transactions involving business worth thousands of dollars,
Section 666 might obviate the concern the Court expressed in McDonnell about casting a “pall of prosecution” over
“commonplace requests for assistance” by constituents to their elected representatives.40

In short, while Section 666 is similar to honest services fraud in many ways, there may be room to argue that the
two statutes differ enough in constitutionally significant ways such that the constitutional concerns discussed in
McDonnell simply do not translate. Indeed, this is the tack that lower courts addressing McDonnell’s effect on Section
666 have taken thus far.

B. Lower Court Treatment of McDonnell and Section 666

Since McDonnell, lower courts have uniformly rejected arguments that the decision created an “official act”
requirement for Section 666 prosecutions. In United States v. Boyland, the Second Circuit reasoned that because the
text of Section 666 is “more expansive” than the statute at issue in McDonnell, it did “not see that the McDonnell
standard applied to [Section 666] counts.”41 Similarly in United States v. Porter, the Sixth Circuit noted that McDonnell
concerned “an entirely different statute” than Section 666 and therefore held that pre-McDonnell circuit precedent
rejecting an official act requirement in Section 666 still controlled because McDonnell had not directly overruled it.42

In United States v. Percoco, the Southern District of New York addressed the constitutional implications of McDonnell,
holding that the constitutional concerns raised by the Supreme Court did not render Section 666 facially
unconstitutional.43 The court explained that McDonnell “in no way state[d] or implie[d] that all federal bribery statutes
that implicate the conduct of government officials are required to have such an [official act] element to be
constitutional.”44 But the court’s decision only rejected an attempt to facially invalidate the entire statute and did not
explicitly address whether an official act instruction was necessary to apply the statute in a constitutional manner. In

35 Devika Singh, et al., Mail and Wire Fraud, 54 AM. CRIM. L. REV. 1555, 1557 n.13 (2017).
36 Id.
37 South Dakota v. Dole, 483 U.S. 203, 210 (1987); see also George D. Brown, Stealth Statute—Corruption, the Spending Power, and the Rise of 18

U.S.C. § 666, 73 Notre Dame L. Rev. 247, 265 (1998).

38 Sabri, 541 U.S. at 608.
39 18 U.S.C. § 666(a)(2).
40 See McDonnell v. United States, 136 S. Ct. 2355, 2372–73 (2016).
41 United States v. Boyland, 862 F.3d 279, 291 (2d Cir. 2017).
42 United States v. Porter, 886 F.3d 562, 565–66 (6th Cir. 2018). The Eastern District of Michigan also distinguished McDonnell from Section

666 on statutory grounds. United States v. Ferguson, No. 10-20403, 2018 WL 1071743, at *4 (E.D. Mich. Feb. 27, 2018) (“McDonnell does
not apply to 18 U.S.C. § 666, which does not include the term “official act” or any similar term.”)
43 United States v. Percoco, 16-CR-776 (VEC), 2017 WL 6314146, at *2–4 (S.D.N.Y. Dec. 11, 2017).
44 Id. at *4.

5
fact, it expressly noted that such an instruction would be the proper cure for any constitutional concerns, should
they exist.45

C. McDonnell’s Application to Section 666

But while the lower courts have thus far avoided applying McDonnell to Section 666 cases by ignoring its
constitutional discussion—apparently treating it as pure dicta—it seems unlikely that this approach would prevail
should the issue reach the Supreme Court. As Judge Carnes put it, “there is dicta, and then there is Supreme Court
dicta.”46 Even if the constitutional discussion in McDonnell is dicta, it provides at least a strong hint that the Court
viewed the government’s broad interpretation of the honest services fraud “official act” requirement as
unconstitutional.47 And if it saw constitutional problems there, it follows that it would most likely see constitutional
problems with prosecuting officials under Section 666 for engaging in conduct that would not satisfy the official act
requirement, given the similarities between the two statutes.

A reading of Section 666 that allowed prosecutors to charge public officials with accepting gifts in exchange for
setting up meetings 48 would run squarely into McDonnell’s concern about casting a “pall of prosecution” over
everyday interactions between politicians and their constituents. 49 While some might argue that Section 666’s
requirement that the bribe be intended to influence an official “in connection with . . . business . . . involving
anything of value of $5,000 or more”50 is enough to limit the statute’s reach to only obvious bribes rather than run-
of-the-mill requests for assistance, this seems unlikely to be true in practice.

Section 666 does not require that the bribe be worth $5,000 or more. Instead, the bribe itself can be “anything of
value,” so long as it is given in connection with government business worth $5,000 or more.51 And it goes without
saying that much of what governments do is worth at least $5,000.52 For example, in McDonnell the Court expressed
concern that a hypothetical union official who had previously donated to an elected official’s political campaign
might be subject to prosecution if he later asked the official to intervene to prevent a plant closing. 53 Given the
tremendous financial consequences that a plant closing could have, it’s easy to imagine a court valuing the official’s
intervention at more than $5,000, and if that is the case, under Section 666, the value of the campaign contribution
would be immaterial—even $25 would be enough to subject the official to potential prosecution. Because the
$5,000 threshold is not enough to prevent the statute from reaching commonplace political interactions, it seems
unlikely that the Court would find it sufficient to render inapplicable its concerns about criminalizing the political
process.

