Collaborative Gatekeepers

Washington and Lee Law Review Washington and Lee Law Review

Volume 73 Issue 2 Article 6

Spring 4-1-2016

Collaborative Gatekeepers Collaborative Gatekeepers

Stavros Gadinis

University of California - Berkeley

Colby Mangels University of California - Berkeley

Follow this and additional works at: https://scholarlycommons.law.wlu.edu/wlulr

Part of the Administrative Law Commons, and the Banking and Finance Law Commons

Recommended Citation Recommended Citation

Stavros Gadinis and Colby Mangels University of California - Berkeley,

Collaborative

Gatekeepers

, 73 Wash. & Lee L. Rev. 797 (2016).

Available at: https://scholarlycommons.law.wlu.edu/wlulr/vol73/iss2/6

This Article is brought to you for free and open access by the Washington and Lee Law Review at Washington and

Lee University School of Law Scholarly Commons. It has been accepted for inclusion in Washington and Lee Law

Review by an authorized editor of Washington and Lee University School of Law Scholarly Commons. For more

information, please contact [email protected].

797

Collaborative Gatekeepers

Stavros Gadinis

∗

Colby Mangels

∗∗

Abstract

In their efforts to hold financial institutions accountable after

the 2007 financial crisis, U.S. regulators have repeatedly turned

to anti-money-laundering laws. Initially designed to fight drug

cartels and terrorists, these laws have recently yielded

billion-dollar fines for all types of bank engagement in fraud and

have spurred an overhaul of financial institutions’ internal

compliance. This increased reliance on anti-money-laundering

laws, we argue, is due to distinct features that can better help

regulators gain insights into financial fraud. Most other financial

laws enlist private firms as gatekeepers and hold them liable if

they knowingly or negligently engage in client fraud. Yet, as long

as gatekeepers maintain deniability, they can accommodate

dubious client requests. Instead, anti-money-laundering laws

require gatekeepers to report to regulators suspicions of

misconduct, even without clear proof of fraud. Because suspicions

arise early in the gatekeeper–client relationship, conflicts of

interest are not likely to be as strong. Moreover, the task of

∗ Stavros Gadinis is an Assistant Professor at Berkeley Law School. We

would like to thank Kenneth Ayotte, Douglas Baird, Robert Bartlett, Omri

Ben-Shahar, Dick Buxbaum, Anthony Casey, Steven Davidoff Solomon,

Dhammika Dharmapala, Holly Doremus, David Gamage, Mark Gergen, Tom

Ginsburg, Todd Henderson, William Hubbard, Saul Levmore, Jonathan Masur,

Richard McAdams, Tom Miles, Randy Picker, Eric Posner, Julie Roin, Andrea

Roth, Jonathan Simon, Matthew Stephenson, Lior Strahilevitz, Stephen

Sugarman, and David Weisbach. We would also like to extend a special thanks

to Luc Thévenoz, Ilias Pnevmonidis, and the staff of the Centre for Banking and

Financial Law at the University of Geneva for providing invaluable assistance

and facilities in support of our case study research in Switzerland. We would

like to acknowledge generous financial support from the Hellman Fund at UC

Berkeley, the Fulbright U.S. Student Research Program, and the Swiss Federal

Commission for Scholarships for Foreign Students. We also would like to thank

Michael B. Davis and Tony Huynh for excellent research assistance.

∗∗ Colby Mangels is a 2015 J.D. Graduate of Berkeley Law School.

798 73 WASH. & LEE L. REV. 797 (2016)

identifying suspicious cases can be more readily outsourced to

compliance departments, lessening dependence on front-line

employees whose future might be tied to specific clients. Finally,

suspicions may arise even in gatekeepers who only have partial

access to clients’ transactions and, thus, cannot come to full

knowledge of the fraud.

Inspired by the collaborative relationship between gatekeepers

and enforcement authorities in anti-money laundering, we develop

a theoretical framework that explains why this approach could

operate as a general template for financial regulation. We then

investigate the implementation of the collaborative model in

practice. Starting from anti-money-laundering laws’ history, we

present new evidence from recently released archival materials to

illustrate that, rather than fighting proposals for expanding their

regulatory obligations, private industry embraced them. Turning

to the present, we discuss how the collaborative model has

reshaped banking oversight in money laundering: It has leveraged

the power of big data, encouraged the creation of dedicated

compliance departments, and spearheaded one of the biggest

inter-agency collaborations in the United States. Finally, we

discuss how the collaborative model could work in the future in

two other areas of financial activity: broker-dealer regulation and

equity issuance.

Table of Contents

I. Introduction .....................................................................800

II. Gatekeepers in Current Theory and Practice .................808

A. The Theory of Gatekeeping—Market-Based

Regulation Through Reputation ...............................808

B. The Harsh Realities of Gatekeeping:

Repeated Collapses Show Weaknesses in

Reputational Model ...................................................812

1. The Conflicting Interests of Gatekeeper

Firms and Their Employees ................................815

2. The Low Probability of Detecting Fraud .............817

3. For Gatekeepers, Knowledge Is Liability ............820

C. Attempts at Reform: Sarbanes–Oxley and

Dodd–Frank ...............................................................824

COLLABORATIVE GATEKEEPERS 799

1. Sarbanes–Oxley: Intensifying Due

Diligence and Increasing Independence

from Management ................................................825

2. Dodd–Frank: Empowering Whistleblowers ........828

3. Academic Thinkers on Gatekeeper Reform .........831

4. Gatekeepers’ Potential Still Unexplored .............834

III. Collaborative Gatekeepers: Elements of a

New Paradigm .................................................................836

A. Background: Collecting Client Information ..............836

B. Key Obligation: Filing an Anonymous

Suspicious Activity Report ........................................838

C. Sanctions for Failing to Report .................................840

D. The Payoff: Immunity About Reported

Actions .......................................................................841

E. How Can Reporting Suspicions Help

Gatekeeper Professionals Overcome Conflicts

of Interest? .................................................................844

F. How Can Reporting Suspicions Help Address

Conflicts of Interest at the Corporate Level? ............845

IV. A Case Study of Collaborative Gatekeeping:

The Anti-Money-Laundering Regime ..............................846

A. The Birth of a New Model: Customer Due

Diligence ....................................................................850

B. Different Model: United States Enacts

Universal Reporting ..................................................858

C. The International Community Adopts and

Extends the Swiss Model: From Due

Diligence to Suspicious Activity Reporting ...............862

D. Customer Due Diligence and Suspicious

Activity Reporting in U.S. Law .................................867

V. The Anti-Money-Laundering Regime in Practice ...........874

A. Volume and Quality of SAR Filings Indicates

Industry Buy-In .........................................................876

B. SAR Filings Besides Money Laundering ..................879

C. Compliance Process and Technology

Behind Suspicious Activity Reporting ......................882

800 73 WASH. & LEE L. REV. 797 (2016)

D. Filing a SAR: Efforts for Investigation and

Drafting by Front-Line Employees,

Compliance Officers, and Management ....................886

E. Regulators in Collaboration ......................................888

VI. Applying the Collaborative Model in Other Areas ..........893

A. Collaborative Gatekeeping in Broker-Dealer

Regulation ..................................................................896

B. Collaborative Gatekeeping for Accountants on

Equity Issuance .........................................................905

VII. Conclusion ........................................................................910

I. Introduction

In his annual letter to shareholders for 2014, Jamie Dimon,

J.P. Morgan’s CEO, made an astonishing revelation.

That year

alone, his firm hired 8,000 new employees just to improve its

compliance with anti-money-laundering laws.

J.P. Morgan’s

recruitment zeal stemmed from a $2.6 billion penalty for

anti-money-laundering violations, due to its failure to spot

Madoff’s Ponzi scheme.

This was hardly an isolated case

Anti-money-laundering laws have played a central part in four

1. See generally Letter from Jamie Dimon, Chairman and CEO, JP

Morgan Chase & Co., to Shareholders (Apr. 8, 2015) [hereinafter J.P. Morgan

Letter], http://files.shareholder.com/downloads/ONE/3844439630x0x820077/8af

78e45-1d81-4363-931c-439d04312ebc/JPMC-AR2014-LetterToShareholders.pdf.

2. Id. at 21.

3. See generally Deferred Prosecution Agreement, Exhibit C, Statement of

Facts, United States v. J.P. Morgan Chase Bank, N.A. (S.D.N.Y. Jan. 6, 2014)

[hereinafter DPA Statement of Facts], http://www.justice.gov/sites/default/

files/usao-sdny/legacy/2015/03/25/JPMC%20DPA%20Packet%20(Fully%2Executed

%20w%20Exhibits).pdf. The U.S. Attorney’s office struck a deferred prosecution

agreement for an indictment of two years. Id. ¶ 12. Under the terms of this

agreement, J.P. Morgan is required to reform its anti-money laundering

compliance in accordance with a consent order issued by the Office of the

Comptroller of the Currency. See generally Consent Order, JPMorgan Chase

Bank, N.A., No. AA-EC-13-04 (OCC, 2013).

4. See Tom Braithwaite, Richard McGregor & Aaron Stanley, Banks Pay

Out $100bn in US Fines, F

IN. TIMES (Mar. 26, 2014), http://www.ft.com/

intl/cms/s/0/802ae15c-9b50-11e3-946b-00144feab7de.html (last visited Feb. 17,

2016) (discussing the key role anti-money laundering laws have played since the

financial crisis) (on file with the Washington and Lee Law Review).

COLLABORATIVE GATEKEEPERS 801

out of the eight biggest fines in the wake of the financial crisis,

becoming a key legal basis in the quest to hold banks

accountable. The newfound prominence of the

anti-money-laundering framework is striking. These laws target

drug cartels and terrorists, the criminal periphery of the financial

system rather than its core weaknesses. But, since 2007, the

anti-money-laundering framework has evolved into a critical

detection and enforcement mechanism for regulators, and a key

priority for private industry compliance.

So far, there is little in

the legal literature that could explain this puzzling shift towards

the anti-money-laundering toolkit.

This Article argues that regulators have turned to

anti-money laundering because of distinct features that set it

apart from other common bases of financial misconduct, such as

the anti-fraud provisions of the federal securities laws. Most

financial laws require financial institutions to identify

misbehaving clients, imposing heavy liability when they

knowingly or negligently fail to shut them out of the financial

5. See Stephen Grocer, A List of the Biggest Bank Settlements, WALL

STREET J. (June 23, 2014, 12:03 PM), http://blogs.wsj.com/moneybeat/2014/06/23/a-

list-of-the-biggest-bank-settlements/ (last visited Mar. 29, 2016) (noting that

cases against J.P. Morgan, BNP Paribas, HSBC, Credit Suisse, and UBS

involved anti-money laundering violations, among others) (on file with the

Washington and Lee Law Review).

6. See infra Part V (discussing the anti-laundering regime in practice).

7. Legal scholarship on anti-money laundering laws focuses mostly on its

criminal law dimensions. See, e.g., William J. Stuntz, Unequal Justice, 121

ARV. L. REV. 1969, 1979–80 (2008) (arguing that money laundering’s impact

has been felt mostly against underprivileged groups); D

OUGLAS HUSAK,

OVERCRIMINALIZATION: THE LIMITS OF CRIMINAL LAW 105 (2008) (arguing that

money laundering is a case of overcriminalization); Samuel W. Buell, The

Upside of Overbreadth, 83 N.Y.U.

L. REV. 1491, 1537 (2008) (arguing that

broader doctrine has assisted the federal government in gaining convictions);

Richard G. Strafer, Money Laundering: The Crime of the ‘90s, 27 A

M. CRIM. L.

REV. 149, 165–75 (1989) (discussing interpretative problems likely to arise after

the criminalization of money laundering). See generally Mariano-Florentino

Cuéllar, The Tenuous Relationship Between the Fight Against Money

Laundering and the Disruption of Criminal Finance,

93 J. CRIM. L. &

CRIMINOLOGY 311 (2003) (examining “the fight against money laundering as a

case study of the separation between an enforcement system’s objectives and

performance”). For a discussion of the impact of money laundering’s

criminalization on the banking industry, see Sarah Jane Hughes, Policing

Money Laundering Through Funds Transfers: A Critique of Regulation Under

the Bank Secrecy Act, 67 I

ND. L.J. 283, 287–97 (1992) (discussing the burden of

fund transfer policing for banks).

802 73 WASH. & LEE L. REV. 797 (2016)

system.

We argue that this liability threshold produces a

perverse effect: Market players are very careful to ensure that

they never reach such knowledge or negligence. Instead of

looking for signals of underlying fraud and investigating

indications, our laws incentivize market players to turn a blind

eye.

In contrast, anti-money-laundering laws require private

industry to share with authorities suspicions of misconduct, even

when these pieces of information fall far short of proving

illegality.

This reporting obligation, we argue, can help financial

intermediaries overcome conflicts of interest and motivates them

to organize more effective compliance operations. In this Article,

we explore this approach as a template for other areas of

financial regulation. We develop a theoretical framework that

explains the advantages of our proposed model, explore its

history and application in anti-money-laundering law, and

discuss how it could operate in other fields.

Enlisting private firms as the primary line of defense against

fraud and misconduct is the dominant strategy in financial

regulation and is known as the “gatekeeper model.”

Gatekeepers are intermediaries whose cooperation is essential for

many financial transactions: bankers, accountants, lawyers,

credit rating agencies, and other professionals.

Gatekeepers

help address the informational asymmetries between investors

and companies by verifying the credibility of contractual

representations, be it the accuracy of financial statements, the

risk profile of bonds, or the enforceability of legal claims.

theory, gatekeepers must maintain a reputation for integrity and

should not be persuaded to participate in fraud just to win one

8. See infra Part II.B.3 (explaining gatekeeper liability).

9. See infra Part IV.D (describing liability under the anti-money

laundering laws).

10. See Reiner H. Kraakman, Gatekeepers: The Anatomy of a Third-Party

Enforcement Strategy, 2

J. L. ECON. & ORG. 53, 54 (1986) (explaining the concept

of gatekeepers).

11. See id. at 53 (defining gatekeepers as “private parties who are able to

disrupt misconduct by withholding their cooperation from wrongdoers”); see also

infra Parts II.A–C (discussing the literature on gatekeepers).

12. See Kraakman, supra note 10, at 62 (noting the role of gatekeepers in

ensuring investors and companies are on the same page).

COLLABORATIVE GATEKEEPERS 803

client’s fees. In many instances, laws supplement gatekeepers’

reputational incentives with the threat of heavy sanctions.

But, as a spate of financial scandals has illustrated,

gatekeepers have often found themselves involved in client fraud

due to conflicts of interest between gatekeeper employees and

their firms.

Gatekeeper employees are likely to have invested

significant time and effort in building client relationships and

may not be willing to sacrifice a loss in compensation.

Acquiescing to dubious client demands can be particularly

tempting because the probability of detecting financial fraud is

notoriously low. Fraudulent schemes are often designed to look

like legitimate profit-making transactions. Victims often do not

realize that they have been misled until their investment has

evaporated and perpetrators have disappeared or are unable to

repay. Moreover, the sheer volume and sophistication of modern

transactions make the financial system very hard to police.

Without an inside tip about fraud, regulators seem reduced to

playing catch-up; they often fail.

Recognizing that information from the inside is essential in

combatting financial fraud, policymakers have tried various

strategies for strengthening gatekeepers’ incentives to come

forward.

Sarbanes–Oxley enacted measures to insulate

accountants from the pressures of corporate executives, but in

most cases, weaknesses in companies’ financial statements

remained unreported.

Dodd–Frank provides some

13. See infra Part III.C (highlighting how economic sanctions can

supplement the threat of reputational harm for gatekeepers).

14. See JOHN C. COFFEE, JR., GATEKEEPERS: THE ROLE OF THE PROFESSIONS

AND

CORPORATE GOVERNANCE 5 (2006) [hereinafter COFFEE, CORPORATE

GOVERNANCE] (noting how gatekeepers can be torn between their legal

responsibilities and their firms’ interests); John C. Coffee, Jr., Gatekeeper

Failure and Reform: The Challenge of Fashioning Relevant Reforms, 84

B.U. L.

REV. 301, 308 (2004) [hereinafter Coffee, Gatekeeper Failure] (same).

15. See infra Part II.B (giving examples of various past failures of the

gatekeeper model, which highlight the importance of inside whistleblowers).

16. See Assaf Hamdani, Gatekeeper Liability, 77 S.

CAL L. REV. 53, 107

(2003) (discussing possibilities policymakers considered to promote gatekeeper

reporting).

17. See John C. Coates IV, The Goals and Promise of the Sarbanes–Oxley

Act, 21

J. ECON. PERSP. 91, 96 (2007) (discussing the motivations behind

Sarbanes–Oxley and reviewing empirical evidence regarding its performance in

practice); infra Part II.C.1 (using Arthur Andersen as an example of an

804 73 WASH. & LEE L. REV. 797 (2016)

whistleblowers a share of the bounty but does not extend this

offer to professionals required to report misconduct to regulators

under other rules.

Academics have also debated how to beef up

the gatekeeper regime.

Prominent proposals involve making

gatekeepers strictly liable for client misconduct up to a capped

amount.

But there are fears that such an expansion of liability

would place too high a burden on the industry and push out

legitimate clients.

This Article explores an alternative design for structuring

gatekeepers’ obligations, which can help overcome many of the

above problems and provide regulators with more information on

potential fraud. Imagine requiring gatekeepers to report not

positive knowledge, but suspicions of illegality. In their dealings

with clients, gatekeepers may come to realize that there are gaps

in a client’s rationale for pursuing a transaction or that the

information the client provides does not add up. Alternatively, a

client’s conduct might be very unusual or have much in common

with past instances of client fraud. Such indications of potential

illegality are information that gatekeepers can collect and pass on

to authorities. In return, regulators could guarantee the

anonymity of the report, so as not to disrupt the gatekeeper–

client relationship should the transaction turn out to be

legitimate.

Setting the reporting threshold at the level of suspicions—

rather than knowledge—can radically change gatekeepers’

incentives and improve the effectiveness of regulatory

interventions. Suspicions are bound to arise at a much earlier

stage in the gatekeeper–client relationship, when the resources

invested in building this relationship are lower and the bond of

accounting firm pressured by its corporate clients).

18. Securities Exchange Act of 1934, § 21F(2)(C), 15 U.S.C. § 78u-6(c)(2)(C)

(2012); see infra Part II.C.2 (discussing Dodd–Frank).

19. See infra Part II.C.3 (providing examples of academics discussing the

role of gatekeepers).

20. See Coffee, Gatekeeper Failure, supra note 14, at 349 (“In this article’s

view, the most direct and practical means to this end would be to convert the

gatekeeper into the functional equivalent of an insurer, who would back its

auditor’s certification with an insurance policy that was capped at a realistic

level.”).

21. See infra Part II.C.3 (discussing the advantages and disadvantages of

strict liability for gatekeepers).

COLLABORATIVE GATEKEEPERS 805

loyalty between the two parties is not as strong. Some client

proposals may trigger suspicions even before gatekeepers actually

start offering any services. Consequently, the conflicts of interest

that gatekeepers are likely to face will be weaker. Moreover,

regulators may be able to combine multiple reports on a single

client or transaction and take preventive actions even if each

gatekeeper has come to know only part of the client’s activities or

dealings.

Because of the close interactions between gatekeepers

and regulators, we term this approach “collaborative

gatekeeping.”

To change industry attitudes towards client cover-ups, the

law should provide for sanctions against gatekeepers who fail to

submit a report. The threat of sanctions can motivate gatekeepers

to establish internal systems for identifying suspicious cases,

train front-line employees to be alert about potential fraud, and

draft reports.

Importantly, sanctions underline that suspicious

activity reporting is a regulatory obligation enforceable against

everyone in the financial industry. As all gatekeepers flock to

submit reports, the stigma associated with providing client

information to regulators is likely to fade away.

Gatekeepers will be more likely to submit reports if they gain

immunity against enforcement actions arising out of the

information they provide.

Immunity will allow gatekeepers to

continue working with the client even after the report without

immediately foregoing all future revenue when a transaction is

legitimate, tempering further the impact of conflicts of interest.

Immunity also shields gatekeepers from the risk of

self-incrimination if an ensuing investigation reveals that their

firm’s involvement in client fraud was greater than initially

suspected. Thus, immunity removes important inhibitions

gatekeepers may have and offers greater benefits upon providing

information. That said, immunity should extend to gatekeepers

only as long as they continue to act in good faith.

22. See infra Parts III.A–B (discussing proposal to require gatekeepers to

report suspicious client activity to regulators based on client information they

collect).

23. Infra Part III.C.

24. Infra Parts III.E–F.

25. Infra Part III.D.

26. Infra Part III.D.

806 73 WASH. & LEE L. REV. 797 (2016)

While these arguments underscore the theoretical appeal of

the model, there might be doubts as to its workability. One set of

worries might focus on the political economy of financial

regulation. Private industry might choose to oppose the

expansion of its regulatory obligations and the additional effort

and resources it entails. Another set of questions might concern

the promised informational upside. The financial industry might

choose to stick with client loyalty and refuse to embrace

suspicious-activity reporting. Or, in the exact opposite scenario,

finance professionals might flood regulators with reports about

clients’ activities, providing incomplete information about

countless cases that authorities could not possibly analyze or

pursue any further.

We can shed some light on these concerns by looking at

anti-money-laundering law, which, due to a historical

happenstance, follows closely the theoretical model outlined

above. While the conventional gatekeeper model is deeply

embedded in U.S. law, the anti-money-laundering regime has its

origins in 1970s Switzerland, from where it spread around the

world, including to the United States.

Taking advantage of

archival materials released publicly from the Swiss central bank

for the first time after forty years, the Article brings to light the

motivations of the regime’s inspirers and the contributions of key

players, such as financial institutions and industry associations.

Our exploration of the historical record reveals collaborative

gatekeeping as a rare case of a regulatory scheme that the

industry chose to embrace, rather than bitterly oppose. Of course,

the industry’s consent did not come without concessions from

regulators, such as standardized, objective criteria for suspicions

and universal application to all market participants. Moreover,

the spread of collaborative gatekeeping as a template for

anti-money-laundering laws around the world shows that stricter

laws do not necessarily place a national market in a competitive

disadvantage, but can trigger parallel developments elsewhere.

A unique constellation of enforcement bodies and market

players have seen the value of intelligence gathered through the

anti-money-laundering regime and launched large-scale

27. See infra Part IV.A (discussing Switzerland’s response to money

laundering scandal in 1977).

COLLABORATIVE GATEKEEPERS 807

collaborations aided by cutting-edge technology.

In the last few

years, over 1.5 million suspicious-activity reports have reached

the Treasury Department annually.

Institutions across the

financial industry have espoused suspicious activity reporting,

from banks to brokerages to wire services, from big financial

powerhouses to small community ventures, from urban centers to

rural areas. To fulfill their obligations, private firms have

revolutionized their compliance operations and introduced

digitalized systems using “big data” approaches. All this private

industry activity shows how the collaborative model can

effectively change market-wide attitudes. In turn, these reports

have helped regulators pursue a wide range of financial and

criminal misconduct beyond money laundering, including tax

evasion, mortgage fraud, and insider trading. Financial

regulators such as the Federal Reserve and the SEC, as well as

government agencies such as the IRS and the Department of

Justice, have expanded their oversight operations to better take

advantage of gathered intelligence.

Over 100 review teams from

these agencies pore over the reports in weekly or monthly

meetings, while the Treasury also operates a central database

that is open to hundreds of local and state authorities.

The first two Parts of the Article outline its theoretical

contributions, presenting collaborative gatekeeping as a coherent

model for enlisting the help of financial intermediaries in

enforcement. Part II below discusses the analytical foundations of

the current gatekeeping regime and the problems associated with

conflicts of interest. Part III delineates collaborative gatekeeping

as a theoretical model, explores its key elements, and discusses

how it could help blunt the conflicts of interest that plague

gatekeepers. The remaining Parts of the Article discuss the

application of collaborative gatekeeping in the case of

28. See generally U.S. DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK,

SAR STATS TECHNICAL BULL. 1 (2014) [hereinafter SAR STATS TECHNICAL BULL.],

http://www.fincen.gov/news_room/rp/files/SAR01/SAR_Stats_proof_2.pdf.

29. Id. at 1.

30. See U.S.

GOV’T ACCOUNTABILITY OFFICE, MONEY LAUNDERING:

TREASURY’S FINANCIAL CRIMES ENFORCEMENT NETWORK 7–9 (1991) [hereinafter

FINANCIAL CRIMES ENFORCEMENT NETWORK], www.gao.gov/assets/220/213999.pdf.

31. See 31 U.S.C. § 310(b)(2)(B) (2012) (explaining what the

government-wide data access service is supposed to contain).

808 73 WASH. & LEE L. REV. 797 (2016)

anti-money-laundering law and explore the potential extension of

the model to other areas of financial regulation. Part IV shows

how regulators and private industry worked together to complete

the anti-money-laundering regime over time, first in Switzerland

and then at the international level under U.S. leadership. Part V

presents some tentative evidence about the operation of the

anti-money-laundering regime in the United States. Part VI

discusses how the principles of collaborative gatekeeping could

work in the context of broker-dealer regulation and securities

issuance. Part VII concludes.

II. Gatekeepers in Current Theory and Practice

A. The Theory of Gatekeeping—Market-Based Regulation

Through Reputation

Financial transactions typically involve a professional

intermediary: Investment bankers, accountants, and lawyers help

companies issue securities to the public; retail banks help

long-term borrowers access funding from short-term depositors;

rating agencies assess bonds’ risk of default on behalf of buyers;

and so on.

For investors wary about their money, these

professionals’ stamp of approval can turn a risky investment into

a legitimate business proposition, rather than a venture into the

unknown.

In turn, businesses seek these professionals’ help to

32. See Sung Hui Kim, Gatekeepers Inside Out, 21 GEO. J. LEGAL ETHICS

411, 415–16 (2008) (“[T]he ideal gatekeeper would be an ‘outsider with a career

and assets beyond the firm,’ such as auditors, investment bankers, securities

analysts, securities attorneys, and, as more recently posited, credit rating

agencies.”); Coffee, Gatekeeper Failure, supra note 14, at 309

Obvious examples of gatekeepers who provide such verification or

certification services would include: (1) the auditor providing its

certification of the issuer’s financial statements; (2) the debt rating

agency certifying the issuer’s creditworthiness (or relative

creditworthiness); (3) the security analyst providing its objective

assessment of the corporation’s technology, competitiveness, or

earnings prospects; (4) the investment banker providing its “fairness

opinion” as to the pricing of a merger; and (5) the securities attorney

for the issuer providing its opinion to the underwriters that all

material information of which it is aware concerning the issuer has

been properly disclosed.

33. See Andrew F. Tuch, Multiple Gatekeepers, 96 V

A. L. REV. 1583, 1595

COLLABORATIVE GATEKEEPERS 809

boost their appeal to investors and gain access to a wider pool of

funds.

By facilitating a transaction, these professionals open the

gates of the financial system to new entrants; thus, they are

termed “gatekeepers.”

Enlisting gatekeepers in the fight against

fraud and misconduct is the most prominent strategy across

many different fields in modern financial regulation.

This

strategy, premised on gatekeepers’ desire to keep their

reputations pristine, has received much attention by scholars and

policymakers alike. The following paragraphs outline the

reputation-based theory of gatekeeping, while subsequent

sections explore the limitations of this reputation-based

approach, and regulatory reforms and academic proposals

intended to place greater liability on gatekeepers.

As intermediaries in many transactions, gatekeepers occupy

a unique place in the market. Because of their close connections

with clients, gatekeepers are better informed than regulators

regarding the goals, intentions, and underlying financial realities

of client firms.

Due to this informational advantage, the market

looks at gatekeepers to verify the disclosures, risk profile, or

general quality of financial instruments offered in a

transaction.

Gatekeepers command markets’ trust because, over

(2010)

Gatekeeper certifications provide a measure of assurance to investors

as to the accuracy of corporate disclosures, reducing the extent to

which investors, fearing they will be sold “lemons,” discount the value

of the asset being sold. In metaphorical terms, gatekeepers are

regarded as renting their reputations to corporations, a function that

economizes on information costs and creates value for the relevant

corporations. Gatekeepers thus function as so called reputational

intermediaries.

34. See id. at 1596 (“[T]he certification function of gatekeepers also favors

corporations relying on the market for gatekeeping services.”).

35. See Kraakman, supra note 10, at 54 (“Accountants and lawyers are

natural gatekeepers for fraudulent securities transactions that require audits or

legal opinions in order to close . . . .”).

36. Supra note 10.

37. See Coates, supra note 17, at 95–96 (stating that, in response to

“information constraints” faced by government enforcers, securities legislation

incorporates “a ‘gatekeeper’ strategy designed to prevent fraud before it

happens”).

38. See Kim, supra note 32, at 423 (“By certifying the issuer’s public

statements to the market, these intermediaries attest that they have evaluated

the issuer and are prepared to stake their reputation on the accuracy of the

810 73 WASH. & LEE L. REV. 797 (2016)

time, they have built significant reputational capital by verifying

transactions and statements that have repeatedly proven

accurate.

Their business model relies on maintaining and

augmenting this reputational capital; without it, they can no

longer perform their verification role.

Certainly, gatekeepers

receive hefty fees for their verifications, which they stand to lose

if they do not cooperate with a client’s fraudulent scheme. But, if

gatekeepers’ role in a fraudulent transaction is found out, they

stand to lose all credibility with the market and, thus, all future

business. This risk is so great, it is thought, that no single client’s

fee payment can possibly compensate for it.

For these reasons,

conventional wisdom holds that gatekeepers will not sacrifice

their reputations and future profits to satisfy any individual

client’s requests, however well the client pays.

Gatekeepers’ verification function is also the basis for their

role in enforcement.

If gatekeepers realize that their client is

violating the law, they can withhold their approval and prevent

this wrongdoer from entering the financial system. Taking

advantage of this power, the law has often required mandatory

gatekeeper participation in transactions plagued by significant

information asymmetries, such as securities offerings.

In those

cases, the law has imposed on gatekeepers a duty to examine

each client’s conduct and credentials.

To back this duty, the law

issuer’s statements.”).

39. Coffee, Gatekeeper Failure, supra note 14, at 308.

40. See Kim, supra note 32, at 423–24 (“They effectively pledge their

reputation to the issuer, which in turn increases investor confidence.”).

41. See Coffee, Gatekeeper Failure, supra note 14, at 308

These two elements—that the gatekeeper is a reputational

intermediary and that it receives only a limited payoff from any

involvement in misconduct—suggest a strategy for law compliance:

the more the law makes the involvement of gatekeepers in sensitive

transactions mandatory, the more it acquires a lever by which it can

effectively discourage law violations.

42. See id. at 302 (“The clearest examples of such reputational

intermediaries are auditors and securities analysts, who verify or assess

corporate disclosures in order to advise investors in different ways.”).

43. See id. at 324 (“[A]uditors continue to be used more because SEC rules

mandate their use . . . .”).

44. See Kraakman, supra note 10, at 53 (“[C]ollateral liability supplements

efforts to deter primary wrongdoers directly by enlisting their associates and

market contacts as de facto ‘cops on the beat.’”).

COLLABORATIVE GATEKEEPERS 811

establishes harsh penalties against gatekeepers that fail to catch

misconduct on their watch.

Gatekeepers must compensate

victims of their clients’ fraud unless they satisfy demanding

defenses.

In addition to investor-driven enforcement,

gatekeepers are also subject to sanctions by regulators, such as

fines or bans from the industry.

The main function of this regime is to deter gatekeepers from

acquiescing to fraudulent activities.

In order to avoid these

penalties, it is hoped, gatekeepers will monitor their clients

closely and steer them away from misconduct through their

advice.

Kraakman aptly called this type of gatekeeper a

“chaperone”: someone who, as the relationship with a client

unfolds, intervenes to prevent a violation from happening,

realigning the client’s actions so that they fall within the

requirements of the law.

In this conception of gatekeeping, regulation plays a limited

role, besides brandishing the threat of sanctions. If everything

works as intended, gatekeepers will either discipline misbehaving

clients or cut them off from the financial system before fraud is

committed. In both cases, harm to third parties is averted, so

there is little reason to involve enforcement authorities. Although

the law expects gatekeepers to monitor their clients and direct

45. See id. at 60 n.18 (“Legal penalties against wayward gatekeepers

generally function as taxes on gatekeeper transactions by imposing costs and

legal risks without compensation. However, penalties can also assume the form

of withdrawing privileges or subsidies, in which case gatekeeping duties may be

compensated ex ante (consider, for example, the withdrawal of professional

licenses).”); see also id. at 70 (penalties include “civil damages, fines, or license

revocations”).

46. See id. at 84 (“The securities acts extend statutory liability to all

‘persons’ who control primary violators, subject to a good faith defense . . . .”).

47. See id. at 70 (“Whenever entry into a gatekeeping market requires

significant capital, including investment in specific human capital or reputation,

simple legal penalties such as civil damages, fines, or license revocations can be

powerful deterrents.”).

48. See Kim, supra note 32, at 423 (“Gatekeeping theory operates on the

assumption that for the gatekeeper to act, the expected benefits should

outweigh the expected costs. The most salient benefit of closing the gate is

avoiding the costs associated with acquiescing in misconduct.”).

49. See Kraakman, supra note 10, at 62–66 (discussing “the likelihood that

gatekeepers will prevent misconduct and the cost of inducing them to detect it”).

50. Id. at 62.

812 73 WASH. & LEE L. REV. 797 (2016)

their activities, it provides neither any guidance nor any tools to

help them achieve these goals. Gatekeepers’ reputational

incentives and fear of sanctions are thought to provide enough

motivation; gatekeepers’ close knowledge of client workings

ensures they are best positioned to deal with attempted

misconduct; finally, gatekeepers’ market power leaves clients

with no other option but to follow gatekeepers’ recommendations.

Essentially, regulators are expected to intervene only ex post, in

the rare event that something in the gatekeeper–client

relationship goes amiss.

B. The Harsh Realities of Gatekeeping: Repeated Collapses Show

Weaknesses in Reputational Model

In practice, neither reputational incentives nor the deterrent

effect of harsh sanctions have managed to discipline gatekeepers,

as the last fifteen years of financial turmoil have shown. In 2001,

Enron collapsed when it was revealed that its management had

successfully pressured its outside accountant, Arthur Andersen,

to overlook liabilities hidden through off-balance-sheet

transactions.

But, if Enron’s deceptions were masterfully

designed, Worldcom’s fraud was much more mundane: Its CFO,

in cooperation with its outside accountant, Arthur Andersen

(again), simply misclassified expenses and inflated revenues.

While Enron and Worldcom are the most famous examples of

large-scale accounting irregularities, they were not the only

51. See Jonathan Weil, How Leases Play a Shadowy Role in Accounting,

ALL STREET J. (Sept. 22, 2014, 12:01 PM), http://www.wsj.com/

articles/SB109580870299124246 (last visited Mar. 29, 2016) (“A big part of the

[Enron] scandal centered on off-balance-sheet ‘special purpose entities.’ These

obscure partnerships could be kept off the books—with no footnote disclosures—

if an independent investor owned 3% of an entity’s equity.”) (on file with the

Washington and Lee Law Review); see, e.g., David Henry et al., Who Else Is

Hiding Debt, B

LOOMBERG (Jan. 28, 2002), www.bloomberg.com/bw/stories/2002-

01-27/who-else-is-hiding-debt (last visited Mar. 29, 2016) (“A suit filed earlier

this month shows that many U.S. finance companies are among 52 partners in

LJM2, an Enron off-balance-sheet entity with over $300 million in assets.”) (on

file with the Washington and Lee Law Review).

52. Simon Romero & Alex Berenson, WorldCom Says It Hid Expenses,

Inflating Cash Flow $3.8 Billion, N.Y.

TIMES (June 26, 2002),

http://www.nytimes.com/2002/06/26/technology/26TELE.html (last visited Feb.

17, 2016) (on file with the Washington and Lee Law Review).

COLLABORATIVE GATEKEEPERS 813

ones.

Nor were accountants the only gatekeepers found to

succumb to such conflicts of interest. In 2003, securities analysts,

who write independent reviews of securities offerings and issue

buy-sell-hold recommendations, were found to overwhelmingly

favor companies that had hired the analyst’s employer to run the

offering

after being promised under-the-table rewards.

The financial crisis of 2007–2008 brought to light a new set of

gatekeeper missteps. To successfully repackage and sell

mortgage-backed securities, investment banks represented to

buyers that their products had a much lower risk profile than the

one eventually revealed once the subprime market collapsed.

Credit rating agencies, hired by investment bankers to conduct

an independent assessment of the securities’ risk profile,

contributed to the collective euphoria by downplaying the

possibility of default.

In the midst of this upheaval, Bernie

53. See, e.g., Daniel J.H. Greenwood, Enronitis: Why Good Corporations Go

Bad, 2004 C

OLUM. BUS. L. REV. 773, 786 (2004) (discussing the Tyco

International Ltd. racketeering scheme); Andrew Ross Sorkin, 2 Top Tyco

Executives Charged With $600 Million Fraud Scheme, N.Y.

TIMES (Sept. 13,

2002), www.nytimes.com/2002/09/13/business/2-top-tyco-executives-charged-with-

600-million-fraud-scheme.html (last visited Feb. 17, 2016) (same) (on file with

the Washington and Lee Law Review); Europe’s Enron, E

CONOMIST (Feb. 27,

2003), http://www.economist.com/node/1610552 (last visited Feb. 17, 2016)

(discussing the Royal Ahold scandal) (on file with the Washington and Lee Law

Review). See generally Claudio Storelli, Note, Corporate Governance Failures—

Is Parmalat Europe’s Enron?, 2005 C

OLUM. BUS. L. REV. 765 (2005) (discussing

the Parmalat scandal).

54. See Jill E. Fisch & Hillary A. Sale, The Securities Analyst as Agent:

Rethinking the Regulation of Analysts, 88

IOWA L. REV. 1035, 1045 (2003)

(“[A]nalysts face firm pressure to issue positive reports because those reports

have greater potential to generate commission revenue for their companies.”).

55. See Ten of Nation’s Top Investment Firms Settle Enforcement Actions

Involving Conflicts of Interest Between Research and Investment Banking, U.S.

SEC. & EXCHANGE COMMISSION (Apr. 28, 2003), http://www.sec.gov/news/press/

2003-54.htm (last visited Feb. 17, 2016) (“UBS Warburg and Piper Jaffray

received payments for research without disclosing such payments . . . . Those

two firms, as well as Bear Stearns, J.P. Morgan and Morgan Stanley, made

undisclosed payments for research . . . .”) (on file with the Washington and Lee

Law Review).

56. See SEC v. Goldman Sachs & Co., 790 F. Supp. 2d 147, 154 (S.D.N.Y.

2011) (presenting SEC allegations that Goldman Sachs misrepresented to

investors in ABACUS vital information about the risk profile and portfolio

selection process).

57. See Charles W. Murdock, The Dodd–Frank Wall Street Reform and

Consumer Protection Act: What Caused the Financial Crisis and Will Dodd–

814 73 WASH. & LEE L. REV. 797 (2016)

Madoff, a well-respected financier and former head of NASD,

shockingly confessed to running an estimated $50 billion Ponzi

scheme.

For decades, Madoff’s fraud continued under the nose

of the nation’s biggest bank, JP Morgan.

In 2014, JP Morgan

agreed to a hefty $2.6 billion in monetary sanctions for failing to

alert regulators about the Madoff case.

These events dominated national headlines, attracted

enormous scholarly attention, and triggered important legislative

and regulatory reforms.

While each of these cases involved

distinct failures, they also underscore fundamental weaknesses

in the conventional gatekeeper model. Gatekeepers, it turned out,

sometimes went out of their way to keep their clients satisfied,

even when this meant disregarding indications of potential

fraud.

After all, the probability that clients’ fraud will be

detected is relatively low. And even when fraud is revealed, our

laws often allow gatekeepers to avoid liability if they were not

aware of it, or at least not negligent in ignoring it. Below we

explain why our current laws have allowed the conflicts of

interest in gatekeeping to burgeon in light of the low probability

of detecting fraud.

Frank Prevent Future Crises?, 64 SMU L. REV. 1243, 1301 (2011) (“Probably the

most reprehensible players in the subprime crisis have been the credit rating

agencies.”); id. at 1302 (“In 2004, Moody’s and Standard & Poor’s ‘eased their

standards’ under pressure from Wall Street to enable more securities to be rated

AAA.”).

58. See Aaron Smith, Five Things You Didn’t Know About Bernie Madoff’s

Epic Scam, CNN (Dec. 11, 2013), http://money.cnn.com/2013/12/10/news/

companies/bernard-madoff-ponzi/ (last visited Feb. 24, 2016) (on file with the

Washington and Lee Law Review) (“He [Madoff] pled guilty to charges of

fraud . . . .”).

59. See Hamilton & Pfeifer, supra note 3 (discussing JP Morgan’s

complacency with Madoff’s actions).

60. Id.

61. For examples of legislators calling for reform following the Madoff

scandal, see generally Liz Moyer, Can New Regulation Stop Another Madoff?,

ORBES (Jan. 5, 2009), www.forbes.com/2009/01/05/madoff-regulation-banking-

biz-wall-cx_lm_0105madoff.html (last visited Feb. 17, 2016) (on file with the

Washington and Lee Law Review); The Securities and Exchange Commission

Post-Madoff Reforms,

U.S. SEC. & EXCHANGE COMMISSION, http://www.sec.gov/

spotlight/secpostmadoffreforms.htm (last visited Feb. 17, 2016) (on file with the

Washington and Lee Law Review).

62. See supra notes 51–53 and accompanying texting (discussing Enron

and Worldcom).

COLLABORATIVE GATEKEEPERS 815

1. The Conflicting Interests of Gatekeeper Firms and Their

Employees

In the eyes of the law, ideal gatekeepers are sizeable

corporations that boast a strong reputation won after decades of

experience, command large market shares in their industry, and

employ thousands of professionals.

With such leverage,

gatekeepers have little reason to acquiesce to devious client

demands. It is individual gatekeeper professionals that strike

deals with client executives. Some gatekeepers, such as certain

investment banks or law firms, operate on the basis of “eat what

you kill” models, where each professional’s compensation or

bonus depends on the billings she brings to the firm.

But even

in firms that do not keep strict tally of the loot, gatekeeper

professionals are in charge of specific client accounts, which they

are expected to cultivate and grow.

Thus, a gatekeeper

professional’s annual pay and future in the firm may depend

entirely on a few clients, or even on a single one. These clients

may only represent a sliver of the firm’s aggregate billings but

mean the world to the individual professional who caters to their

needs. Forging these client relationships requires significant time

and effort from gatekeeper professionals, and building trust with

clients may take years. When such a profitable client relationship

is lost, it could take considerable time for the professional to

63. See DELOITTE, 2015 GLOBAL REPORT 3 (2015) (employing more than

225,000 employees), http://www2.deloitte.com/content/dam/Deloitte/global/

Documents/About-Deloitte/gx-gr15-main-final.pdf; E

RNST & YOUNG, 2015

GLOBAL REVIEW 2 (2016) (employing 212,000 employees), http://www.

ey.com/Publication/vwLUAssets/EY_Global_review_2015/$FILE/EY_Global_revi

ew_2015.pdf.

64. For “eat what you kill” models in the banking industry, see generally

Suzanne McGee, ‘You Eat What You Kill’: Wall Street Bonuses Keep Soaring as

Profits Decline, T

HE GUARDIAN (Mar. 15, 2015), www.theguardian.com/

money/us-money-blog/2015/mar/15/wall-street-bonuses-rise-profits-decline (last

visited Feb. 17, 2016) (on file with the Washington and Lee Law Review). For

“eat what you kill” models in law firm settings, see Milton C. Regan, Jr. & Lisa

H. Rohrer, Money and Meaning: The Moral Economy of Law Firm

Compensation, 10 U.

ST. THOMAS L.J. 74, 115 (2012) (providing a description of

eat what you kill from a Partner’s perspective).

65. See Kim, supra note 32, at 432–33 (“Since a partner’s welfare is now

based almost entirely on her own individual efforts to generate revenue from

her own client base, the potential threat of client defection incentivizes her to

accede to client demands out of a desire to keep the client.”).

816 73 WASH. & LEE L. REV. 797 (2016)

identify and develop a suitable alternative.

Moreover, long-term

relationships between gatekeeper professionals and client

executives often start as staid business transactions but develop

into personal friendships. Sentiments such as trust, loyalty, and

affinity become the basis for lasting and rewarding interactions.

So, if these valued clients and trusted friends ask the

individual professional to cooperate with them and violate the

law, will she be willing to go along, or will she resist? In the cases

examined above, the short-term incentives of personal

enrichment won over the long-term goals of maintaining the

firm’s reputational capital, or even its continued existence.

Apart from specific considerations generated by gatekeepers’

investment in client relationships, market-wide norms paint

cooperating with authorities as betrayal of one’s own clients. In

tightly regulated industries, such as finance, regulators are often

seen, if not exactly as the “enemy,” but as someone whose gaze is

best avoided.

Moreover, informing enforcement authorities

about somebody else’s misconduct carries negative social

connotations: informants are labeled snitches.

In service

industries that prize loyalty above all else, working with

regulators against one’s own clients is often seen as the ultimate

betrayal. Stigmatized by their actions, these gatekeepers are seen

as unfaithful agents who put their own interest ahead of their

66. See Lawrence A. Cunningham, Beyond Liability: Rewarding Effective

Gatekeepers, 92 M

INN. L. REV. 323, 346 (2007) (“A final—and pervasive—

limitation on gatekeeping efficacy is how the enterprise pays the gatekeeper.

That creates an inherent inclination for solicitude simply to retain business.”).

67. See Sung Hui Kim, The Banality of Fraud: Re-Situating the Inside

Counsel as Gatekeeper, 74 F

ORDHAM L. REV. 983, 1065 (2005) (noting the

importance of trust and affinity in the client relationship).

68. See Merritt B. Fox, Gatekeepers Failures: Why Important, What to Do,

106 M

ICH. L. REV. 1089, 1103 (2008) (discussing why people sometimes prefer

short-term incentives over long-term ones).

69. Illustrative of this point is a speech delivered by Xerox Corporation’s

CEO at an SEC conference. The CEO joked that all CEOs feel strange when

entering the SEC building. Anne M. Mulcahy, CEO, Xerox Corporation,

Remarks at the SEC Interactive Data Roundtable Panel on Getting Analyst and

Investors Significantly Better Information 109 (June 12, 2006),

www.sec.gov/spotlight/xbrl/xbrlofficialtranscript0606.pdf.

70. See Naseem Faqihi, Choosing Which Rule to Break First: An In-House

Attorney Whistleblower’s Choices After Discovering a Possible Federal Securities

Law Violation, 82 F

ORDHAM L. REV. 3341, 3351 (2014) (noting that policymakers

historically referred to whistleblowers as “snitches” and “rats”).

COLLABORATIVE GATEKEEPERS 817

client’s, as overly sensitive to their regulatory obligations, as

unpredictable and untrustworthy collaborators who will not

protect their client at moments of crisis. Gatekeepers whose

reputation is thus tainted have trouble attracting business and

may have to abandon the industry.

2. The Low Probability of Detecting Fraud

Modern finance is an exceedingly technical environment,

with layers of intermediaries interacting through complicated

transactions using ever more refined instruments with diverse

motivations. Highly knowledgeable market participants take

advantage of this complexity, typically to make legitimate gains,

but sometimes to devise intricate fraud schemes. However deep

the machinations behind fraud run, the outcome is often similar:

the probability of detecting financial fraud is often very low.

But surely, one might think, at some point the fraudulent

edifice will collapse, the harm done to victims will be revealed,

and enforcement will restore order. While true for most crimes or

torts, this proposition does not necessarily stand for financial

fraud. Some financial crimes take place in order to address

presumably short-term problems, such as a liquidity crunch or a

temporary market downturn, with the intention of replenishing

misappropriated funds once things look up.

Other fraudulent

schemes are designed to continue in perpetuity and can even

withstand some disturbances.

Madoff successfully ran his Ponzi

scheme for decades, and he finally revealed it himself.

In still

other cases, misconduct has a clearly harmful impact for markets’

credibility, but pinpointing specific losses and connecting them to

specific victims might be impracticably difficult because of the

71. See, e.g., Matthew O'Brien, Meet the Most Indebted Man in the World,

THE ATLANTIC (Nov. 2, 2012), www.theatlantic.com/business/archive/2012/11/

meet-the-most-indebted-man-in-the-world/264413 (last visited Feb. 17, 2016)

(discussing Jerome Kerviel, Société Générale rogue trader) (on file with the

Washington and Lee Law Review).

72. See, e.g., supra notes 58–61 and accompanying text (providing Madoff

as an example of a long-term fraudulent scheme).

73. See Smith, supra note 58 (“Some reports say that Madoff’s epic crime

may have started as early as the 1960s . . . .”).

818 73 WASH. & LEE L. REV. 797 (2016)

sheer number of victims harmed just a little.

Finally, some

fraudulent schemes take advantage of a systemic weakness that

affects all market participants in one way or another, as was the

case with the security analysts’ conflicts of interest.

For all

these reasons, identifying or even conceptualizing victims might

be exceedingly hard. When the victims themselves do not have a

clear sense of harm, they are unlikely to point authorities to the

fraud and demand action.

The low probability of detecting financial fraud further

sharpens the conflict of interest between gatekeeper firms and

employees discussed above. Motivated by short-term gains, such

as increased compensation, better reputation, and professional

advancement, employees might be more tempted to disregard

warning signs if they believe no one is going to find out.

Succumbing to misguided client requests is easier if the

probability of being discovered is really low. Moreover, employees

might have pocketed their gains and left the firm by the time the

effects of the fraud fully unfold. Whether these individuals’ assets

will be available for compensating fraud victims depends on

doctrines of gatekeeper fault, which often raise hurdles.

practice, sanctions against individual employees are rare.

Economic theory suggests that, when the probability of

detection is low, policymakers can still deter harmful conduct

successfully with a substantial increase in sanctions.

However,

increasing sanctions might not be a viable option in the case of

gatekeeper misconduct. To start, market reaction to severe

gatekeeper fault can be immediate and overwhelming, as Arthur

74. In the market-timing scandals, the losers were the people who traded

right before the close of the market. See James Surowiecki, Right Trade, Wrong

Time, T

HE NEW YORKER (Oct. 20, 2003), www.newyorker.com/

magazine/2003/10/20/right-trade-wrong-time (last visited Feb. 17, 2016) (“A

Stanford Business School study found that this kind of market timing could be

robbing investors of more than four billion dollars a year.”) (on file with the

Washington and Lee Law Review).

75. Infra Parts III.E–F.

76. Infra Part II.B.3.

77. Stavros Gadinis, The SEC and the Financial Industry: Evidence from

Enforcement Against Broker-Dealers, 67 BUS. LAW. 679, 682 (2012).

78. A. Mitchell Polinsky & Steven Shavell, The Theory of Public

Enforcement of Law, in HANDBOOK OF LAW AND ECONOMICS 403, 427–29 (A.

Mitchell Polinsky & Steven Shavell eds., 2007).

COLLABORATIVE GATEKEEPERS 819

Andersen’s collapse after Enron illustrates.

When the

gatekeeper firm dissipates, there are really no additional

sanctions that policymakers can impose. Moreover, our

regulatory framework utilizes gatekeepers to such an extent that

eliminating them through harsh sanctioning might leave the

market even more exposed to misconduct than before. The global

financial system relies on just four big accounting firms and three

credit rating agencies, and many fear about the already

diminished levels of competition among them.

Yet, it is because of the low probability of detection that

gatekeeper collaboration is absolutely essential in any effort to

uproot fraud. Many of the recent headline financial scandals have

come to light or reached resolution because someone with direct

knowledge of the scheme worked with enforcement authorities. In

Worldcom, it was internal accountants that unearthed the

scheme; in Goldman’s Abacus deal, it was months of interviews

with industry insiders that allowed the Securities and Exchange

Commission to zero in on this particular deal;

Enron’s Andrew

Fastow,

and, of course, Bernie Madoff,

offered up themselves

79. See supra notes 51–52 and accompanying texting (discussing the

Arthur Andersen case).

80. See Adam Jones & Simon Rabinovitch, Accounting: Stalking the Big

Four, FIN. TIMES (Apr. 16, 2013), http://www.ft.com/intl/cms/s/0/cd74664e-9797-

11e2-97e0-00144feabdc0.html#axzz417TnJu9m (last visited Feb. 24, 2016)

(discussing the major accounting firms) (on file with the Washington and Lee

Law Review); Patrick Kingsley, How Credit Rating Agencies Rule the World,

HE GUARDIAN (Feb. 15, 2012), http://www.theguardian.com/business/2012/feb/

15/credit-ratings-agencies-moodys (last visited Feb. 24, 2016) (discussing the

major credit rating agencies and criticizing their “mysterious” power) (on file

with the Washington and Lee Law Review).

81. See Louise Story & Gretchen Morgenson, S.E.C. Accuses Goldman of

Fraud in Housing Deal, N.Y. TIMES (Apr. 16, 2010), http://www.nytimes.com/

2010/04/17/business/17goldman.html (last visited Feb. 17, 2016) (“For months,

S.E.C. officials have been examining mortgage bundles like Abacus that were

created across Wall Street. The commission has been interviewing people who

structured Goldman mortgage deals about Abacus and similar instruments.”)

(on file with the Washington and Lee Law Review).

82. See Jen Rogers, Fastow and His Wife Plead Guilty, CNN MONEY (Jan.

15, 2004), www.money.cnn.com/2004/01/14/news/companies/enron_fastows (last

visited Feb. 17, 2016) (providing details of Fastow’s guilty plea) (on file with the

Washington and Lee Law Review).

83. Scott Cohn, Madoff Says He Provided “Key Information” to Authorities,

CNBC (Dec. 13, 2013), www.cnbc.com/id/101272415 (last visited Feb. 17, 2016)

(“In a new email from federal prison, convicted con artist Bernard Madoff claims

820 73 WASH. & LEE L. REV. 797 (2016)

to authorities. Inside information can save a lot of effort and

resources for regulators by directing them right to the target.

Gatekeepers are uniquely placed to provide such direction,

because they are sophisticated professionals who understand the

intricacies of the financial system and can spot signs of

misconduct. But instead of enlisting their cooperation in finding

out more about fraud, our laws steer them towards knowing as

little as possible about it so as to avoid liability, as the next Part

argues.

3. For Gatekeepers, Knowledge Is Liability

Gatekeeper liability is not strict. Rather, to be liable toward

victims of their clients’ misconduct, or subject to regulatory

sanctioning, gatekeepers must have violated a duty specifically

prescribed by law.

In defining these duties, the law typically

requires that gatekeepers either intend to or at least know that

their actions violate the law.

Take for example Rule 10b–5, the

catch-all definition of securities fraud that has allowed investors

to launch thousands of private claims against investment banks,

accountants, and lawyers.

A key element of 10b–5 fraud is

scienter,

satisfied when the information that the defendant

possesses allows her to know or reasonably foresee the potential

result of her action.

In the same vein, the Securities Exchange

federal regulators ‘eagerly accepted’ information he offered about JPMorgan

Chase, which is in talks with U.S. authorities about a potential $2 billion

settlement over its alleged role in the Madoff fraud.”) (on file with the

Washington and Lee Law Review).

84. See Jennifer Arlen & Reinier Kraakman, Controlling Corporate

Misconduct: An Analysis of Corporate Liability Regimes, 72 N.Y.U.

L. REV. 687,

697 (1997) (defining duty-based liability).

85. See Kraakman, supra note 10, at 76 (explaining the intent requirement

in gatekeeper liability laws).

86. 17 C.F.R. § 240.10b–5 (1975).

87. See Ernst & Ernst v. Hochfelder, 425 U.S. 185, 201 (1976)

(“[Section] 10(b) was addressed to practices that involve some element of

scienter and cannot be read to impose liability for negligent conduct alone.”).

88. See AUSA Life Ins. Co. v. Ernst & Young, 206 F.3d 202, 221 (2d Cir.

2000) (noting that it “is sufficient for a plaintiff to allege and prove that a

defendant could have foreseen the consequences of his actions but forged ahead

nonetheless” to demonstrate scienter for a claim under § 10b).

COLLABORATIVE GATEKEEPERS 821

Act of 1934 authorizes the Securities and Exchange Commission

to suspend or revoke the operating license of broker-dealers when

they willfully violate, or aid and abet in violating, securities

laws.

On the banking side, liability arrangements are similar.

For example, the billion-dollar fines that banks have paid for

misrepresenting the value of mortgage portfolios in the run-up to

the 2008 crisis are based on FIRREA.

Because FIRREA civil

penalties arise in cases of mail and wire fraud that harms

federally insured financial institutions, they also require

scienter.

Tying gatekeeper liability to knowledge of illegality has an

important perverse effect that has been largely overlooked by the

literature.

If a gatekeeper is not held liable unless it knows

about client mischief, then the less it knows about the client, the

more likely it is that the gatekeeper will avoid liability

altogether. By maintaining a position of unawareness,

gatekeepers can continue offering services to and collecting fees

from clients that are engaging in illegalities. In this way,

gatekeepers caught between their client loyalties and their

regulatory obligations can satisfy both masters. Thus, rather

than pursuing their market-monitoring role to the fullest,

gatekeepers are actually better off by averting their gaze, so that

they limit the chances of coming across information that would

compromise their unawareness. Only when they cannot plausibly

89. Securities Exchange Act of 1934, § 15(b)(4)(D)–(E), 15 U.S.C.

§ 78o(b)(4)(D)–(E) (2012).

90. See Financial Institutions Reform, Recovery, and Enforcement Act of

1989, § 951, 12 U.S.C. § 1833a (2012) [hereinafter FIRREA] (listing the civil

penalties for violating FIRREA).

91. See id. § 1833(c)(2) (“This section applies to a violation of, or a

conspiracy to violate . . . section 287, 1001, 1032, 1341 or 1343 of title 18, United

States Code, affecting a federally insured financial institution.”). These sections

include false, fictitious, or fraudulent claims (§ 287); fraud or false claims

(§ 1001); the concealment of assets from a conservator, receiver, or liquidating

agent (§ 1032); mail frauds and swindles (§ 1341); and fraud by wire, radio, or

television (§ 1343). See 17 C.F.R. § 240.10b–5 (including mail and wire fraud);

see also Ernst & Ernst, 425 U.S. at 214 (establishing the scienter requirement

for Rule 10b–5).

92. Note that Professor Jennifer Arlen makes a similar point for corporate

liability regime for employees’ torts and crimes. See generally Jennifer Arlen,

The Potentially Perverse Effects of Corporate Criminal Liability, 23 J.

LEGAL

STUD. 833 (1994).

822 73 WASH. & LEE L. REV. 797 (2016)

deny knowing about clients’ scheming are gatekeepers forced to

abandon their neutrality. As a result, indications that are likely

to signal misconduct, but do not positively prove it, might often be

left unexplored.

In few notable cases, gatekeeper liability departs from

scienter and employs a negligence standard, as is the case for

inaccuracies in a registration statement for underwriters,

accountants, attorneys, and others under § 11 of the 1933 Act.

Courts are left with the challenging task of defining what

constitutes reasonable care in each case, a rather costly and

complicated exercise that typically entails significant uncertainty

for all parties involved.

To reach this determination, courts

typically refer to the standards of due diligence in the relevant

professional setting and invite expert testimony on whether the

defendant took all the appropriate steps to investigate and assess

the situation at hand.

Defendants are found liable if they

continue to work with clients that their peers would have cut off

from the financial system and referred to enforcement

authorities. Recent cases have added more bite to this due

diligence exercise, requiring that gatekeepers consider “red flags”

(i.e. facts that put the defendant on notice that the client is

engaged in wrongdoing, or that would suggest to the average

93. See Securities Act of 1933 § 11, 15 U.S.C. § 77k (2012) (creating a cause

of action for security holders against certain parties if a registration statement

“contain[s] an untrue statement of a material fact or omit[s] to state a material

fact required to be stated therein or necessary to make the statements therein

not misleading”).

94. See Hamdani, supra note 16, at 59 (stating that strict liability for

gatekeepers relieves “courts from entering the thicket of determining what

constitutes reasonable care in a given set of circumstances” (internal quotation

marks omitted)).

95. See, e.g., Wikoff v. Vanderveld, 897 F.2d 232, 235 (7th Cir. 1990)

(stating that expert testimony used at trial to help determine whether

defendant “had breached her warranties concerning MSCI’s financial

statements”); Danis v. USN Commc’n, Inc., 121 F. Supp. 2d 1183, 1191–92 (N.D.

Ill. 2000) (“Plaintiffs fail to present evidence to support their claims of material

misrepresentations in regard to Deloitte’s GAAP and GAAS certification.

Violations of GAAP and GAAS standards are established through expert

testimony.”); In re Discovery Zone Sec. Litig., 943 F. Supp. 924, 935 n.9 (N.D. Ill.

1996) (“This Court finds that whether FASB CON No. 6 constitutes a GAAP is

best resolved by expert testimony, and thus should not be addressed on a motion

to dismiss.”).

COLLABORATIVE GATEKEEPERS 823

investor that she has been defrauded).

Red flags must “strip a

defendant of his confidence” in the accuracy of his clients’

representations.

Although less demanding than scienter, this

negligence standard still requires gatekeepers to turn away their

clients only after evidence starts mounting against them.

Thus,

it still leaves gatekeepers significant space to accommodate client

demands before they run into trouble.

Before information about client misconduct reaches the level

of scienter or negligence, as the discussion above shows,

gatekeepers have no legal obligation to alert regulators, and very

little reason to do so voluntarily. If gatekeepers terminated their

engagement based on unverified suspicions, their relationship

with the clients in question would probably be irreparably

damaged, and their reputation in the market would suffer.

gatekeepers reported their suspicions to regulators, the

transaction would probably stall or be canceled and the client

may suffer as a result. Again, gatekeepers put the opportunity to

collect fees for their services at risk, and might even find

themselves targeted in an enforcement action. As a result,

gatekeepers often find themselves tiptoeing around the red line of

illegality, putting up a shield around their own liability, rather

than worrying about the impact of their clients’ actions for third

parties and the financial system as a whole.

Even when revelation of the fraud is imminent and the

delicate balancing act between satisfying client demands and

complying with regulatory obligations is about to crumble, we

have seen gatekeepers trying to protect their position by

concealing evidence. At least in past financial scandals,

gatekeepers had in fact gone to great lengths to keep authorities

96. See In re Worldcom, Inc. Sec. Litig., 346 F. Supp. 2d 628, 672–73

(S.D.N.Y. 2004) (providing an example of a “red flag”).

97. Id. at 674.

98. See Hamdani, supra note 16, at 83–85 (discussing the negligence

standard for gatekeepers).

99. See Kraakman, supra note 10, at 59–61 (“Whatever their actual

intentions, then, all regulatory targets have a powerful incentive to withhold

information from potential whistleblowers and to refrain from transacting with

anyone of suspect loyalties. Conversely, these responses create a matching

incentive for putative whistleblowers to avoid actual whistleblowing at any

cost.”).

824 73 WASH. & LEE L. REV. 797 (2016)

in the dark. For example, they obstructed evidence,

100

got the

cooperation of other gatekeepers in the transaction

101

and even

created secret code words and reference schemes.

102

misdirecting gatekeepers’ efforts in this way, the regulatory

framework loses the chance to fully capitalize on their role as

intermediaries and collect valuable intelligence that could help

with enforcement.

C. Attempts at Reform: Sarbanes–Oxley and Dodd–Frank

As successive waves of financial scandals drove the economy

into a tailspin, Congress made repeated efforts to reform the

gatekeeper model.

103

While these reforms targeted only certain

types of gatekeepers, each represents a different strategy that

could be applied to gatekeeping more generally. More specifically,

Congress tried two new approaches. First, it sought to instill

greater discipline from the inside; it created procedures that

streamline gatekeepers’ due diligence obligations and supervisory

100. See United States v. Arthur Andersen, LLP, 544 U.S. 696, 701 (2005)

(discussing the shredding of documents by auditor Arthur Andersen in the

midst of the SEC investigation of Enron); see also Samuel W. Buell, The

Blaming Function of Entity Criminal Liability, 81 I

ND. L.J. 473, 473–75 (2006)

(discussing various gatekeepers’ efforts to shield clients and proposing entity

criminal liability).

101. See Stoneridge Inv. Partners, LLC v. Scientific-Atlanta, Inc., 552 U.S.

148, 152 (2008) (discussing the liability of a gatekeeper who had not made a

public misstatement or violated a duty to disclose, but had participated in a

scheme to assist fraudsters); Anthony Sallah, Scheme Liability: Conduct Beyond

the Misrepresentations, Deceptive Acts, and Possible Janus Intervention, 45 U.

TOL. L. REV. 181, 187 (2013) (describing a scenario where multiple directors

allegedly collaborated to present fraudulent financial statements to the SEC).

102. See Collusion in the Stockmarket, E

CONOMIST (Jan. 15, 1998),

www.economist.com/node/111273 (last visited Jan. 20, 2016)

[I]n 70 of the 100 most heavily traded stocks, Nasdaq dealers avoided

quoting prices in odd eighths of a dollar. Buyers were far more likely

to quote shares at 28 1/2 or 28 3/4 than at 28 5/8. This raised the

possibility that the dealers . . . were tacitly colluding to keep the gap

between the price they paid for a share and the price at which they

sold it wider than it would have been in a truly competitive market.

(on file with the Washington and Lee Law Review).

103. See Arthur B. Laby, Differentiating Gatekeepers, 1 B

ROOK. J. CORP. FIN.

& COM. L. 119, 152–55 (2006) (explaining the focus of recent gatekeeper reform

initiatives in Congress).

COLLABORATIVE GATEKEEPERS 825

mechanisms.

104

This set of measures is mostly associated with

the 2002 Sarbanes–Oxley Act,

105

which used this approach in

relation to financial statement audits. Second, Congress offered

increased rewards to gatekeepers who decide to “blow the

whistle” on their employers.

106

Although Sarbanes–Oxley also

included whistleblower provisions, it was the 2010 Dodd–Frank

Act that revolutionized and greatly expanded this approach.

107

The paragraphs below look briefly at these reforms as models for

regulating gatekeepers, arguing that there is still more to be

done.