A broad reading of Section 666 could also raise vagueness issues akin to those noted by the Court in McDonnell. The
statute criminalizes accepting anything of value “intending to be influenced or rewarded in connection with any

45 Id. at *4 n.2.
46 Schwab v. Crosby, 451 F.3d 1308, 1325 (11th Cir. 2006).
47 See United States v. Montero-Camargo, 208 F.3d 1122, 1132 n.17 (9th Cir. 2000) (en banc) (“Supreme Court dicta have a weight that is

greater than ordinary judicial dicta as prophecy of what that Court might hold . . . ” (internal quotation marks omitted)).
48 This would be a possibility under the text of the statute, as long as the meetings were “in connection with” the business of a government

entity. 18 U.S.C. § 666.

49 McDonnell v. United States, 136 S. Ct. 2355, 2372 (2016).
50 18 U.S.C. § 666(a)(1)(B

51 United States v. Duvall, 846 F.2d 966, 976 (5th Cir. 1988) (“[I ]t is clear that the $5000 figure qualifies the transactions or series of

transactions that the recipient of the bribe carries out in exchange for receiving ‘anything of value.’ The wording of the section does not
place a value on the bribe . . .”)
52 See, e.g., United States v. Robinson, 663 F.3d 265, 271–76 (7th Cir. 2011) (finding that the value of the intangible benefit of bribing police

officers to overlook violations of the law was worth at least $5,000, citing similar cases from other circuits).
53 McDonnell, 136 S. Ct. at 2372.

6
business, transaction, or series of transactions” of a government entity. 54 This language could be read to reach
almost anything a public official does, given that almost all of his or her actions can be said to be “in connection
with” government business. If anything, it is even vaguer than the language at issue in McDonnell, which the court
explained was “not defined with sufficient definiteness that ordinary people can understand what conduct is
prohibited,” potentially opening public officials up to “prosecution, without fair notice, for the most prosaic
interactions.”55

The Supreme Court has shown an increasing willingness to strike down statutes on vagueness grounds in recent
years,56 giving teeth to the concerns it expressed in McDonnell. And vagueness concerns drove the Court’s decision in
Skilling, where it narrowed the scope of honest services fraud because a broader reading would be unconstitutionally
vague.57 There is every reason to think that if called upon to interpret the text of Section 666, the Court would
similarly limit its reach to what it views as clearly corrupt transactions—perhaps by adopting a McDonnell official act
requirement, or something similar to it—lest elected officials be left without fair notice of the line between criminal
and non-criminal conduct.

The only constitutional issue raised in McDonnell that may not be directly applicable to Section 666 is the concern
about unduly intruding on state sovereignty. As mentioned above, Section 666 was enacted pursuant to Congress’s
spending power, and the Supreme Court has generally been more solicitous of federal intrusion through spending
conditions than it has been for regulation under the Commerce Clause. But it is not clear that grounding in a
different part of the Constitution is enough to validate a reading of the statute that could give federal prosecutors
oversight over almost anything state officials do. The Supreme Court has repeatedly declined to construe statutes in
ways that would involve “the Federal Government in setting standards of good government for local and state
officials.”58 In short, it is possible that the federalism concerns raised by deep criminal intrusion into the workings
of state and local government are so serious that they cannot justify even legislation enacted pursuant to Congress’s
spending power. And even if federalism concerns are inapplicable to Section 666, the democratic process and
vagueness concerns raised by the Court in McDonnell still appear to apply.

These constitutional concerns carry even more weight given the Court’s recent anti-corruption jurisprudence.
McDonnell came only a few years after Skilling, and taken together, they could be read as a statement that the Court
simply does not view federal regulation of the conduct of state and local officials—beyond explicit quid pro quo
bribery—as constitutionally permissible.

While lower courts interpreting Section 666 have largely ignored the constitutional discussion in McDonnell, this
smacks of wishful thinking. The best interpretation of McDonnell is that the constitutional concerns were an
animating force behind the decision. If that is the case, then it is difficult, if not impossible, to see how they would
not equally doom a broad reading of Section 666.

III. WHAT SHOULD PROSECUTORS DO?

Given the serious constitutional concerns that could arise from a broad reading of Section 666, prosecutors should
exercise caution in bringing cases under the statute.

54 18 U.S.C. § 666(a)(1)(B).
55 McDonnell, 136 S. Ct. at 2373.
56 See Sessions v. Dimaya, 138 S. Ct. 1204 (2018); Johnson v. United States, 135 S. Ct. 2551 (2015).
57 Skilling v. United States, 561 U.S. 358, 412–13 (2010).
58 McDonnell, 136 S. Ct. at 2373; see also McNally v. United States, 483 U.S. 350, 360 (1987); United States v. Enmons, 410 U.S. 396, 410–11

(1973).