1. Sarbanes–Oxley: Intensifying Due Diligence and Increasing

Independence from Management

Troubled by accountants who either actively collaborated

with misbehaving managers, or turned a blind eye toward them,

the Sarbanes–Oxley Act established a new regime to govern the

interactions between public companies and their external

auditors.

108

The Sarbanes–Oxley regime orchestrates

accountants’ due diligence obligations extensively. The Public

Company Accounting Oversight Board (PCAOB), a new

quasi-public regulator, issues Audit Standards that specify steps

and criteria external accountants must follow when auditing

public companies.

109

By directing accountants’ attention to many

104. Infra II.C.1.

105. Sarbanes–Oxley Act of 2002, Pub. L. No. 107-204, 116 Stat. 745

(codified as amended in scattered sections of 11, 15, 18, 28, and 29 U.S.C.

(2012)).

106. See Richard Moberly, Sarbanes–Oxley’s Whistleblower Provisions: Ten

Years Later, 64 S.C. L. REV. 1, 46–47 (2012) (discussing legislated rewards for

whistleblowing gatekeepers).

107. Dodd–Frank Wall Street Reform and Consumer Protection Act, Pub. L.

No. 111-203, § 204, 124 Stat. 1376, 1454 (2010) (codified at 12 U.S.C. § 5384

(2012)).

108. See Catherine Shakespeare, Sarbanes–Oxley Act of 2002 Five Years On:

What Have We Learned?, 3 J.

BUS. & TECH. L. 333, 333–34 (2008) (“The Act was

designed to restore investor confidence in the capital markets by introducing

several sweeping changes and reaffirming other extant rules and regulations.”).

109. See General Auditing Standards, PCAOB, http://pcaobus.org/

Standards/Auditing/Pages/ReorgStandards.aspx (last visited Feb. 24, 2016)

(listing the auditing standards) (on file with the Washington and Lee Law

Review).

826 73 WASH. & LEE L. REV. 797 (2016)

different inquiries, the Sarbanes–Oxley regime makes it harder

for them to claim that they failed to notice problems. Moreover,

Sarbanes–Oxley’s § 404 requires auditors to review not only the

company’s financial statements, but also the systems that the

company has in place in order to collect the information used to

produce the financial statements.

110

Thus, it expands the scope of

accountants’ inquiries even further.

But if these expanded inquiries unearth problems, what are

accountants to do? Sarbanes–Oxley sets up two new channels

that accountants can use to pursue their complaints, one within

the corporate hierarchy and one outside it. First, accountants can

bring up their concerns with the audit committee, a specialized

board committee composed entirely of independent directors.

111

Responsible for hiring and firing auditors, as well as for

supervising the company’s internal systems for financial

information, the audit committee was designed to offer auditors

insulation from pressures by the CEO and the CFO.

112

Second,

accountants are required to attest publicly to the adequacy of the

company’s internal systems for financial information under § 404

and, thus, they must disclose to the market any material

weaknesses they identify.

113

By forcing auditors to put their

reputation on the line through public attestation and public

disclosure, Sarbanes–Oxley sought to boost auditors’ negotiating

position vis-a-vis management while also providing additional

information to investors.

While Sarbanes–Oxley’s stricter due diligence obligations

underlined the monitoring function of gatekeepers, we argue that

they failed to change auditors’ calculations when determining

whether to turn against a client. Accusing a company’s

management of tinkering with its financial statements remains a

bold move, one that auditors are unlikely to make without strong

evidence supporting their allegations. For all their credentials of

110. See Shakespeare, supra note 108, at 335 (“Section 404 requires an

annual audit of management’s assessment of the effectiveness of the internal

control.”).

111. 15 U.S.C. § 78j-1(m)(3)(A) (2012).

112. Id. § 78j-1(l).

113. See Coates, supra note 17, at 102 (“[O]fficers must evaluate and

disclose ‘material weaknesses’ in their firm’s control system . . . outside auditors

‘attest’ to those disclosures.”).

COLLABORATIVE GATEKEEPERS 827

independence, audit committees are likely to be reluctant to start

a fight with management officials unless they can substantiate

their concerns. Thus, a significant amount of evidence will be

necessary before such steps can be taken. Relying exclusively on

their own investigating powers and having to go through

company officials in order to collect their data, gatekeepers may

not be able to surreptitiously collect the necessary evidence.

Similarly, empirical studies of material weakness disclosures

under § 404 suggest that the strategy has only partially worked

and may even have backfired under certain circumstances. On

the one hand, firms with auditor-disclosed material weaknesses

suffer higher costs of capital

114

and tend to remedy the problem

after a year.

115

On the other hand, two-thirds of all public

companies do not disclose a material weakness until after an

earnings restatement.

116

Thus, the disclosure does not operate

preemptively, as Sarbanes–Oxley intended. Not surprisingly,

firms and auditors that do not have material weaknesses

disclosed are less likely to face litigation, possibly because they

can plausibly deny they were aware of the weakness and, thus,

avoid liability.

117

Effectively, the lack of disclosure protects

114. Jacqueline S. Hammersley, Linda A. Myers & Catherine Shakespeare,

Market Reactions to the Disclosure of Internal Control Weaknesses and to the

Characteristics of those Weaknesses Under Section 302 of the Sarbanes–Oxley

Act of 2002, 13 R

EV. ACCT. STUD. 141, 150–62 (2008); Hollis Ashbaugh-Skaife et

al., The Effect of SOX Internal Control Deficiencies on Firm Risk and Cost of

Equity, 47 J.

ACCT. RES. 1, 24 (2009).

115. However, a significant 30% of all firms with disclosed material

weaknesses make no effort to remedy them even after three years. Karla

Johnstone, Chan Li & Kathleen Hertz Rupley, Changes in Corporate

Governance Associated with the Revelation of Internal Control Material

Weaknesses and Their Subsequent Remediation, 28 C

ONTEMP. ACCT. RES. 331,

341 (2011).

116. Sarah C. Rice & David P. Weber, How Effective Is Internal Control

Reporting Under SOX 404? Determinants of the (Non-)Disclosure of Existing

Material Weaknesses, 50 J.

ACCT. RES. 811, 826 (2012).

117. See Sarah C. Rice, David P. Weber & Biyu Wu, Does SOX 404 Have

Teeth? Consequences of the Failure to Report Existing Internal Control

Weaknesses, 90 ACCT. REV. 1169, 1174 (2015) (“[D]isclosure of control

weaknesses serves to acknowledge managements’ and auditors’ awareness of

the existence of those weaknesses, which makes it more difficult for them to

plausibly claim later that they were unaware of the conditions in the control

environment that led to the restatement.”).

828 73 WASH. & LEE L. REV. 797 (2016)

auditors and clients from liability, and probably gains auditors

their clients’ loyalty.

2. Dodd–Frank: Empowering Whistleblowers

Because detecting financial fraud is notoriously hard,

enforcement authorities have an easier job when someone with

privileged access to information alerts them about ongoing fraud.

People who “blow the whistle” on corporate fraud are often

employees of the perpetrator, but can also be its business

partners, subcontractors, or even journalists who collected

evidence on the company’s dealings.

118

Financial intermediaries,

by virtue of their close relationship with the client, are one of the

most important pools of potential informants.

119

But acting as a

whistleblower also comes with significant negative repercussions

that might discourage potential informants.

Once the informant uncovers the fraud and makes her

intentions known to the company, to enforcement authorities, or

to the public, she stands the risk of being fired.

120

Whistleblowers

will find it extremely hard to find another comparable job in the

industry.

121

Exposing an employer to the rest of the world is still

considered morally reproachable, even if the employer’s actions

themselves were illegal or unfair to others.

122

Whistleblowers are

often described in derisive terms such as “rats” and “snitches.”

123

118. Alexander Dyck et al., Who Blows the Whistle on Corporate Fraud?, 65

J. FIN. 2213, 2225–28 (2010).

119. Id.

120. In one dataset, in 82% of the cases involving an employee whistleblower

whose identity became known to management, the individual employee alleged

that she was fired or had to quit her job under pressure. See Dyck et al., supra

note 118, at 2216 (“[I]n 82% of cases with named employees, the individual

alleges that they were fired, quit under duress, or had significantly altered

responsibilities as a result of bringing the fraud to light.”).

121. See Terry Morehead Dworkin, SOX and Whistleblowing, 105 MICH. L.

REV. 1757, 1763, 1763 n.39 (2007) (citing Beverly H. Earle & Gerald A. Madek,

The Mirage of Whistleblower Protection Under Sarbanes–Oxley: A Proposal for

Change, 44 A

M. BUS. L.J. 1, 6 (2007)) (noting that the statute of limitations for a

retaliation claim is only ninety days).

122. See Kathleen F. Brickey, From Enron to WorldCom and Beyond: Life

and Crime After Sarbanes–Oxley, 81 WASH. U. L.Q. 357, 365 (2003) (stating that

whistleblowers are perceived as “disloyal tattletales”).

123. See Faqihi, supra note 70, at 3351 (stating that policymakers

COLLABORATIVE GATEKEEPERS 829

For this reason, whistleblowers risk not only a significant hit at

their finances, but also the loss of friendships with co-workers

and the disdain of their professional and social circle.

124

To counter the negative repercussions of whistleblowing,

Sarbanes–Oxley declared employer retaliation against

whistleblowers a felony.

125

In practice, however, the Sarbanes–

Oxley employee complaint mechanism protected employees in

very limited cases.

126

Moreover, this approach does little to

reverse whistleblowers’ blacklisting from the very industry on

which they depend professionally and have served all their lives.

Dodd–Frank took a more daring approach: to compensate for

the loss of income and the destruction of future employment

opportunities, it offers whistleblowers a share of the bounty.

Section 922 of Dodd–Frank requires the SEC to offer to person(s)

that provided information about the fraud an amount between

10% and 30% of the total monetary sanctions collected, based on

the value of the information provided and the agency’s

policymaking priorities.

127

The SEC established a separate Office

of the Whistleblower that runs the agency’s program for

collecting and assessing information, estimating awards, and

protecting employees from retaliation.

128

In September 2014, the

SEC made headlines by awarding a record $30 million bounty to

a single whistleblower, whose identity has remained unknown to

the public.

129

historically referred to whistleblowers as “snitches” and “rats”).

124. Brickey, supra note 122, at 365.

125. Sarbanes–Oxley Act of 2002 § 1107, 18 U.S.C. § 1513(e) (2012).

126. According to a study of employee retaliation complaints under

Sarbanes–Oxley, of the 677 cases submitted to the Secretary of Labor in the

first three years, the employees won the ALJ’s protection in only six instances.

See Dworkin, supra note 121, at 1764 (describing success as “an uphill battle”

for whistleblowers).

127. Dodd–Frank Wall Street Reform and Consumer Protection Act,

§ 922(a), 15 U.S.C. § 78u–6(b)(1)(A)–(B) (2012).

128. For more information, see generally S

EC. & EXCH. COMM’N, 2014

ANNUAL REPORT ON THE DODD–FRANK WHISTLEBLOWER PROGRAM (Nov. 17, 2014),

https:// www.sec.gov/about/offices/owb/annual-report-2014.pdf. As of the date of

this Article, the 2014 Annual Report was the most recent report.

129. Josh Hicks, $30 Million Award to Tipster Underscores Banner Year for

SEC Whistleblower Program, W

ASH. POST (Nov. 19, 2014), www.washington

post.com/blogs/federal-eye/wp/2014/11/19/30-million-whistleblower-award-under

scores-banner-year-for-sec-program (last visited Jan. 22, 2016) (on file with

830 73 WASH. & LEE L. REV. 797 (2016)

While Dodd–Frank’s bounty program has been met with

apparent success in providing enforcement authorities with

information about ongoing fraud, it also has important

limitations as a model for regulating gatekeepers. To start, Dodd–

Frank’s rule prevents the award of bounty to professionals who

stumble upon indications of fraud when conducting an audit of a

company’s financial statements.

130

As a result, external auditors

and their advisers cannot take advantage of the bounty program.

Internal control officers responsible for internal audits can be

entitled to the bounty only in limited circumstances, typically

when management either fails to take action to remedy the

problem or actively tries to impede the investigation.

131

Another

category of gatekeepers not eligible for bounty awards is persons

who are already under an obligation to report violations to the

SEC.

132

Over the years, the SEC has established many rules that

require regulated professionals, such as broker-dealers or

investment advisers, to provide various reports to the agency.

133

Excluded from the bounty program are also persons who are

otherwise criminally convicted of fraud in the reported case.

134

the extent that gatekeepers find themselves at risk of being

regarded as primary participants in fraud, they will lose the right

to claim bounty. Yet, determining gatekeepers’ criminal exposure

ex ante is not always straightforward.

That Dodd–Frank’s bounty program is not intended as a

measure for regulating gatekeepers is also evident in another of

its key elements: it applies to individuals but does not extend to

corporations.

135

Dodd–Frank envisages the bounty as a reward

Washington and Lee Law Review).

130. Securities Exchange Act of 1934, § 21F(2)(C), 15 U.S.C. § 78u-6(c)(2)(C)

(2012).

131. General Rules and Regulations, Securities Exchange Act of 1934, 17

C.F.R § 240.21F-4(2)(b)(4)(v) (2015).

132. Securities Exchange Act of 1934, § 21F(2)(D), 15 U.S.C.

§ 78u-6(c)(2)(D).

133. See generally S

EC. & EXCH. COMM’N, DIV. OF INV. MGMT., REGULATION OF

INVESTMENT ADVISERS BY THE U.S. SECURITIES AND EXCHANGE COMMISSION,

S.E.C. (Mar. 2013), http://www.sec.gov/about/offices/oia/oia_investman/rplaze-

042012.pdf.

134. Securities Exchange Act of 1934, § 21F(2)(B), 15 U.S.C. § 78u-6(c)(2)(B).

135. See id. § 21(F)(2), 15 U.S.C. § 78u-6(a)(6) (defining whistleblower as

“any individual who provides, or 2 or more individuals acting jointly who

COLLABORATIVE GATEKEEPERS 831

for the individual professional who does not succumb to pressures

from superiors in order to help enforcement authorities.

136

contrast, Dodd–Frank’s bounty program does not include any

provisions that seek to harness the corporate enforcer. Dodd–

Frank creates no incentives for gatekeeper firms to set up

systems that collect and assess information about their clients, to

dig deeper in their due diligence efforts, or to understand clients’

potentially fraudulent intentions. Even if multiple individuals

within a gatekeeper firm possess a piece of the puzzle, the

corporate employer is under no obligation under Dodd–Frank to

put them together. Rather, we argue, gatekeeper firms are left to

navigate the uneasy terrain between the legality of their

participation in clients’ transactions and the disloyalty of

referring to the authorities clients who, in the end, might be

doing nothing illegal.

3. Academic Thinkers on Gatekeeper Reform

As scandals ravaged through the financial markets in the

2000s, many scholars recognize that gatekeeper reputation alone

is not sufficient to deter wrongdoing and have debated whether

and how to expand gatekeeper liability.

137

The most radical

expansion of the current regime would involve holding

gatekeepers strictly liable for any client wrongdoing, and would

thus turn gatekeepers into insurers.

138

In theory, strict liability

provide, information relating to a violation of the securities laws to the

Commission, in a manner established, by rule or regulation, by the Commission”

(emphasis added)).

136. See David Freeman Engstrom, Whither Whistleblowing? Bounty

Regimes, Regulatory Context, and the Challenge of Optimal Design, 15

HEORETICAL INQ. L. 605, 610–16 (2014) (discussing bounty regimes).

137. See generally C

OFFEE, CORPORATE GOVERNANCE, supra note 14

(providing academic thoughts on gatekeeper reform); Frank Partnoy,

Barbarians at the Gatekeepers? A Proposal for a Modified Strict Liability

Regime, 79 W

ASH. U. L.Q. 491 (2001) [hereinafter Partnoy, Barbarians at the

Gatekeepers?] (same); Coffee, Gatekeeper Failure, supra note 14 (same); Frank

Partnoy, Strict Liability for Gatekeepers: A Reply to Professor Coffee, 84 B.U.

REV. 365 (2004) [hereinafter Partnoy, Strict Liability for Gatekeepers] (same);

John C. Coffee, Jr., Partnoy’s Complaint: A Response, 84 B.U.

L. REV. 377 (2004)

(same); Hamdani, supra note 16 (same).

138. For a different system of third party insurance for financial statements,

see generally Joshua Ronen, Post-Enron Reform: Financial Statement

832 73 WASH. & LEE L. REV. 797 (2016)

has major advantages: it would lead gatekeeper firms to put in

place optimal monitoring systems, and would also free courts

from the difficult task of trying to ascertain, ex post, what

gatekeepers knew and did not know about the fraud.

139

However,

a true strict liability regime would be a “draconian response”;

140

would not only increase the price of auditor services, but also

would risk entirely unraveling the market for auditor services.

141

Thus, the most far-reaching proposals currently put forth by

prominent academics involve creating a strict liability regime

with some limits.

142

These proposals remain controversial, with

other scholars arguing instead that gatekeepers should be held

liable only for what they knew

143

or should have known.

144

With the debate about how best to expand gatekeeper

liability to an optimal level far from settled, recent scholarship

highlights other dimensions of the gatekeeper problem that

further complicate matters.

145

Professor Lawrence Cunningham

suggests that the financial industry is likely to fight any

significant expansion of liability tooth and nail.

146

For this

Insurance, and GAAP Revisited, 8 STAN. J.L. BUS. & FIN. 39 (2002).

139. S

TEVEN SHAVELL, ECONOMIC ANALYSIS OF ACCIDENT LAW 9–18 (1987); see

also Hamdani, supra note 16, at 83–86 (comparing the advantages and

disadvantages of strict liability and negligence regimes).

140. Partnoy, Strict Liability for Gatekeepers, supra note 137, at 375.

141. See Hamdani, supra note 16, at 74–76, 89 (“[T]he market will unravel

when the increase in gatekeeper fees is sufficiently large to drive all clients—

whether wrongdoers or law-abiding persons—out of the market.”).

142. See Coffee, Gatekeeper Failure, supra note 14, at 306 (proposing a

combination of strict liability regime with a cap based on a multiple of the

revenue gatekeepers received from wrongdoers); Partnoy, Strict Liability for

Gatekeepers, supra note 137, at 375 (proposing a combination of strict liability

regime with a cap based on a percentage of damages); Partnoy, Barbarians at

the Gatekeepers?, supra note 137, at 367, 370 (proposing a strict liability regime

with damages based on a “minimum percentage of the client’s ultimate

liability”).

143. See Hamdani, supra note 16, at 103–04 (advocating a knowledge-based

regime).

144. See Tuch, supra note 33, at 1628–31 (advocating a fault-based regime);

Juan José Ganuza & Fernando Gomez, Should We Trust the Gatekeepers?

Auditors’ and Lawyers’ Liability for Client’s Misconduct, 27 I

NT’L REV. L. &

ECON. 96, 96–109 (2007) (proposing that under certain assumptions, the

distinction between knowledge and negligence is not significant).

145. See infra notes 146–152 and accompanying text (providing examples of

recent scholarship and arguments).

146. See Cunningham, supra note 66, at 327, 333–35 (recommending that

COLLABORATIVE GATEKEEPERS 833

reason, he suggests that rewarding auditors for performing their

gatekeeping function well rather than punishing them for

gatekeeping failures may be a more realistic way forward.

147

Professor Andrew Tuch highlights that, while most of the

literature is based on the simplifying assumption of a single

gatekeeper, large transactions are typically reviewed by multiple

gatekeepers, including a law firm, an investment bank, and an

accounting firm.

148

Each of these firms may have only a partial

understanding of a client’s business transactions, he notes, and

may “hav[e] incentives to narrow the scope of its activities to

reduce the likelihood that it will acquire knowledge sufficient to

attract gatekeeper liability.”

149

Finally, Professor Asif Hamdani

distinguishes between “speaking” and “silent” gatekeepers;

speaking gatekeepers make statements on which third parties

rely, as in the case of accountants confirming that financial

records are accurate, while “silent” gatekeepers simply fail to

warn third parties.

150

Hamdani argues that market-based

reputation arguments do not work in the case of silent parties, as

by definition, they do not attach their name to a dishonest client’s

activities.

151

He also notes that, while liability for silent partners

in fraud could and should be expanded, it is difficult to hold a

silent party civilly liable under the current securities regime.

152

gatekeepers be rewarded for successfully performing gatekeeping functions).

147. Id. at 381–83.

148. See Tuch, supra note 33, at 1585, 1597–1601 (describing how each of

these entities is involved in a complex web of interactions when executing

transactions within the federal securities laws).

149. Id. at 1586.

150. See Assaf Hamdani, Silent Gatekeepers and Vicarious Liability 12

(May 2012) (unpublished manuscript) [hereinafter Hamdani, Silent

Gatekeepers], http://portal.idc.ac.il/he/schools/law/progs/legalworkshops/

documents/stoneridge%20paper%20idc.pdf (“An actor making a statement,

however, typically faces market incentives to refrain from facilitating fraud.

Those who make statements are often reputational intermediaries that would

suffer substantial market penalty upon the discovery of fraud. Non-speaking

actors, in contrast, do not put their reputation at risk.”).

151. See id. at 14 (“Non-speaking actors’ concealed involvement makes them

less likely to suffer market penalties when issuer fraud is uncovered.”).

152. See id. (“[F]rom a deterrence standpoint, there are convincing reasons

for imposing liability on non-speaking actors who knowingly assist issuers to

mislead investors.”).

834 73 WASH. & LEE L. REV. 797 (2016)

Where does this leave us? The discussion of the literature

above highlights three common themes. First, many scholars

agree that the market-based mechanism of gatekeeper reputation

does not suffice to deter financial fraud.

153

Second, however,

prominent scholars note that, while significantly expanding

gatekeeper liability and turning gatekeepers into insurers, this

type of mechanism would face strong opposition from the

financial industry and might lead to a crisis in the gatekeeping

professions as we know them.

154

Third, important recent work

suggests that any solutions brought forward should take into

account important and unnoticed realities about the market for

gatekeepers—namely, that each transaction involves multiple

gatekeepers, some of which are silent.

155

The proposal that

follows responds to each of these concerns.

4. Gatekeepers’ Potential Still Unexplored

For all the repeated attempts at reform and heated academic

discussions, gatekeepers remain a resource that the current

regime has not managed fully to tap.

156

The prevailing strategy

for gatekeepers seeks to entice their cooperation mostly through

heavy sanctions for failing to monitor their clients.

157

But heavy

sanctions are justified when gatekeepers knowingly assisted in

clients’ fraud or, in a few severe cases, were negligent

monitors.

158

Not surprisingly, gatekeepers have directed their

energy in clearly demarcating their knowledge or negligence, as

153. See supra notes 145–152 and accompanying text (explaining the flaws

in the theory that gatekeeper reputation alone effectively deters financial

fraud).

154. See supra notes 142–152 and accompanying text (discussing

controversies that arise due to expanding gatekeeper liability).

155. See supra notes 142–152 and accompanying text (summarizing

scholars’ proposals to deter financial fraud more effectively with gatekeepers).

156. See supra Part II.B.3 (arguing that, under the current regime,

gatekeepers face liability if they have knowledge of fraud and are thus better off

knowing less about their clients).

157. See supra notes 93–107 and accompanying text (discussing various

theories of negligence and strict liability for gatekeepers and the underlying

incentives for each theory).

158. See supra notes 100–102 and accompanying text (describing

gatekeepers’ assistance, knowledge, or negligence in three well-known cases).

COLLABORATIVE GATEKEEPERS 835

the case may be, so that they can avoid liability.

159

In this effort,

information that does not render gatekeepers knowledgeable or

negligent, but could still offer helpful tips in investigations, never

reaches enforcement authorities.

160

Worse still, gatekeepers have

an incentive to suppress this information, for fear that, if found

out, it might be considered incriminatory in hindsight.

161

Policymakers’ efforts to address this problem by opening up

new channels of communication for gatekeepers and enforcement

authorities have so far fallen short. Sarbanes–Oxley tried to get

external auditors to entrust their concerns either to independent

directors or even to the market itself.

162

Not surprisingly, neither

of these new outputs has yielded much. Before becoming certain

that their clients are violating the law, gatekeepers are unlikely

to turn openly and publicly against them, especially if voicing

concerns can later be used in a lawsuit against the gatekeeper

itself. Dodd–Frank’s whistleblower rules offer a valuable

framework for disillusioned employees and other insiders but do

not fit corporations dedicated to monitoring.

163

Still, in a financial system that grows ever more vast,

complicated, and interconnected, our regulators cannot afford to

ignore the indications of fraud that gatekeepers are bound to

come across. We need a framework that allows gatekeepers to

share this information, while also protecting them from the

negative consequences they might face down the line. If they fail

to do so, their liability should be tied to the severity of their

particular failure rather than the full extent of the underlying

fraud. The following Part proposes such a system.

159. See supra notes 93–99 (examining gatekeepers’ roles in reporting,

considering negligent and strict theories of liability).

160. See supra notes 92–93 and accompanying text (describing the

incentives for gatekeepers averting their eyes from wrongdoing to protect their

own liability).

161. See supra notes 114–117, 130–144 and accompanying text (noting

gatekeepers concerns about criminal liability in light of post-2000 financial

reforms).

162. See supra Part II.C.1 (explaining that Sarbanes–Oxley created the

PCAOB to establish audit standards for auditors and established a regime for

auditors to make complaints).

163. See supra Part II.C.2 (noting that internal auditors are exempt from

claiming a bounty under the whistleblower rules and further that the

whistleblower rules do not reward corporations, only individuals).

836 73 WASH. & LEE L. REV. 797 (2016)

III. Collaborative Gatekeepers: Elements of a New Paradigm

The model proposed here seeks to motivate gatekeepers to

share client information that has so far remained untapped in

enforcement efforts. While interacting with a client hiding

misconduct, the gatekeeper may come across some indications

that raise suspicions but fall far short of confirming problems.

But because these indications might prove extremely useful for

enforcement authorities, this proposal requires gatekeepers to

report suspicions to regulators without informing clients. In

return for submitting their suspicions, gatekeepers gain

immunity with regard to client misconduct. If they fail to submit

their suspicions, gatekeepers are subject to sanctions. This

framework, we argue below, can help gatekeepers overcome

conflicts of interest because it incentivizes them to report as soon

as they realize something is amiss before investing even greater

efforts in building client relationships. Anonymity shields

gatekeepers from clients’ objections, and immunity tempers fears

of self-incrimination. Rather than barricading themselves behind

alleged unawareness of client misconduct, gatekeepers can limit

their exposure to client risk by collaborating with authorities.

A. Background: Collecting Client Information

Existing regulatory obligations already require finance

professionals to obtain, or even actively collect, information about

their clients. For example, client suitability rules require

broker-dealers and investment advisers to understand the risk

profile of their clients.

164

Accountants must attest to the adequacy

of the company’s internal controls, as discussed above.

165

Investment bankers must confirm the accuracy of their clients’

164. See THOMAS LEE HAZEN, TREATISE ON THE LAW OF SECURITIES

REGULATION § 14.16 (6th ed. 2009) (outlining brokers’ obligations to customers

with regard to recommendations, including suitability requirements); see also

Daniel G. Schmedlen, Jr., Note, Broker-Dealer Sales Practice in Derivatives

Transactions: A Survey and Evaluation of Suitability Requirements, 52 W

ASH. &

LEE L. REV. 1441, 1449–51 (1995) (analyzing risks present (e.g., market, credit,

legal, liquidity, funding) for derivatives).

165. See 15 U.S.C. § 7262 (2012) (providing rules for “management

assessment of internal controls”).

COLLABORATIVE GATEKEEPERS 837

statements at the registration stage.

166

Moreover, financial

institutions are subject to a general obligation to supervise their

employees so as to ensure that they are not engaging in illegal

activity, either on their own or in conjunction with clients.

167

More generally, Delaware court rulings set out general rules

requiring all corporations, including gatekeepers, to set up

compliance systems that monitor their employees’ conduct.

168

In the course of tailoring their services to their client’s needs

and particularities, gatekeepers often ask for client information.

As gatekeepers plan client transactions, manage client accounts,

or represent clients in negotiations, they can come across

information on clients’ backgrounds, motivations, and plans. For

example, a client’s financial documents might have

inconsistencies; a client may have abruptly fired its outside

auditors right before a quarter of challenging performance; the

documentation of underlying loans in a securitization might have

gaps; or the timing of trades before or after corporate events

might raise doubts. Under current law, this information does not

reach regulators unless growing indications of misconduct risk

putting the gatekeeper at fault.

169

Collaborative gatekeeping puts this information front and

center. This information could point to potential illegality and

offer a missing piece of the puzzle for enforcement authorities.

Gatekeepers should evaluate this information and, if necessary,

make additional inquiries to supplement their intelligence.

Typically, these inquiries will take place at the beginning of the

client relationship. Through this process, gatekeepers could

166. See id. § 77k (imposing liability on the underwriter for material

misstatements or omissions in an issuer’s registration statement).

167. See Gadinis, supra note 77, at 714–22 (discussing the supervision

obligation using empirical data from recent SEC investigations of large and

small firms for failure to supervise).

168. See In re Caremark Int’l Inc. Derivative Litig., 698 A.2d 959, 971–72

(Del. Ch. 1996) (establishing that corporate directors have an affirmative duty

to monitor their corporations); see also Hillary A. Sale, Monitoring Caremark’s

Good Faith, 32 D

EL. J. CORP. L. 719, 719–20 (2007) (celebrating the landmark

Caremark case and its progeny).

169. For gatekeepers, there is no ledger’s liability. See supra Part II.B.3

(explaining that gatekeepers may be liable if they have knowledge of clients’

misconduct, which may incentivize gatekeepers to remain unaware).

838 73 WASH. & LEE L. REV. 797 (2016)

assess whether their clients’ conduct raises suspicions, as

outlined below.

B. Key Obligation: Filing an Anonymous Suspicious Activity

Report

The central part of this Article’s proposal is a new obligation

for gatekeepers: to file a report alerting regulators to suspicious

activities by their clients. Suspicion of misconduct is, by design, a

particularly low reporting threshold, one that sets this proposal

apart from gatekeeper liability provisions under the current

regime.

170

Suspicions could arise when clients’ rationale for

pursuing a transaction has gaps, when the information they

provide is inconsistent or false, or when their proclaimed strategy

does not fit well with specific actions that they instruct the

gatekeeper to pursue on the ground. Rather than waiting to

gather evidence that fully delineates clients’ illegal actions, the

proposal encourages gatekeepers to come forward at a much

earlier stage. Moreover, the suspicion standard incentivizes

gatekeepers who have only partial information to still alert

regulators about potential client misconduct. While gatekeepers

have no means to collect intelligence on the remaining pieces of

the puzzle, regulators can utilize their investigatory powers to

extract valuable evidence.

Which client actions can give rise to a suspicion of illegality?

Rather than relying on subjective judgments, this proposal’s

suspicion threshold calls for an objective, fact-based inquiry. To

start, suspicions should arise when a client’s conduct resembles

past instances of fraud. Courts, regulators, and self-regulatory

organizations have built a rich jurisprudence that determines the

170. See, e.g., Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 319

(2007) (noting that, “[t]o establish liability under § 10(b) and Rule 10b–5, a

private plaintiff must prove that the defendant acted with scienter, [meaning] ‘a

mental state embracing intent to deceive, manipulate, or defraud’” and

reserving the question of whether recklessness also meets the scienter

standard); see also 5A

CHARLES ALAN WRIGHT ET AL., FEDERAL PRACTICE AND

PROCEDURE § 1301.1 (3d ed. 2011) (discussing the standards for pleading

scienter after the Private Securities Litigation Reform Act of 1995); Gideon

Mark, Accounting Fraud: Pleading Scienter of Auditors Under the PSLRA, 39

ONN. L. REV. 1097, 1102–04 (2007) (arguing that courts have severely limited

auditor liability by setting a high scienter requirement).

COLLABORATIVE GATEKEEPERS 839

elements of many violation types.

171

This wealth of material also

suggests fact patterns that tend to be connected with misconduct.

For example, when a bank suddenly increases its loan granting

supply, it may raise concerns that standards are falling and

obligations to assess carefully borrowers’ profiles are overlooked.

To take a different example, intense trading around corporate

events might indicate abuse of inside information. In another

example, reluctance to provide information about odd accounting

treatments might suggest that something is amiss.

Moreover, gatekeepers might be suspicious when a client

deviates abruptly and sharply from his own past conduct or

departs significantly from the conduct of clients with similar

profiles. For example, sudden large cash deposits might suggest

money laundering; repeated trading in a stock, even as its price is

increasing, might indicate attempts for market manipulation.

Gatekeepers should seek justification for such odd patterns. By

directing gatekeepers’ efforts to similarities with past fraud and

outlier transactions, the new gatekeeper duty can become readily

administrable.

To shield gatekeepers from their clients’ discontent upon a

suspicions filing, the reports must remain anonymous. After all,

gatekeepers are expected to actively alert authorities about their

clients’ conduct; one cannot imagine that this will go down well,

even if no illegal activities are actually occurring. At a minimum,

clients may be displeased by the administrative burden of a

potential regulatory investigation. Clients might also read into

the report’s submission a betrayal of their trust by their closest

advisors. Worse still, clients may threaten to move their business

elsewhere if a report is submitted. But, even if clients understood

that gatekeepers have very little choice in submitting a report,

and even if there is no follow-up by the authorities, the doubts,

concerns, and suspicions expressed in the report’s contents might

still prove unnerving. To reduce acrimony between clients and

gatekeepers and protect the smooth operation of suspicious

activity reporting, it is essential that reports submitted to

authorities remain anonymous. Without anonymity, gatekeepers

would be more likely to hold off reporting as long as possible

because they would not want to see a profitable client

171. See generally Buell, supra note 100.

840 73 WASH. & LEE L. REV. 797 (2016)

relationship destroyed, barring clear evidence that it could not

continue.

If protecting the anonymity of reporting gatekeepers is

essential for the successful operation of the model, how likely is it

that the client finds out who submitted a report anyways? In

many cases, no regulatory action will follow the submission of a

suspicious activity report, and thus the client-gatekeeper

relationship will not be disturbed. Even if an investigation

begins, it is possible that a client will not be able to deduce who

passed the tip to the authorities. This is because most deals

involve a multitude of gatekeepers—bankers, lawyers, and

accountants—assisting multiple parties, and all have a separate

obligation to report.

172

Moreover, regulators are especially likely

to take action when they receive multiple reports and thus, many

investigations may not point toward a single source of

information. For example, regulators might investigate both the

legal and the accounting aspects of a transaction at the same

time, thus muddying the waters for the clients. As an additional

safeguard, the contents and wording of the suspicious activity

report itself will never be made available to the client and, thus,

clients will not be able to connect particularized facts in the

report with certain gatekeepers. With these precautions,

conclusively linking the investigation to a suspicious activity

report, and the report to a specific gatekeeper, might prove hard

for clients. That said, the risk that a client-gatekeeper

relationship will be interrupted once an investigation begins

cannot be excluded.

C. Sanctions for Failing to Report

Backing up gatekeepers’ new obligations with sanctions is

essential to alter gatekeepers’ behavior. Gatekeepers who fail to

report suspicions may be subject to civil penalties at the

corporate level, while individual employees may be subject to

monetary penalties or other disciplining sanctions. Ex ante, the

threat of sanctions might prompt gatekeepers to set up effective

reporting systems. Ex post, these sanctions will help regulators at

172. For example, in Escott v. BarChris Construction Co., 283 F. Supp. 643

(S.D.N.Y. 1968), defendants included lawyers and accountants.

COLLABORATIVE GATEKEEPERS 841

trial, as much less evidence will be needed to punish gatekeepers

who suspected misconduct and failed to act, than to punish

gatekeepers for actively participating in fraud.

The threat of sanctions for failure to report should induce

gatekeepers to submit their suspicions even in cases where

gatekeeper involvement would not otherwise be punishable under

other substantive law provisions. For example, gatekeepers might

have suspicions about clients’ misconduct, but they might not

have any direct involvement themselves, and thus their actions

may fall outside the scope of provisions like Rule 10b–5.

173

But

under the threat of sanctions, gatekeepers now have an obligation

to report even those client activities.

Elevating failure to report to an independently punishable

offense also has a symbolic power, which can help bring about the

cultural and institutional change in the financial industry

envisaged by this proposal. Diverse audiences—clients,

collaborators, even colleagues—might be displeased with having

suspicions filed. To avoid clashes, gatekeepers need to convince

these audiences that the report is mandatory, triggered by factual

considerations over which the gatekeeper has little discretion,

and must be filed even if the gatekeeper believes that the client is

not engaging in misconduct. Moreover, sanctions underline that,

even if a client chose to move its business elsewhere, all

gatekeepers are subject to a uniform regime and would be faced

with the same choices as to the filing of the report.

D. The Payoff: Immunity About Reported Actions

Many regulatory schemes seek to boost implementation

through sanctions for failure to comply ex post, rather than

incentives to comply ex ante. Sanctions are also an essential part

of the collaborative model, as discussed in the previous subpart.

But we also put forward an important incentive: that gatekeepers

gain immunity for reported actions provided they submitted

reports in good faith. This incentive, we believe, is dictated by the

nature of the problem that the collaborative model seeks to

173. See Tellabs, Inc. v. Makor Issues & Rights, Ltd., 551 U.S. 308, 319

(2007) (explaining that liability under Rule 10b–5 requires scienter, or intent to

defraud).

842 73 WASH. & LEE L. REV. 797 (2016)

address, namely to get gatekeepers to share information about

their clients.

To understand why immunity is worth considering in this

context, imagine that you are advising a gatekeeper who has just

come across indications of client misconduct. A comprehensive

report of these indications immediately creates a record of the

extent of gatekeeper suspicions at the time. If it turns out that

the client is indeed committing fraud, victims will ask the court

to evaluate this record ex post. Clearly, there is a risk that the

court will side with fraud victims and hold that the record meets

the fault standard for gatekeeper liability (i.e., knowledge or

negligence, depending on the case). Indeed, the stronger the

indications of fraud, the more likely the court is to side with

victims.

On the other hand, the gatekeeper faces a different outcome

if she holds off from reporting her suspicions, even after fraud is

revealed. Without a record of gatekeeper knowledge or

negligence, the court may be more readily convinced that the

gatekeeper was not at fault and impose sanctions only for failing

to identify suspicions. Still, these sanctions are likely to be less

onerous than damages to fraud victims. Thus, in some cases,

gatekeepers are likely to be better off by not reporting,

particularly when they are uncertain about how courts will

interpret their reports ex post.

The proposed immunity can alleviate this uncertainty for

gatekeepers acting in good faith. With the benefit of hindsight, a

court may examine the reported facts and conclude that, from an

objective perspective, any professional faced with such evidence

should have terminated the client relationship. Yet, if the specific

gatekeeper in question acted in good faith from a subjective

perspective, for example, because she was still trying to sort out

the client’s intentions, the immunity would operate in her favor.

Thus, the immunity would relieve gatekeepers from the need to

tiptoe between legality and illegality and allow them to report all

relevant facts and avoid unwanted legal and regulatory

adventures. Because the immunity comes into effect by law as a

result of the suspicions filing, gatekeepers do not need to

negotiate separate relief with enforcement authorities. Moreover,

the immunity attaches to the actions reported, irrespective of the

specific statutes or rules violated. For these reasons, immunity

COLLABORATIVE GATEKEEPERS 843

helps create a more stable and predictable regulatory

environment for gatekeepers. This increased stability for

gatekeepers comes at little direct cost to their business.

Essentially, the proposal envisages that the gatekeeper, after

reporting, can continue offering its services to the client, to the

extent otherwise allowed by law. The relationship with the client

would have to be interrupted only if the facts outlined in the

report could establish that the gatekeeper is in bad faith.

The proposed immunity can help gatekeepers not only

toward potential victims of clients’ fraud, but also toward clients

themselves. Reporting clients’ suspicious activity to regulators

may clash with gatekeepers’ obligations toward their clients. For

example, there might be professional rules mandating

confidentiality or general privacy laws.

174

By shielding

gatekeepers from such causes of action, the immunity removes

any remaining impediments to reporting.

While the promise of immunity may induce gatekeepers to

submit a report, they still have significant leeway in deciding

what facts to include in their report. Strategically minded

gatekeepers might wish to provide regulators with just enough

facts so as to secure the immunity benefits, while also

discouraging the regulator from actually conducting further

investigations. There may be doubts as to the scope of the

immunity, the sincerity of the reporting gatekeeper, and the

extent to which the information provided actually assisted

regulators’ efforts. To avoid this problem, regulators should be

able to strip away the immunity from gatekeepers who withheld

information from their reports intentionally or recklessly. To

resolve any disputes in accordance with principles of due process,

reporting gatekeepers should have access to a hearing before

regulators, as well as the ability to contest regulators’ decision to

strip them of the immunity in court.

174. See, e.g., MODEL RULES OF PROF’L CONDUCT r. 1.6(a) (AM. BAR ASS’N

2016) (“A lawyer shall not reveal information relating to the representation of a

client unless the client gives informed consent, the disclosure is impliedly

authorized in order to carry out the representation or the disclosure is permitted

by paragraph (b).”).

844 73 WASH. & LEE L. REV. 797 (2016)

E. How Can Reporting Suspicions Help Gatekeeper Professionals

Overcome Conflicts of Interest?

Suspicious activity reporting pays close attention to the

dynamics of the gatekeeper–client relationship, we argue,

because it interjects a regulatory obligation at a very early stage

in the development of this relationship. At that stage, the

conflicts of interest that often burden gatekeepers are less likely

to have gained real strength. To start, the time and effort that

gatekeepers will have invested in building the client relationship

is likely to be much smaller. In some instances, suspicions may

even arise right out of the client’s profile, before any professional

services are offered. The sooner gatekeepers take notice of

suspicions and submit a report, the lower the investments they

will have to make in a client relationship that might be lost if

regulators decide to move forward with an action. Even on a

personal level, the connections between gatekeeper professionals

and clients are not as deep at this early stage, and inhibitions due

to long-standing bonds are unlikely. For these reasons, suspicious

activity reporting’s early kick-start can smooth many of the

dilemmas that gatekeepers face when clients misbehave.

The high volume of suspicious activity reporting has the

potential of bringing about a culture shift in the way financial

professionals understand and perform their regulatory

obligations. Financial executives are likely to file many reports in

their career because many fact patterns can generate suspicions

of misconduct.

175

In fact, gatekeepers will be required to file

reports even when they do not believe that their client is actually

violating any laws. Because the obligation to report would arise

in a similar manner over any gatekeeper faced with similar client

facts, reporting gatekeepers will not see themselves as standing

apart from their competitors. Through these repeated filings,

executives will become acquainted with the process and its

mission and better understand their role as an important link in

safeguarding market integrity. Hopefully, in this way, individual

professionals will come to see the reports as a fulfillment of an

175. For example, some banks file hundreds of SAR reports in a year. See

infra Part V.A (explaining the shift of banks’ attitudes toward preventing money

laundering).

COLLABORATIVE GATEKEEPERS 845

obligation rather than the action of a fiduciary that violates their

clients’ trust. Similarly, the intensity of suspicions reporting can

also shift the attitudes of the financial industry and the public as

a whole. Instead of a rarity that needs to be excoriated,

cooperation with the authorities will become a regular part of

gatekeepers’ continued operation.

F. How Can Reporting Suspicions Help Address Conflicts of

Interest at the Corporate Level?

Suspicious activity reporting can help improve the

performance of existing corporate compliance infrastructure in

two ways. First, the fact-based suspicion standard provides a

more workable reporting obligation because it does not require a

particularly close understanding of the specific violation in

question. Second, the immunity resulting from suspicious activity

reporting provides stronger incentives for corporations to build

effective compliance mechanisms.

Because a client’s transaction is suspicious if it simply

resembles past instances of fraud, the obligation to report carries

a lower evidentiary burden, thus facilitating the work of internal

compliance officers. Once the suspicions threshold is met, the

personal beliefs of the professional handling the client or its

superiors do not really come into play. In fact, the gatekeeper

could also explain in the report why it believes that its client is

not actually violating the law, thus addressing any resistance

that compliance officers may face in reporting clients. Turning

the suspicion standard into a fact-based inquiry delinks it from

the subjective disposition of individual executives and, thus,

reduces the pernicious impact of conflicts of interest between

executives and their corporate employers. Examining whether a

certain set of facts or pieces of information amounts to reportable

suspicions does not require intimate knowledge of the

transactions or parties involved.

Delinking reporting obligations from people with intimate

knowledge of misconduct allows the gatekeeper to build a

separate compliance mechanism oriented towards identifying

suspicious activity and informing regulators accordingly. The

gatekeeper can develop a group of specialists trained to identify

846 73 WASH. & LEE L. REV. 797 (2016)

problematic patterns who will become the chief supervisors of the

institution’s daily activities. These specialists will combine their

constantly developing knowledge of the financial system with a

deep sense of mission to maintain its integrity. The fact that

there is a wider circle of qualified individuals who can safeguard

the company’s compliance is a significant improvement over the

current gatekeeper framework, which relies on the individuals

with the closest relationship to the potentially illicit transaction

to come forward and alert regulators.

Apart from facilitating the task of internal compliance

mechanisms, suspicious activity reporting also boosts the benefits

that the effective operation of these mechanisms can bring to

gatekeepers. In particular, the immunity associated with timely

reporting can help the gatekeepers continue their services with

little disturbance, even after regulators proceed against some of

its executives for misconduct. Under the immunity, direct

consequences—such as civil penalties, monetary awards, or

damages in private lawsuits—are not likely, thus calming

immediate fears about the health of the gatekeeper’s finances. By

pointing to its report filing, the gatekeeper can protect its

reputation by showing that its compliance mechanism works

effectively and has contributed to regulators’ efforts. Moreover, it

can argue more convincingly that the instances of misconduct

within its ranks are limited to the executives already targeted by

regulatory action. Thus, effective reporting and immunity can

help prevent the collapse of the gatekeeper when some of its

executives are found to have been violating the law, as was the

case with Arthur Andersen following Enron’s collapse.

176

IV. A Case Study of Collaborative Gatekeeping: The

Anti-Money-Laundering Regime

Part III above argues that collaborative gatekeeping holds

significant promise as a theoretical proposition.

177

However, bold

176. See supra note 100 and accompanying text (discussing how Arthur

Andersen shredded documents relevant to the ongoing SEC investigation of

Enron).

177. See supra Part III (explaining the elements of the collaborative

gatekeeper paradigm).

COLLABORATIVE GATEKEEPERS 847

policy proposals are often saddled with uncertainties. A first set

of concerns centers on the feasibility of the model. Would

financial institutions balk at the idea, fearful they would pay

dearly for costly compliance systems, only to risk alienating

clients once these were in place? And, in a globalized world,

where capital can easily move from one state to another, why

would a country place itself at a competitive disadvantage by

placing unusually strict regulations on its gatekeepers? But, even

if the collaborative model were adopted, a second set of questions

concerns its effectiveness. Would the resulting suspicious activity

reports prove informative for regulators, or would gatekeepers

provide minimal information to maintain client relationships?

And would regulators be able to analyze all the reports that came

their way, or would they be flooded with data, unable to separate

signal from noise?

To answer these questions, one could start by examining

examples of collaborative gatekeeping in practice. Yet, in almost

all areas of financial regulation, the conventional gatekeeper

model is dominant. For all the inherent variety of policy missions

in finance, from accurate disclosure in securities issuance to best

execution in stock exchange transacting and to diversified

investing in mutual funds, U.S. laws have entrusted the

conventional gatekeeper model with serving investors and the

market. This remarkable uniformity in such a foundational

concept has left little room for experimentation with alternative

regulatory solutions.

However, one area of financial regulation closely follows the

collaborative gatekeeping model developed above:

anti-money-laundering.

178

Anti-money laundering is the singular

area where U.S. policymakers, and policymakers around the

world, have opted to step off the well-trodden path and engage in

a different relationship with financial intermediaries. Therefore,

there are valuable lessons in studying the anti-money-laundering

framework, both as it has been conceptualized in law, and as it

has been implemented in practice. This Part begins our analysis

of the anti-money-laundering regime by studying its history,

which shows how regulators and private industry came to a

178. See infra Part IV (explaining how governmental regulators reached a

compromise with private industry to reach a mutually beneficial result).

848 73 WASH. & LEE L. REV. 797 (2016)

mutually beneficial compromise. Part V completes the analysis of

this regime by exploring how it has worked in practice.

Using historical records that were recently declassified,

following a forty-year embargo, we recount the origins of the

money-laundering current approach in Switzerland.

179

Switzerland, banking secrecy paradise par excellence, pioneered

the modern money laundering approach following a series of bank

scandals in the 1970s.

180

The Swiss were the first to call on

private banks to identify potentially suspicious transactions and

alert regulators accordingly.

181

We underline information

collection as the key challenge that motivated this experimental

regulatory framework. Private banks agreed to collect and report

this information on two conditions: that the process for collecting

information would be standardized, and that the regulatory

obligation would be applicable to everyone in the industry.

182

This historical account of the origins of modern

money-laundering laws upends conventional narratives: U.S.

policymakers

183

and scholars think the United States pioneered

179. Documentation for the support of this Article was acquired in the

Archives of the Swiss National Bank in Berne. The authors would like to thank

the Swiss National Bank archival staff for their assistance. Information on the

Archives, S

WISS NAT’L BANK, https://www.snb.ch/en/iabout/snb/hist/id/

hist_archiv (last visited Apr. 1, 2016) (on file with the Washington and Lee Law

Review). The authors would also like to thank the staff of the OECD Library

and Archives Service. OECD Archives, OECD, http://www.oecd.org/

general/oecdarchives.htm (last visited Apr. 1, 2016) (on file with the Washington

and Lee Law Review).

180. See Thomas D. Grant,

Toward a Swiss Solution for an American

Problem: An Alternative Approach for Banks in the War on Drugs,

14 ANN. REV.

BANKING L. 225, 241 (1995) (noting that a bank scandal involving a Credit Suisse

manager who helped clients evade detection of their assets led to the Inter-Bank

Agreement of 1992, which created a code of conduct relating to due diligence for

Swiss banks).

181. See Swiss Bankers’ Agreement on Due Diligence (Vereinbarung über die

Sorgfaltspflicht der Banken), art. 1 (July 1, 1977) (requiring Swiss-domiciled

banks to “determine that the identity of the bank customer will be reliably

clarified [to] prevent an abuse of banking secrecy [that would permit] activities

otherwise forbidden under the Agreement”) (on file with author). For the latest

version of the Convention, see generally A

GREEMENT ON THE SWISS BANKERS’

CODE OF CONDUCT WITH REGARD TO THE EXERCISE OF DUE DILIGENCE (2008),

http://www.swissbanking.org/en/20080410-vsb-cwe.pdf.

182. Letter from Swiss Bankers Ass’n to Swiss Nat’l Bank, (Oct. 20, 1977)

(on file with author).

183. See, e.g., Russian Money Laundering: Hearing Before the Comm. on

COLLABORATIVE GATEKEEPERS 849

the modern regulatory approaches in this area. Instead, as we

explore below, early U.S. efforts, including the 1970 Bank Secrecy

Act, took a very different approach.

184

After agreeing to back the

Swiss approach internationally, the United States adopted it

domestically as well and redesigned its regulatory infrastructure

to implement it.

185

This Part first traces the development and

international spread of the Swiss approach and then presents

legal mechanisms that U.S. laws have put in place to implement

it.

This historical narrative serves to illustrate the plausibility

of a model that might, at first, encounter significant objections.

Imposing new regulatory obligations on any industry is likely to

meet powerful opposition initially. However, once a few large

countries have adopted the new regulatory requirements, they

have strong incentives to lobby so that all their competitors,

domestically and internationally, are held to the same high

standard. The pages that follow trace this upward regulatory

ratchet in the adoption and spread of money-laundering laws

globally. Similar processes have led to heightened regulation in

many different fields—from the elimination of ozone-producing

chemicals, to the strict regulation of car emissions, to uniform

consumer protection, to harmonized anti-trust regimes.

186

These

historical antecedents make the adoption of the collaborative

model to other areas of financial regulation seem more plausible.

More specifically, this historical narrative helps illustrate that it

Banking and Fin. Servs., 106th Cong. 120 (1999) (statement of James A. Leach,

Chairman, Comm. on Banking & Fin. Servs.), http://commdocs.house.gov/

committees/bank/hba59889.000/hba59889_0f.htm (“[I]t is incumbent upon the

United States to lead in cracking down on money laundering as a technique to

crack down on much more significant crime, and crime that has enormous

implications for the national interest of the United States and world security.”).

184. See infra Part IV.B (choosing to fight money laundering by requiring

banks to flag large transactions).

185. See infra Part IV.C (explaining why the Swiss approach prevailed over

the U.S. approach).

186. See DAVID VOGEL, TRADING UP: CONSUMER AND ENVIRONMENTAL

REGULATION IN A GLOBAL ECONOMY 54, 76, 139 (1997) (mentioning a variety of

international laws and agreements, which have led to increased regulation of

the private sector); Anu Bradford, The Brussels Effect,

107 NW. U. L. REV. 1, 3

(2012) (“The European Union sets the global rules across a range of areas, such

as food, chemicals, competition, and the protection of privacy. EU regulations

have a tangible impact on the everyday lives of citizens around the world.”).

850 73 WASH. & LEE L. REV. 797 (2016)

is possible to overcome industry objections and adopt regulations

requiring gatekeepers to flag suspicious activity early on.

A. The Birth of a New Model: Customer Due Diligence

“Never let a good crisis go to waste”

187

was clearly in the

mind of Leo Schurmann, Vice-Chairman of the Swiss National

Bank (SNB) in the 1970s. As an independent regulator

overseeing the country’s banking sector, SNB had been worried

for a while that intricacies in the financial system could easily be

used to hide illicit gains.

188

Generally, Swiss banking secrecy law

prevented banks from sharing information about their clients

with the authorities.

189

However, the veil of secrecy was to be

187. Recently popularized by Rahm Emanuel, this quote is often attributed

to Winston Churchill, but there is no evidence he really said it. See Gerald F.

Seib, In Crisis, Opportunity for Obama, W

ALL STREET J. (Nov. 21, 2008, 12:01

AM), http://www.wsj.com/articles/SB122721278056345271 (last visited Apr. 25,

2016) (quoting Emanuel when speaking at a conference about using the

financial crisis to accomplish President Obama’s agenda) (on file with the

Washington and Lee Law Review); see also Fred Shapiro, Quotes Uncovered:

Who Said No Crisis Should Go to Waste?, F

REAKONOMICS (Aug. 13, 2009, 12:27

PM), http://www.freakonomics.com/2009/08/13/quotes-uncovered-who-said-no-

crisis-should-go-to-waste (last visited Apr. 1, 2016) (tracing the origin to M. F.

Weiner) (on file with the Washington and Lee Law Review).

188. See Grant, supra note 180, at 241

(explaining the issues leading to

customer due diligence).

189. See Swiss Banking Act, art. 47(1)

[W]hoever intentionally does the following shall be imprisoned up to

three years or fined accordingly: (a) discloses confidential information

entrusted to them in their capacity as a member of an executive or

supervisory body, employee, representative or liquidator of a bank, as

member of a body or employee of an audit firm or that they have

observed in this capacity; (b) attempts to induce such infraction of the

professional secrecy; (c) discloses confidential information to third

parties or uses this information for own benefits or the benefit of

others.

Article 47 was first altered in preparation for the establishment of the Swiss

Mutual Legal Assistance Treaty with the United States, to allow banks to

disclose confidential information requested as part of an ongoing criminal

investigation without breaching Swiss banking secrecy. The Treaty entered into

force on January 1, 1983. See Loi fédérale sur l’entraide internationale en

matière pénale (Loi sur l’entraidepénale internationale, EIMP) [Swiss Mutual

Legal Assistance Treaty], Mar. 20, 1981, Recueil systématique du droit fédérale

[RS] 351.1 (Switz.) (permitting (in relevant part) Swiss banks to disclose

confidential client information without violating secrecy prohibitions under

COLLABORATIVE GATEKEEPERS 851

lifted when banks became aware that their clients were

conducting illegal activities.

190

In effect, knowledge of illegality

required banks to share information with the authorities and

turn away clients to avoid further involvement. This duty

mirrored the conventional gatekeeper model typical of U.S.

regulation and was introduced as a result of U.S. pressure

through a U.S.–Swiss mutual legal assistance treaty.

191

But, as

Swiss central bankers very well knew, evading this duty was far

too easy: banks simply had to avoid becoming aware of clients’

illegalities.

192

As a result, money from doubtful sources could

continue to flow into Swiss banks’ coffers, fueling the central

bank’s concerns.

The opportunity to act upon these worries came in 1977,

when a money laundering scandal hit Credit Suisse, one of the

largest Swiss banks.

193

The manager of the bank’s branch in

Chiasso, a Swiss-Italian border town, was assisting wealthy

Italians to transfer funds out of Italy illegally.

194

Relying on

Swiss banking secrecy and using a Liechtenstein shell company,

the branch official had been able to keep the scheme hidden from

Credit Suisse’s top management for sixteen years.

195

Credit

article 47 of the Swiss Banking Act when such information is requested by U.S.

authorities).

190. See Swiss Federal Law on Banks and Savings Banks of Nov. 8, 1934,

amended on Apr. 22, 1999, art. 47 (Arthur Andersen et al. trans., 1996)

(permitting bank officials to reveal secret information when under a federal

obligation to testify).

191. See Treaty on Mutual Assistance in Criminal Matters, May 25, 1973,

U.S.-Switz., 27 U.S.T. 2019, T.I.A.S. No. 8302 (uniting the United States and

Switzerland to work together to identify money laundering).

192. Memorandum from the Swiss Nat’l Bank Legal Dep’t to the Swiss Nat’l

Bank (May 13, 1977) [hereinafter Swiss Nat’l Bank Legal Dep’t Memorandum]

(on file with authors).

193. See Grant, supra note 180, at 241 (“[The Inter-Bank Agreement of

1992] is the result of a scandal that surfaced in 1977 involving a branch of

Credit Suisse at Chiasso near the Italian border of Switzerland.”).

194. See id. (explaining that the scandal occurred when a manager of the

Credit Suisse branch “employed shell corporations in Liechtenstein to assist

clients in evading official detection of their assets”).

195. In 1977, in what is known as the Chiasso Scandal, Credit Suisse lost

1.4 billion Swiss francs when the bank’s account manager lent money to shell

corporations to help clients evade official detection of their assets. See, e.g.,

Swiss to Vote on Bank Law, N.Y.

TIMES (May 19, 1984), http://www.

nytimes.com/1984/05/19/business/swiss-to-vote-on-bank-law.html (last visited

852 73 WASH. & LEE L. REV. 797 (2016)

Suisse suffered a loss of 1.4 million Swiss francs (equivalent to

over $2 billion in today’s values).

196

More alarmingly, the banking

industry as a whole faced unprecedented public uproar. The

popular press fumed against banks’ practices, and politicians

called for abolishing banks’ privileges and creating a new federal

regulator.

197

Schurmann, an experienced politician and law professor,

seized his moment.

198

Banks might be able to appease public

anger and deflect undesirable regulatory intervention, he

proposed, if they openly agreed to scrutinize their clients more

closely.

199

To lend credibility to this proposal, banks would

willingly subject their information collection efforts to oversight

by the Swiss National Bank.

200

Thus, the vehicle for introducing

Apr. 2, 2016) (discussing an amendment to modify banking secrecy laws

proposed in response to the Chiasso scandal) (on file with the Washington and

Lee Law Review); Grant,

supra note 180, at 241 (discussing the Swiss

Inter-Bank Agreement of 1992).

196. Swiss to Vote on Bank Law, supra note 195.

197. See Memorandum from Swiss Nat’l Bank on Public Handling of the

Swiss Bankers’ Agreement 1–4 (May 25, 1977) (detailing the strategy for public

relations following the publication of the Agreement, and noting the public

pressure for legislative action following the Chiasso scandal) (on file with

author). Previous attempts to repeal Swiss banking secrecy continued to gain

traction after the Chiasso scandal broke. See Motion for a Revision of the Swiss

Banking Act by MP Carobbio (Social Democrats) [AB III 792-95 (1977)]

(representing an attempt to repeal Swiss banking secrecy before the Chiasso

scandal broke). For examples of parliamentary initiatives after the Chiasso

scandal broke, see also Parliamentary Motion of Jean Ziegler (Submitted May 4,

Social Democrats–21 cosignatories) (calling for a limitation of Swiss banking

secrecy’s punitive application and for opening a private right of action for

parties injured by Swiss banking secrecy); Parliamentary Motion of Felix Auer

(Submitted Mar. 9, 1977, Free Democratic Party–47 cosignatories) (suggesting a

revision of Swiss banking secrecy to reduce illegal activities).

198. See Members of the Governing Board from 1907 Onwards, S

WISS NAT’L

BANK (July 2015), https://www.snb.ch/en/iabout/snb/hist/histbio/id/hist_bios_dm

(last visited Apr. 22, 2016) (providing Leo Schürmann’s resume) (on file with the

Washington and Lee Law Review). The Swiss Bank Council is comprised of

members from the political, business, and academic professions. Members are

jointly nominated and elected by the Swiss Federal Department of Finance and

the Swiss National Bank.

199. Bank Committee, Swiss Nat’l Bank, Meeting Minutes (Mar. 31, 1977)

(Comments of Chairman Leo Schürmann) (on file with authors).

200. See Federal Political Department, Finance and Economics Service (Nov.

1977)

The [CDD Agreement] came into effect on July 1, 1977. With

COLLABORATIVE GATEKEEPERS 853

banks’ new obligations, and hopefully regaining public

confidence, would be a private gentlemen’s agreement, to which

each bank could accede, with the Swiss National Bank as a

guarantor.

201

Internal Swiss bank documents outline these banks’

willingness to take on additional compliance obligations as part of

a concerted public relations strategy.

202

With Schurmann in the lead, the central bank began

negotiations with the Swiss Bankers Association and the nation’s

major banks to hammer out the elements of this gentlemen’s

agreement.

203

Minutes of the Swiss National Bank show that they

exception of a British owned bank in Geneva, whose accession should

take place shortly, all members of the Swiss banking association have

joined. Of the non-members are only a few institutions remaining

outside the agreement, however it relates almost entirely to

unimportant, local banks. This practically total accession is due to the

will of the Swiss National Bank and the Swiss banking association to

develop a contractual instrument that from all major banking

institutions would be adopted.

(on file with authors).

201. See id. (“In order to secure a correct, effective and conform application

of the requirements of the Agreement, the Swiss National Bank and the Swiss

Bankers Association have created a working group under the Chairmanship of

the Swiss National Bank Director, which is to work on the practical

implementation with trustworthy banking personnel. The goal is to establish a

unified interpretation and practical application of the Agreement.”).

202. See Press Release, Swiss Nat’l Bank, The Swiss Bankers Agreement on

Due Diligence 1–2 (June 2, 1977) [hereinafter SNB Press Release on Due

Diligence Agreement] (noting that the “Swiss domiciled banks and the Swiss

Bankers Association” cosigned an agreement with the Swiss National Bank

“over the handling of bank secrecy and the due diligence requirements of Swiss

banks when receiving funds”) (on file with authors). Additionally, the Press

Release noted that “the banks commit themselves to abscond from active

assistance to the illegal transfer of capital, as well as falsification to domestic

and foreign authorities, especially tax authorities.” Id.; see also Letter from

Swiss Bankers Ass’n to Swiss Nat’l Bank (Oct. 20, 1977) (providing the Swiss

National Bank with an overview of the policy the Swiss Bankers Association

intended to implement across Swiss banks in regards to the handling of initial

client interviews and information) (on file with authors); Memorandum from

Swiss Nat’l Bank on Public Handling of the Swiss Bankers Agreement 1–4 (May

25, 1977) (detailing the strategy for public relations following the publication of

the Agreement) (on file with authors); S

WISS BANKERS ASS’N, PROTOCOL OF THE

ADVISORY BOARD 25–26 (June 2, 1977) (illustrating the willingness of the Swiss

banking community to adopt measures which would reduce the political fallout

from the Chiasso crisis) (on file with authors).

203. See Bank Committee, Swiss Nat’l Bank, Meeting Minutes (Mar. 11,

1977) (Comments of Chairman Leo Schürmann) (noting that “some operations

carried out under the cover of bank secrecy are not in the long term interests of

854 73 WASH. & LEE L. REV. 797 (2016)

knew that some of the money in Swiss banks was of “doubtful”

origins—i.e., potential connections with crime were suspected

though not positively known.

204

In the future, banks should not

be able to turn a blind eye to such suspected connections. Thus,

the agreement would require banks to conduct due diligence to

ascertain the beneficial ownership of the funds. In the course of

their due diligence, banks could more easily become aware of

illegal connections. With awareness thus forced upon them,

banks would have no choice but to cooperate with authorities.

But what if due diligence did not resolve the question fully,

so that doubts about the funds still remained? At that stage,

banks would be required to ask clients for further documentation,

including a written and signed statement setting out the client’s

representations about the funds’ origins.

205

Even in this scenario,

the SNB suggested, banks should err on the side of caution and

be allowed to provide this evidence to the authorities, without

violating bank secrecy laws.

206

This was a very important victory

for the regulator because it effectively expanded the scope of

banks’ obligations beyond the safe haven of awareness to the

unchartered territory of suspicions and doubts. In many cases,

suspicions and doubts might be the best banks could do. After all,

the regulator knew that banks’ due diligence tools might be

limited because the individuals presenting themselves to banks

would also be the ones to speak as to the origin of their funds.

207

the Swiss banking system”) (on file with authors).

204. See id. (Comments of Fritz Leutwiler) (“There are bank clients with

noticeably doubtful, yes even criminal backgrounds, and therefore should not

have been accepted, nevertheless they were accepted, and there exists related

money in Swiss banks.”).

205. See S

WISS BANKERS ASSOCIATION & SIGNATORY BANKS, AGREEMENT ON

THE

SWISS BANKS’ CODE OF CONDUCT WITH REGARD TO THE EXERCISE OF DUE

DILIGENCE art. 6 (Apr. 7, 2008), [hereinafter SWISS BANKS DUE DILIGENCE

AGREEMENT], http://www.swissbanking.org/en/20080410-vsb-cwe.pdf (requiring

a bank to repeat the identification verification process when doubts arise as to

the truth of its client’s identity, beneficial owner and contracting partner, or

declaration of beneficial ownership).