7
First, when prosecutors are confident that they will be able to prove that a public official agreed to commit official
acts under the definition from McDonnell, they should request that the jury be given an official act instruction.
Requiring the jury to find an agreement to perform an official act in order to convict will eliminate any risk of the
conviction being overturned on the constitutional grounds raised by the Court in McDonnell.

Requesting an official act finding in a Section 666 case when the text of the statute does not appear to require one is
admittedly an odd move. But it is a move that courts appear to have accepted. The honest services fraud statute
does not facially require an official act finding either, but the Court in McDonnell expressed no misgivings about the
parties defining the offense in official act terms—indeed, the Court appeared to hold that an official act instruction
was required to save the statute from being declared unconstitutional.59 And in United States v. Skelos, the Second
Circuit did not object to including an official act instruction for Section 666 counts, holding only that when a jury is
instructed that it must find official acts, the definition of official acts must mirror the definition given in McDonnell.60
Therefore there should be no issue with the government agreeing to meet a more difficult standard than the statute
arguably requires.

Including an official act instruction in the easiest to prove Section 666 cases will avoid the possibility of an appellate
court ruling that such an instruction is required in all cases brought under the statute. This will preserve the
government’s flexibility to use Section 666 to prosecute egregious cases (without using an official act instruction)
when it is not confident that it will be able to prove an agreement to perform an official act.

Of course, any time a jury convicts under Section 666 without having to find an agreement to commit an official act
presents an opportunity for an appellate court—including the Supreme Court—to decide that an official act
instruction is required to save the statute from being struck down on constitutional grounds. Such a decision would
render Section 666 essentially equivalent to the honest services fraud statute, taking one of the government’s
important weapons for fighting state and local corruption out of its arsenal.61

To mitigate this risk, prosecutors should use Section 666 without an official act instruction only in cases where the
facts are sufficiently egregious. While technically irrelevant to the legal question of whether an official act instruction
is constitutionally required, judges might hesitate to overturn a conviction if they are convinced that the defendant
engaged in outrageous conduct that deserves to be punished.

This approach is obviously still risky. All it takes is one appellate court ruling to remove the possibility of charging §
666 without an official act instruction in all future cases in that jurisdiction. But, at the same time, there is little for
the government to lose by taking that risk in an egregious case where it feels that it could not prove an official act.
Even if the conviction is overturned on appeal, the government is in the same place as it would have been if had it
self-imposed an official act requirement in all cases to begin with.

59 McDonnell, 136 S. Ct. at 2372–73.

60 United States v. Skelos, 707 F. App’x 733, 737–38 (2d Cir. 2017) (summary order).
61 Because Section 666 contains some requirements not present in honest services fraud— namely that the official sought to be influenced

is an agent of a government entity that receives more than $10,000 per year in federal funding and that the business the bribe is in
connection with is worth at least $5,000, 18 U.S.C. § 666(a)(1)(B)— prosecutors would have little reason to use the statute if an official act
instruction was required because in almost all cases they could charge honest services fraud instead. The only obvious exception would be
when they could not prove the existence of an interstate wire or mail communication, as is required to convict for honest services fraud. See
18 U.S.C. §§ 1341, 1343, 1346.

8
CONCLUSION
McDonnell has undoubtedly made it more difficult for federal prosecutors to combat state and local corruption. But
through judicious use of the broad language of Section 666, the government may still be able to successfully
prosecute some corruption cases even when it would be unable to prove that the defendant agreed to commit an
official act. This tactic should be used sparingly, to reduce the risk of establishing an unfavorable precedent. But
until the Supreme Court says otherwise, it remains an option for targeting officials like Robert McDonnell who
engage in corrupt behavior that does not clear the official act bar.

9
About:
Author:
This brief was authored by Alex Botoman, Columbia Law School Class of 2018

What is CAPI?
The Center for the Advancement of Public Integrity is a nonprofit resource center dedicated to improving the
capacity of public offices, practitioners, policymakers, and engaged citizens to deter and combat corruption.
Established as partnership between the New York City Department of Investigation and Columbia Law School in
2013, CAPI is unique in its city-level focus and emphasis on practical lessons and tools.

Published: November, 2018 by the Center for the Advancement of Public Integrity at Columbia Law School.
Available at www.law.columbia.edu/CAPI.

This publication is part of an ongoing series of contributions from practitioners, policymakers, and civil society
leaders in the public integrity community. If you have expertise you would like to share, please contact us at
[email protected].

© 2018. This publication is covered by the Creative Commons “Attribution-No Derivs-NonCommercial” license (see http://creativecommons.org). It
may be reproduced in its entirety as long as the Center for the Advancement of Public Integrity at Columbia Law School is credited, a link to the
Center’s web page is provided, and no charge is imposed. The paper may not be reproduced in part or in altered form, or if a fee is charged, without the
Center’s permission. Please let the Center know if you reprint.

Cover Design by Freepik.

10