206. See Press Release, Swiss Nat’l Bank, Swiss Bankers’ Agreement on Due

Diligence (June 2, 1977) (noting that, “[i]n doubtful cases, banks should err on

the side of caution”) (on file with authors).

207. See generally Swiss Nat’l Bank Legal Dep’t Memorandum, supra note

192 (noting that “[t]he problem concerns all those to whom Swiss Banking

secrecy applies. Notably, abuse of Swiss banking secrecy is driven by the

client”).

COLLABORATIVE GATEKEEPERS 855

Upon hearing SNB’s thoughts, the banking community froze

in disbelief. Bankers feared political pressure might derail the

regulator into instituting vague and hard-to-satisfy legal

standards.

208

Yet, bankers’ initial skepticism quickly gave way to

constructive engagement. The Swiss Bankers Association

believed that it would be ineffective to allow each individual bank

to determine whether it had satisfied due diligence obligations.

209

Rather, banks needed to create a uniform approach so as to

ensure high-quality information gathering.

210

More specifically,

standardized forms would guide bank employees in their efforts

to collect information from clients, thus delineating the questions

that banks should ask clients and helping bank employees

evaluate clients’ responses.

211

The Swiss Bankers Association and

208. See Internal Commc’ns between the Swiss Nat’l Bank, the Zurich

Attorneys Ass’n & the Group of Private Bankers of Geneva (revealing that the

two private organizations were wary that political pressure would cause the

Swiss National Bank to rely on vague legal measures) (on file with authors);

Letter from Zurich Attorneys Ass’n (Verein Zürichischer Rechtsanwälte Zürich)

to Swiss Nat’l Bank Legal Dep’t (Sept. 13, 1977) (noting that “the discussion

over the [Swiss National Bank’s Due Diligence Agreement] was the cause of

special concern” within the Zurich Attorneys Association) (on file with authors);

Letter from Groupement des Banquiers Privés Genevois to Dr. Fritz Leutwiler,

Head of Dep’t of the Swiss Nat’l Bank (June 2, 1977) (noting that the GRPT “is

conscious of the political pressure on the banking sector” following the Chiasso

crisis) (on file with authors).

209. S

WISS BANKERS ASS’N, PROTOCOL OF THE ADVISORY BOARD, supra note

202, at 25–26 (illustrating that the members of Advisory Board saw the

Agreement with the Swiss National Bank as only “a part of” an answer to the

Chiasso crisis and that internal controls by banks would also have to be

applied).

210. See Swiss Bankers Ass’n, Meeting Minutes (June 2, 1977) (noting that

an “effective internal monitoring must be adopted for the size, the business type

and the scope of the internal organization of each individual [banking]

institution”) (on file with authors).

211. See generally SNB Press Release on Due Diligence Agreement, supra

note 202

[I]n order to carry out these tasks, the banks are given a qualified due

diligence obligation and there will be procedural guarantees created.

The banks are required to ask about the origin of funds. When the

bank knows, or by using reasonable diligence should know, that

money being brought to their bank is from prohibited or illegal

origins, the bank is not allowed to enter into business. If the client is

handling funds for another person then the banks are required to

provide uniform disclosure in forms. In doubtful cases, banks should

err on the side of caution and fill out standardized forms. The content

should describe whether the client is handling for his own account or

856 73 WASH. & LEE L. REV. 797 (2016)

the SNB created a working group to produce the standard forms

that would streamline the implementation of this requirement.

212

Apart from standardization, bankers believed that another

guarantee was necessary for this scheme to work: Every Swiss

bank should be willing to participate.

213

Bankers justified their

insistence on universal participation by arguing that, if there

were a hole left in this system—however tiny—illegal money

would find it and exploit it.

214

One could imagine that they were

also concerned about competition. As due diligence imposes

burdens both on banks and their customers, banks that simply

stay out of the scheme might immediately gain an advantage over

their competitors who participate. Regardless of bankers’

motivations, the regulator also wanted as broad industry

participation as possible.

215

By using their combined leverage, the

SNB and the Swiss Bankers Association managed to have

practically every Swiss bank enter into the customer due

diligence agreement on July 1, 1977.

216

whose account he is handling if not his own.

212. See Press Release, Fed. Political Dep’t, Fin. & Econ. Serv. (Nov. 1977)

(noting that the goal of the working group established under the Chairmanship

of the SNB Director Ehrsam was to ensure the uniform interpretation and

practical implementation of the Agreement across the Swiss banking sector) (on

file with authors).

213. S

WISS BANKERS ASS’N, PROTOCOL OF THE ADVISORY BOARD, supra note

202, at 25–26.

214. See generally Swiss Nat’l Bank Legal Dep’t Memorandum, supra note

192

[T]he new formalities for opening an account are intended to provide

a deterrent effect directed as doubtful clients. A direct blocking of all

illegally acquired funds is not possible: realistically we have to set

ourselves straight that the individual who assists in the acquisition of

funds is also in the position to speak to their origin. In the US for

example, the heads of organized crime present themselves as

reputable businessmen. However, they are also in the position to

provide very good information over themselves if necessary. Hence

the blocking of doubtful funds is only possible through indirect

means.

See also generally id. at 226 (statement of Fritz Leutwiler).

215. SNB Press Release on Due Diligence Agreement, supra note 202, at 1–

2; see also Memorandum from Swiss Nat’l Bank on Pub. Handling of the Swiss

Bankers Agreement, supra note 202, at 1–4 (detailing the strategy for public

relations following the publication of the Agreement).

216. See SNB Press Release on Due Diligence Agreement, supra note 202, at

1–2 (noting that for the few banks which did not sign “the National Bank is to

COLLABORATIVE GATEKEEPERS 857

For all its private nature and contractual basis, the

agreement also included provisions seeking to cement its

implementation.

217

Upon noticing failures to conduct due

diligence, the SNB could impose significant fines on banks.

218

resolve potential disputes, the agreement set up an arbitration

tribunal.

219

Finally, signatories undertook to avoid participating

in transactions designed to circumvent the agreement, or to

otherwise assist clients in deceiving domestic and foreign tax and

law enforcement authorities.

220

The 1977 agreement established key foundational elements

of a collaborative gatekeeping model. Importantly, the

agreement does not let banks off the hook if they simply fail to

become aware of illegality.

221

Rather, it requires them to assess

whether there are doubts as to clients’ background and gather

evidence and documentation outlining these doubts.

222

As a

result, the agreement introduces a lower threshold that banks

must abide by when determining whether to grant access to the

financial system: Knowledge was not required; suspicions and

doubts would suffice.

223

To implement this regime, the

agreement required banks to abandon their passive complacency

regarding incoming clients and actively ascertain the origin of

clients’ wealth and the goals of clients’ transactions.

224

Thus, it

name them so that they can be seen in the associated documents” distributed at

the time of the Agreement’s publication).

217. See S

WISS BANKS DUE DILIGENCE AGREEMENT, supra note 205, art. 15,

(noting that the agreement becomes active “on July 1, 1977 and is valid for a

fixed period of five years” and that the signatory banks “authorize the

administrative council of the Swiss Bankers’ Association” to undertake

amendments or precisions to the Agreement in coordination with the National

Bank).

218. See id. art. 11 (violating the agreement may result in fines of up to 10

million Swiss francs).

219. See id. art. 13 (allowing a bank imposed with a fine for an alleged

breach of due diligence to arbitrate the matter before paying the fine).

220. Id. art. 8.

221. Id. art. 1.

222. See id. art. 6 (requiring banks to verify clients’ identities again during

the business relationship if doubts arise).

223. See id. (having doubts, not knowledge or certainty, as to a client’s

identity requires banks to confirm the client’s identity again).

224. See Grant, supra note 180, at 245–46 (explaining that bank clients

complete one form (Form A) and then the bank uses an internal recording form

858 73 WASH. & LEE L. REV. 797 (2016)

introduced customer due diligence into the banking world.

Through standardization, it provided banks with clear guidance

about how to satisfy their new obligations but also called for

robust compliance departments to handle the newly required

information collection efforts.

Although the 1977 agreement introduced customer due

diligence as an obligation for banks, it also left a significant

mismatch between banks’ obligations toward clients and their

relationship with law enforcement authorities. If law

enforcement authorities requested information about a client in

the context of an investigation, banks were free to provide it

without violating bank secrecy and privacy laws, even if they

only had doubts about the client.

225

But, the 1977 agreement did

not create an obligation for banks proactively to report their

suspicions to the authorities.

226

Rather, it expected that banks’

newly detailed customer due diligence and recordkeeping

obligations would sufficiently deter them from accepting

criminals’ business because it would be far more

straightforward to determine the extent of a bank’s knowledge

or suspicion of client misconduct. As a result, while the 1977

agreement set banks down the path of collecting information, it

did not establish a mechanism for utilizing this information to

prevent money laundering. This final step occurred when the

international community embraced the Swiss due diligence

obligations, as the Parts below discuss.

227

B. Different Model: United States Enacts Universal Reporting

While Switzerland was pioneering the modern

money-laundering approach, contemporary U.S. regulatory

efforts to fight money laundering followed a different model.

“compound[ing] information about the client investment aims, family

background, and other general matters”).

225. See Grant, supra note 180, at 244–50 (describing the system for

collecting information and when it may be necessary to contact authorities).

226. Because the 1977 Agreement was purely contractual in nature, it did

not have an effect on altering Swiss banking secrecy, as it was defined under

Article 47 of the Swiss Banking Act.

227. See infra Part IV.C (explaining how the Swiss model prevailed over the

U.S. model in the international community).

COLLABORATIVE GATEKEEPERS 859

Instead of requiring banks to screen their clients closely, and

ask in-depth follow-up questions whenever suspicions arose, in

the 1970s, American regulators’ asked banks to flag every large

transaction.

228

As this approach did not bring the desired

results, the United States adopted the conventional gatekeeping

model in the 1980s.

229

In the early 1970s, Congress was looking for new tools to

use in its fight against drugs. Drug dealers, it reasoned, had to

use the financial system to divert profits from illegal operations

into legal activities. If only authorities had a paper trail of all

transactions in which customers deposit or transfer cash in

significant sums, then drug dealers would have far greater

difficulty laundering their profits.

230

Creating this paper trail

was a cornerstone objective of the Bank Secrecy Act of 1970.

231

The Act required financial institutions to file a Currency

Transaction Report (CTR) for all deposits, withdrawals,

exchanges, or transfers of currency in excess of $5,000, as well

as multiple transactions conducted in the same business day by

the same person if reaching that amount.

232

The reporting

threshold was increased to $10,000 in 1984 and has remained

228. See infra notes 231–234 and accompanying text (discussing that the

United States’ approach centered on creating a paper trail of large transactions).

229. See infra notes 240–242 and accompanying text (describing how the

U.S. approach failed because criminals broke transactions up into smaller

transactions).

230. See Peter E. Meltzer, Keeping Drug Money from Reaching the Wash

Cycle: A Guide to the Bank Secrecy Act,

108 BANKING L.J. 230, 231 (1991)

(“Congress reasoned that if it were possible to create a paper trail of all

transactions in which customers deposit, withdraw, mail, or . . . attempt to

transfer cash sums in excess of $10,000 at one time, the money-laundering

process . . . would be made far more difficult.”).

231. See id. at 232 (“The CTR system represents Congress’s attempt to begin

creation of the paper trail at the placement stage of the money-laundering

process.”). The “Bank Secrecy Act of 1970” is the commonly used name for the

Financial Recordkeeping and Reporting in Currency and Foreign Transactions

Act of 1970, Pub. L. No. 91-508, 84 Stat. 1114 (1970) (codified as amended in

scattered sections of 12 U.S.C., 15 U.S.C., and 31 U.S.C.).

232. See 31 U.S.C. § 5316(d)(a) (2012) (allowing the Secretary of the

Treasury to prescribe regulations to further define what amounts to culmination

of closely related events and, specifically, the term “at one time” for the purposes

of that section).

860 73 WASH. & LEE L. REV. 797 (2016)

unchanged ever since.

233

Besides cash, the reporting

requirement was later expanded to wire transfers.

234

The primary objective of the CTR regime is deterrence,

rather than information collection.

235

When filing a CTR,

financial institutions must simply record their identity and

details of the individuals appearing before them accurately but do

not have to make any further inquiries.

236

In the CTR scheme,

financial institutions are passive registers of financial flows,

rather than active investigators.

237

While Congress hoped that

these records might be useful to law enforcement authorities

during ongoing crime investigations, it did not really see them as

jumpstarting new inquiries.

238

Rather, Congress expected that

drug dealers would be loath to have their information recorded.

239

Within the next decade, it became clear that the deterrent

effect of CTRs would not be as prevalent as had been anticipated.

After a spat of high sanctions against banks for failing to comply

with their Bank Secrecy Act obligations, authorities were flooded

with CTRs from around the country.

240

The vast majority of these

233. See Richard R. Cheatham & James W. Stevens, Absent Regulatory

Changes, Hispanic Immigrants Pose an Unbankable Risk,

123 BANKING L.J. 195,

196 (2006) (“Enacted in 1970, the Bank Secrecy Act required financial

institutions to report cash transactions in excess of $5,000 (increased to $10,000

in 1984) . . . .”); Wuliger v. Office of Comptroller of Currency, 394 F. Supp. 2d

1009, 1013 (N.D. Ohio 2005) (explaining the Bank Secrecy Act).

234. See Annunzio–Wylie Anti-Money Laundering Act of 1992, 18 U.S.C.

§ 1960(b)(2) (2012) (“[T]he term ‘money transmitting’ includes transferring

funds on behalf of the public by any and all means including but not limited to

transfers within this country or to locations abroad by wire, check, draft,

facsimile, or courier.”).

235. See generally Cuéllar, supra note 7, at 326.

236. Id. at 359.

237. See F

INANCIAL CRIMES ENFORCEMENT NETWORK, supra note 30, at 1

(“FinCEN does not initiate or carry out any investigations on its own. Rather, it

provides other agencies with tactical and strategic intelligence analyses . . . .”).

238. See id. at 5 (“In line with the concept first proposed by Customs in

1981, FinCEN is not to initiate or carry out any investigations on its own. Its

primary purpose is to serve and assist other agencies . . . .”).

239. Cuéllar, supra note 7, at 352.

240. See John K. Villa, A Critical View of Bank Secrecy Act Enforcement and

the Money Laundering Statutes, 37 CATH. U. L. REV. 489, 497 (1988) (“The most

significant recent event concerning the BSA occurred in United States v. Bank of

New England, N.A., where the Bank of New England was convicted of

committing thirty-one felonies arising out of its failure to file CTRs.”).

COLLABORATIVE GATEKEEPERS 861

transactions were innocent,

241

and law enforcement authorities

rarely used this data for preventive reasons. Moreover, drug

dealers were quick to find ways to evade being reported: they cut

transactions into smaller pieces, spread them over multiple banks

and multiple dates, or used legitimate business fronts, such as

restaurants, to justify cash payments.

242

Congress responded by raising the stakes for violators

through a familiar technique: criminalization. The Money

Laundering Control Act of 1986

243

prohibited structuring

transactions to evade reporting requirements

244

and, more

lastingly, turned money laundering into a criminal offense.

245

both cases, financial institutions would face sanctions if they

assisted their customers in violating the law. This legislation

introduced the conventional gatekeeper model, used throughout

financial regulation in the United States, into

anti-money-laundering law. Effectively, regulators relied on

financial institutions as reputational intermediaries, requiring

them to turn away potential money launderers, or face heavy

sanctions.

Yet, courts interpreted these provisions of the Money

Laundering Control Act as including a “scienter” requirement.

246

241. See Money Laundering: The Volume of Currency Transaction Reports

Filed Can and Should Be Reduced: Hearing on S. 1664 Before the Comm. on

Banking, Hous. and Urban Affairs, 97th Cong. 1 (1994) (statement of Henry R.

Wray, Director, Admin. Justice Issues), http://archive.gao.gov/t2pbat

4/151052.pdf (“[M]any of the reports being filed are of normal business

transactions that could have been exempted from being reported. CTRs that

report normal business transactions are of no value to law enforcement and

regulatory agencies in detecting money laundering activity.”).

242. See Meltzer, supra note 230, at 236 (“Layering is the process of

transferring these funds among various accounts through a series of complex

financial transactions that are intended . . . to separate these funds from their

original sources. Finally, integration is the process of shifting the laundered

funds to legitimate organizations . . . .”).

243. 18 U.S.C. §§ 1956–57 (2012).

244. See 31 U.S.C. § 5324 (2012) (prohibiting persons from structuring or

assisting in structuring transactions to evade the reporting requirements).

245. See 18 U.S.C. § 1956(a)(1) (providing that any person who violates the

statute “shall be sentenced to a fine of not more than $500,000 or twice the

value of the property involved in the transaction, whichever is greater, or

imprisonment for not more than twenty years, or both”).

246. See Duncan E. Alford, Anti-Money Laundering Regulations: A Burden

on Financial Institutions,

19 N.C. J. INT’L L. & COM. REG. 437, 458 (1994) (“These

862 73 WASH. & LEE L. REV. 797 (2016)

As a result, financial institutions would be liable only if they

knowingly or willfully assisted their customers in money

laundering.

247

Although the defendant does not need to know the

specific offense that clients are committing, mere suspicion of

criminality does not satisfy scienter.

248

Yet, money launderers

could devise transactions precisely in order to mask their

criminal intentions. As banks were simply required to report or

analyze what was presented to them, many illicit activities

continued to remain outside the scope of the law. Consistent with

the traditional gatekeeper model, the financial institution’s

liability fell to be determined mostly ex post.

249

That said, some

preventive reporting requirements were also put in place, but

blanket transaction reporting left authorities with too many cases

to analyze fruitfully.

250

In short, by the end of the 1980s, U.S.

anti-money-laundering law included both a reporting

requirement for all large transactions and many elements of the

conventional gatekeeper prototype.

C. The International Community Adopts and Extends the Swiss

Model: From Due Diligence to Suspicious Activity Reporting

Because financial activity—and financial fraud—cross

borders easily, efforts to harmonize regulations around the world

have often been proposed.

251

But at least two models were on the

table in the early 1980s: the Swiss model, requiring banks to

investigate clients’ suspicious activities,

252

and the American

new criminal offenses are broad enough to apply to any person assisting money

launderers. Courts, however, have read a scienter requirement of ‘knowing’ the

source of the property or proceeds into this statute.” (citing United States v.

Massac, 867 F.2d 174 (3d Cir. 1989)).

247. See id. (noting, for example, that a court will find a defendant guilty “if

he knows that the subject property was derived from some criminal activity;

however, the defendant need not know the specific offense”).

248. See id. (emphasizing the breadth of the money laundering provisions).

249. See supra Part II.A (describing the theory of gatekeeping).

250. See infra notes 334–341 and accompanying text (noting that all reports

are not equally informative).

251. See infra Part IV.C (explaining that the Swiss model served as a

template for other countries’ financial regulation efforts).

252. See infra Part IV.C (describing Switzerland’s customer due diligence

and bank recordkeeping requirements).

COLLABORATIVE GATEKEEPERS 863

model, requiring banks to pass on information about all large

transactions.

253

This subpart explores why the Swiss model was

chosen as the template for global financial regulation, how global

regulators extended Switzerland’s efforts, and how hundreds of

countries adopted the resultant regulations.

Once Switzerland started requiring customer due diligence

and bank recordkeeping, the international community quickly took

notice, not least because the Swiss themselves were eager to

advertise their banks’ qualifications.

254

The Council of Europe, an

international organization best known for establishing the

influential European Court of Human

Rights, recommended the

adoption of customer due diligence principles as early as 1980.

255

In 1988, the Basel Committee on Banking Supervision, an

informal meeting of G10 central bankers that issues the prominent

Basel accords on capital adequacy, adopted a non-binding

“Statement of Principles on the Prevention of Criminal Use of the

Banking System for the Purposes of Money Laundering.”

256

In its

253. See infra Part IV.D (outlining U.S. anti-money-laundering law).

254. The Swiss Federal Department of Foreign Affairs called attention to

the Agreement within international organizations as a means of indicating

Swiss efforts to combat financial crime. See Département Politique Fédérale de

la Suisse, Document aux Représentations Diplomatiques Suisses 4 (Nov. 10,

1977) (presenting to Swiss diplomatic representations at the OECD, Bureau of

Observation in New York, Swiss Mission to the European Community in

Brussels, the Swiss permanent representation to the Council of Europe in

Strasbourg, and the Consulates General in Frankfurt, Hong Kong, Milan and

New York) (on file with authors). At the presentation, they stated:

Swiss banking secrecy and numbered accounts are not affected by the

Agreement. In fact, [the Agreement] should help deflect the critiques

that Swiss banks accept funds without regard to their ‘doubtful’

origins because the banks now have an obligation of diligence in these

circumstances. This political element, certainly relative but also

positive . . . could certainly assist you in your delicate task of

defending Swiss interest in a subject particularly controversial.

Id.

255. On June 27, 1980, the Council published Recommendation No. R (80)

10, dealing with measures to combat the transport and sheltering of illegal

capital. See generally Recommendation from the Council of Eur. Comm. of

Ministers to the Council of Eur. Member States (June 27, 1980),

https://www.coe.int/t/dghl/monitoring/moneyval/Instruments/Rec%2880%2910_e

n.pdf.

256. See B

ASEL COMM. ON BANKING SUPERVISION, PREVENTION OF CRIMINAL

USE OF THE BANKING SYSTEM FOR THE PURPOSE OF MONEY-LAUNDERING (1988),

http://www.bis.org/publ/bcbsc137.pdf (“The Committee believes that one way to

864 73 WASH. & LEE L. REV. 797 (2016)

statement, the Basel Committee put its weight behind customer

due diligence requirements.

257

These early endorsements were important in propelling

customer due diligence on the international agenda as one of the

key elements of a comprehensive anti-money-laundering regime

with global reach. To create such a regime, the G7 put together

the Financial Action Task Force (FATF), an informal

international network of officials from treasury departments and

ministries of finance.

258

Over 180 countries around the world

have adopted FATF’s Forty Recommendations on the shape of

national anti-money-laundering, first issued in 1990.

259

To draft

these recommendations, a group of 160 experts from around the

world met in Paris for six months,

260

discussing alternative

approaches. Having already benefitted from the support of

central bankers and other international experts, customer due

diligence became the prototype on the basis of which FATF

framed its recommendations pertaining to the financial

industry.

261

More specifically, Recommendations 12 through 14

promote this objective, consistent with differences in national supervisory

practice, is to obtain international agreement to a Statement of Principles to

which financial institutions should be expected to adhere.”).

257. See id. (“With a view to ensuring that the financial system is not used

as a channel for criminal funds, banks should make reasonable efforts to

determine the true identity of all customers . . . . Particular care should be taken

to identify the ownership of all accounts . . . .”).

258. For more information on the creation of FATF and the spread of its

forty recommendations around the world, see generally Stavros Gadinis, Three

Pathways to Global Standards: Private, Regulator, and Ministry Networks,

109

AM. J. INT'L L. 1 (2015).

259. See F

INANCIAL ACTION TASK FORCE ON MONEY LAUNDERING,

INTERNATIONAL STANDARDS ON COMBATING MONEY LAUNDERING AND THE

FINANCING OF TERRORISM & PROLIFERATION: THE FATF RECOMMENDATIONS 7

(2012) [hereinafter T

HE FATF RECOMMENDATIONS], http://www.fatf-

gafi.org/media/fatf/documents/recommendations/pdfs/FATF_Recommendations

.pdf (“[FATF Recommendations], together with the Special Recommendations,

have been endorsed by over 180 countries . . . .”).

260. See F

INANCIAL ACTION TASK FORCE ON MONEY LAUNDERING, REPORT

1990–1991, 4 (1991) [hereinafter FATF REPORT 1990–1991], http://www.fatf-

gafi.org/media/fatf/documents/reports/1990%201991%20ENG.pdf (“Five series of

meetings were held in Paris. More than 160 experts from various ministries, law

enforcement authorities, and bank supervisory and regulatory agencies, met

and worked together during six months.”).

261. See F

INANCIAL ACTION TASK FORCE ON MONEY LAUNDERING, THE FORTY

RECOMMENDATIONS OF THE FINANCIAL ACTION TASK FORCE ON MONEY

COLLABORATIVE GATEKEEPERS 865

emphasized the need to identify the beneficial owners of bank

accounts and to inquire further until establishing the true

identity of account owners.

262

Moreover, Recommendation 15

called for financial institutions to be vigilant toward complicated

transactions with no apparent financial purpose, as they are

likely to mask criminal activity.

263

These customer due diligence

requirements reflect the influence of Swiss archetypes because

they require banks to move beyond passively reporting

information provided by clients and toward proactively

investigating the truthfulness of client representations, as well as

their background.

Having thus decided in favor of a substantive due diligence

obligation for financial institutions, FATF experts heavily

debated what should these institutions do with the information

they stand to collect.

264

Some countries, like the United States,

which required their financial institutions to report all

transactions above a certain value to authorities, pushed for a

similar recommendation at a global level.

265

Yet, most countries

preferred an alternative system, where only suspicious

transactions would be reported to regulators.

266

They argued that

LAUNDERING 2 (1990) [hereinafter FORTY RECOMMENDATIONS OF FATF],

http://www.fatf-gafi.org/media/fatf/documents/recommendations/pdfs/FATF%20

Recommendations%201990.pdf (enumerating the recommendations falling

under the “Customer Identification and Record-keeping Rules” category).

262. See id.

(“Financial institutions should take reasonable measures to

obtain information about the true identity of the persons on whose behalf an

account is opened or a transaction conducted if there are any doubts as to

whether these clients or customers are not acting on their own behalf, in

particular, in the case of domiciliary companies . . . .”).

263. See id. (“Financial institutions should pay special attention to all

complex, unusual large transactions, and all unusual patterns of transactions,

which have no apparent economic or visible lawful purpose. The background and

purpose of such transactions should, as far as possible, be examined . . . .”).

264. See T

HE FATF RECOMMENDATIONS, supra note 259, at 19 (concluding

that “[i]f proceeds of a criminal activity, or are related to terrorist financing, it

should be required, by law, to report promptly its suspicions to the financial

intelligence unit (FIU)”).

265. See U

NITED STATES: REPORT ON THE OBSERVANCE OF STANDARDS AND

CODES––FATF RECOMMENDATIONS FOR ANTI-MONEY LAUNDERING AND

COMBATING THE FINANCING OF TERRORISM, 6 (2006) (summarizing that the

United States requires people and entities to report suspicious activity to

FinCEN).

266. See FATF REPORT 1990–1991, supra note 260, at 11 (“A large majority

of the participating countries continue to consider that the implementation of a

866 73 WASH. & LEE L. REV. 797 (2016)

they could get similar results with a less burdensome system.

267

Thus, Recommendations 16 through 18 propose that national

laws permit or require financial institutions to report their

suspicions to regulators, and protect them from restrictions on

client confidentiality or privacy.

268

These reports, FATF

recommends, should neither be disclosed nor discussed with

clients.

269

In addition, FATF also suggested that national

authorities establish databases that can easily aggregate and

analyze this information.

270

With the introduction of suspicious activity reporting

through FATF’s Forty Recommendations, all the building blocks

of the modern anti-money-laundering regime took the form they

retain to this date. Most commentators saw suspicious activity

reporting as a mere corollary of customer due diligence, rather

than the veritable link between regulators and the financial

system it was due to become.

271

At the time, no country had

implemented a suspicious activity reporting system.

272

This

would soon change, as governments and regulators turned

FATF’s recommendations into domestic laws.

273

The full potential

system to report all important currency transactions is difficult to envisage.”).

267. See id. (“They feel that at least similar results can be attained through

the less burdensome system of a properly implemented suspicious transactions

reporting scheme.”).

268. See F

ORTY RECOMMENDATIONS OF FATF, supra note 261, at 3 (“If

financial institutions suspect that funds stem from a criminal activity, they

should be permitted or required to report promptly their suspicions to the

competent authorities.”).

269. See id. (“Financial institutions, their directors and employees, should

not, or, where appropriate, should not be allowed to, warn their customers when

information relating to them is being reported to the competent authorities.”).

270. See id. at

4 (“Countries should consider the feasibility and utility of a

system where banks and other financial institutions . . . would report all

domestic and international currency transactions above a fixed amount, to a

national central agency with a computerized data base, available to competent

authorities for use in money laundering cases . . . .”).

271. See Eric J. Gouvin, Bringing Out the Big Guns: The USA Patriot Act,

Money Laundering, and the War on Terrorism, 55

BAYLOR L. REV. 955, 967–70

(2003) (describing Congress’s anti-money laundering initiatives in the 1990s).

272. See id. at 966–67 (“The Money Laundering Control Act of 1986

established money laundering as a crime, making the United States one of the

first countries in the world to criminalize the practice.”).

273. See THE FATF RECOMMENDATIONS, supra note 259, at 7 (“[FATF

Recommendations], together with the Special recommendations, have been

COLLABORATIVE GATEKEEPERS 867

of suspicious activity reporting did not become clear until the

mid-2000s, when developments in information technology

assisted with analyzing large pools of information.

274

This subpart has traced the development and extension of

the Swiss regulatory innovations of extensive due diligence and

record keeping into the modern global system of suspicious

activity reporting. This history of money laundering provides an

additional example of an upwards regulatory ratchet.

275

It shows

how, once leading jurisdictions impose heightened regulatory

requirements, they have incentives to help spread these around

the world, to level the playing field and avoid placing their firms

at a competitive disadvantage. This narrative helps address a

potential concern about the feasibility of the collaborative

gatekeeper model—that countries will put up unyielding

resistance. Instead, this history suggests that, under certain

circumstances, it is possible for crises to trigger far-reaching

national and even global regulatory reforms.

The following subpart and Part discuss the incorporation of

collaborative gatekeeping in U.S. anti-money-laundering law.

Subpart D discusses briefly how the U.S. Congress decided to

experiment with customer due diligence and suspicious activity

reporting, and how regulators implemented these congressional

authorizations. Part V presents the U.S. experience with

suspicious activity reports, which have come to outshine CTRs as

the key method for detecting criminal links in the financial

system.

D. Customer Due Diligence and Suspicious Activity Reporting in

U.S. Law

As they were hard at work in negotiating the FATF

Recommendations, U.S. authorities were becoming increasingly

uneasy with the pitfalls of the CTR model.

276

Not only were

endorsed by over 180 countries . . . .”).

274. See Gouvin, supra note 271, at 970–73 (outlining the many reporting

requirements of the Patriot Act).

275. For a discussion of upwards regulatory ratchets and other examples,

see generally V

OGEL, supra note 186; Bradford, supra note 186.

276. See Gouvin, supra note 271, at 967–70 (describing efforts to reduce the

868 73 WASH. & LEE L. REV. 797 (2016)

regulators inundated with reports of cases that had no connection

to money laundering or other illegal activity, they also felt that

really suspicious cases did not get reported because clients

managed to evade the reporting threshold in one way or

another.

277

This subpart outlines how the United States decided

to move from the conventional gatekeeper model to the

collaborative gatekeeper model in the area of money laundering

and outlines the current U.S. regulatory requirements.

Beginning in 1985, some federal banking regulators asked

financial institutions to guide their investigations by pointing to

particularly suspicious clients.

278

But these forms were available

only to the agency soliciting them and not to other regulators and

enforcement authorities.

279

In 1990, the Treasury asked banks to

use the CTR form to report transactions that they might find

suspicious.

280

However, banks only had to tick a box for

“suspicious” in their form, without a single word of explanation

for the basis of their suspicions.

281

The scope of these suspicions

number of CTRs being filed).

277. See Cuéllar, supra note 7, at 326 (“In some cases, traffickers use the

cover of an existing cash-intensive business, such as a money exchange business

or a restaurant, to help justify large currency deposits—even if they exceed the

reporting threshold.”).

278. See Press Release, U.S. Dep’t of Treasury, The National Money

Laundering Strategy for 2000,

86 (Mar. 2000) [hereinafter National Money

Laundering Strategy], http://www.treasury.gov/press-center/press-releases/

Documents/ml2000.pdf (“Beginning in 1985, the federal bank supervisory

agencies required financial institutions that they supervised to report actual or

potential violations of law and suspicious transactions to federal law

enforcement authorities and the supervisory agencies on what was then referred

to as the Criminal Referral.”).

279. See U.S.

GOV’T ACCOUNTABILITY OFFICE, MONEY LAUNDERING: NEEDED

IMPROVEMENTS FOR REPORTING SUSPICIOUS TRANSACTIONS ARE PLANNED, REPORT

RANKING MINORITY MEMBER, PERMANENT SUBCOMM. ON INVESTIGATIONS,

SENATE COMM. ON GOVERNMENTAL AFFAIRS 3 (1995) [hereinafter MONEY

LAUNDERING: NEEDED IMPROVEMENTS], http://gao.gov/assets/160/155076.pdf

(explaining that the agencies did not share forms with each other and there

were six different agencies collecting different forms).

280. See National Money Laundering Strategy,

supra note 278, at 85

(explaining that the Appunzio–Wylie Money Laundering Act, for example,

amended the BSA to authorize “the Secretary of the Treasury to require bank

and non-bank financial institutions to report suspicious transactions”).

281. See M

ONEY LAUNDERING: NEEDED IMPROVEMENTS, supra note 279, at 3

(“In 1990 the Department of the Treasury modified the Currency Transaction

Report (CTR) form, which financial institutions use to report currency

COLLABORATIVE GATEKEEPERS 869

was also very limited, as banks were under no obligation to

investigate actively clients’ backgrounds or motivations.

282

address the fragmentation of information across agencies, the

Treasury Department established a specialized bureau, the

Financial Crimes Enforcement Network (FinCEN), as an

intelligence unit tasked with aggregating and analyzing

reports.

283

At the same time, the Treasury lobbied Congress for a

redesign of the anti-money-laundering regime that would bring

U.S. law in line with international standards.

284

The desired overhaul came with the Annunzio–Wylie

Anti-Money Laundering Act of 1992,

285

which triggered the shift

toward collaborative gatekeeping.

286

Motivated by a foreign

bank’s collapse for assisting Colombian drug cartels launder

money through its Miami offices,

287

the Annunzio–Wylie Act

paved the way for incorporating FATF’s recommendations into

U.S. law. Realizing that the overly simplistic reporting system of

CTRs could not capture increasingly nuanced money laundering

techniques,

288

Congress introduced suspicious activity reporting

transactions exceeding $10,000 to IRS. A block was added to the form that could

be checked to indicate that the transaction was considered suspicious.”).

282. See id. at 11 (noting that “specific criteria for determining whether a

transaction was suspicious” had not yet been developed).

283. See F

INANCIAL CRIMES ENFORCEMENT NETWORK, supra note 30, at 5

(explaining that FinCEN was created to “improve coordination of financial

crimes enforcement” within the Treasury Department).

284. The congressional hearings for the Annunzio–Wylie Anti-Money

Laundering Act of 1992 expressly refer to Treasury’s efforts to use FATF as an

inspiration for the Act. See Money Laundering Enforcement Amendments of

1991: Hearing Before Comm. on Fin. Inst. Supervision, Regulation, and Ins.,

Comm. on Banking, Fin., and Urban Affairs, 102d Cong. 11–12 (1991) (“In the

Treasury’s view, the most important legislative amendments in this session

relate to implementation of the recommendations of the FATF.”).

285. See Alford, supra note 246, at 460 (“As part of the Housing and

Community Development Act of 1992, the U.S. Congress passed the Annunzio–

Wylie Anti-Money Laundering Act (Annunzio–Wylie Act). The Act implements

many of the recommendations of the G-7 Task Force.”).

286. See Gouvin, supra note 271, at 967 (“[Suspicious Activity Reports] were

an attempt to fine tune the money laundering reporting system in light of the

fact that a bright line dollar amount rule or even a requirement to pick up

smurfing sometimes left suspicious transactions unreported.”).

287. The Bank of Commerce and Credit International (BCCI) collapsed in

1991. See 137 C

ONG. REC. S9461-04 (1991) (statement of Sen. Cranson)

(discussing whether UAE sales linked to international drug money laundering).

288. See Cuéllar, supra note 7, at 358 (“Legislators and law enforcement

870 73 WASH. & LEE L. REV. 797 (2016)

as a general requirement for all U.S. financial institutions. To

implement the requirement, Congress opted for a broad

delegation to the Secretary of the Treasury, who was given the

power to demand reports for any violation of law or regulation.

289

This sweeping authorization forms the foundation of the United

States early reporting system, under which financial institutions

alert U.S. regulators for potential money laundering and other

illegal acts.

290

The statute also prohibits filers from informing

clients about their reports.

291

However, the Annunzio–Wylie Act

did not define what constitutes suspicious activity, nor did it

elaborate on the steps that U.S. financial institutions must take

in order to comply with this obligation.

292

The task of clarifying

these concepts, which form the backbone of customer due

diligence, fell on the Secretary of the Treasury.

293

Based on the Annunzio–Wylie authorization, the Treasury

has developed a definition of suspicious activity that applies

consistently on different segments of the financial system.

294

According to Treasury rules, financial institutions must report

transactions that involve funds either derived from illegal

activities or used to disguise them.

295

Transactions designed to

evade Bank Secrecy Act reporting requirements, such as the

officials have not been entirely blind to the possibility that laundering strategies

are flexible and therefore require more nuanced responses in addition to the

mechanistic currency reporting requirements.”).

289. See Annunzio–Wylie Anti-Money Laundering Act of 1992, §§ 1513,

1517(b), 31 U.S.C. § 5318(g) (2012) (providing the Act’s “Reporting of Suspicious

Transactions” provision).

290. See id. § 5318(g)(3)(a) (granting protections against liability for filers).

291. See id. § 5318(g)(2) (“[N]either the financial institution, director, officer,

employee, or agent. . . of . . . the financial institution or other reporting person,

may notify any person involved in the transaction that the transaction has been

reported . . . .”).

292. See id. § 5318(g) (“The Secretary may require any financial institution,

and any director, officer, employee, or agent of any financial institution, to

report any suspicious transaction relevant to a possible violation of law or

regulation.”).

293. See id. § 5318(a)(2) (authorizing the Secretary to require financial

institutions to maintain appropriate anti-money laundering procedures).

294. See, e.g., 12 C.F.R. § 163.180 (2015) (regarding savings associations); id.

§ 208.62 (regarding institutions that are members of the Federal Reserve

System).

295. See, e.g., id. § 21.11(c) (detailing when banks are required to submit

Suspicious Activity Reports).

COLLABORATIVE GATEKEEPERS 871

$10,000 CTR threshold, are also regarded as suspicious.

296

broadly, financial institutions must report transactions that have

no business or apparent lawful purpose, or is not the sort in

which the particular customer would normally be expected to

engage.

297

This definition of suspicious activity reporting places

financial institutions in a fundamentally different position

compared to their pre-1992 Bank Secrecy Act obligations.

298

Rather than passively recording and reporting transactions over

a certain dollar value, financial institutions are hence expected to

actively seek indications of criminality or illegality.

299

Instead of

their mostly mechanistic role in the past, financial institutions

now have to use their judgment in order to decide whether to

report a client transaction.

300

They also have to describe the

transaction to authorities and include the reasons that give rise

to their suspicions.

301

For all these reasons, financial institutions

have become active participants in the fight against money

laundering and other financial crimes.

As the cornerstone of modern anti-money-laundering

compliance, this definition of suspicious activity reporting works

in conjunction with an extensive body of other statutory rules and

regulatory directives, case law, and informal guidance by

regulators. Over time, subsequent legislative efforts have further

enhanced the information gathering powers of the Treasury and

the obligations of financial institutions to collaborate.

302

Below,

296. See 31 C.F.R. § 1010.311 (requiring reports of each transaction

involving more than $10,000.).

297. See, e.g., 12 C.F.R. § 21.11(c)(4)(iii) (demanding banks report

transactions without an apparent lawful purpose).

298. See supra Part IV.B (describing early American models of

anti-money-laundering rules).

299. See 31 C.F.R. § 1010.312 (2015) (requiring financial institutions to

verify and record the identity of individuals conducting certain transactions).

300. See 12 C.F.R. § 21.11(a) (“This section ensures that national banks file

a Suspicious Activity Report when they detect a known or suspected violation of

Federal law or a suspicious transaction related to a money laundering activity

or a violation of the Bank Secrecy Act.”).

301. See M

ONEY LAUNDERING: NEEDED IMPROVEMENTS, supra note 279, at 11–

12 (discussing the implementation of “know your customer” policies among

financial institutions).

302. See generally Riegle Community Development and Regulatory

Improvement Act of 1994, Pub. L. No. 103-325, 108 Stat. 2160, 2243 (1994)

(Title IV of which is commonly known as the Money Laundering Suppression

872 73 WASH. & LEE L. REV. 797 (2016)

this Article focuses on two elements of this regulatory apparatus

that help illustrate the operation of collaborative gatekeeping: the

standardization of reporting through forms and the sanctions

against financial institutions for failing to comply.

303

The

following Part discusses the implementation of these legal

requirements in practice.

304

To streamline SAR submissions, the Treasury collaborated

with other federal banking regulators and law enforcement

authorities to develop a specific form for submitting suspicious

activity reports.

305

The form is designed to provide more

comprehensive directions to filers as to the information they need

to include.

306

For example, the form requires filers to pick a

specific category or type with which the reported activity

conforms and to include specific identification information for

filers and clients.

307

Institutions must submit a SAR within thirty

days after suspicions arise.

308

SAR submissions under this regime

started in April 1996.

309

FinCEN, tasked with receiving and

Act); Money Laundering and Financial Crimes Strategy Act of 1998, Pub. L. No.

105-310, 112 Stat. 2941 (codified at 31 U.S.C. § 5340 (2012)); International

Counter-Money Laundering and Foreign Corruption Act of 2000, H.R. 3886,

106th Cong. (2000); USA PATRIOT Act, Pub. L. No. 107–56, 115 Stat. 272, 301–

377 (2001) (codified at 31 U.S.C. §5318 (2012)).

303. See infra Part IV (discussing the anti-money-laundering regime).

304. See supra Part V (discussing the practical application of the anti-

money-laundering regime).

305. See 31 C.F.R. § 1010.306 (2015) (“Reports required by § 1010.311,

§ 1010.313, § 1010.340, § 1010.350, § 1020.315, § 1021.311 or § 1021.313 of this

chapter shall be filed on forms prescribed by the Secretary. All information

called for in such forms shall be furnished.”).

306. See M

ONEY LAUNDERING: NEEDED IMPROVEMENTS, supra note 279, at 35

(discussing the evolution of government forms for reporting suspicious

transactions).

307. See U.S. DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK,

SUGGESTIONS FOR ADDRESSING COMMON ERRORS NOTED IN SUSPICIOUS ACTIVITY

REPORTING 3–4 (2007), https://www.fincen.gov/statutes_regs/guidance/pdf/SAR

_Common_Errors_Web_Posting.pdf (providing examples of common errors made

when filing out these forms).

308. 31 U.S.C. § 5318(g) (2012).

309. See U.S.

DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK, THE

SAR ACTIVITY REVIEW: TRENDS, TIPS, & ISSUES 4 (2010),

https://www.fincen.gov/news_room/rp/files/sar_tti_18.pdf (“SAR filing for

depository institutions has only been in place since 1996, and more recently for

other industries.”).

COLLABORATIVE GATEKEEPERS 873

aggregating SARs, regularly issues guidance regarding how to

better complete SARs.

310

These expansive reporting requirements gain strength

through a strict sanctioning mechanism. A willful violation of the

obligation to submit suspicious activity reports may entail civil

penalties,

311

or even criminal penalties of up to five years in

jail.

312

At the same time, financial institutions must maintain

anti-money-laundering programs that oversee compliance, train

employees, and submit reports.

313

The USA PATRIOT Act, passed

shortly after the 9/11 attacks, further tightened financial

institutions’ obligations to collect information about clients’

backgrounds, business purposes, and anticipated account

activities.

314

Moreover, courts have developed a “willful

blindness” doctrine, under which deliberate failure to collect

information amounts to willfulness to conduct money

laundering.

315

This web of provisions and interpretations provides

regulators with the power to enforce anti-money-laundering laws

closely and intently.

This Part has outlined the history of the modern

anti-money-laundering regime. Using newly released archival

data, we have shown how collaborative gatekeeping in the area of

anti-money laundering emerged in an unlikely country—

Switzerland—and quickly spread throughout the world.

316

Contrary to expectations, banks and other financial institutions

did not fight these new regulatory requirements tooth and nail,

310. For an example of filing guidance, see generally FINANCIAL CRIMES

ENFORCEMENT NETWORK, supra note 237.

311. See 31 U.S.C. § 5321(a)(1)–(2) (dictating that the civil penalty ranges

from $25,000 to $100,000).

312. See id. § 5322(a) (noting that imprisonment will not exceed 5 years).

313. See id. § 5318(h) (mentioning that financial institutions must also

create the development of internal policies in compliance with the Act).

314. See id. § 5318 (providing, among other requirements, that the Secretary

of the Treasury prescribe regulations requiring financial institutions to

“consult[] lists of known or suspected terrorists or terrorist organizations”).

315. See Cuéllar, supra note 7, at 344–45 (explaining that “willful blindness

can amount to such knowledge, at least (1) when the defendant claims to lack

such knowledge, (2) the facts suggest deliberate ignorance, and (3) jurors would

not misunderstand the instruction as mandating an inference of willful

blindness”).

316. See supra Part IV.B (discussing universal reporting in the United

States).

874 73 WASH. & LEE L. REV. 797 (2016)

but instead helped create them.

317

More specifically, two critical

elements of the regime, the imposition of regulatory requirements

on big and small gatekeepers alike, and the standardization of

reports, were introduced at the insistence of large banks. That

said, we have not yet explained how this regime has worked in

practice. The next Part explores this question and shows how

gatekeepers and regulators have collaborated in their efforts to

implement the modern anti-money-laundering regime.

V. The Anti-Money-Laundering Regime in Practice

This Part discusses the operation of the

anti-money-laundering regime on the ground. It helps address

two critical concerns about the collaborative gatekeeper model.

First, how might gatekeepers react to the requirement that

clients provide early warning to regulators, and report client

activity that seems suspicious? Will gatekeepers be able to

separate the suspicious from the innocuous, and will they be

willing to pass on this information to regulators? Second, how

might regulators respond? Will they make full use of suspicious

activity reports (SARs), or will they set these aside in favor of

other priorities and sources of information?

As the subpart below discusses, financial institutions across

the United States, representing all segments of the market and

diverse lines of business, are increasingly submitting SARs in

recent years.

318

This widespread embrace of suspicious activity

reporting indicates a shift in the way the industry approaches

money laundering: Instead of withholding information out of

concerns about betraying clients, financial institutions have come

to view reporting as an obligation equally applicable to all. To

carry out this mission, financial institutions created populous

compliance departments, structured under specific regulatory

guidelines and operating under regulatory supervision.

319

They

317. See supra Part IV.D (discussing customer due diligence and reporting

law).

318. See infra Part V.A (discussing the volume and quality of SAR filings).

319. See Bruce Kelly, Firms Pumping Millions into Their Compliance

Departments to Keep Regulators at Bay, INV. NEWS (Oct. 26, 2014),

http://www.investmentnews.com/article/20141026/REG/310269996/firms-pumping-

millions-into-their-compliance-departments-to-keep (last visited Apr. 1, 2016)

COLLABORATIVE GATEKEEPERS 875

have also invested heavily in modern technology for data analysis

and sharing to scout for violations, explore and analyze

surrounding circumstances, and submit and review reports.

320

This compliance infrastructure has greatly expanded

gatekeepers’ information processing capacity, thus boosting their

chances of actually catching misconduct. But it has also changed

dynamics within gatekeepers, blunting the conflict of interest

between gatekeeper firms and their employees. That is, the new

compliance infrastructure utilizes a broad range of employees, as

well as technological infrastructure, to flag suspicious activities,

rather than leaving this task to those employees who courted a

particular client, and who are most likely to suffer from conflicts

of interest. The industry’s embrace of SARs and the related

compliance infrastructure suggest that the proposed theoretical

framework is not entirely impracticable.

Are these investments paying off? Does the information

gathered through suspicious activity reporting have any value for

enforcement authorities? The paragraphs below show that

regulators believe that SARs reveal a lot and thus devote

significant time and resources in reviewing SARs. They review

SARs not only to fight money laundering, but also to combat

diverse types of financial crime and non-criminal fraud.

321

Indeed, since the introduction of the SAR filing obligation,

criminal cases targeting money laundering, as well as

convictions, have generally increased.

322

These developments

illustrate that gatekeepers’ intimate knowledge of their clients’

business models, and their ability to distinguish between

legitimate business proposals and potentially fraudulent ventures

at an early stage, are proving valuable to regulators. They thus

(explaining that firms are spending millions of dollars for their compliance

departments) (on file with the Washington and Lee Law Review).

320. See PETER REUTER & EDWIN M. TRUMAN, CHASING DIRTY MONEY: THE

FIGHT AGAINST MONEY LAUNDERING 100 (2004) (estimating that capital costs

account for two-thirds of anti-money-laundering compliance costs).

321. See infra Part V.B (discussing SAR filings that relate to other issues

other than money laundering).

322. See REUTER & TRUMAN, supra note 320, at 109–13 (noting that money

laundering adjudications increased from 1,159 in 1994 to 1,420 in 2001).

Similarly, money-laundering convictions over the same period increased from

81% to 88%. Id. at tbl.5.2.

876 73 WASH. & LEE L. REV. 797 (2016)

highlight the potential of collaborative gatekeeping as a blueprint

for reforming financial regulation.

A. Volume and Quality of SAR Filings Indicates Industry Buy-In

U.S. financial institutions, though initially apprehensive

about filing SARs, quickly espoused the practice with eagerness.

In 1996, there were about 50,000 SARs filed with FinCEN;

323

2003, the SARs filed per year had risen to over 300,000.

324

Ten

years later, in 2013, filed SARs had exceeded 1,600,000.

325

During

this period, FinCEN has intensified its efforts to police

submission of suspicious activity reports and has imposed fines

some view as extraordinarily large.

326

The Department of Justice

has criminally prosecuted financial institutions for

anti-money-laundering violations, putting some out of

business.

327

To put SARs’ increase in perspective, it is worth contrasting

them to Currency Transaction Reports (CTRs), triggered for

every transfer of over $10,000 through the financial system.

328

The volume of CTRs has remained relatively stable over the same

period, ranging from about 12 million in 1996 to over 14 million

in 2011.

329

The comparison between CTRs and SARs also reveals

that financial institutions are selective about submitting a SAR.

In 2011, there were over 14 million CTRs filed, compared to about

323. U.S. DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK, THE SAR

ACTIVITY REVIEW: BY THE NUMBERS 1 (2004), www.fincen.gov/news_room/

rp/files/sar_by_numb_03.pdf.

324. Id.

325. SAR

STATS TECHNICAL BULL, supra note 28, at 1.

326. See David Zaring & Elena Baylis, Sending the Bureaucracy to War, 92

IOWA L. REV. 1359, 1414 (2007) (noting two instances of $30 million fines).

327. See id. at 1415 (“$43 million in combined criminal and civil fines

against Riggs Bank . . . put the bank of out of business.”).

328. See F

ED. FIN. INST. EXAMINATION COUNCIL, CURRENCY TRANSACTION

REPORTING—OVERVIEW, https://www.ffiec.gov/bsa_aml_infobase/pages_manual/

OLM_017.htm (last visited Feb. 26, 2016) (explaining that multiple transactions

equaling $10,000 or more are treated as a single transaction) (on file with the

Washington and Lee Law Review).

329. U.S.

DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK, ANNUAL

REPORT 2011, 7 (2011) [hereinafter ANNUAL REPORT 2011, http://www.fincen.gov/

news_room/rp/files/ annual_report_fy2011.pdf.

COLLABORATIVE GATEKEEPERS 877

1.4 million SARs.

330

Fears that filers would simply submit a SAR

for every client that crosses their institution’s doorstep and thus

dilute SARs’ signaling value, seem to not have materialized.

Instead, it seems that the suspicions threshold pushes filers to

think hard about when to alert regulators to client activity.

All segments of the market have contributed to the increase

in suspicious activity reporting. Financial institutions from across

the nation, big and small, in one or in multiple lines of business,

are increasingly reporting their suspicions to authorities.

331

Such

diversity in filers indicates that many market participants come

to view reporting suspicions to regulators as their obligation. As

one compliance officer stated: “There has been a cultural change

in the banking industry. Before we were focusing more on the

customer, now we have to focus more on compliance.”

332

Indeed,

finance professionals currently consider SARs as the main

channel through which the U.S. government collects intelligence

about money laundering.

333

The exponential increase in SARs represents a staggering

growth in the amount of tips bank regulators are receiving about

money laundering. How informative are these tips for

enforcement authorities? Generally, regulators treat SARs as an

important source of information about financial misconduct,

suggesting that many disclosures are of high quality. Reports by

regulators that regularly review SARs, such as the Federal

Reserve and the FDIC, have stated that they see little evidence of

defensive filing, such as reports that provide only skeletal

information in an effort to discharge a regulatory obligation

without triggering an investigation.

334

As further discussed

330. See id. (providing a comparison of the volume of the different reporting

methods).

331. See, e.g., U.S.

DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK,

THE SAR ACTIVITY REVIEW: BY THE NUMBERS 4 (2013) https://www.fincen.

gov/news_room/rp/files/btn18/sar_by_numb_18.pdf

(describing filers from

different segments of the market); U.S. DEP’T OF TREASURY FIN. CRIMES

ENFORCEMENT NETWORK, THE SAR ACTIVITY REVIEW: TRENDS, TIPS & ISSUES 15

(2006) [hereinafter TRENDS, TIPS & ISSUES 2006], https://www.fincen.

gov/news_room/rp /files/sar_tti_10.pdf (discussing filings by state).

332. N

EIL KATKOV, TRENDS IN ANTI-MONEY LAUNDERING 2011, at 5 (2011).

333. See R

EUTER & TRUMAN, supra note 320, at 106 (explaining that SARs

are viewed as being more informative).

334. See MONEY LAUNDERING: NEEDED IMPROVEMENTS, supra note 279, at 19

878 73 WASH. & LEE L. REV. 797 (2016)

below, many regulators invest significant time and effort in

reviewing SARs every month, which highlights the importance of

SARs for their agenda.

335

As Andrew Ceresney, Director of the

SEC’s enforcement division, put it:

The SEC receives tens of thousands of tips and referrals every

year from many different sources including investors,

whistleblowers and SROs. But SARs coming from

broker-dealers often stand out from this pack in terms of

reliability because the best ones contain allegations of

wrongdoing that are described clearly and comprehensively,

but also concisely. This reduces the amount of research and

assessment that is needed before determining whether and

how to act.

336

This does not mean that all reports are equally informative.

Unfortunately, examining SAR disclosures themselves is not

possible for researchers, as they are confidential by law.

337

But

regulatory institutions have described their use of SARs in a

variety of annual overviews and statements.

338

This evidence

suggests that the quality of information provided through SARs

varies, with some reports providing important leads for

enforcement actions, and others offering little of substance.

339

FinCEN itself has stated that a number of reports do not fully

sketch the reported suspicious activity, by failing to answer basic

questions such as “who, what, when, were, why, and how.”

340

(noting that “[b]oth banks have a policy of not filing suspicious CTRs”).

335. See infra Part V.E (discussing collaborative regulators).

336. Andrew Ceresney, Remarks at SIFMA’s 2015 Anti-Money Laundering

and Financial Crimes Conference, Feb 25, 2015, http://www.sec.gov/news/

speech/022515-spchc.html (last visited Jan. 21, 2016) (on file with the

Washington and Lee Law Review).

337. See 31 U.S.C. § 5318(g)(2) (2012) (explaining that neither the reporting

institution nor government may disclose the information contained in the SAR).

338. See U.S.

DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK, THE

SAR ACTIVITY REVIEW: TRENDS, TIPS & ISSUES 1 (2013) [hereinafter TRENDS, TIPS

& ISSUES 2013] (“The SAR Activity Review—Trends, Tips & Issues is a product

of continual dialogue and collaboration among the nation’s financial institutions,

law enforcement officials and regulatory agencies to provide meaningful

information about the preparation, use and value of Suspicious Activity

Reports . . . .”).

339. See generally R

EUTER & TRUMAN, supra note 320, at 107; Ceresney,

supra note 336.

340. U.S.

DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK, REPORT

OUTREACH TO DEPOSITORY INSTITUTIONS WITH ASSETS UNDER $5 BILLION 30

COLLABORATIVE GATEKEEPERS 879

light of the enormous volume of reported cases, some variation in

SAR quality is, perhaps, not very surprising. As FinCEN

concludes, the vast majority of filed SARs are generally in line

with regulatory guidance in describing activity as suspicious.

341

An indirect way of assessing SARs’ potential impact is to

explore whether the increase in filings has changed the landscape

for enforcing anti-money-laundering laws. Indeed, U.S. data

suggest that, as SAR filings have increased, so has the number of

money laundering cases brought and the number of convictions

won by criminal authorities.

342

To give one example, regulators

report that between 2003 and 2012, depository institutions

reported over 200,000 cases of suspected insider abuse, such as

cases where bank employees used client funds for personal

gain.

343

In more than half of these cases, the executives involved

were subsequently fired or suspended.

344

These connections are

only tentative because confidentiality rules prevent researchers

from connecting a specific SAR to a specific conviction. That said,

the fact that regulators, the only persons with the full picture,

make extensive use of available SARs suggests that they find

much that is useful in these reports.

B. SAR Filings Besides Money Laundering

Perhaps one of the most staggering aspects of SAR

submissions is that the vast majority of reported cases do not

involve money laundering.

345

Indeed, only 27% of all SARs

submitted to FinCEN in 2014 ended up concerning money

(2011) [hereinafter DEPOSITORY INSTITUTIONS WITH ASSETS UNDER $5 BILLION],

www.fincen.gov/news_room/rp/reports/pdf/Banks_Under_$5B_Report.pdf.

341. See id. (mentioning that, although the SARs are typically consistent

with what is suspicious activity, many of them are filled out incorrectly).

342. See Elod Takats, A Theory of “Crying Wolf”: The Economics of Money

Laundering Enforcement 28 (Int’l Monetary Fund Working Paper No. 07/81,

2007) (explaining that “the number of money laundering convictions measures

how efficient SARs are in providing useful evidence to convict criminals”).

343. T

RENDS, TIPS & ISSUES 2013, supra note 338, at 12.

344. Id. at

13.

345. See generally SAR Stats, U.S.

DEP’T OF TREASURY FIN. CRIMES

ENFORCEMENT NETWORK [hereinafter FinCEN SAR Stats], https://www.fincen.

gov/news_room/rp/sar_by_number.html (last visited Apr. 1, 2016) (on file with

the Washington and Lee Law Review).

880 73 WASH. & LEE L. REV. 797 (2016)

laundering.

346

As regulators quickly realized, money laundering

occurs through actions that are common in many different types

of fraud. Typical activities that trigger money laundering

suspicions involve transactions with no apparent economic

purpose, use of multiple locations or accounts for a common goal,

questionable or false documentation, counterfeit instruments, etc.

All these machinations are not exclusive to money launderers,

but could easily involve tax evasion, insider trading, consumer

fraud, identity theft, and a host of other activities—either

fraudulent, criminal, or both.

347

As a result, SARs have opened a

window into diverse criminal undercurrents in the financial

system.

This extensive review of SAR information has opened

regulators’ eyes to problems they did not clearly see before. For

example, depository institutions clearly identified elder abuse as

a rising trend in financial fraud, as older Americans need to

manage sizeable resources but are not as technologically savvy.

348

A stream of SARs prompted FinCEN to conduct an extensive

report and to identify practices that indicate elder abuse.

349

Moreover, SARs help regulators collect intelligence about new

structures or tools in the financial system, particularly when

these new tools can also facilitate misconduct. For example, when

bitcoins emerged as a successful virtual currency, SARs were

crucial in providing the government with information about the

bitcoin ecosystem and shaping regulatory guidance.

350

346. According to FinCEN, of the 2,413,447 activities reported in 2014 by

depository institutions, only 672,136 involved money laundering. Exhibit 5:

Number of Filings by Type of Suspicious Activity by Depository Institutions,

INCEN (2015), https://www.fincen.gov/news_room/rp/files/SAR02/Section_2-

Depository_Institution_SARs.xls.

347. See generally FinCEN SAR Stats, supra note 345.

348. See, U.S.

DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK,

ADVISORY TO FINANCIAL INSTITUTIONS ON FILING SUSPICIOUS ACTIVITY REPORTS

REGARDING ELDER EXPLOITATION 1 (2011), https://www.fincen.gov/statutes_regs/

guidance/pdf/fin-2011-a003.pdf (crediting financial institutions’ increased

reporting of elder abuse as motivation for advisory).

349. See id. (“Analysis of SARs reporting elder financial exploitation can

provide critical information about specific frauds and potential trends, and can

highlight abuses perpetrated against the elderly.”).

350. See U.S.

DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK,

REQUEST FOR ADMINISTRATIVE RULING ON THE APPLICATION OF FINCEN’S

REGULATIONS TO A VIRTUAL CURRENCY TRADING PLATFORM (2014),

COLLABORATIVE GATEKEEPERS 881

SARs have also contributed a lot of granular information to

well-known weaknesses of the financial system, thus aiding

regulators in addressing long-standing problems with significant

social consequences. Mortgage fraud, which ran rampant in the

period before the 2007 collapse of the subprime market, has been

targeted by FinCEN intelligence gathering efforts.

351

As a result,

the Federal Housing Agency has been able to use nearly 100,000

mortgage loan fraud SARs as the basis for subsequent action.

352

Another example of a well-known regulatory effort where SARs

have made significant contributions is the fight against various

forms of insider abuse, such as insider trading, breach of

fiduciary duties, and other ways of using executive privileges for

personal advantage.

353

These examples of issues that SARs have helped address,

outside of money laundering, are a further indication of the value

of SARs as efforts to gather intelligence. At the same time,

gatekeepers are not obliged to submit SARs for all types of

financial fraud; these diverse SARs are submitted because money

laundering sometimes intersects with other types of crimes.

354

The extension of the suspicious activity reporting requirement

seems likely to draw regulators’ attention to much more

misconduct.

http://www.fincen.gov/news_room/rp/rulings/pdf/FIN-2014-R011.pdf (discussing

FinCEN’s policy and implementation responses to virtual currency issues).

351. See Press Release, U.S. Dep’t of the Treasury, Federal, State Partners

Announce Multi-Agency Crackdown Targeting Foreclosure Rescue Scams, Loan

Modification Fraud (Apr. 6, 2009), https://www.treasury.gov/press-center/press-

releases/Pages/tg83.aspx (last visited Feb. 26, 2016) (“To this end, Treasury and

FinCEN announced an advanced targeting effort already underway to combat

fraudulent loan modification schemes and coordinate ongoing efforts across

agencies to investigate fraud and assist with enforcement and prosecutions.”)

(on file with the Washington and Lee Law Review).

352. A

NNUAL REPORT 2011, supra note 329, at 39.

353. See T

RENDS, TIPS & ISSUES 2013, supra note 338, at 12–13 (describing a

rise in the reporting of insider relationships between 2003 and 2009).

354. See 31 C.F.R. §§ 1010.300–340 (2015) (identifying when financial

institutions must file a SAR).

882 73 WASH. & LEE L. REV. 797 (2016)

C. Compliance Process and Technology Behind Suspicious

Activity Reporting

How are SARs produced? The paragraphs that follow explain

that gatekeeper firms have made major investments in personnel

and technology to comply with their regulatory obligations.

Technological innovations are already allowing computers to flag

many suspicious transactions, so that firms need not rely solely

on front-line employees who might face conflicts of interest.

355

While these investments are sizeable, survey data suggests that

firms do not find these burdens impossibly heavy.

356

The

existence of this compliance infrastructure makes the extension

of the collaborative gatekeeper model to other fields more

plausible.

The resources devoted by financial institutions into staffing

their anti-money-laundering compliance programs show the

extent of private industry participation in this regulatory

effort.

357

To start with a captivating example: J.P. Morgan, the

largest U.S. bank by assets,

358

has 8,000 employees working

solely on anti-money-laundering compliance—more than the

Treasury Department and the Federal Reserve combined.

359

J.P.

Morgan has about 15,000 employees working on regulatory

compliance (including anti-money laundering) and 250,000

employees worldwide.

360

In large banks with multinational

355. See TRENDS, TIPS & ISSUES 2006, supra note 331, at 52 (“While a

computer may flag activity for review, it is the person looking at the screen who

should determine whether a series of transactions is a reportable event.”).

356. See infra notes 379–403 and accompanying text (providing data

suggesting that firms have come to terms with the regulatory obligations).

357. See R

EUTER & TRUMAN, supra note 320, at 100 (estimating the cost of

compliance as exceeding $3 billion per year).

358. See Erik Holm, Ranking the Biggest U.S. Banks: A New (Old) Entrant

in Top 5, W

ALL STREET J. (Dec. 10, 2014), www.blogs.wsj.com/moneybeat/2014/

12/10/ranking-the-biggest-u-s-banks-a-new-old-entrant-in-top-5 (last visited

Feb. 26, 2016) (noting the bank’s reported $2.5 billion dollars in assets as of late

2014) (on file with the Washington & Lee Law Review).

359. See

J.P. MORGAN, ANN. REP. 2013, at 12 (2014) http://investor.share

holder.com/common/download/download.cfm?companyid=ONE&fileid=742266&f

ilekey=2bd13119-52d2-4d78-9d85-a433141c21ae&filename=01-

2013AR_FULL_09.pdf (describing the firm’s “industry leading Anti-Money

Laundering” program).

360. See Monica Langley & Dan Fitzpatrick, Embattled J.P. Morgan Bulks

Up Oversight, W

ALL STREET J. (Sept. 12, 2013), www.wsj.com/articles/SB

COLLABORATIVE GATEKEEPERS 883

presence, anti-money-laundering operations include one large

team that concentrates information and coordinates action, and

specialized officers working on the many different lines of

business in the institution.

361

Big banks can have over eighty

different lines of business, each with a dedicated

anti-money-laundering officer.

362

A medium-sized bank with over

$100 billion in assets would typically have about 200

anti-money-laundering officials.

363

Even the smallest banks, with

up to $1 billion in assets, typically have ten or fewer

anti-money-laundering officials.

364

Many of these officials started

their careers as front-line employees and have a good

understanding of the institution’s client relationships, while

others have worked in other compliance positions or in larger

banks.

365

The expanding size of anti-money-laundering departments

has boosted financial institutions’ compliance firepower, but it is

modern technology that has really revolutionized their

monitoring philosophy. Financial institutions have developed

software that recognizes specific transaction patterns, based on

typologies sketched out on the basis of past investigations and

violations.

366

This software can be enriched and adapted over

time, “learning” more violations and modern techniques for

money laundering. Overall, banks’ newly automated systems are

capable of identifying unusual patterns in transactions by sifting

10001424127887324755104579071304170686532 (last visited Jan. 21, 2016)

(“J.P. Morgan Chase & Co., facing a host of regulatory and legal woes, plans to

spend an additional $4 billion and commit 5,000 extra employees this year to

clean up its risk and compliance problems, according to people close to the

bank.”) (on file with the Washington and Lee Law Review).

361. U.S.

DEP’T OF TREASURY FIN. CRIMES ENFORCEMENT NETWORK, REPORT

OUTREACH TO LARGE DEPOSITORY INSTITUTIONS 5 (2009) [hereinafter

OUTREACH TO LARGE DEPOSITORY INSTITUTIONS], https://www.ffiec.gov/bsa_aml_

infobase/documents/FinCEN_DOCs/FIOI_Bank_Report_Large_DIst_200910.pdf.

362. Id.

363. K

ATKOV, supra note at 332, at 6.

364. Id.

365. D

EPOSITORY INSTITUTIONS WITH ASSETS UNDER $5 BILLION, supra note

340, at 12.

366. See OUTREACH TO LARGE DEPOSITORY INSTITUTIONS, supra note 361, at

15 (“Banks build typologies gleaned from previous investigations into their

investigative strategy, creating risk models that assist the monitoring tools to

identify suspicious activity.”).

884 73 WASH. & LEE L. REV. 797 (2016)

through multiple data points in a manner that manual laborers

would find hard to imitate.

Because software typology relies on past events, it cannot

catch fraudulent transaction structures that appear for the first

time. To improve their alertness to money laundering novelties,

financial institutions have tried an alternative approach: they use

software that observes metrics of client behavior and compares

them to a “peer group” of clients that are expected to behave in

similar ways over time.

367

Peer groups vary by line of business,

client background, geography, and other factors. In addition,

most banks assess not only individual clients, but also a line of

business as a whole, in terms of susceptibility to money

laundering.

368

Financial institutions’ use of technological advances has been

one of the primary drivers of the increase in SARs, according to

finance professionals and regulators interviewed for a GAO

study.

369

That said, referrals from front-line employees continue

to contribute a significant amount of the cases that ultimately

result into a SAR.

370

In some banks, software-generated referrals

amount to 75% of total SAR candidate cases.

371

In other banks,

the picture is reversed: software contributes only 20% of their

referrals.

372

The increased use of software in anti-money-laundering

supervision has allowed financial institutions to outsource a

significant portion of their compliance heavy lifting. This can

reduce conflicts of interest significantly, by empowering many

individuals besides front-line employees to report on suspicious

activities. There are about twenty providers of

anti-money-laundering software in the world.

373

Among financial

367. See id. at 14 (explaining that “peer groups may by segmented by LOBs,

product types, geography and/or account types”).

368. See id. at 6–7 (explaining the various ways banks access suspicious

activity).

369. M

ONEY LAUNDERING: NEEDED IMPROVEMENTS, supra note 279, at 17.

370. See O

UTREACH TO LARGE DEPOSITORY INSTITUTIONS, supra note 361, at 2

(“[B]anks unanimously indicated that they believe their best source of

information . . . comes from referrals by front-line branch personnel and

relationship managers.”).

371. Id. at 16–17.

372. Id.

373. K

ATKOV, supra note at 332, at 3.

COLLABORATIVE GATEKEEPERS 885

institutions, 90% find themselves running their software in house

in collaboration with outside vendors, while 10% outsource their

software management completely.

374

Of course, developing

sophisticated software solutions and staffing populous compliance

departments do not come without a cost. According to a recent

survey, the aggregate global expenditure in anti-money-laundering

supervision in 2011 reached $5 billion per year, with $1.2 billion

spent on software and $3.8 billion devoted to staff and other

operational expenses.

375

These costs are contributing to a growing trend in

structuring compliance departments: increasing integration

between anti-money laundering and general fraud operations.

From a substantive perspective, it is becoming clear that, through

anti-money-laundering supervision, institutions receive alerts

about other types of fraud.

376

Anti-money-laundering technology

can be readily used, with just a few alterations, against financial

fraud more generally.

377

Many institutions find that modern

solutions applied in anti-money laundering deliver superior

results compared to antiquated fraud detection systems.

378

The

pressure to contain costs is particularly strong in smaller

institutions, which have started to use the same staff as both

anti-money laundering and anti-fraud compliance officers.

379

Perhaps because gatekeepers directly reap some of the

benefits of the early detection of client fraud, they have come to

terms with these significant compliance costs. Periodic surveys of

the top global banks suggest that large majorities find the

anti-money-laundering regulatory burden acceptable.

380

The

374. Id. at 12–13.

375. Id. at 4.

376. See

OUTREACH TO LARGE DEPOSITORY INSTITUTIONS, supra note 361, at

10 (discussing the connection between money laundering and financial fraud).

377. See id. at 11 (reporting regulators’ push to encourage financial

institutions to use their fraud resources to combat money laundering).

378. KATKOV, supra note at 332, at 29.

379. See

DEPOSITORY INSTITUTIONS WITH ASSETS UNDER $5 BILLION, supra

note 340, at 2, 12 (mentioning the need for staff at smaller institutions to play

multiple roles).

380. More specifically, KPMG surveyed a wide variety of professionals in the

financial industry involved in anti-money laundering in dozens of countries.

Eighty-four percent of respondents in the 2004 survey believed the regulatory

burden was acceptable, 93% of respondents in the 2007 survey believed the

886 73 WASH. & LEE L. REV. 797 (2016)

significant investments firms have made to combat money

laundering indicate that implementing the collaborative model

has proven feasible in this field. Rather than setting up

compliance systems from scratch, it seems likely that firms would

draw on their existing infrastructure if called on to give early

warning about a broader range of fraudulent activities. In short,

while the expansion of the collaborative gatekeeper model to a

broad range of crimes would undoubtedly involve significant

costs, experience with the AML regime suggests these might not

be insurmountable.

D. Filing a SAR: Efforts for Investigation and Drafting by

Front-Line Employees, Compliance Officers, and Management

To assess how effectively the “suspicious activity” threshold

helps filers distinguish between dubious and harmless

transactions, one can look at the process for filing an SAR.

Compliance officers receive information about many potentially

suspicious situations but proceed with filing in a small subset of

these cases. To draw on one available example, a small financial

institution conducted 439 investigations in 2009 but decided to

file in only thirty-nine cases. Thus, institutions seem to put

serious thought into the potential violations hidden in the

situation at hand, rather than simply filing a report even in

remotely suspicious cases, so as to avoid any regulatory

sanctions.

Investigating a potentially suspicious transaction requires

significant personnel commitment.

381

In many institutions, the

regulatory burden was either acceptable or should be increased, while 85% of

respondents in the 2011 survey found the burden acceptable. KPMG,

GLOBAL

ANTI-MONEY LAUNDERING SURVEY 2014, 7 (2014), https://www.kpmg.com/KY/en/

IssuesAndInsights/ArticlesPublications/PublishingImages/global-anti-money-laun

dering-survey-v3.pdf. This question was not repeated in the 2014 survey. Id.

While most AML professionals found the overall burden acceptable, they also

desired various reforms to the system, notably more guidance and closer

cooperation with regulators. See generally id. at 7;

KPMG, GLOBAL ANTI-MONEY

LAUNDERING SURVEY 2011: HOW BANKS ARE FACING UP TO THE CHALLENGE 9

(2011), https://www.kpmg.com/CN/en/IssuesAndInsights/ArticlesPublications/

Documents/Global-Anti-Money-Laundering-Survey-O-201109.pdf.

381. See generally OUTREACH TO LARGE DEPOSITORY INSTITUTIONS, supra note

361.

COLLABORATIVE GATEKEEPERS 887

compliance officer handling the filing is a former law enforcement

official or experienced investigator.

382

In other cases, it is

front-line employees whose investigative efforts lay the

foundation for the suspicions.

383

In one example, a bank teller

informed the anti-money-laundering officer that one client’s cash

deposits had a strong odor of pepper, often used by drug

traffickers to disorient drug-sniffing dogs.

384

To determine whether the case merits filing, the compliance

officer may present it to an oversight committee.

385

These

procedures demand significant efforts from all employees

involved. In the pepper odor case, the institution’s staff spent

about 160 hours collecting evidence and drafting the report.

386

better coordinate their filing efforts, gatekeepers use case

management software, which incorporates significant details,

timelines, and reminders.

387

At the management level, compliance officers have direct

links with a committee typically composed of independent board

members and can also present reports to the whole board. Thus,

the SAR filing process engages personnel at different levels inside

the institution who become increasingly committed to the

anti-money-laundering effort.

The effort that the institution puts in submitting a SAR often

drives it to reevaluate its relationship with the clients involved.

388

After a SAR is filed, financial institutions monitor the client

much more closely.

389

Some banks may inform clients that they

382. Id. at 17.

383. See generally id.

384. D

EPOSITORY INSTITUTIONS WITH ASSETS UNDER $5 BILLION, supra note

340, at 29.

385. See OUTREACH TO LARGE DEPOSITORY INSTITUTIONS, supra note 361, at 6

(“Governance of the program is typically conducted by key committees

established to ensure the timely escalation and consideration of issues by senior

management.”).

386. D

EPOSITORY INSTITUTIONS WITH ASSETS UNDER $5 BILLION, supra note

340, at 29.

387. See OUTREACH TO LARGE DEPOSITORY INSTITUTIONS, supra note 361, at 6

(discussing web-based case management systems).

388. See id. at 7–9 (exploring SAR-related account closure at large financial

institutions).

389. See id. at 1 (“Generally, once a bank files a second SAR on a customer’s

activity, the account is closely monitored and may be closed, depending on law

enforcement interest.”).

888 73 WASH. & LEE L. REV. 797 (2016)

are under monitoring and will follow up with educational

material on their anti-money-laundering policies, such as

brochures or letters.

390

Once a second SAR is filed, many large

financial institutions are likely to close an account as a matter of

practice and consider terminating the client relationship more

generally.

391

Even smaller financial institutions, for whom

maintaining clientele may be more pressing, did not have

difficulty deciding whether to end the relationship, apart from

few cases with highly idiosyncratic facts.

392

The procedures above suggest that, as a regulatory tool, the

suspicions threshold might be striking a fine balance. On the one

hand, it calls for a well-researched and justified report that

gatekeepers do not seem to be undertaking lightly. On the other

hand, it presents gatekeepers with a workable reporting

obligation, which they can satisfy by using specialized personnel

and technology.

E. Regulators in Collaboration

Collaborative gatekeeping places significant demands not

only on private industry, but also on the state. Will regulators

have the resources to sort through and follow up on the huge

volumes of tips gatekeepers pass along? The discussion that

follows explains that, to process the millions of SARs they receive

as part of the anti-money-laundering regime, regulators have

combined conventional regulatory methods with new approaches.

Significant regulator efforts to review and process SARs suggest

that regulators consider the reports gatekeepers provide

potentially informative.

Enforcement authorities have invested significant resources

into organizing appropriate systems for analyzing the millions of

AML suspicious activity reports per year. The Treasury

Department has established a specialized bureau, the Financial

Crimes Enforcement Network (FinCEN), which receives and

390. Id. at 9.

391. Id. at 1.

392. D

EPOSITORY INSTITUTIONS WITH ASSETS UNDER $5 BILLION, supra note

340, at 8.

COLLABORATIVE GATEKEEPERS 889

maintains financial transaction data.

393

FinCEN operates a

single database accessible to other financial regulators, criminal

authorities, and other state and local bodies investigating

criminals and other violators that may leave footprints on the

financial system.

394

In the first six months of 2014, FinCEN’s

portal received inquiries by over 350 unique agencies, including

federal, state, and local authorities, self-regulatory organizations,

and state attorney’s offices.

395

In that period, FinCEN’s database

received over 1 million inquiries.

396

No other financial regulator

offers comparable access to its information.

Apart from reactively opening its files to specific requests

related to existing investigations, FinCEN also leads efforts to

proactively scan its database in order to discover hitherto hidden

misconduct. For that purpose, FinCEN has spearheaded a

multi-agency task force that systematically reviews SARs to

determine the viability of allegations in the report itself.

397

Task

force participants include criminal authorities such as the DEA

and the FBI, regulators such as the Federal Reserve, the FDIC,

and the SEC, and other agencies such as the IRS.

398

The task

force employs over 100 different SAR review teams covering

different geographical areas across the country.

399

Teams in areas

with higher potential concentration of financial crime may be

393. See 31 U.S.C. § 310(a)–(b) (2012) (describing the structure of FinCEN).

394. See id. § 310(b)(2)(B) (explaining what the government-wide data access

service is supposed to contain).

395. SAR

STATS TECHNICAL BULL., supra note 28, at 1.

396. See Rachel Louise Ensign, Number of U.S. Suspicious Activity Reports

on the Rise, W

ALL STREET J. (July 18, 2014), http://blogs.wsj.com/riskand

compliance/2014/07/18/number-of-u-s-suspicious-activity-reports-continues-to-

rise/ (last visited Jan. 22, 2016) (“Fincen also shed some light on how these

reports are used, saying that in the first six months of 2014, a wide range of

agencies accessed the database that contains the SARs, conducting more than a

million queries.”) (on file with the Washington and Lee Law Review).

397. P

ROTIVITI, GUIDE TO U.S. ANTI-MONEY LAUNDERING REQUIREMENTS:

FREQUENTLY ASKED QUESTIONS 14–17 (2012), https://www.protiviti.com/en-

US/Documents/Resource-Guides/Guide-to-US-AML-Requirements-5thEdition-

Protiviti.pdf.

398. Id.

399. See Kevin Sullivan, The Thin Green Line: The Benefits of a SAR Review

Team, ACAMS TODAY (May 29, 2012), www.acamstoday.org/benefits-of-a-sar-

review-team (last visited Jan. 22, 2016) (noting the breadth of SAR review team

placement) (on file with the Washington and Lee Law Review).

890 73 WASH. & LEE L. REV. 797 (2016)

larger and meet more regularly than other teams. For example,

the New York SAR review team goes over 4,000 reports per

month and, after initial assessment, selects a few hundred for

deeper examination.

400

According to FinCEN, these teams can

cover over 180,000 SARs in a quarter.

401

At this rate, more than

50% of all SARs submitted can get reviewed by an enforcement

official.

402

As a rulemaker, FinCEN’s primary goal is to streamline the

process for composing and submitting SARs.

403

Its rules guide

filers in providing information about all relevant aspects of a case

in a standardized and readily researchable manner.

404

supplements its rules with extensive guidance about handling

SAR submissions. In addition, FinCEN is also providing guidance

to institutions considering how to build their compliance

departments so as to better satisfy their reporting obligations.

405

For this purpose, FinCEN has collaborated with other financial

regulators and created a manual specifying criteria and

procedures for examining an institution’s anti-money-laundering

compliance department.

406

First issued in 2005, the manual

compiles best practices from different regulators and institutions

and seeks to provide consistency in supervision.

407

400. Id.

401. SAR

STATS TECHNICAL BULL., supra note 28, at 1.

402. This calculation is based on 2014 data on submissions and reviews, and

may change if SAR submission changes.

403. See 31 U.S.C. § 310(b)(2)(D) (2012) (empowering FinCEN to administer

the reporting requirements under § 5311).

404. FinCen allows e-filing of SARs in order to streamline submissions,

enhance record keeping, and allow better and faster access to financial

information. See Mandatory E-Filing FAQs, U.S. DEP’T OF TREASURY FINCEN,

http://www.fincen.gov/forms/e-filing/Efiling_FAQs.html (last visited Feb. 26,

2016) (describing reasons for implementing a mandatory e-filing policy) (on file

with the Washington and Lee Law Review); Benefits of Using BSA E-Filing, F

IN.

CRIMES ENFORCEMENT NETWORK, http://bsaefiling.fincen.treas.gov/Why_use_

BSA_002.html (last visited Feb. 26, 2016) (listing benefits of e-filing) (on file

with the Washington and Lee Law Review).

405. See generally Bank Secrecy Act Anti-Money Laundering Examination

Manual, F

ED. FIN. INSTS. EXAMINATION COUNCIL, https://www.ffiec.gov/bsa_aml_

infobasepages_manual/OLM_002.htm (last visited Feb. 26, 2016) (on file with

the Washington and Lee Law Review).

406. See id. (describing the criteria).

407. See generally id.

COLLABORATIVE GATEKEEPERS 891

To further instill compliance, regulators also followed more

traditional approaches, such as bringing highly publicized

enforcement actions against institutions for violating their

reporting obligations. Some of these cases result from supervisory

examinations that reveal weaknesses in the institutions’

compliance system. Regulators will typically be satisfied with an

undertaking by the board of directors that it will take remedial

action, although sometimes they will require a written

commitment from the institution to that effect.

408

In other cases,

an institution’s failure to submit required reports becomes

apparent only after the underlying fraud is revealed, often to

significant losses for victims. Sanctions for failure to report have

become increasingly harsh in recent years.

409

FinCEN has

imposed fines some view as extraordinarily large.

410

And the

Department of Justice has criminally prosecuted financial

institutions for money laundering violations, putting some out of

business.

411

Since the 2007–2008 financial crisis, regulators and criminal

enforcement authorities have used anti-money-laundering law as

a platform for launching some of the most far-reaching actions

against some of the biggest banks, both domestic and

international. Anti-money-laundering violations are at the heart

of four out of the eight biggest fines levied against banks since

2000, as tallied by the Wall Street Journal,

412

including $8.9

billion against BNP Paribas for intentionally hiding transactions

with links to countries targeted by U.S. sanctions, such as Iran

and Cuba; $2.6 billion against J.P. Morgan for failing to identify

the Madoff fraud;

413

$2.6 billion against Credit Suisse for failing

408. MONEY LAUNDERING: NEEDED IMPROVEMENTS, supra note 279, at 18.

409. See Ben Protess & Jessica Silver-Greenberg, JPMorgan Is Penalized $2

Billion Over Madoff, N.Y. TIMES (Jan. 7. 2014, 9:41 AM), www.dealbook.ny

times.com/2014/01/07/jpmorgan-settles-with-federal-authorities-in-madoff-case

(last visited Mar. 29, 2016) (describing the prosecutor’s contemplation of

criminal charges for reporting failures) (on file with the Washington and Lee

Law Review).

410. See Zaring & Baylis, supra note 326, at 1414 (explaining that one

European bank was hit with a $30 million fine).

411. See id. at 1415 (highlighting fines of $43 million and $50 million).

412. Grocer, supra note 5.

413. See supra Part II.B (discussing the weaknesses of the reputational

model in the Madoff context).

892 73 WASH. & LEE L. REV. 797 (2016)

to ensure that U.S. citizens with Swiss bank accounts were not

evading taxes.

414

These cases illustrate the importance of the

anti-money-laundering framework as a readily available tool for

instilling discipline in global banking.

Perhaps the most characteristic example involves the 2012

settlement between U.S. authorities and HSBC, which included

civil penalties of $1.9 billion.

415

HSBC’s U.S. subsidiary developed

close links with HSBC’s Mexican subsidiary and became the

channel through which about $9 billion in cash and $670 billion

in wire transfers entered the United States even though they

could have been illegally acquired.

416

Internal email

correspondence provided clear indications that HSBC chose to

turn a blind eye to clients with potential drug cartel

connections.

417

It was, however, the inefficiencies in HSBC’s

anti-money-laundering compliance that became the focus of its

deferred prosecution agreement that chastised the bank’s

understaffed compliance department and its decision to treat its

Mexican subsidiary as low-risk.

418

414. See Press Release, Dep’t of Justice, Credit Suisse Pleads Guilty to

Conspiracy to Aid and Assist U.S. Taxpayers in Filing False Returns (May

19,

2014), http://www.justice.gov/opa/pr/credit-suisse-pleads-guilty-conspiracy-aid-

and-assist-us-taxpayers-filing-false-returns (last visited Feb. 26, 2016)

(explaining that the fine included “$1.8 billion to the Department of Justice for

the U.S. Treasury, $100 million to the Federal Reserve, and $715 million to the

New York State Department of Financial Services”) (on file with the

Washington and Lee Law Review).

415. See Jessica Silver-Greenberg, HSBC Agrees to Pay Nearly $2 Billion to

Settle Charges of Illegal Transfers, N.Y.

TIMES (Dec. 11, 2012, 2:17 PM),

http://dealbook.nytimes.com/2012/12/11/hsbc-to-pay-record-fine-to-settle-money-

laundering-charges (last visited Mar. 29, 2016) (“‘HSBC is being held

accountable for stunning failures of oversight—and worse—that led the bank to

permit narcotics traffickers and others to launder hundreds of millions of dollars

through HSBC subsidiaries . . . ,’ Lanny A. Breuer, the head of the Justice

Department’s criminal division, said in a statement.”) (on file with the

Washington and Lee Law Review).

416. See Deferred Prosecution Agreement, Attachment A ¶10, United States

v. HSBC Bank USA, N.A., No. 12–CR–763, (E.D.N.Y. July 1, 2013),

http://www.justice.gov/opa/documents/hsbc/dpa-attachment-a.pdf (No. 12–CR–

763) (detailing “four significant failures in HSBC USA’s AML program”).

417. See id. ¶ 39 (describing how HSBC learned of the money-laundering

scheme from employee emails).

418. See id. ¶ 19 (“[F]rom at least 2006 to 2009, HSBC Bank USA rated

Mexico as standard risk, its lowest AML risk category. As a result, wire

transfers originating from Mexico, including transactions from HSBC Mexico,

COLLABORATIVE GATEKEEPERS 893

Legislation that put in place the modern

anti-money-laundering regime raised two significant questions.

First, would banks cooperate with the regime, reporting

suspicious transactions to regulators, or would they try to turn

their eyes away from potential crimes to shield their clients?

Second, would regulators be able to process the information

banks provided and succeed in using it to convict money

launderers? It is impossible to tell exactly how well this regime

has worked, because individual suspicious activity reports remain

confidential.

419

But the high volume of SARs’ gatekeepers

turnover, regulators’ assessments that many of these reports are

revealing, and regulators’ eagerness to access SARs suggest that

money-laundering legislation is no paper tiger and might

significantly dent criminal activity. Anti-money-laundering

efforts offer the one area where the collaborative gatekeeper

model has been put into practice, with at least some success. That

said, is the area of money laundering highly unusual, or could

this model of collaborative gatekeeping be extended to other field?

The next Part turns to this question.

VI. Applying the Collaborative Model in Other Areas

The collaborative gatekeeping template was implemented

with some success in anti-money-laundering law, growing

stronger through the extensive use of information technology in

recent years, as the previous two Parts have chronicled. This Part

explores whether and how collaborative gatekeeping can serve as

a model for other issue areas and identifies features that

facilitate its operation as well as conditions that might constrain

its effectiveness.

In its simplest form, as an obligation to “know your

customer” and report suspicions, the collaborative model could

easily work across many different gatekeeping relationships.

420

were generally not reviewed in the CAMP system.”).

419. See 12 C.F.R. § 563.180(d)(12) (2011) (stating that SARs are

confidential and creating exceptions for law enforcement, filing joint SARs, and

certain employment references or termination notices).

420. See supra Part III (outlining the elements of the collaborative

gatekeeper model).

894 73 WASH. & LEE L. REV. 797 (2016)

various segments of the financial industry, gatekeepers already

conduct some due diligence toward their clients, either because

law requires it or because they need to preserve their

reputations.

421

In the context of due diligence, and their ongoing

client relationships, most gatekeepers can come across potentially

compromising information about their clients.

422

One could easily

imagine requiring gatekeepers to report this information. But

even though the model’s building blocks can be readily

transplanted to other areas, they might not be as successful in

catching misconduct.

The collaborative model works best in areas where financial

fraud follows well-trodden paths. Suspicions can easily arise

when a gatekeeper can compare a client’s conduct with the

actions of other clients. To evaluate whether a client behaves a lot

like past fraudsters, or simply deviates unjustifiably from current

norms, a gatekeeper’s point of reference is necessarily the activity

of others. In turn, comparisons of transactions are easier in fields

that are relatively homogenous. The more standardized the

transactions a gatekeeper sees, the easier it is to identify the

fraudulent ones. Moreover, relatively standardized violation

types are more straightforward to investigate, as front-line

gatekeeper employees can be instructed to ask specific questions

or look for certain indications.

This backwards-looking perspective of the collaborative

model constitutes an important limitation, both conceptual and

practical. No doubt, in referring to past misconduct in order to

identify the future, one might risk leaving innovative types of

fraud off the hook. In many core areas of the financial system,

like securities trading, standardization is quite advanced. Other

areas of the financial system, like derivatives, are becoming more

standardized. Moreover, as the number of financial transactions

continues to grow, the data points that gatekeepers have

available for their comparisons also increase. In these fields,

421. See Cunningham, supra note 66, at 329 (noting that some gatekeepers,

such as auditors and lawyers, have duties arising “initially from contract but

include a regulatory overlay of professional standards” and that “both put

reputations and liability on the line”).

422. See id. at 328 (explaining that some gatekeepers work “directly with

and essentially inside the enterprise”).

COLLABORATIVE GATEKEEPERS 895

there is a lot of currently uncollected information that could

significantly improve enforcement.

In addition, the collaborative gatekeeper model works best in

fields where gatekeepers control an essential component of our

financial infrastructure. The payment system, which modern

banks control with their deposit and wire transfer services, is a

key channel for introducing new money into the financial

universe. As a result, banks are well placed to collect information

that might point to illegal funds. Other components of essential

financial infrastructure include various stock and commodity

exchanges, central securities depositories, and central

counterparties in derivatives. Most of these infrastructures allow

access only to certain finance professionals under strict licensing

requirements, typically imposed through regulatory supervision.

Thus, these finance professionals can gather important

information about types of fraud occurring through their services.

At the opposite end of the spectrum, transactions arranged

privately and tailored to the needs of specific clients would be

harder for gatekeepers to decode. Securitizations, collateralized

debt obligations, or acquisition financing depend a lot on parties’

expectations about the future, which might diverge substantially.

Moreover, these transactions tend to have many moving parts,

combining securities issuance, derivatives, collateralization, and

corporate governance arrangements. As a result, gatekeepers in

these transactions might have greater trouble identifying a

party’s conduct as suspicious, especially at an early stage.

Yet, even in settings of high transaction complexity,

collaborative gatekeeping might prove helpful in one respect.

Typically, complicated transactions tend to include multiple

gatekeepers with different specializations: bankers, accountants,

external auditors, and other technical experts. Some specialists

get only a partial look at the transaction. Thus, although they

may not be able to ascertain fully whether a transaction is

fraudulent, specialists may become suspicious of how the

malfunctions they identify might affect the other parts of the

transaction. These suspicions could form a basis for a report that

might prove helpful for regulators, especially if multiple

gatekeepers flag the same transaction.

To illustrate how these scope conditions of collaborative

gatekeeping would work in the context of specific subject matters

896 73 WASH. & LEE L. REV. 797 (2016)

in financial regulation, the remaining sections of this Part discuss

two possible extensions of the model. The first extension concerns

the regulation of broker-dealers, responsible for trading in

securities. Broker-dealers regulation exemplifies an area of high

transaction traffic, advanced standardization, and gatekeeper

control over an essential market entry point. This extension is

theoretically straightforward and is easy to conceptualize in

practice through the lens of a real life example that hit national

headlines: J.P. Morgan’s indictment for its role as Bernie

Madoff’s chief banker and broker. The second extension examines

how the model might apply, after some adjustments, in a field

involving varied and complicated transactions: equity offerings.

A. Collaborative Gatekeeping in Broker-Dealer Regulation

Just like banks are the essential administrators of our

payment system, brokers are the key operators of our securities

trading venues.

423

Broker-dealers are the only professionals

licensed to access stock exchanges, so investors need to hire them

in order to trade.

424

Moreover, brokers enjoy significant

regulatory privileges, such as the ability to vouch for an investor

so as to exempt transactions from registration requirements.

425

Investors seeking to buy or sell securities through these venues

and private offerings typically engage a broker to act on their

behalf. As a result, broker-dealers have significant information on

the flow of funds in and out of key segments of the financial

system. The nature of this information is very similar to

423. See Trade Executions: What Every Investor Should Know, SEC,

https://www.sec.gov/investor/pubs/tradexec.htm (last visited Feb. 26, 2016)

(explaining that brokers act as an intermediary between the investor and the

stock market) (on file with the Washington and Lee Law Review).

424. See Guide to Broker-Dealer Registration, SEC (Apr. 2008),

https://www.sec.gov/divisions/marketreg/bdguide.htm (last visited Feb. 26, 2016)

(“Most ‘brokers’ and ‘dealers’ must register with the SEC and join a

‘self-regulatory organization,’ or SRO.”) (on file with the Washington and Lee

Law Review).

425. See Securities Act of 1933, 15 U.S.C. § 77d (2012) (exempting many

private placements from registration requirements); Securities Act of 1933,

Regulation D, 17 C.F.R. §§ 230.501–230.508 (2015) (exempting certain primary

securities offerings from federal registration); id. § 230.144A (exempting certain

secondary market transactions from registration requirements).

COLLABORATIVE GATEKEEPERS 897

information that banks collect regarding the flow of funds into

the banking sector when customers deposit funds, make

payments, or cash checks.

Another similarity between banks and broker-dealers is the

importance of client networks. To scout the market for investor

interest in the securities they trade, brokers have established

multiple outposts around the country,

426

instituted online

services that are increasingly popular,

427

and established

branches and subsidiaries across borders.

428

Moreover, brokers

seek to build long-standing relationships with their clients so as

to handle their securities portfolio over time.

429

As a result,

broker-dealers also have information on clients’ holdings,

backgrounds, and trading patterns.

430

Brokers’ unique position at the crossroads of securities

trading and their expansive client networks allow them to gather

intelligence about violations for which securities trading is the

primary medium. For example, insider trading involves sales or

purchases by people who possess non-public information about a

security, sometimes due to their relationship with the security’s

issuer.

431

Brokers possess information both about trades and

426. See, e.g., J.P. Morgan Letter, supra note 1, at 81 (noting the presence of

5,602 branches serving over 36 million customers as of 2014).

427. See, e.g., CHARLES SCHWAB CO., ANNUAL REPORT 2014, at 13, 55 (2015),

http://aboutschwab.com/images/uploads/inline/Schwab_2014_Annual_Report_co

mplete.pdf (noting that over 10 million clients were served with only 325

domestic branch offices due to a large online presence).

428. See Michael Konczal, A Wall Street regulator’s race against time, W

ASH.

POST WONKBLOG (June 22, 2013), https://www.washingtonpost.com/news/wonk/

wp/2013/06/22/a-wall-street-regulators-race-against-time/ (last visited Jan. 16,

2016) (explaining that the financial crises of “AIG, Lehman Brothers, Citigroup

off-balance sheet SIVs, Bear Stearns, Long-Term Capital Management, and the

‘London Whale’ of JP Morgan” all involved exposures to derivatives across

multiple countries) (on file with the Washington and Lee Law Review).

429. See, e.g., Tap Our Resources, BCI F

IN., http://www.broker

ageconsultants.com/jobdetails.php?job_id=1622 (last visited Mar. 29, 2016)

(explaining in a job posting that its firm hopes to attract new clients to

eventually form long-standing client relationships) (on file with the Washington

and Lee Law Review).

430. Id.

431. See United States v. O’Hagan, 521 U.S. 642, 651–52 (1997) (stating

that “[u]nder the ‘traditional’ or ‘classical theory’ of insider trading liability,

§ 10(b) and Rule 10b–5 are violated when a corporate insider trades in the

securities of his corporation on the basis of material, nonpublic information”).

See generally Donald C. Langevoort, “Fine Distinctions” in the Contemporary

898 73 WASH. & LEE L. REV. 797 (2016)

about certain connections between issuers and their key

executives, such as employment or marital relationships.

432

Another example of a fraudulent scheme that relies heavily on

trading is market manipulation, whereby clients seek to inflate

artificially interest in their securities.

433

Again, brokers may be

able to see through clients’ schemes based on the orders they are

called on to execute.

Besides misconduct that centers on trading, broker-dealers’

information could also help illuminate instances of fraud that

happen to occur through their distribution channels. For

example, brokers might buy and sell securities for private hedge

funds, and thus may develop a sense of odd patterns of trading

that may foreshadow the hedge fund operator’s disappearance

into a remote tax haven with the investors’ money. In another

example, a broker who is intermediating a transaction between

two parties might come to realize that one party’s representations

might be duplicitous, and thus potentially misleading.

434

While these indications might be readily apparent or easily

available to brokers, they have little motivation to collect or use

them. Under current law, broker-dealers are not under a general

Law of Insider Trading, 2013 COLUM. BUS. L. REV. 429 (2013); Yesha Yadav,

Insider Trading in Derivative Markets, 103 G

EO. L. J. 381 (2015).

432. See, e.g., O’Hagan, 521 U.S. at 647–49 (explaining how the defendant

used knowledge of a planned tender offer from the company he represented for

his own benefit).

433. See Market Manipulation Fraud, FBI, https://www.fbi.gov/about-

us/investigate/white_collar/market-manipulation-fraud (last visited Mar. 29,

2016) (defining market manipulation as “artificially raising or lowering the price

of stock on any national securities or commodities exchange or in the

over-the-counter (OTC) marketplace”) (on file with the Washington and Lee Law

Review).

434. See Press Release, SEC Charges Goldman Sachs with Fraud in

Structuring and Marketing of CDO Tied to Subprime Mortgages, SEC (Apr. 16,

2010), http://www.sec.gov/news/press/2010/2010-59.htm (last visited Jan. 16,

2016) (explaining that the SEC assessed a $550 million dollar penalty because

“Goldman wrongly permitted a client that was betting against the mortgage

market to heavily influence which mortgage securities to include in an

investment portfolio, while telling other investors that the securities were

selected by an independent, objective third party”) (on file with the Washington

and Lee Law Review); see also Steven M. Davidoff et al., The SEC v. Goldman

Sachs: Reputation, Trust, and Fiduciary Duties in Investment Banking, 37 J.

CORP. L. 529, 530–33 (2012) (discussing this settlement).

COLLABORATIVE GATEKEEPERS 899

obligation to alert regulators about potential fraud.

435

The

general anti-fraud provisions of federal securities laws provide

victims with private rights of action only if brokers were primary

participants in fraud

436

and acted with scienter, i.e., intentionally

or knowingly. Brokers have various obligations of a fiduciary

nature toward their clients, such as to obtain the best execution

for their orders,

437

to avoid trading ahead of clients,

438

and to

disclose their commissions.

439

Most of these obligations aim to

protect clients from brokers’ own overreaching, rather than from

third-party fraud. Thus, information that could be useful in

prosecuting fraud and other misconduct typically remains

scattered and unearthed, out of the reach of regulators.

Collecting this information will probably require some

additional effort both by broker-dealers and by regulators, but the

institutional preconditions for launching this effort are already in

place. Brokers are subject to registration with the Securities and

Exchange Commission, which specifies their licensing

requirements.

440

Just as federal banking regulators oversee the

application of the anti-money-laundering regime, the SEC could

oversee the expansion of suspicious activity reporting obligations

in fraud and misconduct. Broker-dealers already have in place

well-staffed compliance departments and supervisory procedures,

435. See Stoneridge Inv. Partners, LLC v. Scientific-Atlanta, Inc., 552 U.S.

148, 159 (2008) (finding the defendants had no duty to disclose).

436. See id. at 155–56 (affirming that private plaintiffs cannot use § 10(b) of

the 1934 Act to hold brokers liable for aiding and abetting fraud); see also Cent.

Bank of Denver, N.A. v. First Interstate Bank of Denver, N.A., 511 U.S. 164,

191 (1994) (“[W]e hold that a private plaintiff may not maintain an aiding and

abetting suit under § 10(b).”).

437. FINRA Rule 2320(a) (2009). See FINRA Regulatory Notice 09-58, SEC

Approves Amendments Regarding Best Execution and Interpositioning (SEC

Oct. 9, 2009), 2009 WL 3320542 (applying the best execution obligation to all

customer orders, including those involving interposed third parties).

438. See Investment Adviser Codes of Ethics, Investment Company Act, 17

C.F.R. § 275.204A-1 (2005) (adopting a code of an investment-advisor code of

ethics).

439. See 17 C.F.R. § 240.10b–10 (2014) (requiring brokers and dealers

effecting transactions in securities to disclose commissions in writing).

440. 15 U.S.C. §78o(d) (2012). See also Donald C. Langevoort & Robert B.

Thompson, ‘‘Publicness” in Contemporary Securities Regulation After the JOBS

Act, 101 G

EO. L.J. 337, 351–54 (2013) (outlining the different gateways through

which a company can become public, and the attendant requirements).

900 73 WASH. & LEE L. REV. 797 (2016)

because they need to oversee employees’ handling of client

funds.

441

Moreover, broker-dealers are already participating in

the anti-money-laundering regime described above, and thus

have experience with suspicious activity reporting and the

infrastructure necessary to satisfy this obligation.

442

Some

broker-dealers are already using technology developed in the

anti-money-laundering context to keep track of employee fraud.

443

By building on their existing capabilities, brokers could easily

enhance the scope of violations that are subject to greater

scrutiny.

A proposal for transposing a regulatory regime from one

subject matter to another might justifiably generate some

hesitation, as it brings a whole industry into uncharted territory.

To better understand how suspicious activity reporting might

work in brokers’ violations, we next turn to a case that, although

brought under the anti-money-laundering laws, involves fraud

that would be typically associated with trading, not laundering.

444

As discussed above, some suspicious activity that is associated

with money laundering can also lead authorities to other

violations.

445

As a result, there is an overlap between the

anti-money-laundering regime as it currently stands and the

proposed extension of suspicious activity reporting to brokers.

446

By analyzing a case that happens to fall within this small area of

overlap, we can get a fairly good glimpse of how the suggested

441. See Gadinis, supra note 77, at 714–22 (discussing the supervision

obligation using empirical data from recent SEC investigations of large and

small firms for failure to supervise).

442. See David W. Blass, Broker-Dealer Anti-Money Laundering

Compliance—Learning Lessons from the Past and Looking to the Future, SEC

(Feb. 29, 2012), https://www.sec.gov/News/Speech/Detail/Speech/1365171489982

(last visited Feb. 26, 2016) (“[W]ithin a few years of the President signing the

PATRIOT Act into law, broker-dealers went from having basic reporting and

recordkeeping obligations to a robust and complimentary set of obligations

aimed at detecting and deterring money laundering and criminal financing.”)

(on file with the Washington and Lee Law Review).

443. See supra Part V.B (discussing the use of suspicious activity reports).

444. See supra note 3 and accompanying text (introducing the J.P. Morgan

and Bernie Madoff case).

445. See supra Part V.E (concluding that the collaborative model has found

success in the anti-money laundering context).

446. See supra Part V (discussing the current state of the anti-laundering

regime in depth).

COLLABORATIVE GATEKEEPERS 901

extension might operate in practice. Because such cases would

only represent the tip of the iceberg, we can gauge whether a

large mass of undetected fraud might lie beneath the surface.

In 2014, J.P. Morgan paid $2.6 billion in fines to various U.S.

regulators for its role as Bernie Madoff’s primary broker and

bank.

447

The thrust of regulators’ case relied on J.P. Morgan’s

failure to report any suspicions to authorities about the Madoff

Ponzi scheme under the Bank Secrecy Act. J.P. Morgan, one of

the largest U.S. financial institutions, holds both broker-dealer

and banking licenses.

448

Since 1986, J.P. Morgan maintained a

banking relationship with various entities in the Madoff group

and was in charge of the accounts through which money was

directed in and out of Madoff’s Ponzi scheme.

449

As a broker, J.P.

Morgan intermediated on behalf of clients who sought to invest

into Madoff’s funds and developed derivatives based on Madoff

fund returns, selling some to clients and maintaining a

significant portion for itself.

450

In the context of this relationship,

J.P. Morgan came across indications that Madoff’s operations

were fraudulent, as the paragraphs below explain.

451

Although

J.P. Morgan’s compliance systems managed to spot these

indications, the investment bank failed to follow through, conduct

appropriate due diligence, and file the suspicious activity reports

necessary to alert regulators.

452

Thus, the case offers a good

illustration of the indications that brokers may be able to gather,

the actions that they should take to pass on these tips, as well as

the efforts of regulators to impose sanctions that reinforce filers’

regulatory obligations.

The most important red flags that J.P. Morgan failed to

evaluate properly were the consistently high abnormal returns

447. See supra notes 59–60 and accompanying text (discussing J.P. Morgan’s

penalties).

448. See Commercial Banking Disclaimer, J.P. MORGAN, https://www.jp

morgan.com/global/jpmorgan/cb/commercialbankingdisclaimer (last visited Feb.

26, 2016) (explaining the relationship with J.P. Morgan affiliates, which

includes both banking and brokerage) (on file with the Washington and Lee Law

Review).

449. DPA Statement of Facts, supra note 3, ¶¶ 9–10.

450. Id. ¶¶ 33–36.

451. See generally id.

452. See, e.g., id. ¶¶ 70–78 (noting J.P. Morgan’s failure to file a SAR).

902 73 WASH. & LEE L. REV. 797 (2016)

that Madoff funds purported to generate.

453

Individual J.P.

Morgan analysts expressed surprise at these profits, which had

the tendency to arise even under adverse market conditions, and

concluded that they were “possibly too good to be true.”

454

Due to

its role as manager of Madoff’s accounts, J.P. Morgan had already

realized that he was engaging in transactions with no apparent

or very limited economic purpose.

455

The bank received alerts

from its software about unusual third-party wire activity and

Treasury bond redemptions at least twice.

456

Moreover, J.P.

Morgan saw Madoff transferring tens of millions of dollars daily

to another institution’s account, only to have them return shortly

thereafter.

457

None of these observations could render J.P.

Morgan “aware” of the Madoff fraud, as they could also result

from legal activity.

458

Nevertheless, they were sufficient to raise

suspicions of illegality and could have led enforcement authorities

to Madoff earlier.

459

These red flags prompted J.P. Morgan’s employees to start

due diligence, only to be met with Madoff’s unwillingness to

cooperate.

460

Yet, these concerns were never communicated to

J.P. Morgan’s anti-money-laundering department, and no

suspicious activity reports were filed ahead of Madoff’s confession

and arrest.

461

Even after Barron’s, a well-respected business

magazine, published an article about potential fraud at Madoff,

compliance officers disregarded it, thinking that U.S. regulators

would have already examined these concerns.

462

This lack of

453. Id. ¶ 29.

454. Id. ¶ 32.

455. Id. ¶ 25.

456. Id. ¶ 21.

457. Id. ¶ 24.

458. See id. ¶ 24 (quoting a J.P. Morgan employee as concluding he was just

using the money “as a float”).

459. Id. ¶ 25.

460. Id. ¶ 31.

461. Id. ¶¶ 70–79.

462. See Erin E. Arvedlund, Don’t Ask, Don’t Tell: Bernie Madoff Attracts

Skeptics in 2001, B

ARRON’S (May 7, 2001), www.online.barrons.com/

article/SB989019667829349012.html (last visited Feb. 26, 2016) (mentioning

that “[t]hree option strategists for major investment banks told Barron’s they

[could not] understand how Madoff churns out such numbers [using his

strategy]”) (on file with the Washington and Lee Law Review).

COLLABORATIVE GATEKEEPERS 903

appropriate response suggests weaknesses at J.P. Morgan’s

compliance systems, as the bank itself has admitted. One could

similarly chastise U.S. regulators for failing to take action on

Madoff earlier despite urgings by the Press.

463

Essentially,

gatekeepers and regulators were locked into mutual inactivity,

with each party waiting for the other to act, and interpreting the

other’s silence as tacit approval.

464

Yet, if all relevant pieces of

information were combined in a central system, perhaps this

uniform inactivity would have broken down sooner.

While J.P. Morgan’s U.S. operations continued their

relationship with Madoff despite his unresponsiveness and lack of

transparency, its U.K. subsidiary grew increasingly wary.

465

the United Kingdom, J.P. Morgan’s relationship with Madoff was

centered on brokerage services.

466

The bank’s London team had

underwritten approximately $1.14 billion in investments into

Madoff’s funds, including $343 million of J.P. Morgan’s own

money.

467

By June 2007, the U.K. subsidiary’s management

called for additional due diligence on Madoff’s investment

strategy.

468

Although Madoff agreed to answer questions, he

refused to allow full due diligence of his entities.

469

When U.K.

executives tried to analyze Madoff’s strategy, they also failed to

explain the returns he was presenting.

470

As a result, around

September 2008, they decided to unwind their own positions in

463. See id. (noting the unusual amount of secrecy around Madoff’s

strategy).

464. See DPA Statement of Facts,

supra note 449, ¶ 55 (“‘[T]hey were

assured by the claim that FINRA and the SEC performed occasional audits of

Madoff,’ but that they ‘appeared not to have seen any evidence of the reviews or

finds.’”).

465. Id. ¶¶ 54–63.

466. See id. ¶ 58 (explaining that in 2008, Madoff Securities was the main

subject suspect in the United Kingdom).

467. Id. ¶ 32.

468. The U.K. subsidiary had put together a Hedge Fund Underwriting

Committee, who had to approve continuation of this underwriting due to its size.

Id. ¶ 39. This committee discussed the Madoff case on June 15, 2007. Id. ¶ 40.

Present at the meeting were a number of high-ranking executives, risk officers,

and employees dealing in Madoff-related transactions. Id. The written materials

presented to the committee regarded systemic fraud as “extremely unlikely,” yet

the committee decided to investigate Madoff further. Id. ¶ 42.

469. Id. ¶ 43.

470. Id. ¶ 51.

904 73 WASH. & LEE L. REV. 797 (2016)

Madoff.

471

A U.K. analyst explained the misgivings associated

with Madoff in a lengthy memorandum on October 16, 2008,

detailing Madoff’s inability to confirm assets supposedly held in

custody and wondering about the “‘odd choice’ of a small,

unknown accounting firm.”

472

Interestingly, the memorandum

also referred to casual observations of J.P. Morgan front-line

employees in their dealings with Madoff entities: they described

Madoff’s personnel as “defensive and almost scared of Madoff,”

alluding to an atmosphere where “no one dares to ask any serious

questions as long as the performance is good.”

473

Later in October

2008, J.P. Morgan’s U.K. entities filed a suspicious activity report

with the U.K. Serious Organized Crime Agency under the

Proceeds of Fraud Act, reflecting the concerns outlined in the

internal memorandum.

474

However, they never communicated

their report to their U.S. colleagues.

475

J.P. Morgan’s failure in reporting the Madoff fraud provides

a clear illustration of how collaborative gatekeeping could work in

practice. J.P. Morgan had strong suspicions, but it had also

reached the end of its ability to get to the bottom of the problem,

in light of Madoff’s uncooperative stance. If it were required to

report these suspicions and enlist the help of regulators, it could

have precipitated the uncovering of Madoff’s fraud and protected

some investors from falling prey to his deceits. Admittedly, the

SEC had received warnings about Madoff from a disgruntled

former competitor, Harry Markopolos.

476

But Markopolos had no

inside information about Madoff, and many reasons to vilify

him.

477

J.P. Morgan’s submissions would probably have carried a

471. Id. ¶ 53.

472. Id. ¶ 55.

473. Id. ¶ 56.

474. Id. ¶ 58.

475. Id. ¶ 65.

476. See Madoff Whistleblower: SEC Failed to Do the Math, NPR (Mar. 2,

2010, 12:00 AM), http://www.npr.org/templates/story/story.php?storyId=

124208012 (last visited Mar. 29, 2016) (“Madoff didn’t leave any footprints in

the market because he never traded stock, Markopolos explains: ‘It was all made

up and his story was so fanciful and far-fetched that the SEC should have seen

through it immediately. And they didn't.’”) (on file with the Washington and Lee

Law Review).

477. See Andrew Clark, The Man Who Blew the Whistle on Bernard Madoff,

HE GUARDIAN (Mar. 24, 2010, 8:17 AM), http://www.theguardian.com/

COLLABORATIVE GATEKEEPERS 905

different weight in the eyes of the regulator, as they would have

pitted a close collaborator and Wall Street stalwart against the

secretive Ponzi-schemer. J.P. Morgan’s alert would have been

harder to ignore.

More generally, the Madoff example shows the type of

information that brokers can collect regarding their clients.

Trading flows and account movements can be a real data mine for

enforcers looking for indications of fraud, providing brokers with

an unparalleled overview of the financial system. The high

numbers of clients and relatively homogeneous transactions that

a broker supports facilitate comparisons and can readily uncover

odd patterns, such as Madoff’s unrealistic returns. While

regulators sit behind their desks, at a distance from market

developments, brokers are participants in, and entryways into,

the action in the market. Their reports could be a major boost to

enforcement efforts.

B. Collaborative Gatekeeping for Accountants on Equity Issuance

The collaborative model applies most straightforwardly in

fields that represent a good match for its prerequisites, like

broker-dealer regulation, but it also has a lot to offer in issue

areas where conditions might not be as favorable. This Part

focuses on one such area: accountants in equity issuance. Acting

as external auditors to companies issuing and selling securities to

the public, accounting firms are called upon to verify the accuracy

of the issuer’s financial statements. This verification is a

necessary precondition for a company seeking to enter the public

markets, for example, by listing shares on a stock exchange.

Thus, accounting firms play a decisive role in determining access

to a central component of our financial infrastructure. However,

key features of the issuer-auditor relationship, and the regulatory

regime established to govern it, render this a hard case for the

collaborative model.

business/2010/mar/24/bernard-madoff-whistleblower-harry-markopolos (last

visited Mar. 29, 2016) (describing Markopolos as “a belatedly celebrated

whistleblower who was ignored by everybody”) (on file with the Washington and

Lee Law Review).

906 73 WASH. & LEE L. REV. 797 (2016)

Compared to a broker simply executing a client’s orders, the

scope and intensity of accountants’ services is much wider. To

perform an audit of a company’s financial statements,

accountants must spend significant time in learning about the

company, request and review extensive information, and meet

repeatedly with management and key employees. Because each

company has some unique features, comparisons are harder.

Moreover, all the effort and resources accountants must devote

toward an audit suggest that they have much to lose by reporting

their clients to enforcement authorities. As accountants tend to

cultivate client relationships over many years, the conflicts of

interest they face may be particularly strong.

A web of regulation requires accountants, before they finalize

their audit, to clarify any inhibitions they may harbor about the

company’s accounting, or else record them. If they fail to do so,

accountants may be liable toward investors. For example, under

§ 11 of the sSecurities Act of 1933, if a company’s financial

statements are inaccurate or misleading, investors have a claim

for damages against auditors.

478

To best protect themselves from

investor lawsuits, accountants who come across indications of

illegality should make further inquiries and obtain clarifications

from management. Fearing that management might be reluctant

to provide additional information, the Private Securities

Litigation Reform Act of 1995 put in place a framework designed

to strengthen accountants’ bargaining position against their

clients. Section 10A of the Exchange Act

479

requires accountants,

once they come across potentially illegal activity, to notify

management (for example, the CFO) and request that remedial

action be taken.

480

If management fails to satisfy the accountants,

they must submit a formal report to the board, which is also

required to notify the SEC within one day.

481

And if the board

fails to notify the SEC, auditors must resign.

482

Sarbanes–Oxley further enhanced auditors’ exposure to

liability, requiring them to verify not only the accuracy of specific

478. 15 U.S.C. § 77k (2012).

479. Id. § 78j-1.

480. Id. § 78j-1(b)(1).

481. Id. § 78j-1(b)(2)–(3).

482. Id. § 78j-1(b)(4).

COLLABORATIVE GATEKEEPERS 907

statements, but also the adequacy of the procedures that the

company follows to gather the data necessary to draft the

statements.

483

If, in examining these procedures, auditors spot

weaknesses that the company refuses to address, they can

publicly disclose their views to investors.

484

Moreover, for any

issues they identify with regard to management’s handling of

financial statements, external auditors have access to an audit

committee composed entirely of independent directors.

485

At first glance, this dense framework for the regulation of

accountants, anchored in § 10A’s requirement to explore any

indications of illegality, shares many features with the

collaborative gatekeeping proposal we advance in this Article.

However, there is a key difference. Section 10A is designed to

provide the company with an opportunity to avert regulatory

intervention by nipping a complaint in the bud.

486

Notifying

management is the first step that accountants must take in order

to clarify their suspicions, and notifying the board is the second.

These steps ensure that the relationship between accountants

and their clients remains as strong as ever. In reality, § 10A

mostly codifies pre-existing standards of professional conduct and

passed without any objection by the accounting industry.

487

Despite its rhetoric, § 10A leaves accountants with the same

dilemma that gatekeepers typically face. Their first option is to

launch an eponymous attack on management practices, based on

whatever evidence they can gather on their own, to avoid alerting

their client and risk obstruction. The second option is to turn a

blind eye, hoping that fraud will go undetected, or, if revealed,

they will avoid liability. As we explain below, the second option

might seem more appealing in many circumstances.

483. 15 U.S.C. § 7262 (2012); id. § 7213; see also Auditing Standard No. 5,

PCAOB, http://pcaobus.org/Standards/Auditing/Pages/Auditing_Standard_5

.aspx (last visited Apr. 2, 2016) (establishing auditing standards).

484. 15 U.S.C. § 7262; see also Auditing Standard No. 5, supra note 483

(requiring auditors to report on the company’s internal controls).

485. 15 U.S.C. § 78j-1.

486. Gary DiBianco & Andrew M. Lawrence, Investigation and Reporting

Obligations Under Section 10A of the Securities Exchange Act, 40 R

EV. SEC.

COMM. REG. 25, 31 (2007).

487. See Richard W. Painter, Lawyers’ Rules, Auditors’ Rules, and the

Psychology of Concealment, 84 MINN. L. REV. 1399, 1412 (2000) (explaining the

importance of § 10A in auditor defense arguments).

908 73 WASH. & LEE L. REV. 797 (2016)

If accountants decide to pick a fight with management, they

run the risk of losing their client and hurting their professional

reputation. Whether they publicly air their disagreements with

the company’s financial statements, or they simply raise their

concerns with independent directors, they must be able to back

up their fears with some evidence. Even when accountants spot

“red flags,” such as mishandled transactions, mislabeled

accounts, or potentially distorting diversions of funds, they need

further information to determine whether these odd practices

hide a real problem. On their own, accountants do not have the

legal tools to dig deeper against express management wishes, or

simply evasiveness. In practice, the SEC has received fewer 10A

reports than hoped when the provision was passed.

488

This low

reporting rate might reflect that companies actually respond to

the problems pointed out by accountants. But it might also mean

that it is in accountants’ interest to avoid flagging problems in

the first place.

489

In any case, even though accountants might

have indications of potential misconduct, the current regime’s

notification requirements do not really encourage them to come

forward.

By not acting on their suspicions, accountants are left with

the second option and may expose themselves to liability toward

investors. However, they might find hope in that the relevant

liability rules provide them with significant leeway.

490

With

regard to financial statements provided in a public offering of

securities, § 11(b) allows accountants to claim a due diligence

defense if they conduct a reasonable investigation into the

company’s practices.

491

If a company’s financial statements follow

a reasonable interpretation of U.S. GAAP, behind which

accountants can stand in good faith, then they have satisfied

488. According to a 2003 GAO Report, a total of twenty-nine reports had

been submitted in compliance with § 10A between Jan. 1, 1996 and May 15,

2003. See generally U.S.

GOV’T ACCOUNTABILITY OFF., SECURITIES EXCHANGE ACT:

REVIEW OF REPORTING UNDER SECTION 10A (Sept. 3, 2003),

http://www.gao.gov/assets/ 100/92154.pdf.

489. John C. Coffee, The Attorney as Gatekeeper: An Agenda for the SEC,

103 COLUM. L. REV. 1293, 1307 (2003).

490. See, e.g., Securities Act of 1933, 15 U.S.C. § 77k (2012) (setting out

liabilities and exemptions to liabilities).

491. See Escott v. BarChris Constr. Co., 283 F. Supp. 643, 682–83 (S.D.N.Y.

1968) (establishing the contours of the due diligence defense).

COLLABORATIVE GATEKEEPERS 909

their due diligence defense.

492

As far as accountants are able to

show that they submitted their concerns to management and

received a somewhat satisfactory response, they should be off the

hook. Accountants’ risk of liability is even lower in cases where

investors cannot bring a § 11 claim and have to rely on Rule 10b–

5, for example because the misrepresentation occurs in financial

statements subsequent to a public offering. To satisfy the

requirements of Rule 10b–5, plaintiffs must show that

accountants provided misleading information in scienter, i.e.,

that they were essentially aware of the problem in the financial

statements.

493

Thus, as long as accountants can validly claim

that, based on the information available to them following their

inquiries, they did not have knowledge of clients’ misconduct,

they should be able to avoid liability.

Although such information may not prove misconduct, it

could provide a promising start or a helpful boost to

investigations. For example, in In re WorldCom, a $7 billion

fraudulent scheme begun to unravel when an internal accountant

discovered that $400 million held on provision for potential losses

were reclassified as capital expenditures to increase the

company’s income.

494

Although not improper on its face, this

accounting practice was unusual. As long as this information does

not render accountants aware of the problem, they can continue

ignoring it. Given the relatively low probability of detecting

fraud, this might be a plausible strategy but is not without its

risks.

Collaborative gatekeeping offers accountants a new option:

To provide this information directly to regulators by submitting a

suspicious activity report. With the anonymity of the report

preserved, accountants do not risk disrupting their relationship

with their clients. Clients unwilling to cooperate with external

auditors have less flexibility toward regulators, who possess more

extensive legal tools to extract information. If proven correct,

492. Monroe v. Hughes, 31 F.3d 772, 774 (9th Cir. 1994).

493. Ernst & Ernst v. Hochfelder, 425 U.S. 185, 202 (1976).

494. Susan Pulliam & Deborah Solomon, How Three Unlikely Sleuths

Exposed Fraud at WorldCom, W

ALL STREET J. (Oct. 30, 2002),

www.wsj.com/articles/SB1035929943494003751 (last visited Jan. 19, 2016)

(discussing how the WorldCom fraud became exposed) (on file with the

Washington and Lee Law Review).

910 73 WASH. & LEE L. REV. 797 (2016)

accountants can still rely on the immunity in order to avoid

liability. Thus, collaborative gatekeeping can help dislodge the

strong ties between accountants and issuers, and allow the

gatekeepers to overcome the conflicts of interest and alert the

regulators.

VII. Conclusion

Although gatekeepers are key pillars of our regulatory

framework, they often find themselves straddled between their

regulatory obligations and the pressures of building client

relationships in a highly competitive market. In an effort to serve

both masters at once, gatekeepers avert sanctions by making sure

that they remain unaware of “red flags” indicating client

misconduct. As a result, they turn a blind eye to information that

could prove particularly useful in enforcement.

In this Article, we have offered a novel approach out of this

impasse. Our goal is not to expand substantially gatekeeper

liability for clients’ faults but to entice gatekeepers to work more

closely with authorities. We propose that gatekeepers report

suspicions of misconduct to authorities. In return, they stand to

gain immunity from actions arising out of their reports by

regulators and private investors alike, provided they continue to

act in good faith. But if they choose not to report promptly, then

they will be subject to sanctions for failing to report, on top of any

other violations they might be committing. This regime, we

argue, can motivate gatekeepers to cooperate with authorities;

they invest significant time and effort in building a client

relationship, thus fueling conflicts of interest. Moreover, the

routine submission of suspicious activity reports can help change

the market perception about collaborating with regulators, from a

largely stigmatized decision that often costs individual

professionals and firms their reputation for client loyalty, to a

morally sound obligation arising from their gatekeeping function.

Rather than relying exclusively on theoretical

argumentation, our collaborative gatekeeping proposal has also

been tried in practice in the area of anti-money-laundering law.

We trace the beginnings of anti-money-laundering law in

Switzerland to show that banks, rather than opposing the

COLLABORATIVE GATEKEEPERS 911

imposition of these rules, embraced the new regime after

negotiating some compromises. We then show how most countries

in the world, including the United States, have amended and

adopted the Swiss template. The implementation of this regime

in the United States illustrates its promise to increase the flow of

information from market participants to regulators. Especially

with the aid of modern technology, gatekeepers and regulators

were able to aggregate and review information, explore financial

fraud besides money laundering, and identify areas of concern

that would have otherwise gone unnoticed. We have illustrated

how the collaborative model could work in two additional areas in

finance, broker-dealer regulation and accounting in equity

issuance.

We conclude this Article by discussing some concerns that

might arise in connection with the new responsibilities that

regulators and private industry are to assume. Collaborative

gatekeeping seems to trust regulators to read submitted reports,

decide which ones are worth pursuing, investigate further, and

successfully bring an enforcement action. But even if regulators

did possess all the information that collaborative gatekeeping

promises to bring to them, would they be able to fully take

advantage of it? Regulators might lack resources and staff to

pursue every lead and might simply have other priorities. To take

a famous example, the SEC first learned that something was

amiss with Madoff through Harry Markopolos, a securities

analyst who worked for a Madoff rival and studied Madoff’s

revenue stream in order to replicate it.

495

The SEC even

investigated Madoff in 2006 about his management of customer

funds but failed to uncover the Ponzi scheme. So, does the

collaborative model rely too heavily on government intervention,

while authorities may not be up to the task?

While the collaborative model focuses on alerting regulators

early, one should not underestimate the impact of the

information it produces for private plaintiffs. Collaborative

495. Gregory Zuckerman & Kara Scannell, Madoff Misled SEC in ’06, Got

Off, W

ALL STREET J. (Dec. 18, 2008, 11:59 PM), www.wsj.com/articles/

SB122956182184616625 (last visited Mar. 29, 2016) (explaining that the

“Securities and Exchange Commission investigators discovered in 2006 that

Bernard Madoff had misled the agency about how he managed customer

money”) (on file with the Washington and Lee Law Review).

912 73 WASH. & LEE L. REV. 797 (2016)

gatekeeping introduces a very disciplined compliance model,

supported by modern technology and dedicated staff to comb

through clients’ operations repeatedly. This mechanism produces

a detailed paper trail of staff concerns, inquiries, meetings,

reports, and discussions. Without suspicious activity reporting,

all these would have remained unrecorded and, in most cases,

unaired. But once expressed and recorded, these documents can

provide valuable ammunition to private plaintiffs seeking to

establish liability. While gatekeepers would be shielded from

liability under the terms of the immunity, other participants in

the transaction would not have such benefits. Thus, records

produced in the context of preparatory work for suspicious

activity reports empower private plaintiffs as well.

Gatekeeper companies themselves might change their stance

once this information is compiled. A submitted report flags a

client or transaction as potentially harmful for the gatekeeper. As

mentioned above, some retail banks have decided to terminate

relationships with all clients that become the targets of two

SARs, regardless of whether regulators decide to follow up. Thus,

SARs boost the internal monitoring capacity of corporations as

well, and can motivate management to look more closely into

cases that might have otherwise gone unnoticed. Overall,

collaborative gatekeeping’s success or failure does not rest solely

on the shoulders of regulators but spurs private industry and

plaintiffs into action.

But as collaborative gatekeeping seeks to transform the

internal discipline and liability risk for gatekeeper firms, one

might worry that it is likely to face stiff opposition from the

financial industry. Clearly, intensified monitoring requires a

significant investment in infrastructure and resources, which can

pressure corporate profits. However, compared to other regimes

of gatekeeper liability, collaborative gatekeeping has one

advantage: It leaves the initiative of gatekeeping activity with

the intermediaries themselves. Gatekeepers are in charge of

information collection efforts and handling clients, so that they

can continue to manage their relationships at every step. At the

same time, gatekeepers are responsible for spotting and

disciplining misbehaving employees, and even for firing failed

managers, evaluating the risk of fraud at every step and reaching

decisions accordingly. Moreover, gatekeepers can decide what

COLLABORATIVE GATEKEEPERS 913

type of compliance structure works best with their needs, which

compliance staffers to hire, what they want out of their

information technology, and, ultimately, which reports to submit.

As a result, they are also in charge of compliance costs and can

allocate funds in a way that suits the need of their company and

their business. Overall, this continuous monitoring process

translates into greater certainty for gatekeeper firms because it

offers them a way to manage suspicious clients before these

clients grow into an inextricable problem.

Gatekeepers’ autonomy to set up a monitoring system that

fits the needs of their firm and area of activity, which our

proposal offers, is also one of the key advantages that advocates

of strict liability typically underline. At the same time, our

proposal is unlikely to distort the gatekeeper business model to

the extent that strict liability proposals might. Even proponents

of strict liability recognize the possibility of over deterrence and

market distortions arising from the substantially higher fees that

gatekeepers will be forced to charge. Instead, the collaborative

gatekeeping model simply adds a reporting obligation, whose

anonymous character and standardized application over the

entire industry seek to minimize any negative market fallout. For

these reasons, collaborative gatekeeping is far more palatable

politically to the industry, while also bringing about important

changes put forward by other reform proposals.

Calls for increasing gatekeeper liability persist loudly many

years after the financial crisis, as bankers and other finance

professionals continue to attract the ire of policymakers and the

wider public. But, perversely, increased gatekeeper liability ties

the fortunes of gatekeepers with the fate of their misbehaving

clients, at times pushing gatekeepers to ally with fraudsters

rather than working against them. Instead, our goal should be to

offer gatekeepers a way to disassociate themselves from their

clients, provided they contribute enough to the enforcement

process. To reform the conventional gatekeeper model in this

direction, we need to think out of the box. This Article started

from the same premise that gatekeepers’ critics start, namely

that gatekeepers do not always act to prevent potential fraud,

even though they have reasons to suspect that it is about to

happen or is already under way. But rather than simply

punishing gatekeepers for their failings, we reconceptualized this

914 73 WASH. & LEE L. REV. 797 (2016)

problem as one of information retention and dissemination and

proposed ways to help gatekeepers share their information.

Gatekeepers are often eyewitnesses when financial fraud

happens—eyewitnesses with valuable, if partial, information.

Rather than alienating them, we are better off bringing them to

our side.